Renewable Energy & Applications Anglophone Issue

ISSN: 1737-9334
ICID
International Centre for Innovation & Development
-PET- Vol. 51
Proceedings of Engineering & Technology

-PET-
Renewable Energy & Applications
Anglophone issue
Editor:
Dr. Ahmed Rhif
International Centre for Innovation & Development

–ICID–
Editor in Chief Sallam Mebrouk (ALG)
Dr. Ahmed Rhif (Tunisie)
Youcef Soufi (ALG)
Ahmed.rhif@gmail.com
Hassane Mahmoudi (MAR)
Dean of International Centre for
Innovation & Development (ICID) Toufik Madani Layadi (ALG)
Editorial Borad Samira Chader Kerdjou (ALG)
Tounzi Abdelmounaïm (FRA)

Elmati Khoumri (MAR)
Zohra Ameur (ALG)
Driss Youssfi (MAR)
Fabien Delaleux (FRA)
Georges Descombes (FRA)
Hana Boudjedir (ALG)
Fawaz Massouh (FRA)
Nachida K. Merzouk (ALG)
Houria Siguerdidjane
Khenfer Nabil (ALG)
Mohamed Elamine Djeghlal (ALG)
Manar Slimane (MAR)
Abdelfettah Barhdadi (ALG)
Mohamed Gherbi (ALG)
Mohamed Benbouzid (FRA)
Sara Zatir (ALG)
Philippe Guibert (FRA)
Mustapha Hatti (ALG)
Saoussen Hammami (TUN)
Rahmani Lazhar (ALG)
Tahar Bahi (ALG)
Fathi Mohammed (ALG)

Summary
Meeting Bioclimatic Knowledge Needs of Algerian Architects; the Contribution of Expert Systems. Page 1
Mohamed Benabdelfattah, Saida Hamrani, Abdelkrim Makhloufi.
Extraction and valorization of the active ingredients of Saccocalyx satureioides harvested in two Page 5
stations of the province of Djelfa.
H. Aksas, Y. Draoui, F. Amghar.
Fuel Cells and Hydrogen Storage: Challenges Facing Vehicle Manufacturers. Page 14
Bilal Soltani, Nedjem-Eddine Benchouia.
Diagnosis of Defects by Principal Component Analysis a Gas Turbine. Page 23

Fenghour Nadir, Bouakaz Messoud.
The use of the Algerian southwest volcanic rocks as supplementary cementitious materials (SCMs). Page 28
Y. Labbaci, Y. Abdelaziz, B. Labbaci, A. Alouani, A. Mekkaoui.
Group contribution method for the estimation of flammability proprieties. Page 30

Serat fatima zohra, Benkouider Ali Mustapha, Yahiaoui Ahmed.
Numerical investigation of an inclined offset jet. Page 33

Meriem AJMI, Nidhal Hnaien, Salwa Marzouk, Habib Ben Aissia.
Effects of Carrier Mobility and defects on the Recombination Characteristic of P3HT: Graphene Bulk Page 39
Heterojunction Solar Cell.
Chahrazed Dridi, Naima Touafek.
Performances and Robustness Assessment of Sliding Mode Control Applied to a High Frequency Page 43
Switched Parallel Three-Cell DC-DC Converter.
Rihab Hamdi, Amel Hadri Hamida.
Effect of the precursor solution on the structural morphological and optical properties of ZnO thin Page 49
films deposited by Spray Pyrolysis technique for solar cell application.
H. Benzarouk, M. Mekhnache, A. Drici, A. Amara.
PL , EDX , and XRD Analysis of Al2O3 Scales Grown on PM2000 Alloy. Page 54
S. Hakkar, S. Achache, F.Sanchette, Z. Mekhalif, N. Kamoun, A. Boumaza.
Structural, dielectric and piezoelectric properties of (1-x) Pb (Zr0.52 Ti0.48) O3–x Sm Cr O3 cerami Page 59
F. Kahoul, L. Hamzioui, A. Guemache, M.Aillerie, A. Boutarfaia.
Production of 3rd generation biofuel from green microalgae : Oedogonium nodulosum. Page 64
Nadjla Chaib, Sabrina Dzizi, Faiza Noune, Hadjer Kaddeche.
Comparative study of Two and Three-Bladed Wind Turbine’s performances, FAST as Framework. Page 69
Nedia Aouani, Uwe Ritschel, Faouzi Chaabane.
Performance evaluation of Gamma type Stirling engine. Page 74

Houda Hachem, Ramla Gheith, Fethi Aloui, Samia Ben Nasrallah.
Low capacity diffusion absorption refrigeration: Experiments and model assessment. Page 78
Ikram Saâfi, Ahmed Taieb, Yosra Ben Salem, Ahmed Bellagi.
Estimated study of the production of electricity from different types of PV in two Algerian sites. Page 83
Mohamed SALMI, Derradji MEDERREG, Djamal FRIDJA.
Impact of the secondary phase ZnS on CZTS performance solar cells. Page 87
Naïma.Touafek, Ramdane. Mahamdi, Chahrazed. Dridi.
Optical absorption of new n-doped GaN0.38yAs1- 1.38ySby/GaAs QWs suitable for the Page 91
design of 0.8 -1 eV solar cells.
C.Bilel, A. Rebey.
3ème Congrès International sur les Energies Renouvelables et le Développement Durable (ERDD-2019)
Proceedings of Engineering and Technology PET
Meeting Bioclimatic Knowledge Needs of Algerian

Architects; the Contribution of Expert Systems
Mohamed BENABDELFATTAH #1, Saida HAMRANI *2, Abdelkrim Makhloufi⁑3
#
L’université Africaine Ahmed Draia, d’Adrar ; L’école polytechnique d’architecture et d’urbanisme EPAU d’Alger ;
Laboratoire Architecture et Environnement, LAE de l’EPAU, Algérie.
1
Benabdelfattah_med@univ-adrar.dz , Atarchimed@gmail.com
*
Department of Architecture, University of Sciences and Technologies, Oran, Algeria.
2
Hamranisana@gmail.com
⁑
Laboratory on African Studies, Social and Human Sciences, University of Adrar, Algeria
3
amakhloufi22@gmail.com
Abstract— In the architectural design process, architects are necessary knowledge that would enable it find the most
confronted to problems and situations to which they have to find adequate solutions.
the most suitable solutions. Several software programs are Among the various approaches concerned with artificial
developed to enable them find solutions to such problems. intelligence, the Case based reasoning (CBR) can be
However, the majority of the architects in our case study do not
considered as the closest one to the human reasoning based
use any bioclimatic device tools during the architectural design
process. This is mainly due, probably, to the inadequacy of these satisfaction, and is the most adopted in the act of designing
tools to the professionals implied in the designing process, [5]. Indeed, it can help architects abating the lack of
especially the architects. Here, we aimed to give the designer and information related to bioclimatic field issues and find
various actors a practical helper tool in order to understanding optimal bioclimatic solutions.
and integrating the bioclimatic passive concepts regarding the Following this line of thought, this paper concerns itself
desert climate. We developed a bioclimatic sketching book based with:
on decree n0 14-27 and case based reasoning method. This novel - Collecting and conceiving a set of the different
tool offers concrete illustrations to the actors, and assists the conceptual strategies related to earthquake resistance design.
architect in his architectural bioclimatic design. It may
- Finding a way to take them into account during the
contribute to the decisive insertion of building in its desert area
during various stages of designing. early stages of designing.
Finally, developing a support tool to help and assist
architects during the designing process based on the
Keywords— Architectural design process, desert climate, schematization and case based reasoning method.
integration, sketching book, schematisation.
II. ALGERIAN CODES FOR DESERT BIOCLIMATIC DESIGN
I. INTRODUCTION
Since 2000 in Algeria, the elaboration of code for
The architectural project is seen as a problem solving sustainable design was one of the priorities of the public
activity, the result of a long and complex multidisciplinary authorities in order to take into account the natural and
process of planning [1]. It is a non-linear procedure, but it is climatic variables. This is clearly stated in the various
rather a dynamic one. The emergent conceptual solution is the promulgated legislative decrees (laws: 01-20; 02-02; 02-08;
result of various actions. It includes several types of 04-05; 04-09; 09-99; executive decree number 14-27) [6].
interactions between the actors and their fields. [2]. However, the designer’s experience has to be up to the
According to Pierre Fernandez, The major character which level to interpret suitably the spirit of the code, without which
makes the act of modeling the design process difficult is the a contrary result might occur especially in the case of bad
diversity of project elaboration methods. This diversity can be interpretations [7]. Moreover, the majority of current helper
interpreted through several ways of conciliating both the bioclimatic device tools are used as enhancers, checking the
objective (program, site, regulations…) and subjective feasibility of the architectural forms. They intervene at stages
(composition, reference…) parts during the design process [3]. in which the architectural part is already done.
The architect has to get the maximum of information Beside this, the architectural bioclimatic design requires a
concerning the project, and needs to draw from his own close cooperation between the technically and the
experience, knowledge and know-how [4]. architecturally bioclimatic aspects as of the first stages of the
Several approaches are proposed to implement this designing process. Experience shows that this approach
established fact in an intelligent device which would help in allows a kind of reconciliation between the architectural
finding solutions to the problems encountered by architects; aspect and the technical design.
this could be done by providing the machine with the
Copyright - 2019 Page 1

ISSN 1737-9334
Practically, the implementation of such method is the result

of the interaction of extremely complex factors such as: the
current regulation, the economic situation, the technical
know-how, the aptitudes, and even the different actors’
behavior.
Additionally, the majority of designers do not use a
decision advice tool during the early stages of bioclimatic
design process. This is mainly due to the inadequacy of these
tools to the professionals implied in the design process,
especially architects [8]. This absence can be explained on
various levels as:
- They are not used until the last phases of the design
process.
- Designers are heavily short of training in terms of
using these tools.
- Their knowledge about bioclimatic designing
concepts is quite poor.
- Architects consider that bioclimatic active strategies
are in charge of treating the sustainable aspect of
buildings.
- The fact of knowing the existence of an under state
control subsidy for the residences energy
consumption is compulsory make designing
intervenors lazy in considering bioclimatic aspects
during the early stages.
- The current software devices are beyond the mastery
of architects because of their highly specialized
character.
- They require relatively a long training period of time,
which professionals can not actually afford.
The use of these support tools varies according to the mode
of implication of the bioclimatic aspects in the architectural
design process:
In case of post-implication, the support tools will assume
the control, validation, or sanction mission. And hence, the
amendments have to limited with regard to the schedule and
project budget. Indeed, great deal of rectifications during very
advanced state of project design might imply revising a
considerable work of design on various scales. On the other Fig. 1 Implication of current helper bioclimatic device tools in bioclimatic
hand, in case of very early engagement, as of the draft stage, design process (Algerian case study)
they will be involved as assistance in the bioclimatic design,
as shown in Fig. 1. III. PROPOSED DESIGN PROCEDURE
The current simulation tools of bioclimatic aspects of The designing act is a highly complex activity; it is
building available form part of the specialized codes, and they technical and emotional at the same time. It often requests a
are not usually employed in architectural practice. During the search for ideas and information which can assist the designer
early designing phases, architects do not need any validation in the development of his project. This idea has been
but rather an enriching and assistant tool ; as affirmed by the developed by several researchers and designers, such as
head of CLOA-Adrar (Local Council of the Architects Order), Bignon, Halin, and Kacher in his article « A method to index
Mr. LAZRAG A. : «…when conceiving we always seek images in the wooden architecture domain » [9], and S.
nutritive sources for the project and which stimulates the Kacher in her doctoral thesis [10] among others…, they
architectural creativity…» explain that during the conceptual process the need to return
back to visual illustration (images, pictures, drawings, etc.) is
essential. They consider that the information transmitted by
image is easy to understand if compared to written passage.
This results from the fact that the image necessitates a
limited number of interpretations. The image also presents
information which can be directly integrated in the corpus of
Copyright - 2019
Page 2
ISSN 1737-9334
the ideas, constraints or solutions for the project. In other which fixes the various actions defining the bioclimatic cold
terms, the image is that which supports the architectural passive strategy.
creation before and after the design of the project in the
conceptual, architectural and even economic order. It clarifies V. PERSPECTIVE AND DISCUSSION
the various situations “analysis / development” of the The main aim of the experiments was to contribute to our
designing process, and also simulates the comprehension of reflection by supervising two essential points. First is to
the bioclimatic architectural concepts. Moreover, it gives a check that association concept-image allows a correct
concrete illustration to the actors so they can include the interpretation and a performance appropriation of the
formal relationships between the various attributes bioclimatic concepts; second is to validate the practical,
influencing the bioclimatic aspects of the building. [11]. economic and creative benefits of a schematized navigation in
In this paper, our objective is to propose a methodological standard referential of bioclimatic architecture of the Algerian
tool which allows a kind of reconciliation between the southern provinces.
objective and the subjective design parts, and this to facilitate The experiments are supported by 42 architects from
the exchange of communication between the different actors, CLOA-Adrar (Local Council of the Architects Order) who
and hence ensure a tight collaboration between them which is have follow lesson in master academic in architecture within
the fundamental principle of the architectural design. The African university of Adrar in partnership with university of
method suggested consists of two different structures: a linear Sciences and Technologies of Oran. Work proceeds in
structure and in loop one. exercise form of design before and after the use of the
The vertical linear structure defines the requirements, the developed sketching book. All drawn and written traces
bioclimatic design concerns, as well as the various produced during the experimentation are also reserved.
developing phases of the project. The loop structure generates Thanks to this experimentation we managed to improve
alternative solutions and allows the repetition of all activities our sketching book, determine bioclimatic knowledge needs
and phases of linear structure in an iterative way. The loop of architects, and also the main features of the suggest
structure divides the vertical linear one into loops of five approach. They enabled us to check our assumption :
successive stages. The first one examines the project - The presence of several loops of feedback makes it
candidates. The second stage selects bioclimatic concepts possible to return back in order to take into account the new
from the given data. information generated during the designing process.
Next, an analysis phase takes place in order to suggest - The distinction between the various phases of the
possible conceptual solutions. Then a synthesis step generates design process, a beginning and an end with loops between
a candidate solution through the exploration of various them, implies the existence of objective criteria. This
combinations of the primitive elements. And finally at a later illustrates the fact that the objective and its properties guide
stage the obtained results are evaluated and possibly the mechanism of assistance to generate bioclimatic forms as
readjusted and reused. These phases are repeated until a of the early phase
satisfactory designing solution is obtained. The recourse to the CBR concept and to the graphic
demonstrations at each level of the design situation gives the
IV. DESIGNED SKETCHING TOOL BASED ON 14-27 DECREE possibility to express the idea of the project one wishes to
In the design process, the architect always calls upon the carry out, and to communicate it.
references and the schematization preceding cases to advance
and feed his activities. He does not have predetermined ways VI. CONCLUSION
of resolution. But he knows a certain number of procedures The architectural form is incontestably the most significant
and useful methods, and can rely on previous treated similar part in the bioclimatic aspect of buildings. The requirements
projects or existing prototypes, but it is necessary each time of bioclimatic design can appear during a first examination as
to reinvent and to recombine some strategies to elaborate an an impoverishing constraint. In fact, in the professional space,
adequate solution. we should seek shape of buildings adapted as possible with
The need for the bioclimatic design expert rules proves to naturel variables, as well in architectural as of technical
be essential. They can offer to the architect the conceptual aspects. Many successful projects on the aesthetic level prove
bases of his project, and they enable to guarantee an however that a bioclimatic architecture of quality is possible.
integration and insertion of climatic variables as of the early Thus, beyond the "thermomechanical behavior" aspect, we
phases of the design process. The reasoning based case seems can integrate the bioclimatic concerns within the architectural
to be the nearest resolution approach to the problems adopted intention, and it is possible to use the bioclimatic concept as
in the act of designing. Indeed, it implies the artificial an element of architectural expression.
intelligence “AI” which can be used to help the architect in Our model reveals two types of information structuring the
the bioclimatic design and to find optimal solutions. representation and the progress of the architectural project;
The designed bioclimatic sketching book results of the first answers the semantic question: « what is to be done?
superposition of the obtained (urban, architectural, and ». The answer could be considered as a source of bioclimatic
functional) bioclimatic indicators, by analysing of executive conceptual ideas. And the second is an esthetic answer to the
decree n0 14-27, and the synthesis of bibliographical study question: « how it will look like? ». Its role is to seek
Copyright - 2019
ISSN 1737-9334 Page 3
conceivable solution forms. Proposal tool can be considered [6] S. Hamrani, “ L’insertion passive des bâtiments dans un climat
désertique ; État de lieu et recommandations ; Cas de la wilaya d'Adrar
as a tool of assistance and decision-making during early
– Algérie,” M.D. dissertation, University of Sciences and
stages of design process, and even a tool of communication Technologies, USTO, Oran Algeria, 2017.
with the common public. [7] Benabdelfattah M., Kehila Y., Makhloufi A. (2018) Meeting
Paraseismic Knowledge Needs of Algerian Architects. In: Kallel A.,
REFERENCES Ksibi M., Ben Dhia H., Khélifi N. (eds) Recent Advances in
Environmental Science from the Euro-Mediterranean and Surrounding
[1] K. pal SANKAR, SIMON c.k.Shiu, Foundation of soft case-based-
Regions. EMCEI 2017. Advances in Science, Technology &
reasoning, vol 2, 150, 2004.
Innovation (IEREK Interdisciplinary Series for Sustainable
[2] R. Prost, “ conception architecturale une investigation
Development). Springer, Cham.
méthodologique,” l’Harmattan (ed), French, 1992.
[8] L. Khelifi, “ Contribution méthodologique à la conception
[3] P. Fernandez, “ stratégie d’intégration de composante énergétique
bioclimatique en architecture; développement de modèles génériques,”
dans la pédagogie du projet d’architecture,” Ph.D. dissertation, école
M.D. dissertation, Ecole polytechnique d’architecture et d’urbanisme,
des mines de Paris, French, 1996.
EPAU d’Alger-Algérie, 2006.
[4] M. Benabdelfattah, Y. Kehila, H. Khelafi, and N. Djarfour,
[9] S. KACHER, “ proposition d’une méthode de référencement d’images
“Integration du “RBC” à la prise de decision pour assister
pour assister la conception architecturale : Application à la recherche
l’architecture parasismiue; contribution méthodologique,” African
d’ouvrages,” Ph.D. dissertation, l’Institut National Polytechnique de
Review of Science, Technologiy and Development, vol. 3, no. 1, pp.
Lorraine Discipline, Sciences de l’architecture, 2005.
07-17, 2018. [Online].
[10] M. BRET, “ Image de synthèse : méthodes et algorithmes pour la
[5] S. Mazouz and M. S. Zeouala, “The integration of environmental
réalisation d’images numériques,” St-Jean/Richelieu, Dunod (ed), Bo-
variables in the process of architectural design; the contribution of
Pré, Paris, French, 1988.
expert systems,” Elsevier, J. Energy and Buildings, vol. 33, pp. 699-
[11] J. P. JUNGMANN, “ L’image en architecture : de la représentation et
710, 2010.
de son empreinte utopique,” Éditions de la Villette (ed), Paris, French,
1996.

ISSN 1737-9334
Extraction and valorization of the active ingredients of Saccocalyx

satureioides harvested in two stations of the province of Djelfa
H. AKSAS (1), Y. DRAOUI (2) , F. AMGHAR
(1)
Aksas_Ha@yahoo.fr
Number phone : 00 213 5 51 15 81 45

(2)(3)
yashmiine@outlook.fr; amgharfateh@icloud.fr
ABSTRACT:
This study was conducted to analyze the yield, chemical composition and antibacterial and antifungal activities
of essential oils of Saccocalyx satureioides harvested in two stations of the province of Djelfa under different
ecological conditions. The essential oil is extracted by steam distillation using Clevenger-type extraction device.
The yield was 0.87% for the harvested plant to Zaafrane and 1.32% for the Ain Chouhada. Oil Saccocalyx
satureioides exhibits antibacterial activity on bacteria: Escherichia coli, Pseudomonas aeruginosa,
Staphylococcus aureus, Bacillus subtilis, and antifungal activity on the fungus Aspergillus Niger. The results
show that the antibacterial activity of the essential oil of Ain chouhada (12-20mm) is more important than that of
Zaafrane (10-17mm), with a significant antibacterial activity of the antifungal activity. The chemical analysis of
essential oils Saccocalyx satureioides GC-MS revealed the presence of constituents dominated by: thymol, α-
Terpineol, Borneol, p-Cymene, γ-Terpinene, and finally we have the Camphene.
Keywords: Saccocalyx satureioides, essential oils, antimicrobial activity, antifungal activity, Djelfa.

ISSN 1737-9334
Introduction METHODOLOGY :
Algeria is considered among the countries known Drying:

for their taxonomic diversity due to its privileged
geographical position and its spread between the After harvest of saturated Saccocalyx from both
Mediterranean and sub-Saharan Africa. As well as
the Algerian steppe presents a remarkable interest sites by mid-April 2016. The stems and leaves were
from the point of view of its specific vegetal dried in the dark and at room temperature for 8
richness. (Ozenda, 1977). Among this floristic days.
richness we are interested in the plant species
"Saccocalyx Satureioides", harvested in two Hydrodistillation :
communes of high Algerian plains namely,
Zaafrane and Ain chouhada which are located in the It is a method of extraction of essential oils carried
wilaya of DJELFA, in order to carry out the
Extraction of the essential oils of the two sites by out at the level of laboratory of ecology of the
hydrodistillation, the identification of the chemical department of process engineering (UMBB) by a
composition by (GC-MS) of the two extracted distillation apparatus of the modified Clevenger
essential oils, and finally to test their antibacterial type, whose role is to drive the volatile compounds
and antifungal activities, to see if the composition
of the natural products with water vapor. It makes
chemical varies with environmental factors.
it possible to extract the maximum of the organic
compounds (essences) contained in the plants.
We introduce into a 2-liter pyrex glass flask, 40 g of

plant material, then we add a volume of distilled
water that corresponds to 2/3 of the capacity of
the flask. Then we adapt to the balloon condensing
device of the vapor (refrigerant) whose
temperature does not exceed 23 ° C, we supply it
with water and the heating with a balloon heater,
and We close the balloon with a thermometer to
check the temperature. After 15 to 20 minutes,
when the temperature reaches 90 ° C, the water
begins to boil causing steam and essential oil that
Figure 01: Location of the two municipalities in
Djelfa wilaya will liquefy in the cooling column. Finally we
collected a liquid composed of two phases:
Aqueous phase and oily phase, which is the
distillate. the decantation is done automatically, at
the end of the extraction we recover our essential
oil in a bottle of brown glass, sterile, well closed
and kept at a temperature of 0 to 4 ° C, to preserve
it from air and the light, The duration of extraction
is 2 to 4 hours maximum and the operation is
repeated several times (3 times maximum).
Without forgetting to note its weight with a scale

of precision to be able to calculate the yield which
is given by the following formula:
Figure 02: Saccocalyx saturated
Copyright - 2019
Page 6
ISSN 1737-9334
It is an agar diffusion method that aims to

qualitatively evaluate the "in vitro" antimicrobial
activity of the essential oil of saccocalyx
saturioides on the different bacterial and fungal
strains tested by the agar diffusion method
Rd: yield of essential oil expressed as a percentage (aromatogram).
(%).
That of direct contact, which has two methods: the
M ': mass of the essential oil obtained in grams (g). well method, and the disk method. For this study
we used the agar disk diffusion method to test the
M: mass of dry vegetable matter used in gram (g). sensitivity of different strains. (Remmel et al.,
1993).
Figure 03: Installation of a modified clevenger type

hydrodistillator, made at the ecology laboratory
(UMBB). Figure 04: Illustration of the aromatogram method
Gas chromatography coupled with mass

spectrometry:
GC / MS is a separation analysis method that

applies to gaseous compounds or compounds that
can be vaporized by heating without decomposition
(Arpino, et al., 1995). Made in France, it is the
separation technique most used in the field of
essential oils, because it allows the
individualization of constituents from samples of
the order of milligram or microgram.
Aromatogram:
The evaluation of the antimicrobial activity of the
essential oil of Saccocalyx satureioides was made
on 4 bacteria and a mushroom recovered from the
microbiology team of the laboratory VALCOR of
the UMBB. The antimicrobial power of the Figure 05: Preparation bench.
essential oil is obtained by measuring the diameters
of the inhibition zones in mm.
Note that for Ain chouhada we did not test the

antimicrobial activity of the two bacteria P.
aeruginosa (Gram-), B. subtilis (Gram +), because
we did not have enough plant matter.

ISSN 1737-9334
RESULTS:
Hydrodistillation :
Figure 07: Content of the chemical classes

composing the HEs of the two populations of S.
satureoides.
Figure 08: Content of three major constituents of

HEs of two populations of S. satureoides.
The results of the chemical analysis of the

composition of essential oils by (GC / MS)
determined four main chemical groups for the two
sites: Monoterpenes Hydrocarbons, Oxygenated
Monoterpenes, Sesquiterpenes Hydrocarbons, and
Oxygenated Sesquiterpenes.
Figure 06: Comparative yield in essential oils of the And then This analysis made it possible to isolate
two populations of S. satureioides. and identify 37 compounds for the two stations
representing 99.34% for Zaafrane of HE; and
The determination of hydrodistillation yields 98,44% for Ain chouhada.
of essential oils showed a good yield of
The most dominant constituents are: Thymol with a
volatile oil in Saccocalyx satureioides (0.87%
percentage of 34.87% for Zaafrane and 41.09% for
for the Zaafrane sample and 1.32% for that of
Ain chouhada; then there is α-Terpineol with a
Ain chouhada).
Ain chouhada; and finally we have Borneol with a
Comparative chemical composition of the
essential oils of the two S. satureioides
Ain chouhada.
populations:
Evaluation of the antimicrobial activity of the
essential oil:

ISSN 1737-9334
HE with a light yellow color and a strong and

pleasant odor.
The difference between the yields of the two sites is

0.45%, it is an important value that can come back
to several factors that differ from one medium to
another.
This difference in yield between the two sites can

be attributed to many factors such as the stage of
growth, pedoclimatic and edaphic conditions of the
region, extraction technique (Fellah, 2006) plant
genetics, organ of the plant used, the degree of
Figure 08: Antimicrobial activities (zone of
freshness, sunshine, rainfall, these factors among
inhibition in mm) of the essential oils of the two
others that may have a direct impact on HE
populations of S. satureoides.
performance (Vekiari et al, 2002).
The antibacterial and antifungal activity obtained
The yield of the essential oil depends on several
by the aromatogram show that the essential oil of
factors such as species, altitude, geography, harvest
Saccocalyx satureioides has a significant inhibitory
period, cultural practices, extraction technique, etc.
activity on the four bacteria: E. coli, Pseudomonas,
(Silano and Delbo 2008, Marzoukia et al., 2009,
Staphylococcus, Bacillus, and on the fungus:
Olle and Bender 2010).
Aspergillus, with some variation for both sites. The
results show us that the antibacterial activity of the Comparative chemical composition of the
Ain chouhada essential oil (between 12-20mm) is essential oils of the two S. satureioides
greater than that of Zaafrane (10-17mm), with a populations:
significant antibacterial activity of the antifungal
activity. It is found that the class of the Oxygenated
Monoterpenes is the most dominant in the essential
Note that for Ain chouhada we did not test the oil of Saccocalyx satureioides, with a percentage of
antimicrobial activity of the two bacteria P. 80.2% for the site 1 (Thymol = 41.09%, α-
aeruginosa (Gram-), B. subtilis (Gram +), because Terpineol = 14.1% and Borneol = 11.9%); and
we did not have enough plant matter. 72.57% for site 2 (Thymol = 34.87%, α-Terpineol =
27.42% and Borneol = 13.24%), so the essential oil
of Zaafrane has more Oxygenated Monoterpenes
than that of Ain chouhada.
The class Monoterpenes Hydrocarbons follows

with a percentage of 13.2% for site 1 (p-Cymene =
3.69%, γ-Terpinene = 3.26% and Camphene =
3.03%); and 17.57% for site 2 (p-Cymene = 3.12%,
γ-Terpinene = 3.08% and Camphene = 2.01%),
therefore this class is more dominant in the
essential oil of Ain chouhada. After the class of
oxygenated sesquiterpenes with a percentage of
3.23% for site 1 and 4.73% for site 2 therefore it is
dominant in the essential oil of Ain chouhada. And
finally we have the class of Sesquiterpenes
DISCUSSION: Hydrocarbons with a percentage of 2.65% for site 1
and 3.38 for site 2, therefore it is dominant in the
Hydrodistillation : essential oil of Ain chouhada. And a class that
presents the other constituents we have 0.06% for
The extraction of the essential oil from the plant at Zaafrane and 0.19% for Ain chouhada.
both sites by hydrodistillation gave a good yield of

ISSN 1737-9334
We notice a certain variability of the constituents towards the preferential formation of a precise
containing the essential oil of the two saturated constituent (Pellecuer, 1982).
Saccocalyx populations, these observed differences
can be related to the different chemotypes, time of The way in which extractions are carried out (time
harvest, altitude, soil, temperature, rainfall, of operations, packing and crushing of vegetable
humidity, sunshine, neighboring plant populations, matter, number of washes, etc.) can affect the
etc. composition and the organoleptic characteristics of
the essence.
In addition, broadening the comparison within the
framework of the Lamiaceae family. The flowering tops and leaves should be harvested
before flowering, because according to Salle and
The work of many authors has shown that plants Pelletier (1991), after flowering, 70% of essential
react to the surrounding environment and that oils evaporate in the air, however, the whole plant
during their lifetime, the chemical composition of is usually harvested during flowering. A number of
their metabolites could evolve. Extracts of plants,
parameters play a vital role in the chemical
essential oils are very fluctuating in their
composition , which can vary according to the composition of Essential Oil Plants, (Girre, 1979)
geographical and climatic conditions, the and (Anton et al., 2005):
cultivation ground of the plant, the year of culture
(sunshine, hygrometry, etc.), the cultivation mode, • Light: each plant has its requirements for light
the harvesting period, the individual or the organ quality and photoperiodism, most often prefer
considered, the extraction method, etc. (Bignell direct sunlight for at least six to eight hours.
and Dunlop, 1994).
• Temperature: the optimal T ° is decisive for each
Diversity according to the plant organ: For a plant.
species whose different organs may contain an
essential oil, the chemical composition of it may • Fertilization: in general, the fertilization of the
vary from one organ to another. (Richard and Loo, crop should be limited or even absent for dry
1992). Influence of the harvest period: The scrubland plants (thyme, rosemary, savory).
proportion of different constituents of the essence
of a given species can vary considerably during its • Soil: Most aromatic herbs require rich, loose,
development. The maturity or phenological state of well-drained soil with neutral or slightly alkaline
the plant at the time of harvest are difficult to pH.
verify and control (Derbesy, 1997).
In fact, each plant species has its biological and
Indeed, on the same stem, the leaves or the flowers ecological requirements, its own evolution and its
do not appear simultaneously and according to their reactions to the environment in which it lives. As
age, do not have the same composition (Touche, the botanist FLAHAUT wrote: "Every species has
1997). This has been demonstrated in particular on its place in nature by laws to which man can not
mint where differences in chemical composition change anything" (Pellecuer, 1982).
have been observed between the edges and the
Evaluation of the antimicrobial activity of the
center of the sheet.
essential oil:
Existence of chemical varieties or "chemotypes": A
The evaluation of the antimicrobial activity of our
plant species perfectly defined botanically can give
essential oil at both sites was tested on 5 human
species whose chemical composition is different
pathogens, four bacteria: Escherichia coli,
according to the individuals, (Bruneton J., 1993).
Pseudomonas aeruginosa, Staphylococcus aureus,
Bacillus subtilis, as well as its antifungal activity on
a fungus. : Aspergillus niger.
For some authors, chemotypes constitute, within the

species, chemical varieties each possessing a
particular enzymatic equipment, determined
The antimicrobial activity of essential oils of
genetically and which directs the biosynthesis
Saccocalyx satureioides could largely be associated

ISSN 1737-9334
with its main constituents: Monoterpenes properties (Derwich et al, 2010). According to the
Hydrocarbons which have shown antimicrobial results of the GC/MS, the Monoterpenes
properties (Derwich et al, 2010). Hydrocarbons are 13.2% for Zaafrane and 17.57%
for Ain chouhada, so that is why the antimicrobial
From the results mentioned in the two tables, it is activity of Ain chouhada is more active than that of
found that the saturated Saccocalyx essential oil has Zafrane.
antibacterial and antifungal activity on all the
strains tested. The thresholds of its antimicrobial In addition, the positive test organisms justify the
power are between 10 and 20 mm for both sites. popular use of this plant and confirm its anti-
diarrheal therapeutic activity: S.aureus, E. coli and
All strains have been shown to be sensitive to this Bacillus are germs very often implicated in food
essential oil. But Ain chouhada (12-20 mm) poisoning.
saturated saccocalyx essential oil is much more
active than Zaafrane (10-17 mm). Conclusion :
We note that Saccocalyx sematioides HE from both Determination of hydrodistillation yields of

sites does not have a very good activity on the essential oils showed a good yield of volatile oil in
fungus Aspergillus niger (10-12 mm) compared to Saccocalyx satureioides (0.87% for the Zaafrane
its activity on bacteria (11-20 mm). So its sample and 1.32% for that of Ain chouhada). Yield
antibacterial activity is more active than its increases with altitude.
antifungal activity.
The antibacterial and antifungal activity obtained
Several works including those of Hammer et al. by the aromatogram in this study, show that the
(1999); Souza et al. (2006); Derwich et al. (2010); essential oil of Saccocalyx satureioides exhibits
and Bari et al. (2010) confirmed the high resistance significant inhibitory activity on the four bacteria:
of G- bacteria to G +, which is equivalent to the Escherichia coli, Pseudomonas aeruginosa,
presence of a lipopolysaccharide layer (LPS) in G- Staphylococcus aureus, Bacillus subtilis, and the
bacteria that could function as an effective barrier fungus: Aspergillus, with some variation for both
against bacteria (Inouye et al. 2001, Bagamboula et sites, and this may return to the chemical
al., 2004). But this does not get involved effectively composition of the species.
in the case of Saccocalyx satureioides, E. coli
proved to be more sensitive although it is Gram-. The results show us that the antibacterial activity of
the Ain chouhada essential oil (12-20mm) is greater
It is postulated that the different components of than that of Zaafrane (10-17mm), with a significant
HEs show a difference in the degree of antibacterial activity of the antifungal activity.
antimicrobial activity against G + and G- bacteria
(Dorman and Deans, 2000). The yield, chemical composition and antimicrobial
potency of essential oils vary from one individual to
According to Cosention et al. (1999) and Gulfraz et another. This variability may be related to extrinsic
al. (2008), the antimicrobial activity of any and intrinsic factors.
essential oil is assigned to terpenoids and phenolic
compound. According to GC/MS, the Saccocalyx So we can say that Hes coming from the same plant
but coming from different geographical zones,
satureioides HE is dominated by terpene
compounds. imply a variation of the yield in aromatic molecules
as well as a variation of the nature of these
The antimicrobial activity may also depend on the molecules.
composition of the culture medium (Dorman and
Finally, all the results obtained are only a first step
Deans, 2000).
in the search for biologically active natural source
substances. Further tests will be needed to confirm
the multiple biological activities of this essential oil
The antimicrobial activity of essential oils of as well as the factors that increase its effectiveness.
Saccocalyx satureioides could largely be associated
with its main constituents: Monoterpenes Bibliography:
Hydrocarbons that have shown antimicrobial
Copyright - 2019
Page 11
ISSN 1737-9334
[1] : Anton R., A. Lobstein, Teuscher F., (2005). - Salviaofficinalis.L Plucked in Two Different
Aromatic Plant: Spices, Aromates, Condiment And Regions of Tunisia" J.Soc.Alger.Chim., 2006, 16
Essential Oil. Ed: Tec And Doc. Paris: 521. (2), 193-202. Journal of the Algerian Society of
Chemistry.
[2] : Arpino P., Provost A., Serpinet J.,
Tranchant J., Vergnol A., Witier P., 1995. [12] : GIRRE, (1979). -New Guide of Old Natural
Practical handbook of gas chromatography, Remedies. Ed: Rennes. La France. 314 P ISBN 2-
Masson, Paris. 85882-860-1.
[3] : Bagamboula C.F., Uyttendaele M. and [13] : Gulfraz M., Mehmood S., Minhas N.,
Debevere J., 2004. Inhibitory effect of thyme and Jabeen N., Kausar R., Jabeen K. And Arshad
basil essential oils, carvacrol, thymol, estragol, G., 2008. -Composition And Antimicrobial
linalool and p-cymene towards Shigellasonnei and Properties Of Essential Oil Of Foeniculum Vulgare.
S. flexneri. Food Microbiology 21: pp. 33-42. African Journal Of Biotechnology Vol. 7 (24),
Pp.4364-4368.
[4] : Bari M.A., Islam W., Khan A.R. and
Mandal A., 2010. -Antibacterial and antifungal [14] : K. Hammer, Carson C. F., & Riley T.,
activity of Solanum torvum (Solanaceae). Int. J. 1999. -Antimicrobial Activity Of Essential Oils
Agric. Biol., 12: pp. 386-390. And Other Plant Extracts. Journal of Applied
Microbiology Vol. 86, Issue 6: Pp.985-990.
[5] : BIGNELL C.M. and DUNLOP P.J. (1994). -
Volatile Leaf Oils Ofsorne South-Western and [15] : Inouye, S., Tsuruoka, T., Uchida, K.,
Southern Australian Species Ofthe Genus Yamaguchi, H., 2001. -Microbiol. Immunol., 43,
Eucalyptus. Part 1 Subgenussymphyomyrtus, pp.201– 208. In Ahmad I., Aqil F. and Owais M.
Dumaria Section, Seriesincrassatae, Flavourfragr. Modern Phytomedicine: Turning Medicinal Plants
113-117. into Drugs. Ed. WILEY-VCH Verlag GmbH & Co.
KGaA, Weinheim, 405p.
[6] : Bruneton J., 1993.-Pharmacognosy.
Phytochemistry, Medicinal Plants. Tec. & Doc. [16]: Marzoukia H., Elaissib A., Khaldic A.,
Lavoisier, 2nd Edition, Paris. 915P. Bouzidd S., Falconierie D., Marongiu B., Pirasa
A. and Porcedda S., 2009. -Seasonal and
[7] : Cosentino S., Tuberoso C.I.G., Pisano B., geographical variation of Laurusnobilis L. essential
Satta M., Mascia V., Arzedi E. And Palmas F., oil from Tunisia. The Open Natural Products
1999. -In Vitro Antimicrobial Activity And Journal, Vol. 2; pp. 86-91.
Chemical Composition Of Sardinian Thymus
Essential Oils. Letters In Applied Microbiology 29, [17] : Olle M. and Bender I., 2010. -The content
pp131-135. of oils in umbelliferous crops and its formation.
AgronomyResearch 8 (3), pp.687-696.
[8] : DERBESY M. (1997). -Reproductibility of
Industrial Natural Extracts, Pai:! Ums Cosmetics [18] : Ozenda P., 1977. -Florida Sahara. Paris,
News, 132, p. 57-59. C.N.R.S., 622 P.
[9] : Derwich E., Benziane Z. & Boukir A., 2010. [19]: Pellecuer J. (1982). - From the Plant to the
-GC / MS Analysis And Antibacterial Activity Of Medicine: Medicinal Plants and Popular Medicine
The Essential Oil Of Menthapulegium Grown In in Bas Languedoc, Document Published by Odac,
Morocco. Res. J. Agric. & Biol. Sci., 6 (3): Pp. 191- Supplement to Studies on the Hérault, 13 (3), P. 29-
198. 35.
[10] : Dorman H.J.D. And Deans S.G., 2000. - [20]: Remmal A., Tantaoui-Elaraki A.,
Antimicrobial Agents From Plants: Antibacterial Bouchikhi T. & Ettayebi M., 1993. -Improved
Activity Of Plant Volatile Oils. Journal of Applied Method For The Determination Of Antimicrobial
Microbiology 88, pp. 308-316. Activity Of Essential Oils In Agar Medium. J.
Essent. Oil. Res. 5: P.179-184.
[11] : Fellah S, Romdhane M., Abderraba M.,
2006. "Extraction and Study of Essential Oils of

ISSN 1737-9334
[21]: Richard H. And Loo A. (1992). -Nature, foodrelated bacteria to Origanum vulgare L.
Origin And Propriety Of Spices And Raw (lamiaceae) essential oil. Brazilian Journal of
Aromates: In Richard H., Aromatians & Spices, Microbiology 37: pp.527-532.
Tec. & Doc. Lavoisier, Paris, P. 17-69 and P. 213-
238. [25]: TOUCHE J. (1997). -Representativity and
reproducibility of plant extracts of aromatic plants,
[22]: Salle J.L. and Pelletier J., 1991. -Essential RivistaItaliana EPPOS, Special issue, p. 288-294.
oils, aromatherapy synthesis and introduction to
sympathicotherapy. Ed. Frison-Roche, pp.19-45. [26]: Vekiari S.A., Protopapadakis E.F.,
Papadopoulou P, Papanicolaou D., Panou C and
[23]: Silano V. and Delbò M., 2008. -Assessment Vamvakias M., 2002. -Composition and seasonal
report on Foeniculum vulgare Miller. EMEA, variation of essential oil from leaves and peel of a
EuropeanMedicines Agency. London: 23p. lemon variety. Journal of Agrucultural and Food
Chemistry 5 (1
[24]: Souza E.L., Stamford, T.L.M. and Lima
E.O., 2006. -Sensitivity of spoiling and pathogen
), 147-153.
Copyright - 2019
ISSN 1737-9334 Page 13
Fuel Cells and Hydrogen Storage: Challenges Facing Vehicle Manufacturers

Bilal SOLTANI *1 a, Nedjem-Eddine BENCHOUIA*2
*
Department of Mechanical Engineering, Mohamed Chérif Messaadia University, Souk Ahras
MailBox 1553, Souk-Ahras, 41000, Algeria
1
b.soltani@univ-soukahras.dz
2
n.benchouia@univ-soukahras.dz
a
Gestion Maintenance Et Réhalibilitation Des Infrastructures
Urbaines ; infrares@univ-soukahras.dz
Abstract — The main challenge of our time is to find renewable hydrogen that needs to be produced. Its production is
energy sources that meet the needs of society. Nevertheless, therefore a major stake for the installation of the fuel cell.
energy must meet several criteria such as abundance, ease and The evolution of air quality, in terms of pollutant
speed of use, but also the cost of production which must be low. concentration, shows that CO2 emissions are now the main
In addition, the aspect of "clean and renewable energy" is also
transport challenge in the face of air pollution [2]. Transport
to be taken seriously.
Today, fuel cells are a magnet for researchers because they is the sector that generates the greatest increase in CO2
have many advantages. First of all, these batteries have the emissions: a car with a combustion engine releases about one
capacity to produce a significant amount of electricity, water tonne of CO2 every 5000 km into the atmosphere. A break in
and heat. In addition, the reagents of the battery are the oxygen technology seems inevitable.
naturally present in the atmosphere and the di-hydrogen that The fuel cells' environmental qualities and their excellent
needs to be produced. Its production is therefore a major stake efficiency contribute, where they are used to replace
for the installation of the fuel cell. traditional systems, to the improvement of air quality and the
Many hopes rest on the fuel cell. As we have seen, it is one of reduction of greenhouse gas emissions.
the energy sources that have the most to become and application
areas. With a relatively high efficiency, zero pollutant emissions, II. FUEL CONSUMPTION AND EMISSIONS
this fuel cell has many other advantages such as the variety of
batteries that can be adapted to all current needs such as Figure 1 shows the distribution of petroleum consumption
transportation, global electricity generation or digital. by industry sector in the world in 2005
Because of the benefits that flow from its use as an automobile
fuel, hydrogen is expected to take precedence over other
synthetic fuels in the not-too-distant future.
Sector
Keywords— Fuel cell; Hydrogen storage; vehicle

Other
14.5 %
I. INTRODUCTION Sector *
All nations across the globe face a severe energy crisis.
Industry 9.4 %
Our current energy systems are based on fossil fuels, which
not only have adverse effects on the environment, but are also No-
depleted and running out. With the growing demand for energy…
15.8 %
energy predicted to increase by more than 50% by 2025 due
to the rising world population, pressure is mounting to fi nd Transport
60.3 %
alternative, renewable sources of energy.
Oil covers 95% of the needs in this sector, energy 0
20
consumption related to transport in France represents a 40
60 %
quarter of total consumption and this share is one that 80
increases faster. Automakers are making thermal vehicles
cleaner by adding new, cleaner engines, catalytic converters Fig.1: Distribution of global oil consumption by industry sector in the world
and particulate filters. The introduction of these new in 2005
*including agriculture, public and commercial service, residential and
technologies has led to a significant decrease in emissions of unspecified [3]
pollutants such as SO2, CO, total hydrocarbons (HC), NOx,
particulates and has made it possible to comply with Euro1, The transport sector accounts for a large share of global oil
Euro2 and Euro3 standards issued by the European Union [1]. requirements (60.3%), but other sectors are not to be
Today, fuel cells are a magnet for researchers because they neglected.
have many advantages. First of all, these batteries have the Oil is irreplaceable today in transport; it covers 97% of
capacity to produce a significant amount of electricity, water needs. The globalization of the economy and trade implies an
and heat. In addition, the reagents of the battery are the important development of the transport sector. It is estimated
oxygen naturally present in the atmosphere and the di- that transport currently accounts for 60.3% (1860 Mtoe) of oil
consumption; this proportion should rise to 65% (3194 Mtoe
Copyright - 2019
Page 14
ISSN 1737-9334
by 2031 [4]. In other words, oil consumption for the transport

sector alone is expected to increase by about 71% by 2031. Although it represents the largest share of global oil
consumption, the transWe can see that the industry sector is
Although it represents the largest share of global oil still very strong (34%), but it is now joined by the residential
consumption, the transport sector is not the main contributor sector (heating) (34%). The transport sector contributes 25%
to the increase in global greenhouse gas emissions (mainly of emissions.
CO2) as shown in Figure 2.
In the transport, industrial and residential sectors, the
current technology will not be able to last very long because
of its contribution to the greenhouse effect, the scarcity and
the increase of the fossil fuels (for the transport and the
industry).
For the automobile, a probable evolution is the hybrid
vehicle equipped with both a heat engine and an electric
motor coupled to a battery or super-capacitor.
In the longer term, fuel cell technology seems to be very

promising for the future. Indeed, it theoretically offers the
advantage of electric propulsion and heat generation (zero
local emission and silence) without its major drawbacks
(limited autonomy and problems related to the massive
production of electricity).
III. FUEL CELLS AND THEIR APPLICATIONS
Fig.2: Distribution of greenhouse gas emissions by activity sector in the The principle of the fuel cell was stated in 1839 by Sir
world in 2004 (%) [3]
William GROVE [6] [7] [8].
It should be noted that the bulk of CO2 emissions (74%)
come from different sectors of the industry, namely the
energy industries, industrial processes and deforestation. This
is explained by the fact that the electricity producers, the
refineries and a part of the means of transport, correspond to
these sectors. On the other hand, if these numbers are
attributed to electricity in the sectors that consume it, another
trend is observed as shown in Figure 3.
Figure 4: The Sir William Grove experience 1839
A fuel cell makes it possible to directly convert chemical

energy into electrical energy. Typically, a stack is composed
of the assembly of elementary cells, the number of which
depends on the desired voltage and current.
Each cell comprises an anode compartment supplied with
fuel and a cathode compartment supplied with oxidizer
separated by two electrodes loaded with the catalyst and an
intermediate electrolyte providing ionic bond. Since their
discovery, enormous scientific and technological progress has
Fig.3: Distribution of greenhouse gas emissions by activity sector in 2004 been made.
(%) in the world, by attributing electricity to the consumption sector [5]

ISSN 1737-9334
Hydrogen powered vehicles emit only water; they are "Zero

There are currently 6 types of fuel cells: Emission" vehicles. Nevertheless as with electric vehicles, the
real environmental impact of the fuel cell vehicle is
 AFC (Alkaline Fuel Cell),
dependent on the origin of the hydrogen.
 PEMFC (Polymer Exchange Membrane Fuel Cell),
 DMFC (Direct Methanol Fuel Cell), The hydrogen fuel cell can be used for various applications
 PAFC (Phosphoric Acid Fuel Cell), other than automotive. Indeed, the transport sector, as shown
 MCFC (Molten Carbonate Fuel Cell), in Figure 3, shares the responsibility for increasing
 SOFC (Solid Oxid Fuel Cell). greenhouse gas emissions with the agriculture, residential-
service sector, industry, etc.
These fuels cell differ according to:
 The nature of their electrolyte; A. The polymer exchange membrane fuel cell (PEM type):
 Their operating temperature;
 Their architecture;  Operating principle:
 Their field of application.
The fuel cell allows the direct conversion into electrical
In addition, each cell has different requirements in terms of energy of the free energy of a chemical oxidation-reduction
fuels. Their characteristics are summarized in Table 1. reaction. Like a battery or accumulator, a fuel cell has two
electrodes. The negative electrode is the seat of the oxidation
Operatin
Fields of
drafting of the fuel, usually hydrogen. On the side of the
Power g positive electrode is the reduction reaction of the oxidant,
Name Electrolyte applicatio
range temperat
ure
n usually the oxygen of the air. The two electrodes are
DMFC(« separated by an electrolyte. In the case of a PEM cell, a solid
Direct Polymer 1mW to membrane performs the function of electrolyte; the faces of
60-90 ° C Portable
methanol membrane 100 kW the electrodes are covered by a catalyst, platinum. The
fuel cell ») negative electrode - electrolyte - positive electrode assembly
PEMFC(«
Proton
Transport constitutes the heart of the cell. The supply of the latter in
Polymer 100W to Portable reagents is via distributing plates. The fuel and oxidant are
exchange 60-90 ° C
membrane 500kW Stationar
Membrane
y provided to the cell in conditions of pressure, temperature,
FC fuel cell ») humidity and purity defined continuously to ensure the
at AFC Aqueous
low («Alkaline alkaline
10kW to 50-250 ° Transport production of the current.
100kW C Spatial
tem fuel cell ») solution
pera PAFC («
tures Phosphoric Phosphoric Up to 10 160- Stationar
acid fuel acid MW 220 ° C y
cell »)
MCFC («
Hig
Molten Molten Up to Stationar
h 650 ° C
carbonate carbonate 100 MW y
tem fuel cell »)
pera
ture SOFC («
Up to 750- Stationar
Solid oxide Solid Oxide
FC 100 MW 1050 ° C y
fuel cell »)
Table 1: The different types of fuel cells and their characteristics
Low temperature fuel cell (PAFC, PEMFC, AFC and

DMFC) operate at temperatures below 250 ° C. Their fast
start, compact size and low weight are particularly relevant to Figure 5: Principle of operation of a PEM fuel cell
the transport sector.
The PEMFC fuel cell seems for the moment the most Depending on the type of cell, the intermediate chemical
appropriate for the on-board application because it best meets reactions involved vary but the general principle remains
the criteria of temperature and speed of operation. unchanged. The overall reaction of a PEM type fuel cell can
be written as follows:
The major problem with implementing this type of energy
source in a vehicle is its supply of hydrogen. Although  Reaction to the anode
hydrogen has high qualities, storing this gas in a vehicle
remains a delicate operation for users. H2 → 2H+ + 2e- (1)

ISSN 1737-9334
 Reaction to the cathode: Property Gasoline Methane Hydrogen

Density (kg/m3) 4.40 (*) 0.65 (**) 0.84 (***)
Diffusion 0.05 (*) 0.16 (**) 0.610 (***)
½ O2 + 2H+ + 2e- → H2O (2) coefficient in
air (cm2/s)
 By combining (1) and (2) the overall reaction is: Specific heat at 1.20 (*) 2.22 (**) 14.89 (***)
constant
pressure (J/kg
H2 + ½ O2 → H2O + heat (3) K)
Flammable 1.0 – 7.6 (***) 5.3 – 15.0 (**) 4.0 – 75.0 (*)
limit in air
Electricity is produced by the 2 electrons released by the
(vol%) + (LEL-
hydrogen molecule (1). The H+ hydrogen protons pass UEL)
through the membrane separating the anode from the cathode Flammable 0.24 (**) 0.29 (***) 0.02 (*)
and recombine at the cathode with the electrons and the energy in air
(MJ)
oxygen atoms. Ideally, the only side products.
Flammable 501–744 (*) 813 (**) 858 (***)
temperature
The reaction is water and heat. Figure 5 summarizes the (K)
principle of the PEM fuel cell. The potential difference across Flame 2470 (*) 2148 (***) 2318 (**)
each cell is small; in operation, it is less than volt. For most temperature in
air (K)
applications, it is therefore necessary to build a stack of cells, Explosion limit 1.1 – 3.3 (*) 6.3 – 14.0 (**) 13.0 – 59.0
commonly called stack, to have a sufficient level of voltage. in air (vol%) (***)
Fuel toxicity (*) (**) (***)
But before hydrogen entering our daily lives and exploitation
in feeding fuel cells, progress must be made at every stage of Combustion (*) (**) (***)
products
the industry stages: production, transport, storage and use. toxicity
Flame (*) (**) (***)
emissivity
IV. HYDROGEN STORAGE FOR FUEL CELL VEHICLES Safety level: (***) More safety; (**) Intermediate safety; (*) Less safety
+ (LEL: Volume concentration of a gas, from which it can be ignited;
A. History and characteristics of hydrogen UEL: Maximum concentration in volume of a gas, above which it cannot be
ignited)
The few important dates in the history of hydrogen are:
Table 2: H2 compared with traditional fuel
 1766: The British chemist Henry Cavendish manages
to isolate a strange gaseous substance, which by Hydrogen has many advantages. Indeed, this molecule has
burning in the air gives water. the advantage of being particularly energetic: 1 kg of
 1781: Until now called "flammable gas", hydrogen hydrogen releases about 3 times more energy than a kilo of
owes its name to the French chemist Antoine-Laurent gasoline (120 MJ / kg against 45 MJ / kg for gasoline). It is
de Lavoisier, who performs the synthesis of water. an abundant element, but it does not exist in pure form on
 1804: Frenchman Louis-Joseph Gay-Lussac and Earth. It is a component of water or organic compounds
German Alexander Von Humboldt jointly demonstrate (biomass ...).
that water is composed of a volume of oxygen for two Hydrogen is colorless, odorless, non-toxic and has no
volumes of hydrogen. environmental impact (its combustion produces water and
heat) and is lighter than air (it dissipates 4 times faster than
Some figures concerning the physicochemical properties of natural gas).
hydrogen are reported in Table 2. Despite all these advantages, it has many disadvantages.
For example, it has a low density. Thus, to produce as much
energy as with 1 liter of gasoline, 4.6 liters of compressed
hydrogen are required at 700 bars. It is highly flammable
(0.02 mJ), has a wide flammability range (5 times that of
CH4) and has a very fast flame spread rate. Moreover, his
flame is invisible.
Hydrogen has a negative image with the public since it is
considered a dangerous gas and its acceptability is not
acquired.
However, hydrogen has long been used as a raw material

in the chemical and petrochemical industry as shown in Table
3 and table 4.

ISSN 1737-9334
 In Europe:
Industrial Europe
Sector
Consumption (Million %
tons / year)
Ammonia 3,2 39
Production
Other 1,15 14
Chemicals
Refining 3.9 47
Total 8.25 100
Table 3: European distribution of hydrogen consumption in industry (2014)

[9]
 In the World:
Industrial World
Sector
Consumption % Fig.6: Graphic representation of global distribution of hydrogen
(Million tons / year) consumption in industry
Ammonia 22.8 38
Production Today, hydrogen is little used in the field of energy, with
the exception of space propulsion.
Other 4.8 8 It should be noted that from the late 19th to the beginning
Chemicals of the 20th century, hydrogen was used (mixed) for lighting
lamps and as city gas for heating.
Refining 26.4 44
B. Fuel Cell vehicle
Various 6 10
(space, food
A fuel cell vehicle (FCV) or fuel cell electric vehicle
industry, glass,
(FCEV) is a type of electric vehicle which uses a fuel cell,
etc.
instead of a battery, or in combination with a battery or super-
Total 60 100
capacitor (secondary source of energy) to power its on-board
electric motor.
Table 4: Global distribution of hydrogen consumption in industry (2014) [9]
Fuel cells in vehicles generate electricity to power the
The sector where hydrogen consumption is the most motor, generally using oxygen from the air and compressed
important is oil refining since it requires 26.4 million tons per hydrogen. Most fuel cell vehicles are classified as zero-
year (44%). In the manufacture of ammonia the consumption emissions vehicles (ZEV) that emit only water and heat. As
of hydrogen is 22.8 million tons per year (38%). Hydrogen is compared with internal combustion vehicles, hydrogen
also used to manufacture other chemicals (8%) such as vehicles centralize pollutants at the site of the hydrogen
amines, methanol, and hydrogen peroxide. production, where hydrogen is typically derived from
reformed natural gas. Transporting and storing hydrogen may
also create pollutants [10].
The engine schematic for a fuel cell vehicle is represented in

the following figure (Figure 7):

ISSN 1737-9334
The main producers of FCVs are the United States and

Japan. In 2017, 2298 vehicles were sold in the United States
and 849 in Japan. South Korea is the third largest market with
61 vehicles sold in 2017. Annual sales of these vehicles by
country of sale are shown in Figure 9.
Fig.7: Engine schematics for a fuel cell vehicle
C. The early adopters of fuel cell vehicles

Several car manufacturers promote FCVs to consumers.
Vehicles are often compared to battery electric vehicles
(BEVs). Both types of vehicles emit no exhaust, can be
powered by renewable energies and are powered by electric
motors. The main difference between these vehicles is their Fig.9: Worldwide annual sales of fuel cell vehicles between 2013 and 2017
range and style of refueling. FCVs have a range of over 300 by country [12]
miles and can be refueled in less than 10 minutes at a
hydrogen refueling station [11].
V. THE MAIN PRODUCTION METHOD OF DIHYDROGEN :
Three FCVs are currently available to consumers. These
include the Hyundai Tucson / ix35 FCEV, the Toyota Mirai Hydrogen is one of the fuels needed to operate a fuel cell.
FCEV and the Honda Clarity FCEV. These vehicles are However, dihydrogen is not an energy source strictly
currently on sale in North America, Europe and Asia. In 2014 speaking but rather an energetic vector. Compared to oxygen,
sales of these vehicles began. The Toyota Mirai was sold in it is necessary to produce it using energy. Until now,
record numbers with 5233 units delivered to consumers from hydrogen production has been carried out using fossil fuels
2013 to 2017, including 2944 in North America. By the end such as vapo-forming, which consists of reforming natural
of 2017, 637 Honda Clarity FCEVs had been delivered to gas. The gaseous fuel (methane) is subjected to water vapor
consumers, including 440 in North America. Finally, 727 to achieve different chemical reactions. It is necessary to
Hyundai Tuscon / ix35 FCEVs were delivered, the most carry out this experiment at very high temperature, at
important market for these vehicles being Europe with 373 atmospheric pressure and in the presence of a catalyst. Figure
units delivered. Figure 8 provides an overview of the annual 10 shows hydrogen production and applications [13].
sales of these vehicles and the total sales of the three vehicle
models.
Fig.10: The "hydrogen" chain [13]
Hydrogen production today represents 630 billion cubic

Fig.8: Global annual sales of fuel cell vehicles between 2013 and 2017 meters, used mainly to produce ammonia, methanol, HNO 3;
by vehicle model [12] this production will have to increase significantly to meet the

ISSN 1737-9334
new energy needs. World demand for primary energy was VI. STORAGE MODES
18
443 EJ (1 Exajoule (EJ) = 10 joules) in 2003, dominated by
oil and coal followed by natural gas, meaning that current The major obstacle to the implementation of fuel cell
hydrogen production would cover only 1.5% of demand [14]. technology in the various fields is the supply of hydrogen and
There is therefore a lot of work to be done in the field of mass its storage. A lot of research then looked into the problem in
hydrogen production to achieve a significant share (especially order to fix it and develop reliable storage systems. The
considering a growing global energy demand). According to evaluation of these performances is done mainly by
the reference scenario of the IEA (International Energy volumetric and gravimetric densities (of hydrogen and of the
Agency), global demand for primary energy is expected to whole with the means of storage) and the general conditions
grow by 53% between 2004 and 2030 [14], an average annual of storage and destocking of the hydrogen (efficiency, speed,
rate of 1.6% [14]. Developing countries will be responsible apparatus necessary annexes). This is in addition to the
for more than half of this global increase [14]. determining criteria of safety and cost.
Hydrogen can be produced from many different sources. The introduction of hydrogen as fuel in the automotive
Currently, however, most hydrogen (97%) is produced from sector requires that a suitable means of storage on board the
hydrocarbon reforming [14]. The choice of hydrogen vehicle be developed. Currently, the three storage
manufacturing process is based on many parameters (type of possibilities, namely:
available primary energy, purity, flow rates,).
 Compressed gas under high pressure, [15]
To produce hydrogen, several possibilities are studied, some  liquefied hydrogen
have already reached technological maturity and others are
still at the stage of development:  Solid: - in metal hydrides in which the hydrogen is
absorbed (chemisorption), or in carbon compounds
 from fossil fuels: (activated carbon, nano fibers and carbon nanotubes)
 thermal decomposition in which hydrogen is adsorbed (physisorption).[15]
 catalytic reforming
It is interesting to examine the three possibilities of hydrogen
 steam cracking
storage and to analyze to what extent they are able to meet the
 partial oxidation
requirements of this application and which include:
 auto-thermal reforming
 High mass and volume storage capacity
 from the electrolysis of water or by thermodynamic
 Storage pressure and temperature are as close as
decomposition
possible to the ambient.
 from biomass (gasification, biological transformation)
 from green algae or bacteria  Handling conditions that do not impose too stringent
 from the nuclear safety standards.
 Almost unlimited storage time
The vapor-forming and partial oxidation techniques  Investment fees and acceptable operating costs
correspond to about 95% of the total production, only 2% of
hydrogen being produced by electrolysis. A. Storage in liquid form:
The use of fossil fuels leads to the formation of a gas rich in This process can store hydrogen in liquid form at -253 C °
H2 and CO. But a fuel cell such as PEMFC is very sensitive at pressures between 1 and 10 bar (cryogenic tank). Its
to CO. It is therefore necessary later to eliminate the advantages are the low pressures, its weight reduction
maximum CO (<10 ppm) to obtain a fairly pure H2. potential and moreover this process has been demonstrated on
a vehicle (BMW year 1988) [16].
To be used in fuel cells, existing technologies for purifying
hydrogen are: But this process requires high liquefaction energy (30-40%
 the oxidation of CO by water (Water Gas Shift); lost energy), moreover it requires a constant thermal
 molecular sieve purification with PSA (Pressure Swing insulation and there is loss of hydrogen by evaporation (1 to
Adsorbed) 2% per day)
 purification by methanation
 Purification by metal membranes (Pd, Pd-Ag).
 the preferential oxidation of CO (CO-PROX)
Copyright - 2019
ISSN 1737-9334 Page 20
B. Storage in compressed form The storage of gaseous hydrogen under high pressure is
essentially penalized by the low value of the mass capacity
This process makes it possible to store hydrogen in gaseous and volume relative to this system. Added to this are the risks
form (in composite tanks) at high pressures (up to 700 bars). that accompany the use of high pressures, this solution is not
This process has various advantages such as the lightness of indicated on board a vehicle.
the tanks (when they are in polymer liner), the speed of filling The cryogenic storage of hydrogen is also not advisable in
and it does not need thermal management. this case, because of the dangers it presents (constant losses)
and the cost of its liquefaction. The permanent surveillance of
which it must be the object, limits its use to very particular
C. Hydrogen storage in hydrides (Indirect storage) uses (aeronautical and space applications). Add to the risks
associated with high pressures
Several processes involve embedding hydrogen in the form
of molecules or atoms bound to other materials or more easily
transportable molecules. In this area, several avenues are VIII. DISCUSSION:
being explored: It is finally metal hydrides that seem to be able to best meet
 metal hydrides (main disadvantages: mass capacity, the requirements that this type of storage requires.
cost, autonomy ...), A decisive advantage of this storage method lies in greater
safety of use, because, in case of rupture of the reservoir, the
 Organic hydrides, the Hydrogen on Demand TM process
endothermic of desorption reaction causes a rapid cooling of
and embedded production from a fuel. the hydride mass, which reduces both the equilibrium
pressure and the rate of the decomposition reaction.
We can also note the carbon nanotubes (principle of gas
adsorption on the surface of a solid). Given that none of the hydrides already known meets all the
necessary qualities, it is therefore essential to modify the
The on-board production of hydrogen by a fuel seems to be a characteristics by adapting them to the specifications imposed
promising way. On the other hand, it will be necessary to find on them, in particular as regards:
a mode of production and an adequate raw material.
 Their mass storage capacity
VII. COMPARISON OF DIFFERENT STORAGE MODES  Their desorption temperature
 Their enthalpy of decomposition
The characteristics of the possible hydrogen storage  The exchange kinetics
methods are summarized in Table 5:
Storage mode
IX. CONCLUSION
H2 H2 liquefied Condensed in The theme of the hydrogen economy has become in the
Characteristics compressed a hydride last five years one of the most debated issues in specialized
Mass capacity 1.6 (steel)
(kg H2 / 100 kg 2.2 18 4 to 5 energy backgrounds.
system) (aluminum) The production and transport of hydrogen is not without
Volume problems. The bulk of current production is done by
capacity 1.4 5.5 ~5 reforming fossil sources and releases as much CO2 or more
(kg H2 / 100 dm3
system)
than the direct use of these sources. Hydrogen can also be
Storage pressure produced by electrolysis, with renewable energies such as sun
(atm) at 200 1 to 8 1 to 10 and wind, which allows the energy produced to be stored
ambient according to natural cycles to be used when needed.
temperature
Storage
temperature (° Ambient -253 Ambient Because of the benefits that flow from its use as an
C) automobile fuel, hydrogen is expected to take precedence
Storage losses over other synthetic fuels in the not-too-distant future.
(% per day) null 0.5 to 1 null In order to make better use of the energy content of
stored hydrogen, it seems more advantageous to move
Energetic cost
high very high low towards the use of fuel cells, which make it possible, from the
same quantity of hydrogen to produce a quantity of hydrogen
energy 4 to 5 times more than internal combustion engines.
Table 5: Comparative characteristics of different modes of hydrogen storage
Copyright - 2019
ISSN 1737-9334 Page 21
Among the possibilities to store it in a motor vehicle, it REFERENCES

seems that it is the metal hydrides that will be the best [1] https://fr.wikipedia.org/wiki/Norme_europ%C3%A9enne_d%27%C3
compromise solution. %A9mission
[2] R. Joumard – INRETS ‘’Les enjeux de la pollution de l'air des
transports’’ 5ème Colloque C-VELEC – Grenoble, 25 et 26 juin 2002.
However, this attractive solution which is the combination [3] C-P. Hwang, C-T. Yeh, Q. Zhu, ‘’ Rhodium-oxide species formed on
of the storage of hydrogen in a hydride with its combustion in progressive oxidation of rhodium clusters dispersed on alumina’’
a stack cannot be put into practice immediately. Catalysis Today, vol 51, p 93-101; June 1999.
Significant research and development efforts are still needed, [4] D. Martin, D. Duprez, ‘’The effects of support and of particle size on
the redox properties of rhodium’’ Applied Catalysis A: General, vol
both with regard to hydrides and their implementation as well 131, p 297-307, October 1995.
as with regard to fuel cells. [5] J. C. Vis, H.F. Van't Blik, T. Huizinga, J. Van Grondelle, R. Prins,’’
The hydrogen is widely used in automotive applications, The morphology of rhodium supported on TiO2 and Al2O3 as studied
along with materials discovery; efforts are needed to optimize by temperature-programmed reduction-oxidation and transmission
electron microscopy’’ Journal of Catalysis, vol 95, p 333-345, October
the overall performance of the fuel cell system to overcome 1985.
the relevant technical barriers for all hydrogen storage [6] H.C. Yao, S. Japar, M. Shelef, ‘’Surface interactions in the system
approaches. RhAl2O3‘’ , Journal of Catalysis, vol 50 p 407- 418, 1977
[7] Se H. Oh, Joyce E. Carpenter, ‘’The oxidation state and catalytic
activity of supported rhodium’’, Journal of Catalysis, vol 80, p 472-
Finally, to see if current and future hydrogen technologies 478, April 1983.
have the opportunity to replace fossil fuels, it is not only [8] C-B. Wang, C-T. Yeh, ‘’ Calorimetric study on oxidation of alumina
necessary to consider the entire production-storage-transport- supported rhodium by dioxygen‘’, Journal of Molecular Catalysis A:
use chain, but also to assess the time and capital needed to Chemical, vol 120, p 179- 184, 13 June 1997.
[9] Association Française de l’hydrogène et les pile à combustible,
install the necessary infrastructure. ‘’PRODUCTION ET CONSOMMATION D’HYDROGENE
AUJOURD’HUI’’ Mémento de l’Hydrogène FICHE 1.3, February
2016.
In the future: http://www.afhypac.org/documents/tout-savoir/Fiche%201.3%20-
%20Production%20et%20consommation%20d'hydrog%C3%A8ne%2
In the field of hydrides, research programs should be oriented 0rev.%20fev2016%20PM.pdf
[10] "How Do Hydrogen Fuel Cell Vehicles Work?" Union of Concerned
in particular towards an improvement: Scientists, accessed July 24, 2016.
[11] Scott Hardman, Gil Tal ‘’who are the early adopters of fuel cell
 Their mass storage capacity; vehicles?’’ Institute of Transportation Studies, University of
California, Davis, USA, International Journal of Hydrogen Energy
 Their adsorption-desorption conditions (temperature, Volume 43, Issue 37, Pages 17857-17866, 13 September 2018.
pressure, amount of heat put into play, exchange [12] EV Volumes. EV data center. 2018. http://www.ev-volumes.com/
[13] Association Française de l’hydrogène et les pile à combustible,
kinetics);
‘’PRODUCTION D'HYDROGENE PAR ELECTROLYSE DE
 Their chemical insensitivity to the different impurities L'EAU’’, Mémento de l’hydrogène, FICHE 3.2.1, Janvier 2017 ;
that can contain hydrogen. http://www.afhypac.org/documents/tout-savoir/Fiche%203.2.1%20-
%20Electrolyse%20de%20l%27eau%20revjanv2017%20ThA.pdf
Although currently in a much more advanced stage of [14] S. H. Chan, H. M. Wang ‘’Thermodynamic analysis of natural-gas
development, fuel cells can only be generalized in their use if: fuel processing for fuel cell applications’’. Int J Hydrogen Energy, Vol
25, 441-449, 2000.
[15] K. Reddi, A. Elgowainy, E. Sutherland ‘’Hydrogen refueling station
 Their power density compression and storage optimization with tube-trailer deliveries’’
 Their lifespan and reliability International Journal of Hydrogen Energy, Vol 39, p 19169-19181, 11
November 2014.
 Their low temperature operating characteristics (fast [16] Association Française de l’hydrogène et les pile à combustible, ‘’ LA
start-up capability). DISTRIBUTION DE L’HYDROGÈNE POUR LES VÉHICULES
AUTOMOBILES’’, Mémento de l’hydrogène, FICHE 4.5.1, February
2016 ;
Significant improvements can be made, provided that their http://www.afhypac.org/documents/tout-savoir/Fiche%204.5.1
cost of production can reach a reasonable value. distribution%20H2%20automobile_rev.fev2016ThA.pdf

ISSN 1737-9334
Diagnosis of Defects by Principal Component

Analysis a Gas Turbine
Fenghour Nadir #1, Bouakaz Messoud #2
#
Electromechanical Department, Badji Mokhtar University
Annaba, Algeria
1fen.nadir@gmail.com
2messaoud.bouakkaz@univ-annaba.org
Abstract— The fault detection and isolation have a great will be used to determine whether the system is in a failed
practical importance in engineering system. Early detection of state or not.
the fault occurrence is critical in avoiding product deterioration,
performance degradation, major equipment damage and human Multi-variable statistical methods are the most effective for
harm health. The aim of this work is to study the fault detection treating the generation of residues. Among them, methods
using linear principal component analysis (PCA) which can be based on Principal Component Analysis (PCA), are very
applied to the complex industrial processes. In the first step, we effective for highlighting significant linear correlations
study the fundamental principles of linear PCA are presented, between the variables of the process without explicitly
the PCA is used to model normal processes behaviour. In the formulating the model of the system. Thus, all the correlations
second time, we study different detection methods have been between the different variables are taken into account in the
discussed. Finally, we do the application of this study concern PCA model. In this paper we will study this particular model
the fault detection an electric power production process. for diagnosed (detection equipment or subassembly fault by
precision) a very complicated system of a gas turbine.
Keywords— Process monitoring; fault detection; linear
principal component; electric power production process
II. PRINCIPAL COMPONENT ANALYSIS
I. INTRODUCTION
Principal Component Analysis (PCA) is part of the group
In all industrial systems, breakdowns cause considerable of multidimensional descriptive methods called factorial
economic losses, it is therefore essential to implement methods. These methods, which appeared in the early 1930s,
monitoring and diagnostic systems to avoid unexpected were mainly developed in France in the 1960s, in particular by
shutdowns and then increase reliability and ensure the safety Jean-Paul Benzécri, who made extensive use of geometric
of the systems. The industrial diagnosis whose primary aspects and graphic representations.
function is the detection and localization of system
failures, the detection relates to the detection of events that It has been adapted to the detection and isolation of faults
affect the evolution of a system, so to assess then consists in that can be manifested at the system level, it is an analysis
comparing the actual functioning of the system. System with technique for multi-variable data [3].
what it should be under the assumption of normal This technique is based solely on the use and operation of
operation. Originally, the diagnosis was limited to high-risk the input and output measurements of a system. A data matrix
industrial applications for the community such as nuclear or is then constructed from these measurements. By a
aviation [1] [2], as well as advanced industries such as, decomposition into singular values of the matrix of the data,
industry armament or aerospace. Since the last three decades, the PCA splits this matrix in two parts, one containing the
the diagnosis has attracted increasing interest both in the dominant singular values, representing the useful data, and the
industrial world and in scientific research. other the remainder of the singular values supposed to be
In the field of diagnostics, methods based on the concept negligible and representing the noises. [4]
of redundancy of information have been developed. Their Several studies have recently appeared in the literature
principle is generally based on a consistency test between an dealing with the use of PCA in the field of defect detection [5,
observed behavior of the process provided by sensors and an 6, 7, 8, and 9].
expected behavior provided by a mathematical representation
of the process. Analytical redundancy methods therefore In Fig. 1. We present the PCA algorithm that illustrates the
require a model of the system to be monitored. This model principle of principal component analysis.
includes a number of parameters whose values are assumed to
be known during normal operation. The comparison between
the actual behavior of the system and the expected behavior
given by the model provides a quantity, called residue, which

ISSN 1737-9334
III. DETECTION BY THE PCA

The PCA is defined as a linear transformation of the
original correlated variables to a set of uncorrelated variables,
however, the original variables can be represented by a small
number of the main components because of the existing
analytic redundancy between the variables, and therefore,
instead of analyzing all the variables, the PCA analyzes these
components.
The presence of a fault affecting one of the variables
causes a change in the correlations between the variables
indicating an unusual situation because the relationships
between the variables are no longer verified. In this case, the
projection of the measurement vector in the residue subspace
will increase relative to its value under normal conditions.
To detect such a change in the correlations between the
Fig. 1. PCA algorithme. variables, the PCA uses several detection indices that are used
for the detection of abnormal operation by the PCA: Hoteling
To find a model based on the linear PCA one needs a
Statistics T2 and the squared error of prediction SPE.
database contains variables to monitor for a set of
measurements made on the normal operation of the A. SPE Statistic (Q-Statistic)
system. Generally, the procedure of model identification A typical statistic for detecting these unnatural conditions
consists, after the normalization of the data matrix, to estimate is the SPE statistic, also called Q (Squared Prediction Error)
the parameters of the model then to choose a fixed structure which is given by the equation:
and finally to validate this model.
The basic idea of the PCA is to reduce the size of the data SPE(k )  eT (k )e(k ) 
matrix. This reduction will only be possible if the initial
variables are not independent and have correlation coefficients To detect the presence of an anomaly, the SPE must check
between them non-zero. These initial variables are the following condition:
transformed into new variables, called principal
SPE ( k )   2
components. The variance approach of the reconstruction error 
was chosen to determine the number of principal components
to be retained. The variance of the reconstruction error (VNR) Or is the threshold of confidence after Jackson and
can be calculated as follows: Mudholkar [10] that develop the expression of the confidence
threshold for the SPE.
~ T  ~

ui  var  T ( x  xi )   var  f i   ~i T ~ 2i
( i  i ) (1) B. Hotelling T2 Statistic
The T2 statistic can be applied, on the first principal
2
~
~iT   i ~io ~

components [2], thus one obtains:
ui  ~ T 2  (2)
( i  i ) ~i

t i2
T (k )  tˆ T (k )1tˆ(k )  
2
2

i 1 i2
Where is an estimate of the magnitude of the defect f which
   diag (1 , 2 ,...... ) (7)
measures the displacement in the direction , with ∑ is the
correlation matrix and is the variance of the reconstruction Or (7) is a diagonal matrix containing the ℓ largest eigenvalues
error estimating using . So the PCA is used as a modeling of the correlation matrix, detect the presence of an anomaly
technique of the relationships between the different variables the statistic T2 must check the following condition:
of the process. The estimation of the parameters of the PCA
model is performed by estimating the eigenvalues and T 2 (k )   2, (8)
eigenvectors of the matrix of correlations of the data.
However, for the determination of the structure of the model, Where the upper limit for a confidence level α.
it is necessary to determine the number of components to We have seen that the use of different tests for detection
retain in this model (number of eigenvectors). can affect by the modeling errors that are capable of
generating false alarms, the EWMA filter
(exponentially weighted moving average ) [11] is used to
improve the detection.

ISSN 1737-9334
Ⅳ. APPLICATION OF PCA ON THE GAS T URBINE UNIT. 21 temperature purge lube oil bearing 1
LTB1D1
turbine
To verify and illustrate the effectiveness of the PCA
22 temperature purge lubrication oil
technique for defect detection, data provided by a real process LTB2D1
bearing 2 turbine
will be used. This part is dedicated to the application of Linear
LTOT 23 lubricating oil tank temperature
Principal Component Analysis for the detection of faults in an
LTTH 24 lubricating oil collector temperature
industrial installation (a single cycle power generation unit 25 fuel purge air temperature
PAT
(gas turbine) 26 collector pressure lubricating oil
GAP2
The group, as it is designed for most installations, consists TNH 27 tree rotation speed
of a single cycle "single shaft" gas turbine designed for TNRCOR 28 velocity shaft rotational%
continuous operation and designed to drive an alternator. The 29 inter-wheel temperature first stage Rear
TTWS1AO1
combustion of an air-fuel mixture serves to produce the power intake turbine
necessary to drive the compressor shaft, some auxiliaries, and, 30 inter-wheel temperature first stage
TTWS1FI1
mainly the alternator. before intake turbine
31 inter-wheel temperature second stage
TTWS2FI1
before exhaust turbine
32 inter-wheel temperature third stage rear
TTWS3AO1
exhaust turbine
TTXM 33 mid turbine exhaust temperature
TTXP1 34 first difference real control combustion
WTAD 35 temperature evacuation cooling water
WTTL1 36 temperature cooling water leg TG
Ⅵ. DETERMINATION OF THE OPTIMAL NUMBER OF

PRINCIPAL COMPONENTS
To obtain an optimal number of principal components
retained in the PCA model, we chose the reconstruction
approach, after using this reconstruction method (PCA model
Fig.2. Principle of operation of a gas turbine
structure), Figure (3) presents the evolution of the
reconstruction variance depending on the number of principal
Ⅴ. THE VARIABLES USED TO CONSTRUCT THE LINEAR components, this result indicates that the first seven major
PCA MODEL components are sufficient to construct the PCA model.
To build the PCA model, we chose 36 essential sensors, 100
Evolution de la variance de reconstruction VNR en fonction du nombre de composantes l
measurements are taken during the normal operation of the gas 90
turbine, represented on the table. 1. 80
70
TABLE I
60
DESCRIPTION OF THE VARIABLES OF THE GAS TURBINE
Vnr
50
1 temperature of the air main 40
AAT
atomizations 30
AFPDB 2 compressor inlet pressure

20
10
RFA 3 mass flow air intake compressor 0

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
BBMAX 4 maximum vibration turbo-alternator Composantes Principales
BBMAX GT 5 maximum vibration gas turbine Fig.3. Evolution of non-reconstruct variance VNR as a function of number
CMHUM 6 humidity of components.
CPD 7 compressor outlet pressure
CPR 8 compressor compression ratio Ⅶ. DETERMINATION WITH THE PCA E VOLUTION OF
CSGV 9 IGV angle in degrees THE VARIABLES AND THEIR MODELS
CDT 10 compressor discharge temperature
CTIM 11 compressor inlet temperature The six figures (4), (5) and (6) show, respectively, the
DF 12 alternator frequency measurements and their estimates of : compressor inlet
CFO 13 power factor pressure, humidity and compressor compression ratio by using
DV 14 alternator voltage in% the already obtained PCA model.
PSUs 15 reactive power alternator
DVX 16 alternator voltage
DWATT 17 active power alternator
FPG2 18 inter-valve fuel pressure
PCF 19 gas fuel mass flow rate
FTG 20 gas fuel temperature
Copyright - 2019
ISSN 1737-9334 Page 25
2.5 90
variable 80
2
estimation
1.5 70
60
1
50
0.5
T2
x2
40
0
30
-0.5
20
-1
10
-1.5
0
0 200 400 600 800 1000 1200
100 200 300 400 500 600 700 800 900 Echantillon
Echantillon
Fig.9. Evolution of T2 "faultless case".

Fig.4. Evolution of X 2 and its estimate.
3
ⅨFAULT DETECTION WITH LINEAR PCA
Variable
2
Estimation
To test and verify the effectiveness of the PCA model
1
obtained for detection from defects, we collected a
database with four real defects of the gas turbine.
X6
-1 Based on the model PCA already obtained, they are

-2 four figures (10), (11), (12), (13) have, respectively, the
-3
0 200 400 600 800 1000 1200
evolution of the PES detection index filtered in the presence of
Echantillon
the four scenarios of defects:
Fig.5. Evolution of X 6 and its estimate.  Triggering of the gas turbine by lower pressure
gas (fuel),
2

Variable
1
Estimation
Tripping of the electric default gas turbine
0
(maximum current 51TP),
-1
 Tripping of the gas turbine default of the group

X8
-2
-3
-4
circuit breaker synchronous coupler,
-5
-6
 Triggering the gas turbine default gas
0 200 400 600
Echantillon
800 1000 1200
compressor.
Fig.6. Evolution of X 8 and its estimate.
Ⅷ. THE EVOLUTION OF DETECTION INDEX CASE

WITHOUT DEFECT
Figures (7), (8) and (9) illustrate the evolution of the three
statistical tests; SPE, SPE filtered and T2, respectively, from
data representing normal operation, they are adapted to choose
the detection thresholds for each test (red lines).
90
80
70
60
50
SPE
40
30
20 Fig.10. Detection defect lowering gas pressure by PCA.

10
0
0 200 400 600 800 1000 1200
Echantillon
Fig.7. Evolution of SPE "faultless case".
14
12
10
SPE filtré
0
0 200 400 600 800 1000 1200
Echantillon
Fig.8. Evolution of SPE filtered "faultless case". Fig.11. Electrical type fault detection by PCA.
Copyright - 2019
ISSN 1737-9334 Page 26
process variables. Another recommendation is to study

nonlinear PCAs that extract both linear and nonlinear
relationships.
REFERENCES
[1] S.ASHISH, E .DALE, “Evaluation of pattern matching method for the

Tennessee Eastman challenge process,“ Journal of Process Control.
Vol 16.pp.601-613, 2006.
[2] S.W.CHOI, J.H.PARK, LEE In-Beum, “Process monitoring using a
Gaussian mixture model via principal component analysis and
Fig.12. PCA synchrocoupler fault detection. discriminant analysis”, Computer and Chemical Engineering, Vol
28.pp.1377-1387, 2004.
[3] S.J.QUIN, R.DUNIA, “Determining the number of principal
components for best reconstruction”, Journal of process control, Vol
10.pp 254 – 2050, 2000.
[4] N.L.RICKER, “Optimal steady-state operation of the Tennessee
Eastman challenge process”, Computers and Chemical engineering,
Vol 19.pp.949-959, 1994
[5] M.TAMURA, S.TSUJITA, “A study on the number of principal
components and of fault detection using PCA”. Computers and
Chemical engineering, 2006.
[6] L. WEIHUA,S.SIRISH, “Structured residual vector – based approach
to sensor fault detection and isolation”, Journal of Process Control,
Vol 12.pp.147-160, 2002.
[7] J.A.WESTERHUISE, S.P. GURDEN, “Generalized contribution plots
in multivariate statistical process monitoring of residual in MSPC”,
Fig.13. Gas compressor fault detection by PCA.
Journal of Intelligent Laboratory systems. Vol 51. pp. 95-114, 2000.
[8] B.M. WISE, “Adapting Multivariate analysis for monitoring and
According to the above figures, it is noted that the detection modeling of dynamic systems. PhD. Dissertation”, University of
index (black curve) has a large excess relative to the detection Washington. Seattle,1991.
threshold (red line) calculated by the PCA model for proper [9] G.Zwingelstein, “Diagnostic des défaillances, théorie et pratique pour
les systèmes industriels“ Edition Hermès, 1995.
functioning, thereby conclude an unusual operating condition [10] J.E.JAKSON, G.S.MUDHOLKAR, “Control procedures for residuals
of the turbine has occurred. associated with principal component analysis”. Technometrics, Vol
21, N° 3, 1979.
Ⅹ. CONCLUSION [11] T. KOURTI, J.F. Macgregor, “Recent Developments in Multivariate
SPC methods for Monitoring a diagnosing Process and product
The work presented in this paper focuses on the diagnosis Performance”. Journal of Quality Technology, Vol 28 N° 4.pp. 409-
of PCAL based systems, which is used as a modeling 428, 1996
technique for relations between different process
variables. The estimation of the parameters of the linear PCA
model is performed by estimating the eigenvalues and
eigenvectors of the matrix of correlations of the
data. However, for the determination of the structure of the
model, it is necessary to determine the number of components
to retain in this model (number of eigenvectors).
Once the model is identified, the defect detection

procedure can be performed by generating the defect
indicators (residuals) by comparing the observed behavior of
the process given by the measured variables and the expected
behavior given by the PCA model. Most PCA-based
detection methods use the SPE (squared error) statistic for
defect detection, and the T2 statistic of Hoteling. Thus, we
used the EWMA filter to improve detection, but with die delay
detection.
Finally, we presented the application of the model PCAL

for the detection of defects of a production process electric
energy based on a gas turbine, to complete this work we
recommend to use the data of this memory in the study of fault
localization methods. As we used in this work the PCAL
model that seeks to identify the linear relationships between
Copyright - 2019
ISSN 1737-9334 Page 27
The use of the Algerian southwest volcanic rocks

as supplementary cementitious materials (SCMs)
Y. Labbaci1, Y. Abdelaziz2*, B. Labbaci3, A. Alouani4, A. Mekkaoui5
1
Department of civil engineering, University of Bechar 08000, Algeria.
2
3
LMS laboratory, Department of civil engineering, University of Bechar 08000, Algeria
4
5
FIMAS laboratory, Department of civil engineering, University of Bechar 08000, Algeria.
E-mail: abdelaziz970@yahoo.fr
Abstract— This study is part of a sustainable

development policy that is dictated by the I. Experimentation
growing needs of material resources and the In the first stage, several volcanic rocks were
requirements of environmental protection. It characterized from the mineral and chemical
addresses an investigation on the possibility of viewpoint. In order to make the petrographic
using volcanic powders as supplementary identification, a number of thin sections of volcanic
cementitious materials for environmental- rocks were examined under a Nikon Polarized
friendly durable concrete. The work attempts to Light Microscope (Eclipse LV100Pol). The XRD
characterize several volcanic rocks (basalt, powder pattern was recorded on a Siemens D-5000
olivine andesite, amphibole-biotite andesite, X-ray diffractometer, with Kα cobalt anticathode (λ
amphibole andesite, hyodacite and scoria) from = 1.789 Å), at a current of 40 mA and voltage of 40
the mineral and chemical viewpoint and kV. The scans were performed in the 2θ range from
evaluate their pozzolanic activity. Furthermore, 0° to 70° with a scanning speed of 2°/min. The
a supplementary cementitious material, used by chemical compositions of raw materials were
many cement plants in Algeria, has been determined by X-ray Fluorescence (XRF) using a
included in order to establish a comparative Philips PW 1404 X spectrophotometer.
study.
In the second stage, the evaluation of pozzolanic
activity of various volcanic rocks by means of
mechanical strength tests was established. The
preparation of the mortars was carried out according to
Keywords—, mineral characteristics, chemical the norm NFP 15-403 (NFP 15-403 2006). A control
characteristics, pozzolanic activity. mixture was produced with a constant binder/sand/water
proportion of (1/3/0.5). In the test mixtures, 20% of the
mass of Portland cement used in the control mixture
was substituted by the same mass of the test
supplementary cementitious materials. The strength
activity index with Portland cement as follows: SAI =

ISSN 1737-9334
(tm /cm) x 100, wheretm is the average compressive REFERENCES

strength of test mixture specimen and cm is the average
Al-Chaar, G. K., Al-Kadi, M., & Asteris, P. G. (2013). Natural pozzolan
compressive strength of control mixture specimen. as a partial substitute for cement in concrete. The Open Construction and
Technology Journal, 7, 33–42.
ASTM Standard C618-00 (2000).Specification for Coal Fly Ash and
Raw or Calcined Natural Pozzolan for Use as a Mineral Admixture in
Portland Cement Concrete, ASTM International, West Conshohocken.
Lothenbach B, Scrivener K, Hooton D. (2011). Supplementary
cementitious materials. Cement and Concrete Research 41(12), 1244–1256.
Meindre M. (1952). Geology of massive Boukaïs. Scientific Economic
Bulletin, Mining Research Office of Algeria 2, 8-45.
NFP 15-403 (2006).Normal and normal mortar sand. AFNOR.
Owaid H, Roszilah B, Taha M. (2012). A Review of Sustainable
Supplementary Cementitious Materials as an Alternative to All-Portland
Cement Mortar and Concrete. Australian Journal of Basic and Applied
Sciences 6(9), 2887-303.
Seddiki A, Remaci-Benaouda N, Cottin J, Moine B, Ménot R, Perrache
C (2004). The volcano-sedimentary Boukais inlier (south-western Algeria):
evidence for lithospheric thinning during the Late Neoproterozoic. Journal of
African Earth Sciences 39, 257-266.
Seddiki, A. (1997). Petrological and geochemical study of magmatic
rocks of Boukaïs (Northwestern Bechar, Algeria sudoccidentale).
Fig.. Pozzolanic reactivity of the additions at 28 days. Contribution to a geodynamic study of the region to the Infra-Cambrian. PhD
Thesis University of Science and Technology.
Tazghout-Graine K. (2010). Geodynamics and metallogenic framework
of massif of Boukaïs and mountains Ougarta (Bechar / southwest Algeria).
PhD Thesis, University of Science and Technology.
Yang K, Jung Y, Cho M , Tae S. (2015). Effect of supplementary
cementitious materials on reduction of CO2 emissions from concrete. Journal
of Cleaner Production 103(15), 774-783.
Zerrouki A. (2000). Lithostratigraphy of the Proterozoic Boukaïs region
and study of associated mineralization (Bechar, south-western Algeria).
Geological Survey of Algeria 11(2), 163 -183.
Copyright - 2019
ISSN 1737-9334 Page 29
Group contribution method for the estimation of

flammability proprieties
Serat fatima zohra #1, Benkouider Ali Mustapha *2, Yahiaoui Ahmed#3
# 1,2 3
, Laboratoire de chimie Organique, Macromoléculaire et des Matériaux (LCOMM). Université de Mascara. BP 763,
29000 Mascara, Algérie.
Abstract-This Work presents a new group contribution method these previous works. This is due to several factors such as
development for estimation of flash point of organic compounds accuracy of the model, simplicity of the model, range of
values while it is one of the important variables used for review applicability covering a large number of compounds with
the fire and explosion hazard. The proposed method is based on different chemical compounds.
the molecular structure of organic compound, and the
development of experimental data set base including 1685 TABLE I
organic compound. The result are discussed and compared with DIFFERENT APPROACHES FOR THE PREDICTIVE VALUE OF AVAILABLE FP
PREDICTION METHODS FOR PURE COMPOUNDS, AS CHARACTERIZED BY
precedent works. This model shows a very low average relative
AVERAGE ABSOLUTE ERRORS (AAE), AVERAGE RELATIVE ERRORS (ARE)
error of 2, 43 % and an average absolute error of 9.22 K.
AND R2
Keywords— flash point, group contribution.
Ref year Dat method AAE ARE R
set
Introduction [6] 2014 1170 GC 13.3 3.5 /
[7] 2011 1471 Correlation / 2.4 0.97
The flash point is one of the most important properties in
[8] 2008 1030 QSPR 10.2 / 0.96
chemical process safety. It is used to evaluate the
[9] 2008 1378 ANN 8.1 / 0.97
flammability and combustibility of combustible material and
also to characterize the fire and explosion hazard. The flash [10] 2011 1062 Correlation 4.65 1.32 /
point gives the knowledge needed for the handling, storing
and transporting of any combustible material- solid, liquid or
Based on both simplicity of the model and accuracy of the
gas. The flash point is the lowest temperature at which a model, the group contribution approach is more convincing
liquid gives off sufficient vapor to form an ignitable mixture than the others approaches. This approach which is widely
with air near the surface of the liquid or within the vessel used to estimate and predict thermodynamic and other
used[1]. A compound with a higher flash point is less properties of pure organic compounds and mixtures [11 ,12],
flammable or dangerous than one with a lower flash point.
needs mainly tabulated groups (molecular fragments) where it
The experimental determination of flash point can be made is assumed that the property value of the molecule is a
generally by the open cup and the closed cup methods [2]. It function of the contribution of all the groups and that a
is one of the ways to have accurate and reliable data. property value of any group has the same contribution in all
However, the experimental determination of flash point can molecules. The group contribution approach does not need
be expensive and very difficult especially if the compound any other physical properties, does not require specialized
concerned is metastabl, toxic, or has other hazardous software to implement (simple procedure of calculation of
properties. Moreover, the large gap between the number of group from SMILES of chemical structure) and has the
compounds registered and the available experimental values advantage of quick predictions without requiring significant
of flash points of compounds is constantly increasing. Thus, a computing power.
model that can provide reliable estimates of FP for a wide In this study we developed a simple straightforward method
range of compounds is needed. for the estimation of the flash point of organic compounds
using the third order of group contribution method with two
Different approaches have been presented in the literature for models one based on only the molecular structure and the
the prediction of FP of pure compounds. A good review on second base on molecular structure.
flash point estimation has been already published [3, 4, 5].
Tree major approaches which include the group contribution Materials and methods:
(GC)strategy, the quantitative structure−property Data base: The quality of a predictive method depends on a
relationship(QSPR) approach and empirical correlations have variety factors such as quality of data set used in its
been developed for the estimation of the FP. Table 1 development and validation. In order to provide this fact, a
summarizes these previous works. It must be note that it is wide variety of classes of organic compounds containing
difficult to provide a comprehensive comparison between hydrocarbons, halogenated hydrocarbons and compounds
Copyright - 2019
ISSN 1737-9334 Page 30
contain O, N and S atoms were selected in data set base. The A: is obtained with the addition of group contributions for
description of this data-set based on various classes of each molecule compositions:
organic compounds is given in Figure1. A=∑iNiCi
Ni : frequency of occurrences of n group in a molecule
Ci : contribution of n group
K1 =546,65 ; K2 =-0,179 ; K3=226,66.
50 group contributions are defined corresponding to each
group. For readability of the display the last are not displayed.
Results and discussion

The Performance of this model for the prediction of FP values
of organic compounds was evaluated, analyzed and discussed.
Generally, in the case of large data set it is necessary to
estimate the detail results of the proposed model.
The statistical performance indicators used in this work are
coefficient of determination (R2) standard deviation (SD)
average deviation (AAD) and the average relative deviation
Fig. 1: class wise description of data sets base. (ARE) for this model based on molecular structure is shown
in Table 3.
The data set base involves reliable experimental FP
TABLE 2
temperature of 1685 pure compounds have been employed for RESULTS OF THE PREDICTED MODEL OF FP, AS CHARACTERIZED BY AVERAGE
developing and testing an accurate group contribution model, ABSOLUTE ERRORS (AAE), AVERAGE RELATIVE ERRORS (ARE) AND R2 AND
80% of the main data set are selected for training set (1348 STANDARD DEVIATION SD
compounds) and 20% are used for the test set (337
compounds). The first one is used to develop the model and Total compound 𝑁° ARE(%) AAD𝐾) SD (𝐾) 𝐷𝑚𝑎𝑥(𝐾) 𝑅2
the second to evaluate the predictive power of the model. In
addition, ebullition temperature TB is given for each Training 1328 2,34 8,29 8,87 81,30 0,9877
compound in Kelvin (K) to facilitate the comparison.
Test 332 3,02 10,75 12,89 113,30 0,9757
Method development: Total 1660 2,68 9,52 10,88 97,30 0,9817

The group contribution method is base on the molecular
structure of a compound. Is considered to be divided on three
categories of groups: first, second and third order groups. The The prediction deviations according to their experimental
first one is used to describe a variety of organic compounds values are displayed in the figure 1. It is noted that the
while the second and the third categories are to prove more prediction error does not exceed 12, 4 K, these results
precise structures information about the fragments of the confirm the credibility of our model over the results of the
compounds. After finishing the data set base of FP defined by methods found in literature
the collection of structural groups. And the fragmentation of 100
the molecular structure was performed by an automatic ERR Train
program using Marrero and Gani [13]. In this work the data ERR Test
set was divided into two sub-data sets using semi randomly 50
choice. Indeed if a compound is selected and if this

compound is described by one of the irrelevant groups: group 0
having their contribution value generated by less than three
compound , this compound is removed from the test set , this
is important to avoid the lack of some necessary data for -50
training step and to allow better regression of same model

parameters. -100
In this work a simple practical multivariable
nonlinear model is proposed. In fact, in a several
mathematical functions were tested to represent the FP model. -150
0 200 400 600 800 1000 1200 1400 1600 1800
From many researches in the literature and a variety functions
tests, two of the best forms to be a simple nonlinear and more Fig 2: plot the error test and training
practical ones. The first model is based only on the group’s
contribution of the molecular structures. It is as follows:
The following figure Fig2 shows the dispersion of the
average relative deviation in the four intervals. So it is seems

ISSN 1737-9334
that over 70% of the values of flash point are predicted with REFERENCES
an instrumental precision 2,5% .This means that our [1] [1]-AICHE 2013. Project 801, Evaluated Process Design Data, Public
objective is achieved. Release Documentation, Design Institute for Physical Properties
(DIPPR). American Institute of Chemical Engineers (AIChE).
[2]
[3] [2]-NFPA30 2012. Flammable and Combustible Liquids Code,
National Fire Protection Association
[4]
[5] [3]-VIDAL, M., ROGERS, W. J., HOLSTE, J. C. & MANNAN, M. S.
2004. A review of estimation methods for flash points and
flammability limits. Process Safety Progress, 23, 47-55
[6]
[7]
[8] [4]-PHOON, L. Y., MUSTAFFA, A. A., HASHIM, H. & MAT, R.
2014. A Review of Flash Point Prediction Models for Flammable
Liquid Mixtures. Industrial & Engineering Chemistry Research, 53,
12553-12565.
[9]
[10] [5]-LIU, X. & LIU, Z. 2010. Research Progress on Flash Point
Prediction. Journal of Chemical & Engineering Data, 55, 2943-2950.
[11]
[12] [6]- Didier Mathieu, Thibaud Alaime. Insight into the contribution of
individual functional groups to the flash point of organic compounds
Journal of Hazardous Materials 267 (2014) 169– 174
Fig 3: Percentages relative error range for FP [13] [7]- Gharagheizi, F., Eslamimanesh, A., Mohammadi, A.H., Richon,
D.,2011. Empirical method for representing the flash-pointtemperature
CONCLUSIONS of pure compounds. Ind. Eng. Chem. Res. 50,5877–5880.
A new method has been developed for simple and reliable [14] [8]- Gharagheizi, F., Alamdari, R., 2008. Prediction of flash
pointtemperature of pure components using a quantitativestructure–
predictions of flash point of organic compounds by GC model. property relationship model. QSAR Comb. Sci. 27,679–683.
This model is based only on molecular structure of the [15] [9]- Gharagheizi, F., Alamdari, R.F., Angaji, M.T., 2008. A new
compounds neuralnetwork-group contribution method for estimation of flashpoint
The reliability of the methodology presented here is relatively temperature of pure components. Energy Fuels 22,1628–1635.
[16] [10]- )_ Rowley, J.R., Rowley, R.L., Wilding, W.V., 2011. Prediction
good compared to the using the training test data sets of 1348 ofpure-component flash points for organic compounds. FireMater. 35,
compounds and was tested using test data set of 337 343–351.
compounds. This method produces an average relative error [17] [11]-ARGOUB, K., BENKOUIDER, A. M., YAHIAOUI, A.,
(ARE) 2,68 %, coefficient of determination (R2 0,9817) KESSAS, R., GUELLA, S. & BAGUI, F. 2014. Prediction of standard
enthalpy of formation in the solid state by a third-order group
standard deviation (SD) 10,88 K and the average deviation contribution method. Fluid Phase Equilibria, 380, 121-127.
(AAD) 9,52 K Investigation of the capability of the proposed [18] [12]-BENKOUIDER, A. M., KESSAS, R., GUELLA, S., YAHIAOUI,
method for evaluated of the flash point of various compounds A. & BAGUI, F. 2014. Estimation of the enthalpy of vaporization of
demonstrates its wide range of applicability compared with organic components as a function of temperature using a new group
contribution method. Journal of Molecular Liquids, 194, 48-56.
the available method in the literature. [13]-J. Marrero, R. Gani, Fluid Phase Equilib. 183–184 (2001) 183–208.
Copyright - 2019
Page 32
ISSN 1737-9334
Numerical investigation of an inclined offset jet

Meriem AJMI#1, Nidhal HNAIEN#2, Salwa MARZOUK#3, Habib BEN AISSIA#4
#
University of Monastir, National Engineering School of Monastir, Energy Engineering Department, Laboratory of Metrology
and Energy Systems
5000.Monastir. Tunisia
1 meriem.ajmi@enim.u-monastir.tn
1 ajm.meriem@gmail.com
Abstract— A numerical investigation was conducted in order jet is still considered as one of the complicated flows in
to study the heat transfer process of an inclined turbulent term of its dynamic and turbulent features, that is why this
offset jet. The inclined bottom wall was subjected to a constant
heat flux. Ansys Fluent was the used CFD solver for work aims to study the thermal behavior of one specific
conducting the numerical simulations. Actually, the offset jet configuration. The investigated flow is a two-dimensional
was considered two-dimensional with an offset ratio (OR) of 9, offset jet attaching a heated wall, that makes a varying
while the inclination angle (α) was varying from -20° to +20°
inclination angle with the horizontal, from -20° to +20°.
and the Reynolds number( Re) was between 20,000 and
40,000. The effects of these two last parameters were examined Moreover, the fluid, the air, is ejected from a nozzle with an
in details. In this paper, a validation with old experimential OR of 9 and with various inlet velocities, that correspond to
data is provided aiming to prove the efficiency of the mesh size Re numbers going from 20,000 to 40,000. Thus, the effects
and the turbulence model choices. Furthermore, results have
shown that the thermal behaviour of such flow is mainly of these changing parameters will be discussed and
governed by its velocity, which is represented by a presented. Among the first scientific works on offset jets
corresponding Re number. Regarding the effect of the are Bourque and Newman [1]'s investigation, in 1960. In
inclination angle, it was found that the decrease of α intensifies fact, they examined experimentally offset jet aiming to
the convective heat transfer. However, α' s effect on the heat
transfer turned out to be not significant comparing to the Re' s understand its mean flow characteristics. Besides, in 1977,
effect, which was seen to be quite important and dominant. Patankar [2] solved numerically a three dimensional
Keywords— Offset jet, Turbulente, Inclination, heat transfer, turbulent jet flow field using the Reynolds Average Navier
numerical investigation Stocks, RANS, method. Additionally, Pelfrey and Liburdy
[3,4] conducted two experimental studies in order to
I. INTRODUCTION examine the flow field characteristics as well as the
Jet flows have been extensively investigated, either through turbulent features of such a flow. Furthermore, Nasr et al
experimental or numerical studies, and still they are the [5,6,7,8], studied extensively offset jet flows. First [5], they
subject of recent and current scientific works. This fact is compared experimentally, between the dynamic behavior of
owing to the multiple and diverse industrial applications of a two parallel plane jet and a simple offset jet to see which
this such particular type of flow. In fact, offset jets may be are the common and different features they present. It was
found in combustion chambers, for the mixing process, in found that the size of the recirculation region is
evaporator, in cooling and heating systems and in other considerably larger for a single offset jet compared to the
different applications. Hence, in this present work, we two parallel plane jets flow case. In addition [6], they
adopt a two-dimensional Offset Jet flow as the study's proved that the Standard k-ε present a good efficiency in
subject. First, offset jet is defined when the fluid is predicting this flow and the use of the side plates is
discharged from a nozzle, that is situated at certain distance, mandatory to enhance the two-dimensionality of the flow.
h, from the horizontal. Besides, it is characterized Finally [7,8] and most important, they investigated,
essentially by the "Coanda" effect, which makes the fluid numerically and experimentally, an inclined offset jet,
deflect towards the bottom wall, due to the creation of sub- mainly in the recirculation region. As results, the
atmospheric pressure zone that promotes the deviation of reattachment point was found to move downstream while
the fluid. Further, the fluid attaches the bottom wall at one the increasing of the inclination angle and the Reynolds
specific point, called "The reattachment point", Fig 1, Stress turbulence Model it was judged as the most adequate
which represents the limit point of the recirculation region turbulence model owing to its agreement with experimental
and the starting point of the impingement one. In fact, offset data. Understanding the dynamic flow field on an offset jet,

ISSN 1737-9334
in terms of its velocity field and its turbulence

characteristics, was the main purpose of the previously
mentioned scientific works. However, as the offset jets exist
in several thermal systems, it seemed mandatory to examine
the thermal characteristics of different application of offset
jet flows. That is why, other several studies of this
particular type of flow were done aiming to understand its
thermal behavior under certain conditions. In this regard,
Holland and Liburdy[9], studied experimentally a heated
and turbulent offset jet attaching an adiabatic wall. The
obtained temperature field measurements gave evidence on
Fig. 1 Adopted Geometry
the dependency of the thermal behavior on the offset ratio
value, essentially near the wall region. Further, Song [10]
examined the simultaneous effects of the wall inclination
and the offset ratio variations on the flow's dynamic
behavior and the flow's thermal characteristics. In this
study, the reattachment point positions was detected to be in
the near field of maximum local Nusselt number values,
which seemed interesting to find. More recently, Gao[11]
examined the Reynolds number and the offset ratio effects
on the measurements of the convective heat transfer
coefficient for a planar attaching jet. The results showed
that the local heat transfer coefficient presented a peak
almost at the reattachment point, similar to Song [10].
To conclude, offset jets are still an interesting research

Fig. 2 Working configuration
subject owing to their complex flow field behavior , their
sensitivity towards several parameters and most important • The flow is stationary and turbulent fully
their wide range of applications. That is why in this present developed.
work an inclined heated and turbulent offset jet is • The jet is considered two-dimensional.
examined. This paper aims mainly at studying the • The jet is ejected in the downstream direction.
simultaneous effects of two different and independent
• The air thermo-physic properties are assumed to be
parameters (Re and α ). The main reason of this scientific
constant.
work is to determine which is the most significant
• The heat flux is assumed to be constant.
parameters that affect the heat transfer phenomenon and
enhances it. Thereby, a two-dimensional and turbulent The Reynolds-Average Navier-Stocks and energy
offset jet will be investigated. The Re is in the range from equations are modeled in a Cartesian tensor as follows:
20,000 to 40,000, the OR is fixed to 9 and the inclination
angle from -20° to +20°, Fig. 2. 
( u i ) = 0 (1)
x i
II. MATHEMATICAL FORMULATION
 p    u u j 2 u   
During this work we adopted the same geometric
configuration as Kim and Yoon[11]. Air is issued from a x j
( u i u j ) = − +   i + − ij l   +
x i x j   x j x i 3 x l   x j
(
−u i u j ) (2)
two-dimensional nozzle, having an aspect ratio of 40, which

    T 
proves the two-dimensionality of this configuration.
x j
( Tu j ) = 
x j  Pr x j
− T ''  (3)
Besides, the lower wall is subjected to a constant heat flux. 
The chosen domain dimensions guaranteed no effect on the Equation (2), presents an additional term that reflects the
flow propagation and development. Moreover, we adopted effect of turbulence: It is the Reynolds stress constraint,
the following hypothesis to accurately solve the governing which should be modeled to close (2).
equations.

ISSN 1737-9334
The Boussinesq approximation gives a relationship between

the shear stress and the mean velocity gradient in order to
model the Reynolds stress constraint. The relationship is
presented as follows:
 u u j  2  
−u i u j =  t  i +  −  k +t u k  ij (4)
 x j xi  3 x k
 
Furthermore, Figure. 3 represents the different adopted

working conditions and the mesh distribution. As for the
emission conditions, a uniform velocity profile was applied
at the nozzle inlet, U=28.47 m/s, with a constant turbulence
intensity of 0.01% and a Reynolds number of 39,000. With
regards heat flux, as it was assumed to be constant, 1951
W/m², These conditions were used by Kim and Yoon[11].
III. RESULTS AND DISCUSSION
A. NUMERICAL VALIDATION
Validating our numerical code is the first done step in order
to show the efficiency of the chosen turbulence model and
the selected mesh size and type. In this regard, three sizes
of meshes, Mesh1, Mesh2, and Mesh3 each having 11,875,
33,600 and 52,332 non-uniform and quadratic cells,
respectively, were compared in terms of their capability of
detecting the local Nusselt number on the heated plate.
Figure 4 shows a mesh of 33,600 quadratic cells is quite
efficient. In addition, four different turbulence models,
RNG k-ε, Realizable k-ε, Standard k-ω, and SST k-ω, were
tested, Fig. 5, and the SST k-ω proves its great agreement
with the experimental result of Kim and Yoon[11].
Fig. 4 Mesh size test: (a): OR=5, (b): OR=6.5, (C): OR=7.5
Fig. 3 Adopted working conditions and the mesh distribution, 's' is the
spacing value and 'e' is the expansion ratio

ISSN 1737-9334
Fig. 5 Turbulence model test
B. THERMAL CHARACTERISTICS
The studied geometric configuration is presented in Fig.2.
This figure shows that the fluid is issued from a two-
dimensional nozzle, with different velocities that
correspond to Reynolds numbers between 20,000 and
40,000, besides the OR is set at 9. Moreover, the heated
wall inclination angle is between -20° to +20°, Fig.2.
During this work, the mathematical equations, RANS, that
govern this flow, are solved numerically in order to
investigate the bottom wall inclination and the Re
simultaneous effects on the heat transfer phenomenon.
Figure 6 illustrates the local variation of the Nusselt number,

Fig. 6 Local Nu distribution: (a): Re=20,000 (b): Re=40,000
detected on the heated plate, in the whole range of
inclination angle and two different Reynolds number,
On the other hand, the Re number variation seems to
20,000 and 40,000.
govern more, than the α variation, the convective heat
This figure demonstrates the effect of the wall inclination, transfer and its effect appear almost throughout the whole
as well as the Re number, on the convective heat transfer plate's length. In fact, the comparison between these two Re
between the ejected air and the heated wall. As for α's number outlines that the heat exchange between the
effect, it is clearly seen that every increase of α' value, from ambient air and the heated plate is intensified with nearly
-20° to +20°, leads to the enhancement of the heat transfer 45%. For instance, for α=+20° and Re=20,000, Numax is
process. However, this amelioration is mainly noticeable in around 55 while for the same inclination angle and
the limit of the recirculation region, where Numax presents a Re=40,000 it is detected to be nearly 100.
remarkable rise, that reaches 40% from α=-20° to α=+20°,
In fact, the forced convection process depends strongly on
with the increase of the inclination value. In fact, this result
the fluid velocity, in other words: the Re number, and as the
makes sense, as the fluid reaches the heated bottom wall
flow is turbulent fully developed, the kinetic forces are
with a higher Kinetic Energy that promotes the convective
important relative to the viscous forces and this fact
heat transfer phenomenon when α is increasing. Moreover,
ameliorates the convective heat transfer exchange between
throughout the impingement and wall jet regions, the
the fluid and the solid wall.
variation of the bottom wall inclination seems to have no
signifiacant effect.

ISSN 1737-9334
Further, in the recirculation region, the turbulence intensity

increases with α, for both Re number. While in the
impingement and wall jet regions and for low Re, I is
maximum for α=10°.
IV. CONCLUSION
This paper represents a numerical investigation of a two-
dimensional and turbulent offset jet. The offset ratio (OR)
was fixed in 9 while the Re number was varied from 20,000
to 40,000 and the bottom wall inclination was between -20°
and +20°. Besides the bottom wall was subjected to a
constant heat flux. The simultaneous effects of Re and α
were studied in order to see which parameter governs the
most heat transfer process. The local Nusselt distributions
on the bottom wall have shown that the enhancement of the
heat transfer, between the heated wall and the ejected fluid,
is more guaranteed with the increase of the fluid velocity,
which corresponds to the Re number. Furthermore, the
turbulence intensity highlighted the presence of a
relationship between the turbulent features of this particular
flow and the amelioration of the heat exchange
phenomenon.
REFERENCES
[1] S. V. Patankar, D. K. Basu and S. A. Alpay, "Prediction
of the Three-Dimensional Velocity Field of a Deflected
Turbulent Jet," Journal of Fluids Engineering, vol. 99,
pp. 758-762, 1977
[2] C. Bourque and B. G. Newman, "Reattachment of a
Two-Dimensional, Incompressible Jet to an Adjacent
Flat Plate," The Aeronautical Quarterly, vol. 11, pp.
201–232, 1960.
[3] J. R. R Pelfrey and J. A Liburdy, "Mean Flow
Characteristics of a Turbulent Offset Jet," Journal of
Fig. 7 Local distribution of the turbulence intensity I Fluids Engineering, vol. 108, pp. 82–88, 1986.
[4] J. R. R Pelfrey and J. A Liburdy," Effect of curvature on
Moreover, in order to investigate the turbulent intensity the turbulence of a two-dimensional jet," Experiments in
Fluids, vol. 4, pp. 143–149, 1986.
related to the same (Re, OR, α) combinations, Fig. 7 gives [5] A. Nasr and J. C. S Lai,"Comparison of flow
information on the local distribution of I. characteristics in the near field of two parallel plane jets
and an offset plane jet," Physics of Fluids, vol. 9, pp.
The first thing to note is that Nux and Ix present almost the 2919–2931, 1997.
same local paces, which highlights the dependecy [6] A. Nasr and J. C. S Lai, "A turbulent plane offset jet
with small offset ratio," Experiments in Fluids, vol. 24,
relationship betwwen the heat transfer and the turbulent pp. 47–57, 1998.
features. Moreover, I and Nux peak almost at the same [7] A. Nasr and J. C. S Lai, "The effects of wall inclination
downstream positions, which means that the maximum heat on an inclined offset jet," 2006.
[8] A. Nasr, J. C. S Lai and O. Siah, "Young, J. Flow
transfer is along with a maximum turbulence intensity.
Characteristics of an Inclined Offset Jet " in 36th AIAA
Fluid Dynamics Conference and Exhibit, Fluid
In addition, unlike the Nux evolutions, the Re has partically Dynamics and Co-Located Conferences American
no effect on I. In fact, only a slight increase of I is noted Institute of Aeronautics and Astronautics, San
with the rise of Re number. Meanwhile, the inclination's Francisco, USA, 2006.
[9] J .T. Holland and J. A. Liburdy, "Measurements of the
effect turns out to be more significant as it is more thermal characteristics of heated offset jets"
noteworthy. International Journal of Heat and Mass Transfer,vol.
33, pp. 69–78, 1990.

ISSN 1737-9334
[10] H. B. Song, S. H. Yoon , D. H. Lee, "Flow and heat

transfer characteristics of a two-dimensional oblique
wall attaching offset jet," International Journal of Heat
and Mass Transfer, vol. 43, pp. 2395–2404, 2000.
[11] D. S. Kim, S. H. Yoon, D. H. Lee, K. C. Kim, "Flow
and heat transfer measurements of a wall attaching
offset jet," International Journal of Heat and Mass
Transfer, vol. 39, pp. 2907–2913, 1996.

ISSN 1737-9334
Effects of Carrier Mobility and defects on the

Recombination Characteristic of P3HT:
Graphene Bulk Heterojunction Solar Cell
Chahrazed Dridi#1, Naima Touafek*2
#
Laboratoire de physique mathématique et subatomique LPMS ,Fréres Mentouri Constantine1 University ,Algeria,2500
chazed2011@gmail.com
*
Higher National School of Biotechnology “Toufik Khaznadar” (ENSB), Constantine, Algeria University
ntouafek@yahoo.fr
Abstract— Several extensive studies have been carried out for Also it has properties such as high electrical conductivity
poly(3-hexylthiophene) (P3HT) mixed with carbon derivatives and high solubility in different solvents [6] mixed with
such as phenyl-C61-butyric acid methyl ester (PCBM) in organic graphene, which is an excellent acceptor [7],[8] its mobility is
solar cells, but few searches found on the blend of p3HT remarkably high 1E4cm² / Vs, it is highly transparent [9] and
:graphene.
its electrical conductivity is very high [10],[11] .
We have used the Solar Cell Capacitance Simulator (SCAPS)
and we have choose the effective medium model (EMM) to This blend of P3HT:Graphene constitutes the active layer
simulate the active layer lightly doped P, the appropriate of the studied organic cell.
parameters have been introduced in a platform of (SCAPS) We have used the effective medium model and the
Simulator. First we optimized the thickness of the active layer, appropriate parameters have been introduced in a platform of
where we model the characteristics of the solar cell: efficiency the Solar Cell Capacitance Simulator (SCAPS).
(η), open circuit voltage (Voc), short circuit current (Jsc) and fill
factor (FF), next we varied the holes mobility and the defects
density and we calculated the characteristics of the II. THEORETICAL MODEL
recombination rate according to these two variables.
The bulk heterojunction organic solar cell is composed of
two materials which each have two orbitals called highest
Keywords—Organic Cell, Conjugated Polymer , poly(3- occupied molecular orbital (HOMO) and lowest unoccupied
hexylthiophene) (P3HT), Graphene, SCAPS simulator. molecular orbital(LUMO).
The P3HT is a first material, is an electron donor and has
the lowest ionization potential (Ei).
I. INTRODUCTION
The second material is a Graphene , is an electron
In an attempt to solve energy and environmental problems, acceptor and has the highest electron affinity (χa) [12].
as an alternative feasible to oil combustion, the world of After the donor material has absorbed light there is
photovoltaics has evolved, mainly inorganic semiconductor creation of electron-hole pairs called excitons, the latter
cells [1]. diffuse towards the interface between the electron donor and
Inorganic photovoltaic devices have been widely used acceptor material where an electron can easily be transferred
since the 1970s because of their high power conversion between the LUMO of the donor material (higher in energy)
efficiency [2]. and the LUMO of the acceptor material, the most important
Compared to inorganic photovoltaic devices, organic is that the exciton reaches an interface.
photovoltaic (OPV) devices have many advantages, including For this, it is necessary that the distance between the
mechanical flexibility, the possibility of treatment in solution donor material and the acceptor material be less than 10nm.
and a low specific weight [3],[4]. Once the interface is reached the exciton of the electron-hole
As a result, the exploration of organic photovoltaic devices pair dissociates into a polaron.
has developed, especially solar cells with conjugated After the dissociation, the charges are transported to the
polymers. electrodes The last step is the extraction of the charges Fig.1.
In this work we chose P3HT (poly (3-hexylthiophene) a p-
type semiconductor conjugated polymer is one of the most
studied conjugated polymers that is suitable for organic solar
cells because of its appropriate absorption spectrum and low-
cost manufacturing potential [5].

ISSN 1737-9334
There are several recombination mechanisms in the

organic cells in our work we took into account that of
recombination by traps
III. METHODOLOGY
Several considerations must be taken into account when
simulating the organic solar ones such as the dielectric
constant ε which depends on the weight / donor / acceptor
ratio of the materials, the effective mobility of the carriers
Fig. 1 Energy diagram of HOMO and LUMO orbitals of donor and (μn in the acceptor and μp in the donor) in the BHJ
acceptor materials. [12 ] configuration is different from the value in their initial
phase, which is closely related to the nano-morphology of
The equations that govern the studied model are the the disordered bulk heterojunction organic cell.
Poisson equation and the continuity equations for electrons The nanostructure p-n is represented by a single, lightly
and holes given by: doped semiconductor layer P; we have considered the
effective medium model (EMM) Fig.2,this effective medium
will be characterized by an average of the properties of the
materials n,p, it's supposed that the two materials are very
well interpenetrated in the active layer, with a distance
between all the points of the mixture and a donor / acceptor
interface being less than the diffusion length of the excitons.
To create a driving force for the separation of the electron-
hole pairs the contacts are considered to be selective, ie a
contact accepts only the electrons and the other accepts only
the holes[15].
where 𝜑 describes electrical potential, q is the unit charge,

𝜀0 , 𝜀𝑟 is the dielectric constant of vacuum and the
semiconductor respectively ; n and p are electron and hole
density, respectively; 𝑁+ −
𝐷 𝑎𝑛𝑑 𝑁𝐴 the density of ionized
donors and acceptors, 𝜌𝑡 the charge density of defects , 𝐽𝑛 and
𝐽𝑝 the electron and hole current density; G is the generation
rate and Rn(p) is the recombination loss.
Equation 4 define the transportation of carriers:

Fig.2 Schema of the effective medium model (EMM).
Parameters necessary for the simulation of the active layer

are extracted from the literature such as the permittivity, the
band gap as well as the absorption coefficient of the P3HT:
graphene blend [6], subsequently are introduced on the
platform of the simulator capacity of solar cells in one
dimension (SCAPS-1D). Fig.3 and are given by Table 1.
where 𝜇𝑛(𝑝) is the electron/hole mobility and Dn(p) is the
diffusion coefficient .
The generation rate is given by:
where 𝜑0 is the incident photon flux , x is a depth and α the

absorption coefficient of the active material.
Fig.3 Solar cell active layer P3HT:Graphene
Copyright - 2019
ISSN 1737-9334 Page 40
charge concentration on the recombination rate,that increases

Table 1 Parameters used in simulation
proportionally with defect s density.
Parameter Symbol Value Unit The recombination near the electrode can be regarded as
dark carrier recombination[14].
Thickness d 250 nm This mixture of materials P3HT(donor) and
Band gap Eg 2.065 eV GRAPHENE(acceptor), allows to increase the interfacial
Electron affinity χ 3.4 eV zones between the two, and consequently to reduce the
Effective densit of states Nc,Nv 2E18 𝑐𝑚−3 recombination problems of photogenerated excitons far
Electron mobility µn 1E4 𝑐𝑚2 /𝑉𝑠 from the interface
Hole mobility µp Vari 𝑐𝑚2 /𝑉𝑠
Dielectric constant ε 6.3
IV. RESULTS AND DISCUSSION

Thickness is a very important parameter used to improve
the properties of semiconductor devices, such as photovoltaic
devices. The thickness of the active layer has been changed
from 80 to 900 nm in order to determine the optimum
thickness of the active layer according to the simulation
shown in Fig.4.
The optimized value of the layer is indicated at ~300
nm. In conclusion, with the increase in the thickness of the
active laye r beyond 300nm, the values of the effeciency
"eta", short-current circuit "Jsc" remain almost constant,
whereas the open-circuit voltage "Voc" continues to increase
with increasing active layer thickness and the fill factor 'FF' (a)
gradually decreases. All the performance parameters vary in a
better direction with thickness increasing.
(b)
Fig.4 Optimizing the thickness active layer Fig.6 Recombination rate:(a) effect of mobility ;(b) effect of defects
In this research paper, we found the recombination rate of

V. CONCLUSION
solar cell at different charge mobility and different charge
defects concentration. In BHJ solar cell, in which the blend is
treated as one effective medium with an effective band- gap The ITO / P3HT: GRPHENE / Al architecture P3HT solar
given by LUMO (lowest unoccupied molecular orbit) of cell is designed and analyzed using a solar cell capacity
acceptor and HOMO (highest occupied molecular orbit) of simulator.
donor [13]. Absorbent layer thickness, defect density and hole mobility
The recombination occurs mostly near semiconductor and influence solar cell performance
electrode interface and there is less recombination in the Hence this simulation work will provide the handy
bulk. information in fabricating P3HT:Graphene solar cells to
Figure6(a) shows the variations of total recombination in reasonably choose material parameters and to achieve the
P3HT/Graphene solar cell with the mobility changing from good performances.
1E-3 to 0.2 cm²/Vs ,that basically decreases with increasing
mobility. It can be seen in fig.6(b) the influence the defects
Copyright - 2019
ISSN 1737-9334 Page 41
ACKNOWLEDGMENT [9] Z.in, S.Sun, T.Salim, S.Wu, X.Huang, Q.He, Y.M. Lam, H.Zhang,
"Organic photovoltaic devices using highly flexible reduced graphene oxide
The authors would like to thank Marc Burgelman and his films as transparent electrodes", ACSNano vol.4 ,pp.5263–5268, 2010.
team at Department of Electronics and Information Systems [10]G. Jo, S. Na, S. Oh et al., “Tuning of a graphene-electrode work function
(ELIS) of the University of Gent, Belgium for the access of to enhance the efficiency of organic bulk heterojunction photovoltaic cells
with an inverted structure,” Applied Physics Letters, vol. 97, no. 21, pp.
simulator SCAPS . 213301–213303, 2010.
[11] X. Du, I. Skachko, A. Barker, and E. Y. Andrei, “Approaching ballistic
REFERENCES transport in suspended graphene,” Nature Nanotechnology, vol. 3, no. 8, pp.
491–495, 2008.
[1] A. Goetzberger, C. Hebling, and H.-W. Schock, "Photovoltaic materials, [12] L. Darchy," Nanotubes de carbones semi-conducteurs pour cellules
history, status and outlook", Materials Science and Engineering R solaires organiques,"these doctorat, Ecole Polytechnique X, France,Oct.
vol.40,pp.1-46, 2003. 2013.
[2] Deibel, C. and V. Dyakonov, "Polymer-fullerene bulk heterojunction [13] S.Boukli Hacène,"Modélisation et simulation des cellules solaires à
solar cells", Reports on base de P3HT : PCBM (effet de la mobilité et la vitesse de recombinaison en
Progress in Physics, vol.73(9), 2010. surface)" thèse Doctorat, Université Abou Bakr Bekkaid,Tlemcen,Algerie,
[3] G.Li, R. Zhu, and Y. Yang, "Polymer solar cells", Nature Photonics, 2014.
vol.6(3), pp. 153-161,2012. [14]Ying-Xuan Wang, Shin-Rong Tseng, Hsin-Fei Meng,a_ Kuan-Chen
[4]A. Facchetti, "Polymer donor-polymer acceptor (all-polymer) solar cells", Lee,Chiou-Hua Liu, and Sheng-Fu Horng," Dark carrier recombination in
Materials Today, vol.16(4),pp. 123-132, 2013. organic solar cell " , APPLIED PHYSICS LETTERS ,vol.93, 133501 ,2008.
[5] Jarmila Müllerová , Michal Kaiser , Vojtech Nádazˇdy , Peter Šiffalovicˇ , [15]C.Bendina,"Etude et Simulation des Proprietes Electriques et Optiques
Eva Majková , des Cellules Photovoltaiques organiques multicouches",thése Doctorat,
"Optical absorption study of P3HT:PCBM blend photo-oxidation for bulk Université Abdelhamid ibn Badis Mostaganem,Agerie,Fev.2019.
heterojunction solar cells", Solar Energy ,vol.134 pp.294–301,2016.
[6]V. Saini, O. Abdulrazzaq, S. Bourdo, E. Dervishi, A. Petre, V. G. Bairi,
T. Mustafa, L. Schnackenberg, T. Viswanathan, and A. S. Biris, “Structural
and optoelectronic properties of P3HT-graphene composites prepared by in
situ oxidative polymerization,” J. Appl. Phys., vol. 112, no. 5, pp. 54327–
54329, 2012.
[7]Q.Liu, Z.Liu, X.Zhang, N.Zhang ,L.Yang, S.Yin, Y.Chen,"Organic
photovoltaic ccells based on an acceptor of soluble graphene",
Appl.Phys.Lett., vol.92, 223303, 2008.
[8] Z. Liu,Q.Liu,Y.Huang,Y.Ma,S.Yin,X.Zhang,W.Sun,Y.Chen,"Organic
photovoltaic devices based on a novel acceptor material:graphene",Adv. .
Mater,vol20,2008
Copyright - 2019
ISSN 1737-9334 Page 42
Performances and Robustness Assessment of Sliding

Mode Control Applied to a High Frequency
Switched Parallel Three-Cell DC-DC Converter
Rihab Hamdi#1, Amel Hadri Hamida#2
#
LMSE Laboratory, Department of Electrical Engineering
University of Biskra BP 145 RP, 07000 Biskra, Algeria
1rihab.hamdi2012@gmail.com
2am_hadri@yahoo.fr
Abstract— In the field of strong currents with high switching inherent advantages that makes it an increasingly attractive
frequencies, new structures based on the combination of solution which can be considered for any application.
components have been developed. Among them, we find the It provides more degrees of freedom to the designer, but it
parallel multicellular converters. also makes design more complicated since many options can
This paper features a variable structure non-linear sliding mode
controller (SMC) for closed loop control of DC-DC three-cell
be considered. Indeed, a significant increase in the current
buck converter feeding parameters uncertainties and load required by the load can be satisfied by driving several
disturbance, in order to overcome the design constraints and to branches of the power supply simultaneously.
mitigate the undesired transient response. Thus, when the phase-shifted currents add up to form the
First, a nonlinear system modelling is derived for open-loop output current, the ripple ratio is significantly reduced with a
Buck converter with state variables of the input current and the frequency n times greater than the switching frequency [4].
output voltage. The model is extended for the closed-loop system For this reason, it seems important to develop an approach
while employing a Sliding Mode Control (SMC) solution. After that allows fast evaluation of several solutions with different
the initial analysis of this converter, we utilize numbers of series and parallel cells, the particular case of one
MATLAB/Simulink packages to analyse detailed performances
as important parameters are varied.
cell in series and one cell in parallel corresponding to the case
Simulations are carried-out on the three-cell Buck converter to of standard two-level conversion [5].
assess the performances and the robustness of the proposed Besides, modelling is a very important step for control
control approach. Indeed, featuring robustness and high laws and since DC–DC converters are nonlinear systems,
accuracy, they cope with system uncertainty keeping a properly they represent a big challenge for control design. The control
chosen constraint by means of high-frequency control switching. methodologies for DC-DC converter are designed for tight
regulation of out-put voltage amidst varying input and
Keywords— DC-DC converter, High switched devices, loading conditions, keeping in mind the unpredictable nature
Multicellular converter, Sliding mode control, Buck converter, of such diverse plants. Meanwhile, the fixed-frequency PWM
Robustness control is by far the most popular control technique used for
the regulation of DC–DC converters that assume two specific
I. INTRODUCTION forms, namely voltage feedback control and current-
Power electronics are currently embedded in all electric programmed control and operate optimally only for a specific
devices; they play an important role in the modern-day condition. The limit of this strategy results in the small-signal
electronics and electrical applications. The development in assumptions, the converter’s behaviour cannot be adequately
the power electronics field is the result of the current energy captured when there is a large transient.
challenges that require power converters working with higher The failure of conventional linear and partial nonlinear
voltages and currents. Among the most characteristic control schemes, namely, PWM voltage and PWM current-
elements of this development, the conception of new power mode control, to operate satisfactorily in large-signal
converters topologies is an important topic. operating conditions, is the main motivation for investigating
Design of such converters is a challenge for electrical alternative methods of controlling DC–DC converters.
engineers due to the use requirements and the inherent An overview of the various control methodologies being
nonlinear nature. All the converters have few requirements considered and have been found to have potential applications
like high density which leads to smaller size, high efficiency in DC–DC converters that require very high performance in
which leads to low losses and robust to any changes in the dynamical response such the sliding mode control that offers
input or output [1]. For surpassing the problem of high excellent large-signal handling capability.
frequency, a new class of power converter appeared: the A variable structure non-linear control scheme like the
multicellular converter [2], [3]. Multicellular conversion has sliding mode control acts as a perfect fit for controlling a
Copyright - 2019
ISSN 1737-9334 Page 43
highly non-linear and time varying system like the DC-DC the output capacitor; RL is the equivalent load. Then the
switched mode converter. SMC is a powerful method that is system dynamics of the buck converter can be expressed as:
able to yield a very robust closed-loop system under plant di
uncertainties and external disturbances [6]. This controller L L = −iL − Vout + u.Vin (1)
ideally operates at an infinite switching frequency such that dt
the controlled variables can track a certain reference path to dV V
achieve the desired dynamic response and steady-state C out = iL + out (2)
operation [7], ant it is adopted to suppress the input dt RL
disturbance and reduce the effects from the load variation [8]. Where iL is the inductor current; Vout is the output
Moreover, the sliding mode control may reduce the capacitor DC voltage; u is the control input.
complexity of feedback control design through decoupling of
system into the independent subsystems of lower dimension.
Because of these properties, the diverse applications of the
sliding mode control methodology can be found in the areas
of electric motors, manipulators, power systems, mobile
robots, spacecraft and automotive control. However, the main
drawback of the sliding mode control is undesirable
oscillations having finite amplitude and frequency due to the
presence of unmodelled dynamics or discrete time
implementation. This destructive phenomenon, so-called Fig. 1 Basic topology of the buck converter
‘chattering’, may lower control accuracy or incur an
unwanted wear of a mechanical component. In the literature, The considered buck converter is a second order converter,
various solutions to reduce the chattering have been studied with Single Input Single Output (SISO) structure, where, the
[9], [10], [11]. Several publications have appeared in recent cells are associated in parallel with R load, realized in
years have shown interest in utilization of the relative design MATLAB/SIMULINK.
simplicity and the robustness of SMC to deal with DC-DC
A. Multicellular Converter
converter control problem [12]-[16].
A DC-DC converter controlled using sliding mode theory Multicellular converter consists of phases, where each cell
yields large signal stability counter to the linear conventional contains two complementary power electronics components
controllers which guarantee desired performance against only and it is controlled by a binary switch.
small perturbations in input voltage or load current [17], [18]. As shown in Figure.2, a 3-phase converter feeds a single
Hereinafter, section 2 deals with the model of the converter load assumed to be resistive.
and the concept of the conventional sliding mode control
methodology for DC-DC buck converter. While the following
section presents the simulation results and brief discussion of
the same. Finally, the conclusion is reported in section 4 that
draws comprehensive inferences from the work performed
and the observations made.
II. CONVERTER MODELLING AND SMC CONCEPTION

DC-DC converters constitute a particular class of nonlinear,
time-varying systems. The periodic repetition of the sequence
absorption-transfer results in a cyclical variation of the
converter topological configuration, which leads to a
repetitive sequence of two or three circuit structures whose
time durations are not constant due to the modulation of TON
and/or TOFF. Therefore, DC-DC switching converters can be
considered as variable structure systems and the voltage
regulation has been traditionally undertaken by means of
linear control based on a small signal model of the converter
in the frequency domain. The sliding-mode approach based
on the equivalent control method is particularly appropriate Fig. 2 Studied p-cells Buck Converter
for designing the regulation loop of a DC-DC switching
Our studied multiphase step-down power converter is
converter [19].
composed of several DC/DC Buck converters with an output
Figure.1 depicts a basic topology of the buck converter, in
which Vin is the input DC voltage; L is the filter inductor; C is capacitor. A 3-phase converter can be modelled by the
following set of differential equations:

ISSN 1737-9334
 0 
 x1  
dik  0 1  0 
L = −ik − Vout + Vin .uk ; k = 1, 2, 3 (3)  x1    
= 1    + Vin  u + Vref (7)
 x2  −
dt 1
−   x  −   
L
dVout 3
=  ik −
Vout
(4)
 LC RC  2  LC   LC 
dt k =1 R
Where ik is the circulating current through the 3-phase, Vout B. Conventional Sliding Mode Control
the output voltage and uk the k-phase control input which Sliding modes are obviously adopted as the three-cell
takes values in the set {0,1}. Identical phase parameters are converter has at least one discrete control variable. We start
assumed, namely, inductances (L), capacitors (C) and input by choosing the sliding surfaces whose number and shape are
voltages (Vin). determined to ensure a convergence of output values to the
Each cell comprises a chopper DC-DC converter and a reference values [20]. The control aims to ensure the
single protection transistor connected in a high-voltage convergence of switching surfaces Si to zero, to allow the
portion of the converter thereby enabling any of the cells to reaching of the state variables to their references.
be taken out of service independently of the other cells while For an SM voltage controller, the switching function u can
minimizing power consumption in normal operation. be determined from the control parameters x1 and x2 using the
Figure 3 illustrates the configuration adopted of the three- state trajectory computation:
cell step down DC-DC converter associated in parallel and
realized in MATLAB/Simulink environment. S =  x1 + x2 = Jx (8)
Where β is a positive parameter, J =[β,1]; and x =[x1,x2]T.
By enforcing S =0, a sliding line with gradient β can be
obtained. The purpose of this sliding line is to serve as a
boundary to split the phase plane into two regions.
The choice of the surface is based on the objectives to be
achieved, which generally represent the desired behaviour of
the quantities to be regulated.
S = k P1 (Vref − Vout ) + k P 2iC (9)
Where kP1=1/RC and kP2= -α/C. In the above equation, the

terms (Vref − α Vout) and iC are the feedback state variables
from the converter that should be amplified by gains kP1 and
kP2, respectively, before a summation is performed.
Synchronization among phases can be attained through the 3
switching surfaces.
It is simpler to reconfigure the switching function to the
following description:
C C
S=  x1 + x = Q. x (10)
  2
Fig. 3 Three-cell Buck converter associated in parallel
C C  T
To illustrate the operating principle, the state-space where Q =
     and x =  x1 x2  then we get
description of the buck converter under SM voltage control is
represented.
− V ) − i
1
x1 = Vref − Vout S= (V (11)
(5)  R ref out C
dVout  Vout u.V − Vout Thus, the practical implementation of S becomes

x2 = x1 = − = ( −  in dt ) (6) independent of C, thereby reducing the amplification of the
dt C R L feedback signals. The control law generated by a sliding
where L, C and R are the inductance, capacitance and mode controller is composed of two essential parts. The first
instantaneous load resistance, respectively; Vref, Vin, and is called equivalent control variable denoted Ueq, obtained by
αVout are the reference, input voltage, and output voltage, the exact linearization and serves to maintain the state on the
respectively; u = 1 or 0 is the switching state of power switch sliding surface. The second discontinuous control quantity Un
SW. Therefore, the model can be obtained as developed to verify the convergence condition. To determine
the two control quantities, calculate the derivative of the
expression of the sliding surface according to the parameters
Copyright - 2019
Page 45
ISSN 1737-9334
of the system model given by (7). The derivative of the where ξ is an arbitrarily small positive quantity.
sliding surface is: Substituting (7) and (18) into (20), the inequalities become
S S x C 1 1 Vref −Vout
S ( x) = = (12) 1 = ( − )x − x + 0 (21)
t x t  R 2 L 1 L
S S Vref
S ( x) = ( A( x, t ) + B ( x, t ).U eq ) +
B ( x, t ).U n (13) 2 = (
C
−
1
)x − x +
1
0 (22)
x x  R 2 L 1 L
Assuming that during slip mode and during steady state
establishment the slip surface is zero, this implies that its Where
derivative and the discontinuous control magnitude are also 1=Q.x
 0 S 
zero. The expression of the equivalent order is then the  (23)
following: 2 =Q.x
 - S 0
S −1 S Hence, to ensure that β is high enough for fast dynamic
U eq = −( B ( x, t )) . A( x, t ) (14) response and low enough to maintain a large existence region,
x x it is sufficient to set
According to (14), the existence of the equivalent order
depends on the term ∂S/∂x B(x,t) which must not be zero. In =
1
(24)
this case, by replacing the expression of the equivalent RC
command in (13), we can write:
On the other hand, if a faster dynamic response is desired,
U n = −U Max .sat ( S ) (15) β can be reduced, as long as the existence condition of the
system is met. For this purpose, it is necessary to take into
S
 if S  1 consideration the operating condition of the converter.
sat ( S ) =  1 (16)

sgn(S) if S  1
III. SIMULATION RESULTS
A. Simulation Results and Robustness Assessments
−1 if S 1
sgn( S ) =  (17) In order to illustrate the performance of the proposed
1 if S 1 control, we considered a three-cell converter connected to an
UMax: Positive constant and l the threshold width of the R load. The performances of the developed sliding mode
saturation function. control were verified through simulation using MATLAB
The specification of the switching state for each sector in software.
the case of a second-order system like the buck converter can
be considered as for u=1 must be employed so that the 1) Robustness versus power reference variation
trajectory is directed toward the sliding line. Conversely, Robustness test versus the power reference value changes
when the phase trajectory is at any position below sliding line, are illustrated by figures 4 and 5. Indeed, at t = 0.2 s, Vref is
u=0 must be employed for the trajectory to be directed toward slightly changed. According to these results, it is obvious that
the sliding line. This forms the basis for the control law: the control objective is fulfilled.
1=' ON '
 when S 0
u= (18)
0=' OFF '
 when S 0
The system trajectory will eventually reach the sliding line,
and the control law in (18) provides the general requirement
that the trajectories will be driven toward the sliding line.
To ensure that the trajectory is maintained on the sliding
line, the existence condition, which is derived from
Lyapunov’s second method to determine asymptotic stability,
must be obeyed:
lim S .S 0 (19)
S→0
Thus, by substituting the time derivative of (8), the
condition for SM control to exist is:
Fig. 4 Simulation result for output voltage due to a step change in Vref

 Jx 0 for 0 S 
S= (20) from 8V to 12V

 Jx 0 for - S 0

ISSN 1737-9334
Fig. 5 Simulation result for load current due to a step change in Vref
Fig. 7 Load current response with a variation of 50% of the load
from 8V to 12V
resistance
2) Robustness versus load resistance variations According to these results, it can be noticed that the
To test the performance of the proposed controller, a performances of the proposed control for load variation are
comprehensive simulation under step load change (variation satisfactory.
of 50%) with constant input voltage was tested. The reference Then, the above two tests confirm the robustness of the
output voltage was set to 12 V and the load demand was set proposed sliding mode control strategy for the three-cells
to 3A. converter. It rejects external perturbation and some controlled
The situation where the load changes suddenly from one system parameters variations. The robustness of the proposed
value of load resistance to another is considered. This SMC was tested with a load resistance variation of 50%.
particularly interesting because it is a typical problem for
power electronics, where the power supply is supposed to 3) Robustness versus input voltage variations
compensate quickly for the load variation. Finally, the input voltage was varied to corroborate the line
Robustness test versus the load resistance are illustrated by regulation of the converter. To test the robustness of the
figures 6 and 7. developed control, a first disturbance is assigned to the level
These figures show that the state variables exhibit a of the supply voltage. The converter is initially powered by a
transient but the promptly converge to their respective input voltage Vin = 11.5V which varies at instant t = 0.1s to
reference. reach 12V
The input voltage was fixed to 24 V, a step load change Figures 8 and 9 summarize the results of this test. There is
occurs, The load resistance changes from 2Ω to 4Ω. The drop a very brief transient that lasts a few ms followed by a steady
in the regulated output voltage went up by 0.5 V and it took state giving voltage values Vout proportional to the new value
about 0.003s for it to settle down at 12 V. of Vin.
Fig. 8 Simulation results of the Output voltage response with a input

Fig. 6 Output voltage response with a variation of 50% of the load voltage variation
resistance

ISSN 1737-9334
[8] A. Hadri Hamida, A. Ghoggal, Fatiha Khelili and S. Zerouali, Second-

Order Sliding
[9] Mode Control Scheme with a Non-Linear Phenomenon Analysis of a
DC-DC Power Converter Dedicated to Distributed Power Systems”,
International Conference on Electronics, Energy and Measurement,
Algiers, (2018)
[10] Hoon Leea, Vadim I. Utkinb and Andrey Malininc. Chattering
reduction using multiphase sliding mode control, International Journal
of Control Vol. 82, No. 9, (2009), 1720–1737
[11] Aroudi, A., Robert, B., and Leyva, R. (2005), ‘Sliding Mode Control
of a High Voltage DC-DC Buck Converter in Circuit Theory and
Design, Proceedings of the 2005 European Conference (Vol. 3,28),
Cork, Ireland, pp. III/55–III/59.
[12] Lee, H., and Utkin, V.I. (2006), ‘The Chattering Analysis’, in
Advances in Variable Structure and Sliding Mode Control, Lecture
Notes in Control and Information Sciences, eds. C. Edwards, E. Fossas
Colet, & L. Fridman, Berlin: Springer.
[13] Daniel González Montoya; Carlos Andrés Ramos-Paja; Roberto Giral.
Improved Design of Sliding-Mode Controllers Based on the
Requirements of MPPT Techniques, IEEE Transactions on Power
Fig. 9 Simulation results of the load current response with an input Electronics, (2016)
voltage variation [14] Mahdi Salimi, Jafar Soltani, Adel Zakipour, Navid Reza Abjadi.
Hyper-plane sliding mode control of the DC–DC buck/boost converter
IV. CONCLUSIONS in continuous and discontinuous conduction modes of operation, IET
Power Electron., pp. 1–10, (2015)
A variable structure non-linear control technique has been [15] Junxiao Wang; Shihua Li; Jun Yang; Bin Wu; Qi Li. Extended state
presented and implemented successfully for a high switched observer-based sliding mode control for PWM-based DC–DC buck
three-cell DC-DC converter associated in parallel. The power converter systems with mismatched disturbances, IET Control
performance of dynamic evolution under step load change Theory & Applications (2015).
[16] Zhao, Yue; Qiao, Wei; and Ha, Daihyun, "A Sliding-Mode Duty-Ratio
and step reference output voltage change condition has been Controller for DC/DC Buck Converters with Constant Power Loads"
tested, the proposed dynamic evolution controller (2014).
outperforms a conventional sliding mode controller with [17] Oswaldo López-Santos, Luis Martínez-Salamero, Germain García1.
disturbance rejection. Hugo Valderrama-Blavi3, Daniel O. Mercuri. Efficiency analysis of a
sliding-mode controlled quadratic boost converter, IET Power
The main advantage of the SM converter is the simplicity Electron, Vol. 6, Iss. 2, pp. 364–373, (2013]
in the controller’s design and implementation. Unlike [18] S. Tan, Y. Lai, C. Tse. Sliding Mode Control of Switching Power
conventional current-mode and voltage-mode controllers, Converters. CRC Press, (2011)
which require special techniques. [19] J.F. Tsai, Y.P. Chen. Sliding mode control and stability analysis of
buck DC-DC converter. International Journal of Electronics. 94 (2007).
The carried-out simulations show very promising results in [20] L. Martinez-Salamero, A. Cid-Pastor, A. El Aroudi, R. Giral, J.
terms of reference tracking performances and robustness. Calvente, and G. Ruiz-Magaz. Sliding-Mode Control of DC-DC
They prove the appropriateness of sliding mode control for Switching Converters, Proceedings of the 18th World Congress. The
such kind of the system. International Federation of Automatic Control Milano (Italy) August
28 - September 2, (2011)
[21] Ben Said, Ben Saad K, Benrejeb M. Sliding mode control for multicell
converters. In: International conference on control, engineering and
REFERENCES information technology, vol.2, pp39-44 (2013)
[1] Hamit Erdem, Comparison of Fuzzy, PI and Fixed Frequency Sliding
Mode Controller for DC-DC converters. IEEE International Aegean
Conference on Electrical Machines and Power Electronics, pp 684-689
(2007)
[2] Ben Said S, et al., HIL simulation approach for a multicellular
converter controlled by sliding mode, International Journal of
Hydrogen Energy (2017).
[3] Al-Hammouri AT, Nordstrom L, Chemine M, Vanfretti L, Honeth N,
Leelaruji R. Virtualization of synchronized phasor measurement units
within real-time simulators for smart grid applications. In: Power ang
Energy society general meeting. IEEE; 2012. P. 1-7.
[4] A. Hadri Hamida, A. Ghoggal, Fatiha Khelili and S. Zerouali, “Étude
qualitative des convertisseurs multicellulaires entrelacés par la théorie
de bifurcation”, Proc. Conf. SGE 2018, Nancy, France, (2018).
[5] Thierry. Meynard, Analysis and Design of Multicell DC/DC
Converters using Vectorized Models, ISTE Ltd and John Wiley &
Sons, Inc. (2015)
[6] V. Utkin, J. Guldner, and J. X. Shi, Sliding Mode Control in
Electromechanical Systems. London, UK: Taylor and Francis, 1999
[7] Yin, Y. Liu, J. Vazquez, S. Wu, L. Franquelo, Leopoldo.G.
Disturbance observer based second order sliding mode control for DC-
DC buck converters. IEEE Industrial Electronics Society, 43rd Annual
Conference. (2017)

ISSN 1737-9334
Effect of the precursor solution on the structural

morphological and optical properties of ZnO thin
films deposited by Spray Pyrolysis technique for
solar cell application
H. Benzarouk#1, M. Mekhnache#2, A. Drici*3, A. Amara*4
#1,2
20 Aout1955 University; Skikda 21000, Algeria
#1
h.benzerouk@univ-skikda.dz
#1
b_hanin08@yahoo.fr
*3,4
Department of Physics, Badji Mokhtar University, BP 12, Annaba 23000, Algeria
Abstract—The objective of this research is to study the structural, coating as it absorbs the UV radiation such that lower energy
morphology and optical properties of zinc oxide thin films photons are able to reach the absorber layer [6].
deposited by spray pyrolysis technique. The ZnO thin films were ZnO thin films can be produced by several techniques,
prepared and grown on glass substrate at 400°c, using two including: electrodeposition [7], spray, thermal evaporation,
different precursor solutions namely zinc acetate, and zinc
reactive spraying, sol gel [8] and laser ablation etc. [9].
chloride with a molarity of 0.2M/L. Our interest is on the
investigation of the precursor solution on structural; The technique of spray pyrolysis is a simple, less expensive
morphological and optical properties of ZnO thin films. The X- and allows the production of highly pure and ultrafine
ray diffraction patterns indicated that the thin films were almost nanoparticles with a high degree of transparency [10], [11]-
pure and are relatively uniform the preferential orientation of [12].
ZnO growth was (002). The nanostructure images obtained by Different morphologies can be produced including nanorods
scanning electron microscopy show the whole samples have a [13], [14], nanowires [15], [16], nanotubes [17] and
homogenous distribution of ZnO product on the substrates. nanostructured thin films [18].
The spectrophotometer UV-Visible confirms that it is possible to Accordingly, the preparation of ZnO films by spray an
get good transparent ZnO films with a transmission of 80 to 95%
aqueous solution has been seen to attract increasing interest
in the visible. The values of optical gaps Eg deduced from the
spectra of UV-Visible transmissions vary between 3.209 and over the last few years.
3.288 eV. Most research is based on the study of the influence of
different precursor solution on physical and chemical
Keywords—ZnO; Spray Pyrolysis, Thin film, XRD, OTC properties of ZnO thin films [19], [20].
I. INTRODUCTION II. EXPERIMENTAL
Transparent conducting oxide (TCO) thin films have been ZnO thin films were grown on glass substrates by the spray
very important concerning the design of devices with pyrolysis technique. Two different precursor solutions were
immediate applications in sensors, photovoltaics, catalysis, used. In the precursor solution 1, the zinc acetate
micro- and opto-electronics, among others [1], [2]. (Zn(CH3COO)2.2H2O) dehydrate was dissolved in ethanol and
ZnO thin films is one of the most studied TCO materials, the solution was magnetically stirred at 60°c temperature for
since it exhibits distinct advantages like abundance in nature, 20min. The resulting solution was highly transparent and
non-toxicity, high optical transmittance in the visible region stable. For the precursor solution 2, zinc chloride (ZnCl2) was
and high chemical and thermal stability, among others [3]. dissolved in distilled water and the solution was stirred with a
Zinc oxide (ZnO) is a binary, n type- semiconductor magnetic stirrer at 60 °C for 20min.
material with a large direct gap varying between 3.20 and 3.35 Glass slides (2.2 * 2*0.08) cm were used as substrates, Prior
eV [1] and large exciton binding energy (60 meV) [4]. to the deposition of the films, the glass substrates were cleaned
Its properties make it an important base material for the rinsing them in acetone, and then in ethanol.
synthesis of many technological devices such as: solar cells, The precursor’s solutions are ejected onto glass substrates
gas sensors, piezoelectric sensors, guides of waves, etc. [5] positioned onto a substrate holder. The deposition temperature
and transparent layers in photovoltaic solar cells due to the (Ts) (temperature at the substrate surface) was maintained at
high transmittance in the visible region and its low electrical 400 °C with an accuracy of ±5°C. The latter is measured using
resistivity Also, a ZnO layer may work as an antireflection a Chrome–Nickel thermocouple.

ISSN 1737-9334
Two series of thin films were grown on glass substrates, one The lattice constants can be calculated using the formula
series for each solution. (4) [9].
During the pyrolytic process, the following possible
reactions take place. 1 4 𝑙2
2 = (ℎ2 + ℎ𝑘 + 𝑘 2 ) + (4)
𝑍𝑛𝐶𝑙2 + 𝐻2 𝑂 → 𝑍𝑛𝑂 + 2𝐻𝐶𝑙 (1) 𝑑ℎ𝑘𝑙 3𝑎2 𝑐2
(𝑍𝑛(𝐶𝐻3 𝐶𝑂𝑂)2 . 2𝐻2 𝑂) + 2(𝐶2 𝐻5 𝑂𝐻) → 𝑍𝑛𝑂 +

5 Where;
2(𝐶𝐻3 𝐶𝑂𝐶2 𝐻5 ) + 3𝐻2 + 𝑂2 ↑ (2)
2 d: Plane spacing, obtained by the Bragg law (5)
The crystal structure of the ZnO thin film was determined
by X-ray diffraction (XRD) using (Siemens D-5000) 𝜆 = 2𝑑𝑠𝑖𝑛𝜃 (5)
diffractometer with CuKα X-ray source with Cu anode
(wavelength λ=1.54056 Å). h k l: Miller indices.
The morphology and composition of the sprayed films were a and c: Lattice parameters(nm).
analyzed by scanning electron microscopy (SEM) and energy The calculation results for the crystallite sizes (D) and the
dispersive spectrometry (EDS) using a JEOL JSM-6400 model values of lattice parameters (a and c), are presented in Table 1.
operating at 7 kV. The lattice parameters give a = 0.3243nm, and c = 0.5209
The optical transmission spectra are used to study the nm for all the samples. These values are in good agreement
optical properties of deposit thin films and have been analyzed with the reported values in the literature [24].Similar results
using UV–Visible spectrophotometer with double beam were also obtained by [25], [26].
‘‘UV–visible SHIMADZU’’ in a wavelength range of 300– The texture coefficient (TC) was calculated using Eq. (6).
1100 nm at room temperature.
𝐼(ℎ𝑘𝑙)/𝐼0 (ℎ𝑘𝑙)
𝑇𝑐 (ℎ𝑘𝑙) = 1 (6)
III. RESULTS AND DISCUSSION 𝑛
∑𝑛 𝐼(ℎ𝑘𝑙)/𝐼0 (ℎ𝑘𝑙)
A. Structural characterization
Where:
The XRD patterns of the samples grown by the spray I0: Represents the standard intensity.
technique on glass substrates using different precursor I: The observed intensity of (hkl) plane.
solutions are presented in Fig. 1. It can be seen that the films n: The reflection number
are polycrystalline with a hexagonal wurtzite structure. The value of TC calculated for the diffraction peaks of ZnO,
The samples grown using precursor solution 1 present four are shown in Table 1. It was observed that the (002) peak has
peaks at: 31.83°, 34.33°, 56.55° and 56.65° that correspond to the highest TC value for the sample 2
the diffraction planes of hexagonal ZnO (1 0 0), (0 0 2), (1 0 The volume of the ZnO hexagonal cell was calculated using
1) and (1 1 0) respectively. In the patterns of the samples 2, the formula [3], [27] (7).
three peaks can be clearly seen at 34.84°, 36.78° and 47.87°
this peaks correspond to the (002), (101) and (102) diffraction 𝑉 = 0.866𝑎2 𝑐 (7)
planes of ZnO wurtzite structure. Where:
The films were in good agreement with those reported in a and c: Lattice parameters (nm).
JCPDS data cards [36–1451]. All the calculated parameters are summarized in Table 1.
Fig. 1(a) shows that the (100) direction is the preferential
Table 1 Different parameters obtained from XRD (Bragg’s angle (2θ),
growing orientation for the ZnO film. However, one can see in lattice parameters (a and c), crystallite size (D), the texture coefficient
Fig. 1.b for ZnO films that the preferential orientation is (002). (TC) and the volume (V).
Sagar et al. [21], [22] reported preferentially oriented films in
the (0 0 2) direction obtained from Zinc Acetate precursor. Zinc Chloride Zinc Acetate
Thus Zinc Acetate promotes the growth of crystals in the 2θ(°) 34.84 31.83
36.78 34.33
(0 0 2) direction. 47.87 36.55
The mean crystallite size of the sample is estimated by using 56.65
the Scherrer formula (3)[23]: hkl (002) (100)
(101) (002)
𝑘𝜆 (102) (101)
𝐷= (3) (110)
𝛽𝑐𝑜𝑠𝜃
a(nm) 0.3250 0.3245
c(nm) 0.5206 0.5190
Where:
c/a 1.60 1.59
D: Crystallite size (nm). D(nm) 33.11 15.28
K: Crystallite shape factor a good approximation is 0.9. TC 2.738 1.266
λ :Wavelength (1.5406 Å, 0.15406nm). V(nm)3 0.0476 0.0473
ẞ: Full width at half maximum (FWHM) of the peak (in V(Å)3 47.6 47.3
radian).
θ: Bragg angle (in radian).
Copyright - 2019
Page 50
ISSN 1737-9334
The ratio of lattice constants c/a was found to be around

1.60 for all layers. This is in good accord with values of the
bulk wurtzite ZnO. Similar results were also obtained by [28]
Fig.2 Optical transmittance spectra for ZnO thin films using different
precursor solutions (1) Zinc acetate, (2) Zinc Chloride with M=0.2M/L
and Ts=400°C
Fig.1 X-ray diffractograms of ZnO thin films using different precursor
solutions (1) Zinc Acetate, (2) Zinc Chloride with M=0.2M/L and
Ts=400°C
B. Optical characterization
Fig.2 shows the transmittance spectra of the samples. As it

can be seen the films with the highest transmission are the
samples grown using precursor solution 1 with a value
of about 96%. These results are consistent with the observed
in the precursor solutions. An absorption edge is observed
between 300 and 400 nm, this edge could be related to the
band gap value of the ZnO films. The glass substrate was
measured in order to determine the absorption edge of glass,
this was found to be near 300 nm, and thus the observed edge
in the films at 300 nm corresponds to the absorption of the
glass substrate used in the experiments.
The relationship between the absorption coefficient (α) and
photon energy (hυ) for direct transition [29], [30] is:
(𝛼ℎ𝜈)2 = 𝐴(ℎ𝜈 − 𝐸𝑔 ) (8)
Where:
A: Constant
Eg: The band gap value (eV).
Eg can be determined by extrapolating the straight line
Fig.3 The plots of (αhυ)2 vs. photon energy for ZnO thin films using different
portion at a = 0 as indicated in Fig.3.
precursor solutions (1)Zinc acetate, (2) Zinc Chloride with M=0.2M/L and
The band gap (Eg) of the samples was calculated from the Ts=400°C .
absorption spectra also known as optical density of the films
Fig.3. An extrapolation of the linear part of the curves
intercepting the hν axis gives the band gap value [32], [32]. The refractive index (n) of a semiconductor decreases with
increasing band gap energy Eg.
The refractive index (n) is calculated using the model of M.
Berruet and M. Va´zquez using Eq. (9)[33] and it is found to
vary from 2.29 to 2.42, which is within the values reported in
the literature [33].

ISSN 1737-9334
𝑛2 −1 𝐸𝑔 It is observed that for the solution of zinc chloride the

= 1−√ (9) chemical reactions are not complete hence the presence of Cl,
𝑛2 +2 20
but for the solution (zinc acetate) the composition is almost
We summarize the average values of the transmission (T%) stoichiometric. The presence of Si is due to the glass substrate
the band-gap energies (Eg) and the refractive index in Table 2.
Table 3 Atomic and weight percentage of the chemical composition
Table2 Different parameters obtained from transmittance spectra for ZnO thin films
transmission (T%), band gap (Eg) and refractive index (n) Precursor solution Zinc chlorure Zinc Acetat
T(%) Eg(eV) n Wt% At% Wt% At%
Zinc Chloride 64 3.228 2.33 Zn 82.23 55.77 81.38 51.86
Zinc Acetate 96 3.209 2.34 O 13.57 37.61 18.32 47.72
Cl 2.22 4.10 - -
C. Morphology Si 1.98 2.52 0.30 0.54
O/Zn 0.67 0.92
Fig. 3 displays a typical SEM micrographs series of the

surface morphologies of the ZnO thin films deposited by spray IV. CONCLUSION
pyrolysis. The zinc oxide film surface is rough and is
composed of fine particles. Furthermore, the surface Two different precursor solutions were used to obtain ZnO
morphologies of the ZnO films are intimately correlated to thin films. The influence of the precursor solution on the
precursor solution. All images show homogeneous and dense physical properties of the thin films was studied. Transparent
surfaces. All the films deposited in the optimized conditions polycrystalline thin films were obtained.
exhibit a good adherence [31]. the structural properties such as crystallite size, grain size
The results show that the different solvents used in the and surface morphology depend on both the precursor
preparation of ZnO solution strongly affect the properties of solution, thus the choice of the proper solution depends on the
the ZnO thin film surface. The same results were remarked in application wanted for the films.
the literature [34]. The calculated band gap of the films is in agreement with
the expected value for ZnO thin films. It was shown that the
band gap can be modified when using different solutions, this
is important because the use of ZnO thin films for solar cell
applications requires the control of the band gap.
Both precursor solutions give good quality ZnO thin films.
Thus two simple and effective ways to produce ZnO thin films
were developed. Based on the results from XRD the thin films
grown using Zinc Acetate based solution have better
crystallinity than the ones grown with the Zinc Acetate based
solution.
UV–Vis transmittance spectra results show that the films
obtained from the Zinc Acetate solution are more transparent
than the films obtained from Zinc Chloride solution.
ACKNOWLEDGMENTS
The authors greatly acknowledge the technical staff of Msila
University for the X-ray measurements.
REFERENCES
[1] H. Abdelkader, Y. Fayssal, W. Daranfed, N. Attaf and M S. Aida, Les
propriétés structurales, optiques et électriques des couches minces de ZnO:Al
élaborées par spray ultrasonique, Revue « NATURE & TECHNOLOGIE ».
N° 06, JANVIER 25 - 27 (2012).
[2] S. Kurtaran, S. Aldag, G. Ofofoglu, I. Akyuz and F. Atay, On the role of
Al in ultrasonically sprayed ZnO films, Materials Chemistry and Physics 185
Fig.4. SEM micrographs visualization of ZnO thin films using different (2017) 137e142
precursor solutions (1)Zinc acetate, (2) Zinc Chloride. [3] M. Ding, Z. Guo, L. Zhou, X. Fang, L. Zhang, L. Zeng, L. Xie and H.
Zhao, One-Dimensional Zinc Oxide Nanomaterials for Application in High-
Performance Advanced Optoelectronic Devices, Crystals (2018) 8(5), 223;
https://doi.org/10.3390/cryst8050223.
[4] O. Slimi, L. Hammiche, D. Djouadi, A.Chelouche and M. Meddouri,
The chemical composition of the elements of two ZnO Elaboration and characterizations of ZnO aerogel nanoparticles dispersed in
samples is presented in Table 3 silica matrix, Revista Română de Materiale Romanian Journal of Materials
(2017) 47 (3), 303 – 308
Copyright - 2019
ISSN 1737-9334 Page 52
[5] S. Mondal, K. P. Kanta and P. Mitra, Preparation of Al- doped ZnO (AZO) [28] S. Benzitouni, M. Zaabat, A. Mahdjoub, A. Benaboud and B. Boudine,
Thin Film by SILAR. Journal of Physical Sciences. 12 (2008) 221-229 High transparency and conductivity of heavily In-doped ZnO thin films
[6] Y. K. Mishra, R. Adelung, ZnO tetrapod materials for functional deposited by dip-coating method, Materials Science-Poland
applications, Materials To day Volume 21, Number 6d July/August 2018. http://www.materialsscience.pwr.wroc.pl/ DOI: 10.1515/msp-(2018)-0037
[7] F. E. Tazangi and A. Rahmati, Characteristics of Electrodeposited ZnO [29] J.J. Tauc, Amorphous and Liquid Semiconductor, Plenum, New York,
Nanostructures: Size and Shape Dependence, Armenian Journal of Physics, 1976.
(2016), vol. 9, issue 1, pp. 20-28 [30] T. Saidani, M. Zaabat, M. S. Aida, A. Benaboud, S. Benzitouni and A.
[8] Z. N. Kayani, F. Naz, S. Riaz and S. Naseem, Characteristics of Al-doped Boudine, Influence of annealing temperature on the structural, morphological
ZnO:Ni films grown on glass by sol–gel dip coating technique, Journal of and optical properties of Cu doped ZnO thin films deposited by the sol–gel
Saudi Chemical Society (2017) 21, 425–433 method, Superlattices and Microstructures 75 (2014) 47–53
[9] M. Meddouri., L. Hammiche., O. Slimi and D.Djouadi and A.Chelouche, [31] H. Benzarouk, A. Drici , M. Mekhnache, A. Amara, M. Guerioune, J. C.
Effect of cerium on structural and optical properties of ZnO aerogel Bernède and H. Bendjffal, Effect of different dopant elements (Al, Mg and Ni)
synthesized in supercritical methanol, Materials Science-Poland, 34(3), (2016) on microstructural, optical and electrochemical properties of ZnO thin films
pp. 659-664 deposited by spray pyrolysis (SP), Superlattices and Microstructures 52 (2012)
[10] Z.EL Khalidi, B. Hartiti, S. Fadili, P. Thevenin and M. Siadat, Propriétés 594–604
structurales, optiques et électriques de ZnO préparé par Spray Pyrolysis, [32] S. Islam, K. M. A. Hussain and M. J. Rashid, Deposition and Optical
Afrique SCIENCE 13(2) (2017) 250 - 258 Characterization of ZnO Thin films on Glass Substrate, IOP Conf. Series:
[11] F.A. Garcés, N. Budini, J.A. Schmidt and R.D. Arce, Highly doped ZnO Journal of Physics: Conf. Series 1086 (2018) 012009
films deposited by spray-pyrolysis. Design parameters for optoelectronic [33] M. Berruet and M. Va´zquez, Electrodeposition of single and duplex
application, Thin Solid Films 605 (2016) 149–156. layers of ZnO with different morphologies and electrical properties, Materials
[12] A. Hafdallah, A. Azzedine, H. Belhani, M. S. Aida and N. Attaf, Effect Science in Semiconductor Processing 13 (2010) 239–244.
of the Nozzle-Substrate Distance on the Structural and Optical Properties of [34] K.L. Foon, M.Kashif, U. Hashim and Wei-WenLiu, Effect of different
ZnO Thin Films Deposited by Spray Pyrolysis Technique, American Journal solvents on the structural and optical properties of zinc oxide thin films for
of Nano Research and Applications (2017); 5(6): 87-90. optoelectronic applications, Ceramics International 40 (2014)753–761
[13] A.E. Rakhshani. Optical and electrical characterization of well-aligned
ZnO rods electrodeposited on stainless steel foil, Applied Physics A:
Materials Science and Processing 2008;92:303.
[14] M. Guo, C. Yang, M. Zhang, Y. Zhang, T.Ma, X.Wang, et al. Effects of
preparing conditions on the electrodeposition of well-aligned ZnO nanorod
arrays, Electrochimica Acta (2008);53:4633.
[15] J. Elias, R. Tena-Zaera and C. Le´ vy-Cle´ment. Effect of the Chemical
Nature of the Anions on the Electrodeposition of ZnO Nanowire Arrays,
Journal of Physical Chemistry C (2008);112:5736
[16] J. Elias, R. Tena-Zaera and C. Le´vy-Cle´ment. Electrodeposition of ZnO
nanowires with controlled dimensions for photovoltaic applications: Role of
buffer layer, Thin Solid Films (2007);515:8553.
[17] X. Ren, C.H. D.D. Jiang and L.Li, He, Effect of Dimensionality and
Polymorphism on the properties of ZnO Materials Letters 2008;62:3114.
[18] Z. Chen, Y. Tang, L. Zhang and L. Luo, Electrodeposited nanoporous
ZnO films exhibiting enhanced performance in dye-sensitized solar cells.
Electrochimica Acta (2006) ;51:5870.
[19] J.G. Quiñones-Galván, I.M. Sandoval-Jiménez, H. Tototzintle-Huitle,
L.A. Hernández-Hernández, F. de Moure-Flores, A. Hernández-Hernández, E.
Campos-González, A. Guillén-Cervantes, O. Zelaya-Angel and J.J. Araiza-
Ibarra, Effect of precursor solution and annealing temperature on the physical
properties of Sol–Gel-deposited ZnO thin films, Results in Physics 3 (2013)
248–253
[20] N. Lehraki, M.S. Aida, S. Abed, N. Attaf, A. Attaf and M. Poulain, ZnO
thin films deposition by spray pyrolysis: Influence of precursor solution
properties, Current Applied Physics 12 (2012) 1283e1287
[21] P. Sagar, M. Kumar, R.M. Mehra, H. Okada, A. Wakahara and A.
Yoshida, Epitaxial growth of zinc oxide thin films on epi-GaN/sapphire
(0001) by sol–gel technique, Thin Solid Films (2007);515:3330–4.
[22] P. Sagar, P.K. Shishodia, R.M. Mehra, H. Okada, A. Wakahara and A,
Yoshida, Photoluminescence and absorption in sol–gel-derived ZnO films, J
Lumin (2007);126:800–6.
[23] S. Kurtaran, S. Aldag, G. Ofofoglu, I. Akyuz and F. Atay, Transparent
conductive ZnO thin films grown by chemical spray pyrolysis: the effect of
Mg, Article in Journal of Materials Science Materials in Electronics April
(2016) DOI: 10.1007/s10854-016-4862-1
[24] M. Mekhnache, A. Drici, L. Saad Hamideche, H. Benzarouk, A. Amara,
L. Cattin, J.C. Bernède and M. Guerioune, Properties of ZnO thin films
deposited on (glass, ITO and ZnO:Al) substrates, Superlattices and
Microstructures 49 (2011) 510–518.
[25] U. Alver, T. Kılınc, E. Bacaksız and S. Nezir, Materials Chemistry and
Physics 106 (2007) 227–230.
[26] R. Robert. Reeber, Lattice parameters of ZnO from 4.2° to 296°K,
Journal of Applied Physics 41, 5063 (1970)
[27] A. Chelouche, T. Touam, D. Djouadi and A.Aksas, Synthesis and
characterizations of new morphological ZnO and Ce-doped ZnO powders by
sol–gel process, Optik, 125 (2014), 5626.
Copyright - 2019
ISSN 1737-9334 Page 53
PL , EDX , and XRD Analysis of Al2O3 Scales

Grown on PM2000 Alloy
a*
, S. Achache , F.Sanchette , Z. Mekhalif , N. Kamoun , A. Boumaza a
b b c d
S. Hakkar
a
: Laboratoire des structures, propriétés, interactions interatomiques (LASPI2A), faculté des sciences et technologie, université
Abbes Laghrour, Khenchela 40000, Algérie.
b : Université de technologie de Troyes (antenne de Nogent).
c
: Laboratoire de chimie et électrochimie des surfaces, faculté Notre Dame de la Paix, rue de Bruxelles 61, B-5000 Namur, Belgique.
d
:Université de Tunis El-Manar, Faculté des Sciences de Tunis Département de Physique, Laboratoire de Physique de la Matière
Condensée.
*
: Corresponding author : samihahakk@hotmail.com
Résumé ІI . Etude expérimentale

Les changements microstructuraux lors de l’oxydation du
superalliage PM2000 ont été étudiés après 48 heures en isothermes à L'alliage à base de fer PM 2000 renforcé par la dispersion
haute température (entre 600 et 1200 ° C). Les oxydes (-alumine, d'oxyde (ODS) provenant de Schwarzkopf Plansee a été
alumines de transitions, et oxydes mixtes) ont été caractérisés en
utilisé; la composition chimique en pourcentage en poids est
utilisant les techniques d’observations par microscopie électronique
à balayage MEB, les analyses EDX, la spectroscopie Raman, la la suivante: A1-5.0, Cr-22.0, Mn-0.2, C-0.02, Si-0.3, Y-0.1,
spectroscopie de photoluminescence et la diffraction des rayons X. Zr-0.1 et le reste étant le Fe. Des échantillons rectangulaires
La diffraction des rayons X résulte, dans des conditions d'incidence de dimensions 1 × 1 cm2 et 1 × 2 cm2 et d’épaisseur de 1,0 ±
classiques et de en conditions rasantes, permettent d'estimer la taille 0,1 mm, ont été sectionnés (figure 1).
des cristallites et les surfaces spécifiques de l'alpha-alumine. La
spectroscopie Raman (λexc = 632,18 nm) donne les différentes
signatures des différentes alumines générées (γ, δ / θ, θ et α-Al2O3).
Les analyses de photoluminescence à diverses longueurs d'onde
d'excitation (200 nm) fournissent des informations sur les défauts
existants (intrinsèque et extrinsèque) en fonction de la température
d'oxydation. Les résultats PL permettent de caractériser les défauts
intrinsèques et extrinsèques des divers alumines développées.
Mots clés : Alliage PM2000, MEB/EDX, DRX, Spectroscopie

Raman, Photoluminescence.
І. Introduction Fig .1 Les échantillons du Pm2000 utilisés (1x1x0.1 et 1x2x0.1 cm3)
L'alliage ferritique PM2000 a fait l'objet de nombreuses

recherches en raison de leur importance technologique et de
leur intérêt scientifique. Leur capacité de résistance est due,
entre autres, au fait qu'ils développent à haute température ІI .1 - Observations MEB et analyses EDX, et observation par
une fine couche d'alumine hautement protectrice. Α-Al2O3 FEG
est devenu un matériau indispensable en raison de ses
propriétés mécaniques, chimiques et thermiques supérieures.
Cet oxyde agit comme une barrière de diffusion tout en Le microscope électroniques à balayage (MEB) Bruker at été
conservant une bonne adhérence. Avant d'atteindre l'alumine utilisé pour les observations des surfaces d’alliages oxydés, et
la plus stable, se développent de alumines de transition, dont aussi pour les analyses EDX de ces surfaces.
les plus courantes sont les phases γ, δ et / ou θ. Dans cette
étude, nous utiliserons le MEB/EDX, la diffraction des rayons
X, la spectroscopie Raman, et la photoluminescence [1-2-3-4].
Copyright - 2019
ISSN 1737-9334 Page 54
discontinues.
A 900°C, les cristaux ressemblent à ceux des alumines de

transition présentées en littérature. A 1000°C, et 1100°C, les
cristaux d’alumine alpha se présentent clairement [5].
Figure 3 : Analyses EDX des divers alliages de PM2000 oxydés sous air
durant 48h (a) 600°C, (b) 800°C, (c) 1000°C, et 1200°C.
Fig . 2 (a) Analyses EDX, et (b) par diffraction des rayons X de l’alliage
PM2000.
Les températures d’oxydations ont été choisies entre 600 et

1200°c, durant un temps de 48h et sous air. Les analyses
EDX en figure 2, permettent de suivre les diverses analyses
des différents oxydes selon la température utilisée. Nous
pouvons remarquer que les éléments essentiels qui rentrent
dans la formation des oxydes formés sont essentiellement le
fer, le chrome, et bien sur l’aluminium. Il est à remarquer que
le Magnésium est aussi présent pour les plus hautes
températures (supérieures à 1000°C). Les oxydes riches en
chromes et ceux riches en fer ont tendance à diminuer au fur
et à mesure que la température augmente, laissant place aux
oxydes riches en aluminium qui s’installent progressivement
à mesure que la température augmente. Ainsi, vers 1200°C,
l’alumine devient dominante.
Les images MEB en figure 4, permettent de suivre les
diverses morphologies d’oxydes qui s’installent
progressivement au fur et à mesure que la température Figure 4 : Images MEB des surfaces du PM2000 Oxydé à 800°c, 900°C,
1000°C, et 1100°C.
augmente. En effet, la surface oxydée vers 800°C ; s’avère
plutôt mal cristallisée présentant des nodules, et des parties
Copyright - 2019
ISSN 1737-9334 Page 55
Les observations par FEG de la surface de l’oxyde formé à

1200°C, sur l’alliage PM000 sont représentées sur la figure 5.
Les grains clairement définis. La morphologie denses et
équiaxes avec présence de zones poreuses.
Fig.6 Spectres de diffraction des rayons X de l’alliage PM2000 oxydé à

1000°C (48h), en conditions rasantes (2.5°) en en conditions classiques
(rouge et bleu respectivement), et comparaison avec un spectre de la poudre
d’alumine alpha (de référence).
Le spectre en rouge correspond aux rayons X en incidence

rasante (2.5°); et le spectre en bleu correspond aux rayons X
réalisés en conditions classiques. Il est constaté que les
largeurs à mi-hauteur des pics sont plus importantes dans le
cas des DRX rasants. Les largeurs à mi hauteur permettent de
refléter les tailles des cristallites qui par la formule de
Scherrer permettent d’estimer cette taille à 14.6±2.9 nm dans
le cas rasant et 28.3±2.6 nm dans le cas des incidences
classiques. Ces valeurs permettent une estimation des
surfaces spécifiques qui sont de 177.9 ± 37.6 m2/g (en DRX
Fig.5 Images FEG de la surface du PM2000 oxydé 48h sous air à 1200°C. rasants) et 88.4 ± 8.3 m2/g (en DRX classiques) [6].
ІI .2 – Diffraction par rayons X

ІI .3 – Spectroscopie Raman, et Photoluminescence
L'analyse par diffraction des rayons X a été réalisée avec un
diffractomètre PANalytical X’Pert ProMRD avec un Les échantillons oxydés ont fait l’objet d’analyses par
rayonnement CuKα (λ(Kα1) = 0.15406 nm, λ(Kα2) = spectroscopie Raman , ainsi que par photoluminescence.
0.15444 nm). Les spectres en figure 7a permettent de comparer les résultats
Les spectres de diffraction par rayons X présentés en figure 6, d’un échantillon oxydé à 900°C durant 7h et un autre durant
permettent de caractériser les oxydes formés à 1000°C, en 48h. Les résultats permettent de voir que durant une période
effet on distingue clairement les pics relatifs à l’alumine de 7h d’oxydation on peut dire par comparaison avec des
alpha qui sont comparés au spectre d’une alumine alpha de résultats de la littérature que nous avons coexistence des
référence (en gris). On retrouve aussi des pics relatifs à alumines (γ, δ, θ, et aussi α-Al2O3), par contre durant 48h
l’alumine thêta et aussi des pics qui correspondent à FeAl2O4. nous avons coexistence des alumines (δ, θ, et aussi α-Al2O3)
[7-8].
Copyright - 2019
ISSN 1737-9334 Page 56
La figure 7b donne le spectre PL (avec une λexc = 200 nm) Tableau 1 : Affectation des valeurs PL et défauts correspondants
d’un échantillon PM2000 oxydé à 1000°C durant 48h permet
d’avoir des raies PL caractéristiques des défauts intrinsèques
et extrinsèques [9-10-11] dont les affectations sont
énumérées dans le tableau 1 ci dessous
Longueur d’onde, et affectations (défauts intrinsèques et
extrinsèques)
F+ 343.5
395
F
405
422.5
448
F2
459.5
484.5
F22+/Ali+ 529
Oi 597
4+
Ti
Mn2+ 620.5
Fe3+ 669.5
Mn4+ 681.5
IV. Conclusion
Le superalliage PM2000 a fait objet de notre étude, de par sa

tenue à l’oxydation exceptionnelle surtout à haute
température. Les observations MEB, analyses EDX, analyses
Raman et par photoluminescences permettent de montrer que
cet alliage développe des oxydes complexes vers 600-700-et
800°C. à 900°C, nous avons essentiellement les alumines
thêta et alpha. Pout des températures supérieures à 1000°C,
l’alumine alpha devient dominante. La présence de faibles
quantités d’oxydes mixtes ainsi que de l’alumine thêta ont été
Fig7. : Spectres Raman, et PL des l’alliage PM2000 oxydé. constatés vers 1200°c, on parlera d’oxydes résiduels qui sont
localisés particulièrement en surface externe.
Copyright - 2019
ISSN 1737-9334 Page 57
.
[11].V. S. Kortov, V. A. Pustovarov, T. V. Spiridonova, and S. V. Zvonarev ,.
Acknowledgments Journal of Applied Spectroscopy, Vol. 80, No. 6, January, 2014 (Russian
Original Vol. 80, No. 6, November–December, 2013) .
This work was partly supported by the National Project
Research (PNR) and LASPI2A laboratory of Khenchela
University,
Algeria.
Références
[1]- R. Molins and A.M. Huntz, Materials Science Forum, vols.461-464 ,

pp.29-36, 2004 ,
[2] D.M. Lipkin, D.R. Clarke, M. Hollatz, M. Bobeth, W. Pompe, Corrosion
Science,vol.39,pp.231-242,1997
[3]- S. Chevalier, R. Molins, O. Heintz, J.P. Larpin, Science reviews,
pp.365-372 , 2005
[4]- H. El Kadiri, R. Molins,Y. Bienvenu, and M. F. Horstemeyer. Oxidation
of Metals, Vol. 64, Nos. 1/2, August. 2005 .
[5]- P. Y. Hou, X. F. Zhang, and R. M. Cannon, Scripta Materialia, vol. 50,
no.1,pp.45–49,2004 .
[6]- P. Scherrer, Nachr. Ges. Wiss. Göttingen , vol.26, pp .98-100 , 1918
[7] V.K. Tolpygo, D.R. Clarke, Materials at High Temperatures, vol.17,pp.
59-70,2000
[8] A. B. Kulinkin, S. P. Feofilov, R. I. Zakharchenya: Physics of the Solid
State , Vol.42, p. 857, 2000
[9] B. D. Evans and J. Nucl. Mater., 219, pp. 202–223,1995 .
[10] A. I. Surdo, V. A. Pustovarov, V. S. Kortov, A. S. Kishka, and E. I.

Zinin, Nucl. Instrum. Methods Phys. Res., Sect. A, 543, 234–238 2005 )

ISSN 1737-9334
Structural, dielectric and piezoelectric properties of

(1-x) Pb (Zr0.52 Ti0.48) O3–x Sm Cr O3 ceramics
F. Kahoul*1,2, L. Hamzioui1,2, A. Guemache1, M.Aillerie3,4, A. Boutarfaia2
1
Université de M’Sila, Département Socle Commun ST, Faculté de Technologie, M’Sila 28000 Algérie
2
Université de Biskra, Département de Chimie, Laboratoire de Chimie Appliquée, B. P. 145, RB-Biskra 07000 Algérie
3
Université de Lorraine, LMOPS, EA 4423, 57070 Metz, France
4
Centrale Supelec, LMOPS, 57070 Metz, France
fares.kahoul@univ-msila.dz
Abstract— This work reported the structural, dielectric, and In general, PZT system ceramics should be sintered at high
piezoelectric behavior of (1-x) Pb (Zr0.52 Ti0.48) O3 – x Sm Cr O3 temperatures between1100 and 1300 °C in order to obtain
ceramics (abbreviated as PZTSC, where x = 0.005, 0.01, 0.015, complete densification. Accordingly, environmental pollution
0.02 and 0.025), were prepared by the traditional solid-state due to its PbO evaporation and the use of expansive Pd rich
reaction method. The phase transition, microstructure,
Ag/Pd internal electrode in case of manufacturing multilayer
dielectric, piezoelectric properties, and the temperature stability
of the ceramics were investigated. X-ray diffraction analysis ceramic actuator are inevitable.
indicated that as-prepared ceramics were of pure perovskite Hence, to reduce its sintering temperature, various kinds of
phase and the possible morphotropic phase boundary (MPB) material processing methods such as hot pressing, high
between the tetragonal and rhombohedral phase compositions energy mill, liquid phase sintering, and using ultra fine
were located near the SC content of x ≥ 0.02, confirmed by their powder have been performed. Among these methods, liquid
corresponding dielectric and piezoelectric properties. All phase sintering is basically an effective method for aiding
specimens present high relative density above 97%, indicating a densification at low temperature. The theoretical explanation
wide sintering window for this system. Microstructural for liquid phase sintering was already reported over 40 years
investigations of all the samples reveal that SC doping inhibits
ago.
grain growth. The dielectric and piezoelectric properties show a
maximum response at x ≥ 0.02, which corresponds to the In this work, the phase structure, density, electrical
morphotropic phase boundary (MPB). properties, low-temperature sintering and temperature
characteristics of PZTSC piezoelectric ceramics were
described systemically. The aim of the work was to find out
Keywords— XRD, Dielectric Properties, Piezoelectric
the optimized content of Sm2O5 and Cr2O3, which can make
Properties, MPB, Density
this system have higher electrical properties for multilayer
I. INTRODUCTION piezoelectric device applications.
The high power characteristics of piezoelectric materials II. EXPERIMENTAL PROCEDURE
have been investigated for device applications in ultrasonic
PZTSC ceramics having the chemical formula of (1-x) Pb
motors, piezoelectric actuators, piezoelectric transformers,
(Zr0.52 Ti0.48) O3 – x Sm Cr O3, where x = 0.005, 0.01, 0.015,
ultrasonic vibrator, filter, blue luminescence and resonator,
0.02 and 0.025), were prepared by the mixed oxide route.
and medical applications [1–5].
An appropriate mixture of pure oxides (purities ˃99 %)
For high-power multilayer piezoelectric device
were ball milled for 24 h using ZrO2 media with distilled
applications, piezoelectric materials are electrically driven to
water. The mixture was then dried and calcined at 850 °C for
high mechanical vibration near the resonance frequencies,
3h. The calcined powder was again ball milled to achieve1.2
leading to a temperature rising and deterioration of
μm particle size. This was followed by mixing of 10 wt.% of
piezoelectric properties with the increase of their vibration
PVA as a binder in the calcined powder and the specimens
velocities [6,7]. Therefore, the lead-based piezoelectric
were pressed using uniaxial press. Specimens were
ceramics should have high piezoelectric constant (d 31), high
subsequently sintered in a PbO rich atmosphere between 1150
electromechanical coupling factor (kp), high mechanical
°C for 2 h and final dimension obtained after lapping was12
quality factor (Qm), and good temperature stability [8,9].
mm diameter and 1mm thickness.
The Pb(ZrxTi1-x)O3 system and its modified solid solutions
The density of the sintered samples was measured using
are known to exhibit excellent dielectric and elastic properties
Archimedes’ principle. Microstructural features such as a
at the “Morphotropic Phase Boundary (MPB)” [10–13]. This
grain size and pores were characterized by means of atomic
MPB is believed to be a coexistence region of two phases
force microscopy (AFM). The sintered pellets were
namely, tetragonal and rhombohedral phases and still is a
electroded with high-purity silver paste and fired at 700 °C
topic of great debate [14–22].
for 1 h. The samples were pooled in a silicon oil bath at 120
°C for 1 h under a static DC electrical field of 4.5 kV/mm.
Copyright - 2019
ISSN 1737-9334 Page 59
Powder X-ray diffraction (XRD) measurements were equal or greater than 0.02 suggests that this composition
carried out using a Rigaku X-ray diffractometer with Cu Kα. (x ≥ 0.02) corresponds to MPB.
The XRD patterns were recorded at a scan rate of 2°/min for
2θ varying from 10º to 80°. Temperature-dependent dielectric
constant (εr) and tangent loss (tanδ) were obtained as the
functions of temperature using a inductance capacitance
resistance meter (LCR, HIOKI, Model-3532-50) in a x=0.025
proportional integral derivative (PID) controlled heating
Intensity (a.u.)
chamber. x=0.02
Twenty four hours after poling, the piezoelectric
properties: piezoelectric constant (d31), electromechanical x=0.015
planar coupling factor (kp) and mechanical quality factor (Qm)
were measured by a method similar to that of the IRE x=0.01
standard. The resonance and anti-resonance frequencies were
obtained by using the maximum and the minimum of spectra x=0.005
admittance.
10 20 30 40 50 60 70 80
III. RESULTS AND DISCUSSION 2-Theta / Degree

A. X-ray Diffraction Profile Fig. 1 XRD patterns of PZTSC ceramics with different SC contents in the 2θ
range of 10º–80°, sintered at 1150 °C
Fig. 1 shows the XRD patterns of sintered (1-x) Pb (Zr0.52
Ti0.48) O3 – x Sm Cr O3 (where x = 0.005, 0.01, 0.015, 0.02 B. AFM Analysis
and 0.025) ceramics. The tetragonal, rhombohedral and Fig. 2 shows AFM images of ceramic samples of doped
tetragonal-rhombohedral phases are identified by analysis of solid solutions: (1-x) Pb (Zr0.52 Ti0.48) O3 – x Sm Cr O3 with
peaks [002 (tetragonal), 200 (tetragonal), 200 x=0.005, x=0.01, x=0.015, x=0.02 and x=0.025.
(rhombohedral)] in 2θ range of 10º–80°. Microstructural details, such as grain size and morphology as
The XRD patterns confirm that all the PZTSC powders are well as domain wall arrangements, can be observed from
pure and monophasic in nature, and free from the pyrochlore these micrographs. With the increase of SC content, the (1-x)
phases. As evident from the XRD patterns, the structure is Pb (Zr0.52 Ti0.48) O3 – x Sm Cr O3 ceramics become much
tetragonal at SC concentrations up to x = 0.015. The co- denser, and their average grain size gradually decreases, as
existence of rhombohedral and tetragonal phases at x values shown in Fig.2.

ISSN 1737-9334
Fig. 2 AFM images of the (1-x) Pb (Zr0.52 Ti0.48) O3 – x Sm Cr O3 solid solutions with various compositions: x=0.005, x=0.01, x=0.015, x=0.02 and x=0.025
densification behavior, because of the enhanced sintering

activity. This may be due to mass transfer caused by oxygen
C. Density Measurement vacancies via partial substitution of Cr3+ for Zr4+ and Ti4+,
Fig. 3 shows the relative density (i.e. the measured density which is used as a transferred carrier to accelerate mass
divided by the theoretical density of PZT of 8.0 g/cm3) of the mobility.
ceramics as a function of the SC content. With increasing SC
content, the density increases, reaching a maximum value of
7.91 g/cm3 in samples with 0.025 wt.% Sm2O5 and Cr2O3.
In addition, the density results in Fig. 3 are consistent with
the porosity variations shown in Fig. 2 which lead to the
decreasing density. These results show that adding an
appropriate amount of Sm2O5 and Cr2O3 can promote the
Copyright - 2019
Page 61
ISSN 1737-9334
50
7,92 0.005
0.01
7,90
40
(b)
0.015
7,88 0.02
Density (g/cm3)
0.025
7,86 30
tan
7,84
20
7,82
7,80
10
7,78
7,76 0
0,005 0,010 0,015 0,020 0,025 0 100 200 300 400 500
SC content (mol.) Temperature (°C)
Fig. 3 Density of ceramics as a function of SC content Fig. 4 Temperature dependence of (a) ɛr and (b) tanδ of the PZTSC
ceramics with different SC contents
D. Dielectric Properties
E. Piezoelectric Properties
The temperature dependence of the (a) dielectric constant
(ɛr) and (b) losses (tanδ) for (1-x) Pb (Zr0.52 Ti0.48) O3 – x Sm Fig. 5 indicates the mechanical quality factor (Qm),
Cr O3 ceramics with different SC contents at the frequency of electromechanical coupling factor (Kp) and piezoelectric
1kHz is presented in Fig. 4. It is observed that with the constant (d31) as a function of SC content. As can be seen,
increase in Sm and Cr substitution, there is systematic Kp, d31 and Qm show a similar variation with increasing SC
decrease in phase transition temperature (Tc). The decrease in content. Kp, d31 and Qm both increase with increasing SC
phase transition temperature is mainly due to dis-appearance contents. Kp, d31 and Qm reach their maximum
of long range order co-operative phenomenon. (0.58, 293 pC/N and 745) at 0.025 (mol.) SC content.
The dielectric constant increases with increase in Sm and Cr The increase of Qm may be due to the low-valence
substitution and the losses (tanδ) are smaller for the denser substitution of Cr3+ for Zr4+ and Ti4+, which will lead to the
samples (x=0.02 and 0.025). appearance of oxygen vacancies to maintain electrovalence
The decrease in dielectric constant (ɛr) may be due to the balance. The oxygen vacancies will produce a “pinning effect”
dominance of 90° domains contribution to the dielectric on domain rotation; thereby, the material becomes “hard”
properties after poling. Here in our case the aligned ∼90° [23, 24]. The increase of Qm may be caused by the increasing
rotated domains after poling cause more mechanical stress and density.
enhance the anisotropy that results in decrease in dielectric In addition, the changes of Qm, Kp and d31 are complex and
constant. slightly different from traditional acceptor doping effects,
because two factors act together to cause the variation of the
electrical properties, i.e.,Cr3+ modification resulting in the
14000 x=0.005 development of hard piezoelectric behavior, and densification
x=0.01 of the ceramics with Cr2O3 addition.
12000 x=0.015 (a)
Therefore, under the combination of both of these effects,
Dielectric constant (r)
x=0.02
10000 x=0.025 some electrical properties do not agree with the expected hard
characteristics of this system.
8000
6000
4000
2000
0
0 100 200 300 400
Temperature (°C)
Copyright - 2019
ISSN 1737-9334 Page 62
[7] C. Cheng, D. YiZheng, G. Peng, S. Hu, H. Zhang, and J. Zhang, “The

effects of Sm2O3 doping on properties of PNN–PZT ceramics near
morphotropic phase boundary,” J. Mater. Sci: Mater. Electron, vol. 28,
800 pp. 1624–1630, Sep. 2017.
300 d31 0,58 [8] C. P. Fernández, F. L. Zabotto, D. Garcia, and R. H. G. A. Kiminami,
280 Qm 700 “In situ sol-gel co-synthesis at as low hydrolysis rate and microwave
kP 0,56 sintering of PZT/Fe2CoO4 magnetoelectric composite ceramics,”
260 Ceramics International, vol. 43, pp. 5925–5933, Jan. 2017.
600
240 0,54 [9] A. G. Znamenskii, A. M. Ionov, and V. A. Marchenko, “Gas Release in
(pC/N)
k
500 the Process of Thermal Treatment of Sputtered Pb(Ti 0.48Zr0.52)O3
P
Q
220
0,52 Films,” Physics Solid State, vol. 58, pp. 1239–1246, Jan. 2016.
m
200 400 [10] P. Zhou, K. Liang, Y. Liu, Z. Zheng, and T. Zhang, “Efect of interface
31
0,50 coupling on magnetoelectric response

d
180
300 of Pb(Zr0.52Ti0.48)O3/La0.67Sr0.33MnO3 thin film under different strain
160 0,48 states,” Applied Physics A, vol. 124, pp. 670–675, 2018.
140 200 [11] J. R. Gatabi, S. Rahman, A. Amaro, T. Nash, J. Rojas-Ramirez, R. K.
0,46 Pandey, and R. Droopad, “Tuning electrical properties of PZT film
0,005 0,010 0,015 0,020 0,025
deposited by Pulsed Laser Deposition,” Ceramics International, vol. 43,
SC content (mol.) pp. 6008–6012, Jan. 2017.
[12] X. Chao, J. Wang, J. Pu, S. Zhang, and Z. Yang, “ Aging behavior
Fig. 5 Qm, Kp, and d31 of ceramics as a function of SC content and electrical properties of low-temperature sintered (Ba, Ca)(Ti,
Zr)O3-Ba(Cu, W)O3 ceramics and plate loudspeaker,” Sensors
IV. CONCLUSIONS Actuators A, vol. 237, pp. 9–19, Jan. 2016.
[13] Z. Xiao, X. Li, X. Dong, J. Tang, C. Wang, T. Zhang, S. Li, and L. B.
Ceramic samples of (1-x) Pb (Zr0.52 Ti0.48) O3 – x Sm Cr O3 Kong, “Sintering and electrical properties of commercial PZT powders
(where x = 0.005, 0.01, 0.015, 0.02 and 0.025) were prepared modified through mechanochemical activation,” J. Mater. Sci., vol. 53,
by the conventional oxide-mixed method. The structure, pp.13769–13778, Apr 2018.
microstructure, dielectric and piezoelectric properties were [14] D. Bochenek, P. Niemiec, M. Adamczyk, Z. Machnik, and G. Dercz,
“Dielectric properties of the multicomponent PZT-type solid solution,”
investigated systematically. XRD analysis reveals the co- Eur. Phys. J. B, vol. 88, pp. 279–283, Oct. 2015.
existence of tetragonal and rhombohedral phases (MPB) were [15] O. Namsar, S. Pojprapai, A. Watcharapasorn, and S. Jiansirisomboon,
located near the SCX content of x ≥ 0.02. The grain size in the “Polarization fatigue in ferroelectric Pb(Zr0.52 Ti0.48)O3-SrBi2O9
ceramics,” Electron. Mater. Lett., vol. 11, pp. 881–889, Feb. 2015.
modified microstructure decreases with the increase of Sm and
[16] J. Yi, M. Shen, S. Liu, and S. Jiang, “Effects of PbO-B2O3 glass doping
Cr substitution. The SC addition promoted densification of on the sintering temperature and piezoelectric properties of
ceramics and the highest density of 7.91 g/cm3 was obtained at 0.35Pb(Ni1/3Nb2/3)O3-0.65Pb(Zr0.41Ti0.59)O3 Ceramics,” Journal
0.025 wt.% SC addition. The room temperature dielectric Electronic Materials, vol. 44, pp. 2846–2851, Sep. 2015.
[17] X. Zong, and Z. Yang, “Effects of Fe2O3 Additive on the structure and
constant (ɛr) and tangent loss (tanδ), piezoelectric coefficient
piezoelectric properties of PZT-PFW-PMN Ceramics,” Journal
d31, electromechanical coupling factor kp, and mechanical Electronic Materials, vol. 44, pp.258–262, Jan. 2015.
quality factor (Qm) of 0.975 Pb (Zr0.52 Ti0.48) O3 – 0.025 Sm Cr [18] J. Zeng, K. Zhao, X. Shi, X. Ruan, L. Zheng, and G. Li, “Large strain
O3 ceramics are 998, 1.405 %, 293 pC/N, 0,58, 745, induced by the alignment of defect dipoles in(Bi3+,Fe3+) co-doped
respectively, which mean it has a great promise for ultrasonic Pb(Zr,Ti)O3 ceramics,” Scripta Materialia, vol. 142, pp. 20–22, Aug.
2018.
motors. [19] Y. Zhang, J. Roscow, M. Xie, and C. Bowen, “High piezoelectric
sensitivity and hydrostatic figures of merit in unidirectional porous
REFERENCES ferroelectric ceramics fabricated by freeze casting,” Journal European
[1] J. Xu, M. Yang, K. Gana , Y. Qu, X. Zhang, N. Ma, Y. Wanga, and J. Ceramic Society, vol. 38, pp. 4203–4211, Apr. 2018.
Yang, “Enhanced piezoelectric properties of PZT ceramics prepared by [20] T. A. Babu, K. V. Ramesh, V. R. Reddy, D. L. Sastry, “Structural and
direct coagulation casting via high valence counterions (DCC–HVCI),” dielectric studies of excessive Bi3+ containing perovskite PZT and
Ceramics International, vol. 42, pp. 2821–2828, Nov. 2016. pyrochlore biphasic ceramics,” Materials Science Engineering B, vol.
[2] G. Tian , W. Deng , Y. Gao , D. Xiong , C. Yan , X. He , T. Yang , 228, pp. 175–182, Jan. 2018.
L. Jin , X. Chu , H. Zhang, W. Yan, and W. Yang, “Rich lamellar [21] Y. Yan, Y. Feng, and Z. Li, “The effect of uniaxial pressure,
crystal baklava-structured PZT/PVDF piezoelectric sensor toward temperature and electric field, on the internal bias field in the lead-
individual table tennis training,” Nano Energy, vol. 59, pp. 574- based piezoelectric ceramics,” Materials Letters, vol. 164, pp. 248–251,
581, Mar. 2019. Jan. 2016.
[3] Z. Zhang, J. Xu, L. Yang, S. Liu, J. Xiao, X. Li, X. Wang, and H. Luoa, [22] A. Limpichaipanit, S. Somwan, and A. Ngamjarurojana, “Dielectric
“Design and comparison of PMN-PT single crystals and PZT ceramics properties of PFN–PZT composites: From relaxor to normal
based medical phased array ultrasonic transducer,” Sensors Actuators A, ferroelectric behavior,” Ceramics International, vol. 44, pp. 14797–
vol. 283, pp. 273–281, Sep. 2018. 14802, May. 2018.
[4] J. Wu, Y. Mizuno, and K. Nakamura, “Ultrasonic motors with poly [23] O. M. Zhigalina, D. N. Khmelenin, Y. A. Valieva, V. Y. Kolosov, A. O.
phenylene sulfide/alumina/PZT triple-layered vibrators,” Sensors Bokuniaeva, G. B. Kuznetsov, K. A. Vorotilov, and A. S. Sigov,
Actuators A, vol. 284, pp.158–167, Oct. 2018. “Structural Features of PLZT Films,” Crystallography Reports, vol. 63,
[5] S. Gholami, M. M. Mohebi, and B. Janipour, “Template-free formation pp .646–655, 2018.
of Pb(Zr0.52Ti0.48)O3 (PZT) nanorods,” J. Mater. Sci: Mater. Electron, [24] K. Guo, Q. Mou, T. He, H. Kong, and R. Zhang, “Ferroelectric,
vol. 26, pp. 10087–10094, Sep. 2015. ferromagnetic, magnetodielectric and in-plane ME coupling properties
[6] S. Hajra, P. Sharma, S. Sahoo, P. K. Rout, and R. N. P. Choudhary, of Pb(Zr0.52Ti0.48)O3–Bi0.9Nd0.1FeO3 bilayer nano-films prepared via
“Processing and electrical properties of gallium-substituted lead Sol–gel processing,” J. Mater. Sci: Mater. Electron., vol. 28,1 pp. 971–
zirconate titanate ceramics,” Applied Physics A, vol. 123, pp. 786 –795, 1975, Jan. 2017.
2017.
Copyright - 2019
Page 63
ISSN 1737-9334
Production of 3rd generation biofuel from green

microalgae : Oedogonium nodulosum
Nadjla CHAIB 1, Sabrina DZIZI 1,2, Faiza NOUNE 1,3, Hadjer KADDECHE 1,3
Faculty of Technology, University of 20 august 1955 – Skikda (Algeria)
1
1
n.chaib@univ-skikda.dz
2
Laboratoire de recherche sur la physico-chimie des surfaces et interfaces (LRPCSI), Université de Skikda (Algérie)
3
Faculty of sciences, University of 20 august 1955 – Skikda (Algeria)
2
nadjlachaib@gmail.com
Abstract— The socio-economic and environmental situation Biofuels are intended to replace petroleum-based fuel.
strongly marked by the increase in oil production and global They are a sustainable alternative to fossil fuels as they are
warming has led researchers towards an alternative path such as renewable and less toxic to the environment. For the time
renewable energies. The purpose of this study was to synthesize
biofuel from oil extracted from green microalgae by the being, these biofuels are of vegetables origin and come from
transesterification process using ethanol as alcohol and agricultural crops. A major disadvantage is the debate
potassium hydroxide (KOH) as homogeneous catalyst for the between the use of land to produce food or biofuel [9]. As a
reaction. Odogonium nodulosum was the selected green
result, many studies have been conducted to use an alternative
microalgae for our Biodiesel production. This species exhibited
high capacity of lipid (45.38% in dry weight) which could be of energy source, many of which focus on algae.
high potential as raw materials for biofuel production, with an Indeed, the biomass from algae is convertible via chemical
overall efficiency of 80%. Some physic-chemical properties and or enzymatic transformations into biodiesel or bioethanol.
characterization of the obtained products were also verified such
relative density, refractive index, kinematic viscosity, cloud Almost all biomass can be transformed into transport energy.
point, color and Biodiesel flash point measurements. The The lipids are converted into biodiesel and the rest of the
comparison of our results with the properties of Diesel and biomass can be used to produce bio-methanol. The production
Gasoline showed us that our biodiesel has properties very similar
of biofuel from algae has many advantages that go beyond
to those of diesel but more ecological.
replacing oil as transport energy [10, 11].
Keywords— Green microalgae, Odogonium nodulosum, Lipids In fact, the purpose of this study is to synthesize a
extraction, Transesterification, Biofuel biodiesel from green microalgae oils by transestérification
I. INTRODUCTION using a basic homogenous catalyst.
With the continued development of industry and transport, The most effective basic catalysts used for the
energy demand in our world is increasing day after day. Oil, transesterification of triglycerides include NaOH, NaOMe,
or even all fossil fuels (coal, natural gas and oil), currently KOH and KOMe [12, 13]. In addition, the basic catalyst
provide almost all of this energy. However, these resources technique in a homogeneous medium in the presence of
are not renewable and their consumption at the current rate methanol (MeOH), is the most commonly used because it is
will lead to the disappearance of reserves in a few decades [1]. the cheapest and can be carried out at low temperatures and
On a global scale, the solar energy received represents about pressures [14].
10,000 times humanity's annual energy consumption. It is
II. METHODS AND MATERIALS
clearly the most abundant source of renewable energy [2].
Photosynthesis is an excellent natural way to convert and store Microalgae samples were collected from a freshwater
channel covered with filamentous algae according to the
solar energy [3]. The world of aquatic photosynthetic
CIPEL protocol [15]. The samples were grown at laboratory
microorganisms, including microalgae, represents a scale in an algal photobioreactor in batch mode. A dominant
considerable and almost untapped potential for bioenergy green algae species has been selected for the production of
production [4, 5]. Indeed, these microorganisms are of great biofuel. It is Oedogonium nodulosum (Fig. 1). The most
importance, due to their ability to produce high-value and important physical (light, agitation, temperature) and chemical
energy-interesting compounds such as: fats, hydrogen, (various nutrients) factors necessary for its growth and the
carbohydrates and proteins [6, 7]. Their use makes it possible culture medium used were considered [16]. This microalgae
was initially grown in batch mode in 1.5 liter photobioreactors
to consider the development of innovative biofuel production
under sterile and aseptic conditions, while considering that
processes [8].

ISSN 1737-9334
light and carbon dioxide have an important influence on the extraction cycles are thus carried out, the duration of the
cell growth of Oedogonium nodulosum. operation is fixed at 4 hours according to the standard [18]. It
is then considered to have reached total solute depletion of the
solid substrate and concentrated the extract.
Fig. 1 Oedogonium nodulosum (x400 magnification)
A biomass-concentration monitoring was carried out in

order to recover the optimal amount of our selected species
under optimal growth conditions (temperature (22°C), light
(120µmol.m-2.s-1), agitation (100 rpm), dissolved CO 2 level
(13%), pH ≈ 6 to 8, and nutrients). The maximum biomass
growth rate µmax is calculated as follows:
𝑙𝑜𝑔2 𝑁𝑛 − 𝑙𝑜𝑔2 𝑁0
µ𝑚𝑎𝑥 =
𝑡𝑛 − 𝑡0
With : Fig. 2 Soxhlet diagram (source: Engineering Tech.) R : Refrigerant, E :
Extractor, B : Balloon
µ max : maximum biomass growth rate (mn-1) ; log2 : decimal
logarithm ; N0 : Initial concentration of dry algal mass (mg.L- The extracted oil was then placed in the Rotavapor flask to
1
) that corresponds to the initial time t0 (mn); Nn : Highest
remove organic solvents in excess (Fig. 3) by distillation
Biomass Concentration (mg.L-1) corresponding to time tn
(mn). process at 70°C with 100 rpm [19].
II.1. Physico-chemical analyses

II.1.1. Determination of dry matter
To determine dry matter (DM) content of wet algal "pastes",
a standard MB35 moisture meter was used. This equipment
works on the principle of thermogravimetry, which consists in
measuring the mass variation of a sample according to the
temperature, which is fixed to 110°C until a constant mass
was obtained [17]. Once the stabilized weight has been
reached, drying is completed and the result displayed indicates
the percentage of moisture present in the matrix and the final
weight of the sample. Dry matter calculation of the algal paste
was carried out as follows: Fig. 3 Rotavapor type R3000-Gemini BV for the distillation of extracted oil
𝐷𝑀% = 100 − 𝐻𝑀% II.2.1. Homogeneous transesterification reaction

A basic homogeneous transesterification reaction was
Avec : DM%: percentage of dry matter ; HM%: pourcentage conducted to transform the extracted oil into an ester [20]. 200
d’humidité donné par l'humidimètre ml intake of oil extracted from Oedogonium nodulosum is
placed in a tricols flask, to which 341 ml of methanol (MeOH)
II.1.2. Lipid extraction and 1.85 g of KOH were added. The proportion of the catalyst
Lipid extraction from dry Oedogonium nodulosum algae amount to be used varies within a range of 0.5 to 1% of the oil
was performed using the Soxhlet system (ISO 659-1988) [18]. mass, for a conversion efficiency of 94 to 99% [14, 20]. The
This method consists in extracting oils contained in the algal mixture is heated up rapidly to 70°C, for the reaction under
cells below 70°C using for our case a combination of two mechanical agitation of 100 rpm [21]. The standard
organic solvents: n-hexane and chloroform (2: 1). The solvent, transesterification reaction is as follows:
contained in the flask, is boiled, which then transfers it to the
upper part of the Soxhlet (Fig. 2). There, it is condensed by a 𝑇𝑟𝑖𝑔𝑙𝑦𝑐𝑒𝑟𝑖𝑑𝑒 + 𝐴𝑙𝑐𝑜ℎ𝑜𝑙 ↔ 𝐸𝑠𝑡𝑒𝑟 𝑎𝑙𝑘𝑦𝑙𝑖𝑞𝑢𝑒 + 𝐺𝑙𝑦𝑐𝑒𝑟𝑜𝑙
refrigerant at the top of the installation and accumulates
around and inside the cartridge. When the solvent reaches the However, the molar ratio alcohol/oil 9/1 chosen for our
upper level of the siphon, the mixture is returned to the tank case of methanolic transesterification of the extracted oil,
by pressure difference, where it is evaporated again. Several favors a better separation of glycerin because it increases its
Copyright - 2019
ISSN 1737-9334 Page 65
solubility [22]. Methanol is the alcohol used in our mass of the extracted oil from this dry algal mass is 112.58g
transesterification reaction, due to its better reactivity than with an extraction efficiency rate of 43.32%.
ethanol [19, 22]. This efficiency advantage is that the use of
methanol does not pose any difficulties during separation [20]. C. Transesterification Reaction Efficiency
The water content of the extracted oil is 0.5 (% by weight), The influence of various transesterification reaction times
and the total phosphorus content was 0.8 ppm. (1, 2 hours) at temperature (70°C) was studied (mass ratio
After 120 minutes of stirring, the resulting mixture is MeOH / oil of 9 /1 and 1% by mass KOH). The evolution of
poured into a separating funnel and left to stand for 24 hours, biodiesel yield (g biodiesel /g oil) over the time of reaction is
in order to promote the separation of the biofuel from shown in (Fig. 5).
Glycerol/methanol.
The transition from an oil state to an alkyl ester state, 80%
commonly known as biodiesel, makes it possible to: reduce
the molecular weight to one-third of that of the oil, reduce the
viscosity by a factor of 8, reduce the density and increase its
volatility [24, 25].
70%
A purification step of the ester consists in removing
impurities such as residual glycerin, excess alcohol, traces of
catalysts, soaps and salts formed by the homogeneous
catalysis. This purification operation is carried out by flash
distillation under vacuum. t = 1h t=2h
After washing the biodiesel with water, the drying
operation allows the water present to be removed by heating
Fig. 5 Evolution of biodiesel yield according to reaction time (mass ratio
the product to a high drying temperature (120°C) in the oven. MeOH/oil: 9/1 KOH : 1% mass, 70°C)
III. RESULTS & DISCUSSION The transesterification reaction results show that the
A. Biomass Growth maximum biodiesel yield of 80% is obtained after 2 h reaction
Based on microscopic examination, we mainly observed is complete. An efficiency of 70% is obtained after 1 hour of
the abundance of chlorophyceae, and the culture conditions in reaction at room temperature. Therefore, the reaction time
the photobioreactor in batch mode showed a concentration of seems to have an influence on biodiesel yield.
algal biomass of the Oedogonium nodulosum species, with the
higher growth rate (µ max = 4.9 days-1) (Fig. 4). D. Physicochemical Properties of Biodiesel
The physico-chemical properties of biodiesel produced
70 from Oedogonium nodulosum oil, compared to those of diesel
fuel, are presented in Tab. 1.
60
50 Tab. 1 Physicochemical properties of biodiesel from Oedogonium nodulosum

oil compared to diesel oil
Acceleration
40
Log2 N
phase
Exponential -3 2 -1
Deceleration
30 phase
phase
Compounds pH Density g.cm Viscosity at 40°C mm .s Flash point °C Cloud point °C
20 Oedogonium nodulosum oil 8.59 0,9236 0,663 155 -10
10
Biofuel 9.77 0,8484 1,8984 138 -12
Diesel Fuel 9.39 0,8393 2,8620 8 -14
0
0 2 4 6 8 10 12 14 16 18
Biodiesel has a lower density, viscosity and flash point
Time (days)
than extracted oil. These lower physico-chemical properties of
Fig. 4 : Biomass growth curve of the selected species Oedogonium nodulosum biodiesel compared to that of the oil correspond to those of
methyl esters by glycerol removal. However, the density and
From the maximum growth rate obtained, we can deduce flash point of biodiesel are higher than those of diesel. These
that the maximum number N of Oedogonium nodulosum was results show that the combustion of the latter, particularly at
reached from the 11th day of pure algal culture. start-up, is easier than that of biodiesel.
Log2Nn = X = 41.5 where : N = 2X = 241.5 ≈ 1018 cells.L-1. E. Optimization of Biodiesel Yield
In order to determine the optimal biodiesel yield, a series
B. Characterization of the Extracted Oil
of transeterification reactions of the extracted oil were carried
This cultured algae paste was used for lipid extraction.
out under the following conditions: reaction time: 1h,
After drying, we obtained a dry mass of Oedogonium
temperature 70°C, mass ratio Methylene / oil: 12/1, 9/1, 6/1
nodulosum equal to 259.87g from a wet paste of 1.55kg. The
with a mass percentage KOH / oil: 1%.

ISSN 1737-9334
Tab. 2 Biodiesel production results according to the alcohol/oil mass ratio chain alcohol (methanol) for the reaction and KOH as a
catalyst. Parameters affecting alkyl ester yield such as catalyst
1st trial 2nd trial 3rd trial mass percentage (1% KOH) and catalyst/oil mass ratio were
Alcohol / oil mass ratio 12 / 1 9/1 6/1 studied over time.
Yield (%) 79 80 68 The highest biodiesel yield obtained is 80% and corresponds
to a mass ratio of 9/1 Methanol / oil, at a temperature of 70°C
The best separating result was obtained with a ratio of 9/1 and a reaction time of 2 hours.
(Fig. 6). V. REFERENCES
[1] A Préat (2012). Paniquer sur les réserves du pétrole ? Université libre de
Bruxelles. Available :
http://www.academieroyale.be/academie/documents/AlainPreatPaniquere
servespetrole15844.pdf
[2] Keniar, K., Ghali, K., & Ghaddar, N. (2015). Study of solar regenerated
membrane desiccant system to control humidity and decrease energy
consumption in office spaces. Applied Energy , Volume: 138:121-132.
[3] D. Kramer (2017). Biology lead the race to turn sunlight into fuels.
Physics today , 70, 4, 30.
Biofuel from [4] A. Demirbas and M. Demirbas (2009). Algae Energy: Algae as a New
Oedogonium nodulosum Source of Biodiesel. Springer , 224.
[5] B. S. SAHARAN, D. SHARMA, R. SAHU, O. SAHIN and A. WARREN
(2013). TOWARDS ALGAL BIOFUEL PRODUCTION: A CONCEPT
OF GREEN. Innovative Romanian Food Biotechnology , 12: 1-21.
[6] B. SANDESH KAMATH (2007). Biotechnological production of
microalgal carotenoids with refrece to astaxanthin and evaluation of its
Glycerol + Methanol biological activity. Thèse de doctorat, Univ of Mysore, India
[7] R. Voloshin, M. V. Rodionova, S. K. Zharmukhamedov, S. I.
Allakhverdiev (2016). Review: Biofuel production from plant and algal
biomass. International Journal of Hydrogen Energy 41(39).
DOI: 10.1016/j.ijhydene.2016.07.084
[8] W. RYAN and A. Z. HUNT (2009). Das Electromagnetic Biostimulation
Fig. 6 Biodiesel produced by transesterification separated from the organic of Living Cultures for Biotechnology, Biofuel and Bioenergy
phase Glycerol/methanol Applications 10, 4515-4558. Int. J. Mol. Sci. , 10: 4515-4558.
[9] OFID study prepared by IIASA (2009). Biofuels and food security.
Glycerol is essentially insoluble in biodiesel, so almost all Implications of an accelerated biofuels production. Vienna, Austria.
Available: http://www.globalbioenergy.org/uploads/media/0903_OFID_-
glycerol is easily removed by settling. Alcohol can act as a co- _BiofuelAndFoodSecurity.pdf
solvent to increase the solubility of glycerol in biodisel. In our [10] J. T. Teck Chye, Y. J. Lau, L. S. Yon, M. K. Danquah, (2018). Biofuel
case, the yield obtained is less than 100 (Tab. 2). This can be production from algal biomass. In book: Bioenergy and Biofuels Edition:
explained by the fact that glycerol has not been solubilized in First Chapter: 3 Publisher: CRC Press, Boca Raton (Taylor & Francis
Group) Editors: Ozcan Konur. DOI: 10.1201/9781351228138-3
biodiesel due to the transformation of glycerol into fatty acid [11] Y. Chisti (2007). Biodiesel from microalgae. Biotechnology Advances ,
which in turn has undergone saponification in the presence of 25: 294–306.
KOH [26, 27, 28]. Indeed, during transesterification, parallel [12] E. I. Bello, T. S. Mogaji and M. Agge (2011). The effects of
parasitic reactions occur at the same time [28, 29, 30]. Fatty transesterification on selected fuel properties of three vegetable oils.
Journal of Mechanical Engineering Research , 3(7): 218-225.
acids react with a basic catalyst to form soap [31, 32]. This [13] S. Nasreen, M. Nafees, L. A. Qureshi, M. Shahbaz Asad, A. Sadiq and
will inevitably reduce the conversion efficiency [33]. A. S. Danial (2018). Review of Catalytic Transesterification Methods for
Biodiesel Production. Open access peer-reviewed chapter , DOI:
IV. CONCLUSION 10.5772/intechopen.75534.
[14] A. P. Vyas, P. H. Shukla and N. Subrahmanyam (2011). Production of
Biodiesel using Homogeneous Alkali Catalyst and its Effect on Vehicular
From the standpoint of renewable energy and sustainable Emission. INTERNATIONAL CONFERENCE ON CURRENT TRENDS
development, third-generation biofuels are attracting IN TECHNOLOGY , 1-7.
increasing investment interest from researchers and some [15] J.C. DRUART et F. RIMET. Protocoles d’analyse du phytoplancton de
green industries. Biofuels are emerging as an essential l’INRA : prélèvement, dénombrement et biovolumes. INRA-Thonon,
Rapport SHL 283 - 2008, 96 p.
transport energy source. One of the interesting sources of [16] A. E. Simosa (2016). Factors affecting algal biomass growth and cell
biofuels are algae, which are capable of producing biodiesel wall destruction. University of New Orleans Theses and Dissertations.
and bioethanol that can be used in different types of engines. 2277. Available: https://scholarworks.uno.edu/td/2277
Biodiesel, which is derived from triglycerides methyl or ethyl [17] C.J. Zhu and Y.K. Lee (1997). Determination of biomass dry weight of
marine microalgae Journal of Applied Phycology (1997) 9: 189. Available:
alcohol by transesterification, has attracted considerable https://doi.org/10.1023/A:1007914806640
attention over the past three decades due to its benefits: [18] ISO 659: 1988 (1988). Oilseeds -- Determination of hexane extract (or
renewable, biodegradable and non-toxic fuel. The objective of light petroleum extract), called "oil content". International
this work was to synthesize a biodiesel from green Organization for Standardization. Available:
https://www.iso.org/standard/4815.html
filamentous microalgae "Oedogoium nodulosum" oils, by [19] C. Dejoye (2013). Eco-extraction et analyse de lipides de micro-algues
transesterification, which is the most widely used technique pour la production d’algocarburant. Université d’Avignon, NNT :
for biodiesel synthesization. For this purpose, we used a short- 2013AVIG025.

ISSN 1737-9334
[20] M. K. Lam, K. T. Lee, and A. R. Mohamed (2010). Homogeneous,

heterogeneous and enzymatic catalysis for transesterification of high free
fatty acid oil (waste cooking oil) to biodiesel: A review. Biotechnology
Advances , 28 : 500-518.
[21] S. PAZ, P. Meneghetti, MR Meneghetti, CR. Wolf (2007).
Transformation of triglycerides into fuels, polymers and chemicals: some
applications of catalysis in oleochemistry. Quim Nova 2007; 30:667–76.
[22] Y. Tang, Y. Zhang, J. Rosenberg, F. Wang (2016). Optimization of One-
Step In Situ Transesterification Method for Accurate Quantification of
EPA in Nannochloropsis gaditana. Applied Sciences 6(11):343.
[23] G. R. Stansell, V.M. Gray and S.D. Sym (2012). Microalgal fàtty acid
composition: implications for biodiesel quality. Journal of Applied
Phycology, volume 24, numéro 4, p.791-801.
[24] A. GERAR TCHAKBLO (2009). Production de Biodiesel par
Tansestérification Alcoolique : étude sur réacteur pilote, pp 24.
[25] M. Hunsom, P. Saila, P. Chaiyakam and W. Kositnan (2013).
Comparison and Combination of Solvent Extraction and Adsorption for
Crude Glycerol Enrichment. INTERNATIONAL JOURNAL of
RENEWABLE ENERGY RESEARCH , Vol.3, No.2, 2013.
[26] B. K. Uprety (2017). Conversion of crude glycerol from the biodiesel
industry to value added products. Lakehead University, Thunder Bay,
Ontario, Canada, 215 p.
[27] J. A. Dutton (2015). Alternative Fuels from Biomass Sources. e-
Education Institute. https://www.e-education.psu.edu/egee439/node/684
[28] T. Rachmanto, D. Allanson, C. Matthews and I. Jenkinson (2014).
Monitoring of Biodiesel Transesterification Process Using Impedance
Measurement. International Journal of Materials, Mechanics and
Manufacturing, Vol. 2, No. 4.
[29] A. Rodrigues, J. C. Bordado and G. dos Santos, (2017). Upgrading the
Glycerol from Biodiesel Production as a Source of Energy Carriers and
Chemicals— A Technological Review for Three Chemical Pathways.
Energies, 10, 1817.
[30] F. Trejo-Zárraga, F. Herernádez-Loyo, J. C. Chavarría-Hernández and R.
Sotelo-Bovás (2017). Kinetics of Transesterification Processes for
Biodiesel Production. Intech Open Chapter Book. DOI:
105772/intechopen.75927.
[31] E. Suyenty, H. Sentosa, M. Agustine, S. Anwar, A. Lie and E. Sutanto,
(2007). Catalyst in Basic Oleochemicals. Bulletin of Chemical Reaction
Engineering & Catalysis, 2(2-3), 22-31.
[32] K. Amri, I. Paryanto, A. Kismanto, Adiarso and T. H. Soerawidjaya
(2011). Esterification Reaction of High Free Fatty Acid Content of Crude
Palm Oil and Glycerol Using Zinc Glycerolate and Zinc Soap Catalysts.
Conference: Sriwijaya International Seminar on Energi Science and
Technology. DOI: 10.13140/RG.2.1.2431.4409
[33] G. Yoo, M. S. Park and J. W. Yang, (2015). Chemical Pretreatment of
Algal Biomass. In book Pretreatment of Biomass. DOI: 10.1016/B978-0-
12-800080-9.00012-8.

ISSN 1737-9334
Comparative study of Two and Three-Bladed Wind

Turbine’s performances, FAST as Framework
Nedia Aouani#1, Uwe Ritschel*2, Faouzi Chaabane#3

#
National School of Advanced Sciences and Technology, Borj Cedria, Université de Carthage, Tunisia
aouani_nedia@yahoo.fr
faouzi.chaabane@enstab-b.rnu.tn
*
Stiftungslehrstuhl für Windenergietechnik (LWET) , Universität Rostock, Germany.
uwe.ritschel@uni-rostock.de
Abstract— In this paper we compare dynamic behaviours of model. However, the two bladed prototype have taken
two and three bladed 5MW Wind Turbine. advantage of many attention. Numerous works have dealt
with performance differences between both patterns [3] , [5] .
There is huge number of software and frameworks that have
Keywords— Two-bladed wind turbine, Three-bladed wind
been developed in this context. Among the tools of
turbine, 5MW-NREL wind turbine, FAST. Torque,
Performance Simulation for load cases dedicated to Wind Turbines, there
are : Bladed, Flex 5 [6], FAST [7], [8].
I. INTRODUCTION In this paper, we focus on the Class of HAWT and will
Wind Energy has become lately, one of the most important compare the performance of two and three bladed Wind
sustainable resources one can investigate in order to get Turbine using FAST Simulation tool.
electricity [1]. The prototype investigated for the simulation process is the
Wind is a clean, economic and renewable energy source with 5MW NREL model given in [9]. The results show the
a great availability in the environment. The main role of change in torque coefficient, startup speed and power
Wind turbines is to convert the kinetic energy available in generation.
the wind in order to get to produce electric energy [2]. The paper is organised as follows. In the next section,we give
Wind turbines are classified into two broad classes: Vertical the model development of the Wind Turbine. The section that
Axis Wind Turbine (VAWT) and Horizontal Axis Wind follows presents the BEM Theory where we express the
Turbine (HAWT) [3] , former one being the most popular Torque, the Power and other characteristics relative to an
device for electricity generation for its higher efficiency due element ring of the Blade. The section IV gives an overview
to high rotational speed. of the FAST Framework. Finally in section V, we provide
In terms of modelling, wind turbines are complex mechanical the simulation results, the 5MW NREL’s model Wind
systems exposed to uncontrolled wind profiles which make Turbine is used and experiments are carried out under a
turbine design a challenging task [4]. steady wind of 18m/s.
Load Simulations are considered as a very important part in
II. MODEL DEVELOPMENT
the Design Process of a Wind Turbine.
Before the assembly of Wind Turbine Components and its A. The Wind Turbine Model
installation, one needs to build a numerical model besides of The following well-known algebraic equation (Eq. 1) gives
the test procedure. the relation between wind speed and mechanical power
Instead of testing real model Wind Turbines that cost Million extracted from the wind [10].
euros per Megawatt, virtual prototypes are used and this 1 3
Pwr = ( )rCp(l , q)A V wind (Eq.1)
charges less for the industries. 2
Several studies have been conducted in order to simulate
behaviours of Wind Turbines, especially the three bladed where Pwr is the power extracted from the wind in watts;

ISSN 1737-9334
 is the air density (kg/m3); B is the number of Blades

Cp is the performance coefficient or power coefficient; r is the air density (1.225 kg/m³)
Vt w is the relative speed of the wind expressed such that:
 is the tip speed ratio ( ) , the ratio between blade tip
V wind
w2 = v2 + u 2 (Eq.4)
speed Vt (m/s), and wind speed at hub height upstream of the
rotor Vwind, given by :
Where
• v is the axial wind speed in the rotor plane
Vt 2.R .wr • u is the tangential wind speed
l = = hGB (Eq.2)
V wind pV wind (See Fig.2)
c is the chord length. r and the thickness dr are defined in
p is the number of poles; the previous paragraph (see Fig.1).
r is the electric speed of the rotor;
 is the pitch angle (in degree) C L is the “Coefficient of Lift”,
R is the radius of the turbine (Blade length) in m. g is the angle of relative wind to rotor axis.
And A is the area covered by the wind turbine rotor (m2). w is the angular speed of the rotor given by
B. The Blade Element Momentum Theory (BEM Theory) w = 2p n [rad/s] (Eq.5)

where , n is the rotational speed of the rotor in round per
In the blade element momentum (BEM) method the flow area second.
swept by the rotor is divided into a number of concentric ring
elements. The rings are considered separately under the The blade, as shown on the Fig.2, is moving up wards, thus
assumption that there is no radial interference between the the wind speed, seen from the blade, is moving down wards
flows in one ring to the two neighbouring rings [11]. with a speed of u.
In this section, it is demonstrated by equations, how does the
number of blades in a Wind Turbine influences the electric or
aerodynamic parameters , be it the generated power, the rotor
speed, the torque or others.
If we look at one blade element in the distance r from the

rotor axis with the thickness dr, one Wind Turbine can be
schematized as follows [11]:
Fig.2: Velocities and angles on a chord Line
For more details about the notations on Fig.2, we mention:

• v1 represents the wind speed upstream rotor (m/s)
• φ represents the angle of relative wind to rotor plane
• β represents the pitch angle of the blade
The angle of attack is given by : α=φ-
Fig.1: Blade section
FL and FD are respectively the Lift and the Drag Forces
applied to the airfoil.
R is the length of the Turbine.
Heneceforth; the next equations can be edited:
Power equations [12] Torque equations
If we have B blades, the power produced is expressed such If we have B blades [12], the Torque equation in one ring is
given by
that
1
1
dP = B r w 2cdrC L cos( g )r w [W]
(Eq.3) dU = B ( )r w 2cdr C x [N.m] (Eq.6)
2 2
Where
dP is the Power produced on a ring element dU is the Torque on a ring element

ISSN 1737-9334
C x is such that
C x = C L sin(j ) - C D cos(j ) [-] (Eq.7)
Where
C L and C D are respectively the “coefficient of Lift” and
the “coefficient of Drag”.
φ is previously defined as the angle of relative wind to rotor

plane.
The other elements of the equation are previously defined in
the power expression.
For further details about these equations, see [2], [11], [13],
[14].
III. WORK DESCRIPTION AND SOFTWARE USED
A. The FAST Framework

The FAST (Fatigue, Aerodynamics, Structures, and
Turbulence) Code is a comprehensive aeroelastic simulator
capable of predicting both the extreme and fatigue loads of
two- and three-bladed horizontal- axis wind turbines
(HAWTs) [15].
So Far, Jonkman have developed numerous versions of
FAST. The one that is actually in use is the FAST V8.
The application in this paper, uses the design tool FAST
with AeroDyn and ElastoDyn and ServoDyn—developed and
verified by Jonkman and Buhl at NREL [7]—to create a
model of a wind turbine concept. The time-domain simulation Fig. 3 FAST V8 [17]
tool FAST can model the aero-hydro-servo-elastic response
of a variety of onshore, besides of offshore floating wind
turbines. Elastodyn Module, [18]
Is the Structural-Dynamic module for horizontal-axis wind
turbines. It includes structural models of the rotor, drivetrain,
B. FAST Modules
nacelle, tower, & platform
There are several modules used in the design tool FAST such
that : AeroDyn, HydroDyn, ServoDyn, ElastoDyn, SubDyn,
Servodyn Module [19]
MAP, IceFloe…
Is the Control and electrical-drive module for wind
Each (see Fig.3) considers a specific part of the Wind
turbines and is used to be a fundamental part of FAST . It
Turbine, besides of its environment, either be it onshore
Includes control & electrical-drive models for blade pitch,
farms or offshore ones.
generator torque, nacelle yaw, high-speed shaft (HSS) brake,
In the present work, it is needed the use of the three
& blade-tip brakes
modules, that we define next.
IV. NUMERICAL SIMULATIONS
Aerodyn Module
The aerodynamics are calculated in the FAST module A. NREL 5-MW Baseline Turbine Properties
AeroDyn, which uses a state-of-the art blade element- The NREL 5-MW baseline wind turbine represents a
momentum approach (BEM) with empirical corrections to typical state-of-the-art multi-megawatt turbine. This section
calculate the rotor aerodynamics. The empirical corrections gives an overview of its properties without providing details
consider the losses caused by AeroDyn also is able to apply that are out of the scope of this work. An in-depth description
the generalized dynamic wake (GDW) theory to account for of the turbine can be found in Jonkman [20], [21]. The
the effects of dynamic inflow [16]. baseline turbine’s properties presented in Table 1 are derived
from publicly available data on the Multibrid M5000 and
REpower 5-MW machines and from conceptual models used
in the WindPACT, RECOFF, and DOWEC projects [16].
To perform a thorough loads analysis in the design code
FAST, the floating wind turbine model was first created
[16 ]..
Copyright - 2019
ISSN 1737-9334 Page 71
TABLE I
5MW-NREL WIND TURBINE SPECIFICATIONS
Property Specifications
Related Power 5MW
Rotor Orientation Upwind
Rotor Configuration 61.5m length
Rotor, hub diameter 126m, 3m
Hub height 90m
Wind Speed
Cut- in Vin 3m/s
Rated 11.4 m/s
Cut-out Vout 25m/s
Cut-in rotor speed 6.9 rpm
Drivetrain concept Geared
Gearbox ratio 97:1
Rated generator speed 1173.7 rpm Fig. 5 A Two Bladed Wind Turbine performances (respectively the
Generator Speed, the Generator Power and the Generator Torque)
Generator efficiency 94.4 %
Rated tip- speed 80m/s C. Three Bladed WT
Overhang 5m For this simulation, in the ElastoDyn Module file , we set
Shaft tilt 5° specifically the Number of Blades in the Turbine
Precone 2.5° Configuration NumBl=3.
Rotor mass 110,000kg Shapes of the Rotor speed, the Power and the Generator
Nacelle mass 240,000kg Torque, are depicted in Fig. 6
Tower mass 347,460kg
Tower Diameter base 6m
Tower top diameter 3.87m
Structural Damping Ratio 0.47%
Control system Variable speed generator
torque and collective active
pitch (PI)
The simulation examines the onshore wind turbine subject

to a steady 18m/s wind without shear (see Fig.4). We
consider both cases: Wind Turbine with two blades, then
wind turbine with three blades.
Fig. 6 A three Bladed Wind Turbine performances (respectively the

Generator Speed, the Generator Power and the Generator Torque)
Based on Figures 5 and 6, we notice the differences between

Two and Three Bladed Wind turbines. In Fig.5, it is clear
that the startup speed of a WT with Two Blades is lower than
the one with three blades. There is experimental evidence that
for smaller scale turbines or at higher speeds, adding more
blades can actually increase efficiency and other things like
Fig. 4 Wind Profile : steady 18m/s without shear startup speed. We can note the value of 1140 rpm for a two-
bladed Wind Turbine, whereas it is 1160 rpm for a three-
B. Two Bladed WT
bladed one.
For this simulation, in the ElastoDyn Module file , we set
Also, from the analysis it is found that rotor with less blade
specifically the Number of Blades in the Turbine
number produces higher Torque (Fig.5). Initial value of the
Configuration NumBl=2.
torque is 45.5 KN.m towards 43.9 KN.m for three bladed
Shapes of the Rotor speed, the Power and the Generator
WT.
Torque, are depicted in Fig. 5
The value of Power Generated is the same for any blade
numbers.
Copyright - 2019
Page 72
ISSN 1737-9334
V. CONCLUSION [7] Jonkman, J.; Buhl, M. FAST User’s Guide. NREL/TP-500-38230.

Golden, CO: National Renewable Energy Laboratory, October 2005.
In this work, we have provided that number of bladed in an [8] P. Schünemann, “First Steps with OpenFAST”, LWET , Universität
HAWT influences well the aerodynamic behavior of the Rostock, Germany, 25 April 2017.
Turbine. [9] J. Jonkman, S. Butterfield, W. Musial, and G. Scott, “ Definition of a 5-
MW Reference Wind Turbine for Offshore System Development”,
Technical Report , NREL/TP-500-38060 , February 2009.
ACKNOWLEDGMENT [10] N. Aouani, F. Bacha and R. Dhifaoui (2009) ‘Control strategy of a
The The present document is resulting on a cooperation variable speed wind energy conversion system based on a Doubly Fed
between the Stiftungslehrstuhl für Windenergietechnik Induction Generator’, IREACO International Review of Automatic Control,
Vol. 2, No. 2, pp.197–204.
(LWET) , Universität Rostock, Germany and the National [11] S. Gundtoft, “Wind Turbines”, University College of Aarhus, June 2009.
School of Advanced Sciences and Technology in Borj [12] N.Aouani, “Wind Energy Lectures”, Chapter 4: Aerodynamics of Wind
Cedria , Université de Carthage Tunisia. Turbines, National School of Advances Sciences and Technologies,
We would like to express our sincere gratitude to the the December 2018.
[13] H. Bredmose, « Wind Energy Lectures : Flow and Forces around a
Stiftungslehrstuhl für Windenergietechnik (LWET) , Wind Turbine Blade » Université technique du Danemark (DTU), Coursera
Universität Rostock, Germany and the National School of 2018.
Advanced Sciences and Technology in Borj Cedria , [14] P.M. Mohan Das & R.S. Amano, “ Basic Theory for Wind Turbine
Université de Carthage Tunisia. for the financial support. Blade Aerodynamics”, WIT Transactions on State of the Art in Science and
Engineering, Vol 81, 2014.
[15] Jason M. Jonkman, Marshall L. Buhl Jr. , “FAST User’s Guide” ,
National Renewable Energy Laboratory, Technical Report , August 2005.
REFERENCES [16] D. Matha, “Model Development and Loads Analysis of an Offshore

Wind Turbine on a Tension Leg Platform, with a Comparison to Other
[1] N.Aouani, “Wind Energy in the context of Tunisian Environement”, Floating Turbine Concepts”, April 2009, University of Colorado – Boulder,
Linkedin Posts, December 15, 2016. Thesis Dissertation.
[2] U. Ritschel , “Modern Wind Turbines”, Wind Turbine Lectures, Tunis, [17] J. Jonkman « Overview of FAST » , NREL Wind Turbine
December, 13-15, 2017. Modeling Workshop September 11-12, 2014 Bergen, Norway.
[3] W. U. Ahmed, M. J. H. Zahed, M. A. Rahman, M. Mamun, [18] J. Jonkman, “ Overview of the ElastoDyn Structural-Dynamics
“Numerical study of two and three bladed Savonius wind turbine”, 2nd Module ”, NREL Wind Turbine Modeling Workshop September 11-12,
International Conference on Green Energy and Technology, September 2014. 2014 Bergen, Norway.
[4] N.Aouani, S. Salhi, G. Garcia, M.Ksouri, « Robust control strategy for a [19] J. Jonkman, “Overview of the ServoDyn Control & Electrical-Drive
wind energy conversion system », Int. J. Computer Applications in Module”, NREL Wind Turbine Modeling Workshop September 11-12,
Technology, Vol. 55, No. 3, 2017. 2014 Bergen, Norway.
[5] M. Predescu1, A.Bejinariu, O.Mitroi, A. Nedelcu, “ Influence of the [20] J. Jonkman, Dynamics Modeling and Loads Analysis of an Offshore
Number of Blades on the Mechanical Power Curve of Wind Turbines”, Floating Wind Turbine. NREL/TP-500-41958. Golden, CO: National
International Conference on Renewable Energies and Power Quality Renewable Energy Laboratory, November 2007.
(ICREPQ’09) , Valencia (Spain), 15th to 17th April, 2009. [21] Pedro A. Galvani , Fei Sun and Kamran Turkoglu, ‘Aerodynamic
[6] SIMPACK News, “Wind Turbine Aeroelastic Simulation by Modeling of NREL 5-MW Wind Turbine for Nonlinear Control System
Embedding FLEX5 into SIMPACK”, Florian Stache, Birte-Marie Ehlers, Design: A Case Study Based on Real-Time Nonlinear Receding Horizon
Suzlon Energy GmbH, November 2009 Control’, Aerospace 2016, 3, 27.
Copyright - 2019
ISSN 1737-9334 Page 73
Performance evaluation of Gamma type Stirling

engine
Houda HACHEM#1, Ramla GHEITH#, Fethi ALOUI*, Samia BEN NASRALLAH#
#
University of Monastir, High Engineering School of Monastir (École Nationale d’Ingénieurs de Monastir- ENIM), Laboratory
LESTE, Avenue Ibn El Jazzar, 5019 Monastir, Tunisia
houdahachem@yahoo.fr
*
LAMIH CNRS UMR 8201, Department of Mechanical Engineering, University of Valenciennes (UVHC), Campus Mont Houy,
F-59313 Valenciennes Cedex 9 - France
Abstract— Present paper is performed to investigate the effect of Gheith et al. [6] proved that the SE power and torque increase
working parameters on Gamma type Stirling engine with hot source temperature. Shuttle loss rises with
performances. An experimental evaluation of an engine with a temperature difference between the two ends of the displacer
maximum power of 500 W has been done. The impact of four which is related to temperature gradient between the two heat
operating parameters (load pressure, rotational speed, hot
sources. Sowale et al. [7] concluded that the performance of
source temperature and cooling water flow rate) on the power
produced by the Stirling engine was discussed. Particular care is the Stirling engine is very sensitive to the heat transfer by
subsequently given to the effect of each parameter on the conduction and the temperature of the working gas. Previous
produced power. Results show that an optimum value of the numerical investigation of 500 W Gamma type Stirling
rotation speed must always be respected to ensure the proper engine [3] showed that the evolution of output brake power
functioning of the Stirling machine. It is also shown that the versus rotation speed reaches an optimal value. In fact, the
brake power is more sensitive to a change in rotation speed and rotation speed increase reduces heat loss by external
hot temperature for high charge pressure values than for lower conduction through engine’s walls, reduces heat exchange
values. time and increases friction loss inside each heat exchanger [3].
This result is approved by Karabulut [8] when investigating
Keywords— Initial charge ressure, hot temperature, speed, the performances of a Beta type Stirling engine driving
cooling water flow rate and power mechanism by means of a lever. Cheng et al. [9-10]
developed a dynamic model incorporated with
I. INTRODUCTION thermodynamic model to study the thermal-lag Stirling
The need of reducing the fossil fuels use; the need of engine start process. They found that the optimal engine
reducing energy bills, the need of searching alternative ways speed leading to maximum shaft power is significantly
to produce cleaner energy and the possibility of waste heat influenced by the geometrical parameters (bore size, stroke,
recovery encourages us to gain from Stirling engine and volume of working spaces). According to Chen et al. [9-
benefits[1]. Nowadays, Stirling installations are patterned and 10] engine speed pose strong positive effects on power but
commercialized such as Microgen installation, Whispergen exert weak effects on efficiency. Optimal frequency depends
unit, Stirling Dish, etc. The Stirling technology provides a on the Stirling engine geometric parameters. Gheith et al. [11]
good solution, since it can work with any external heat source has shown that the heat exchanged in the heater increases
such as solar, combustion of any fuel, nuclear and even with the initial charge pressure until an optimum value. The
warmth. The Stirling engine is characterized by a silent mode regenerator thermal irreversibility decreases with the initial
of operation, easy maintenance, lower NOx, HC, and CO charge pressure. However, the Stirling engine output power
emission and a good efficiency [2]. and efficiency are enhanced until an optimum value.
The Stirling engine is mainly composed by five
compartments: 2 working spaces and three heat exchangers The aim of this experimental work is focused on the
(Heater, cooler and regenerator). The regenerator (porous dependence of four operating parameters (load pressure,
medium) characteristics are determinant for the whole engine rotational speed, hot source temperature and cooling water
performances [3]. Compared to the Ericsson engine, at nearly flow rate) on the power produced by the Gamma type Stirling
the same working conditions, the Stirling engine presents engine.
higher global performances (specific indicated work,
thermodynamic and exergetic efficiencies), due to the II. DESCRIPTION OF THE GAMMA TYPE STIRLING ENGINE (GSE)
presence of a regenerator [4]. The Gamma type Stirling engine is a motor sized by the
German company Viebach ST05 G. This engine is kinematic
Many researches focus numerically and experimentally on type having a rod-crank drive system. Like all Stirling
working parameters effect on the output power of the Stirling engines, this engine is composed of three heat exchangers
engine (SE). Walker experiments [5] show that the SE (the heater, the regenerator and the cooler) and two work
mechanical output power depends on the expansion hot spaces. A section of this engine, on which appear the five
temperature. The dimensionless output power increases elementary spaces of the engine, is provided in Fig. 1. The
gradually with the expansion hot chamber temperature.
Copyright - 2019
ISSN 1737-9334 Page 74
initial charging pressure Stirling Gamma engine is imposed The Stirling's ideal thermodynamic cycle is composed of
by an external compressor. The latter makes it possible to two isochoric and two isotherms. It is similar to the Carnot
keep a constant pressure inside the engine during the cycle; the only difference between them is that the two
experimental measurements. The GSE uses air as a isothermal processes in the Carnot cycle are replaced by two
compressed working fluid at a maximum load pressure of 10 isochoric processes in the Stirling cycle. As shown in fig.2,
bar. Its maximum rotational speed is about 600 rpm and it can the working gas trapped in the engine undergoes the
provide a maximum mechanical power of 500 W. The following transformation: During the first transformation
compression and expansion spaces are arranged at an angle of (34), the volume of the gas decreases and the pressure
90 °. They are animated by a classic system of crank-rod. increases as it gives up heat Qk to the cold source. During the
Table 1 summarizes the different geometric characteristics of second transformation (41), the volume of the gas remains
this prototype. constant as it passes back through the regenerator and regains
TABLE I
some of its previous heat. During the third transformation
GEOMETRIC PARAMETER VALUES OF THE VIEBACH STIRLING (12), the gas absorbs energy Qh from the hot source, its
volume increases and its pressure decreases, while the
setup temperature remains constant. Finally, during the fourth and
Parameters Values Parameters Values last transformation (23), the volume of the gas remains
constant as it transfers through the regenerator and cools.
Working gas air External 134 mm Tmax
diameter
Crank radius 66 mm Internal 98 mm Th
diameter
Length of the 130 Height 50 mm a b
connecting rod mm
Compression space Porous media Cooper Tk q2b
Diameter 80 mm Porosity 85% q4a
Tmin
Height 145 mm Cooler
Course 75 mm Diameter 12mm
Expansion space Length 650 mm Fig.2. Theoretical cycle in the P-V and T-S diagrams of a Stirling engine
Diameter 95 mm Heater d’un moteur
The actual efficiency of Stirling engines Stirling [11
is obviously lower
Height 120 mm Tube number 20 than the theoretical efficiency. The first reason is that heat
Compression 3.5 Tube length 349mm transfer requires keeping a temperature difference between
ratio the sources and the working fluid. Thus, in particular, the
Regenerator Tube inside 6mm internal transfer only partially supports heating 4-1 and
diameter cooling 2-3. We denote by α the fraction of heat exchanged:
q4a = αq41 = −q2b = −αq23 (1)
The rest heat exchanged quantities qa1 and qb3, must be
performed with the sources and significant thermal
differences, which consumes extra energy at the hot source
(qa1), increases irreversibilities, and therefore decreases the
efficiency.
The second reason is that, despite the technological efforts
made, transfers 1-2 and 3-4 with sources are never isothermal
and transformations 4-1 and 2-3 do not strictly correspond to
polytropes. All transformations have a certain level of
irreversibility.
In the Stirling cycle (Fig.2) the thermal exchanges with
sources also correspond to transformations with mechanical
energy exchange, since, being isochoric, the two other
transformations take place without work: W 41 = W23 = 0. We
can therefore write, considering the law of perfect gases P V
= r T:
W12 = −rTmax lnε et W34 = rTmin lnε (2)
W = W12 + W34 = r lnε (Tm − TM ) (3)
With ε is the volumetric compression ratio:
V V
ε= 2= 3 (4)
V1 V4
The temperature and the internal energy being constant, the
amount of heat received from the hot source is:
qmax = q12 = −W12 = rTmax lnε (5)
Let us examine the influence of thermal gradients in
Fig. 1 Gamma type Stirling engine Geometry exchanges (Fig.2) on the yield and, firstly, that of the
coefficient α. In this case, the expression of the job does not
III. STIRLING ENGINE PERFORMANCES
Copyright - 2019
ISSN 1737-9334 Page 75
change. On the other hand, the heat that hot source must 8
(a) Th=300°C
supply is increased by the quantity (1-α) .q41. As: 22
24
0
0
r 7.5
q41 = ∆u41 = cv (Tmax − Tmin ) = (Tmax − Tmin )
γ−1 7
18
Charge pressure (bar)
20 22
(6)
0
0 0
6.5
Thermal efficiency of the Stirling cycle is expressed as:
16
0
6
W 1
ɳth = − = ɳc 1−α ɳc
20
(7) 5.5
0
qM +(1−α)q41 1+ 18
γ−1 lnε 0
5 160
With ɳth is the Carnot yield calculated with the extreme 14 0
4.5 16 0 18
temperatures of the cycle. 12 0
14 0 0
4
10 0 12 0
3.5 14
80 10 0 0
16
IV. RESULTS AND DISCUSSIONS
0
3
1 2 3 4 5 6 7 8
Cooling water flow rate (l/min)
A. Hot temperature and pressure dependence
(b) Th=500°C
Fig.3 shows the contour of the power produced by the 8 32
0
34 0
GSE as a function of the temperature of the hot source and
28
7.5
0
the load pressure of the GSE. The range of temperature varies 7
30 0
28
from 300°C to 500°C and that of the charging pressure varies

0
6.5
26
from 3 to 8 bars. It should be noted that the same power can
0
30 0
6
be obtained with different conditions. For example a power of 26
28 0
5.5 0
300 W can be obtained when P i = 8 bar and Th = 390°C or
24
0
5
when Pi = 7 bar and Th = 410 ° C. The engine power increases 26 0
with the temperature of the hot source. Raising the hot end 4.5
22
0
24 0
temperature provides additional heat to the working fluid 4

24 0
inside the heater. As a result, the temperature difference 3.5 20 0 22 0
between the two workspaces also increases; which improves 3
1 2 3 4 5 6 7 8
the power produced by the Stirling engine. However, the Cooling water flow rate (l/min)
increase in temperature of the hot end is limited by the heat
Fig. 4. Contour of power as a function of pressure and flow rate of cooling
capacity of the working fluid. water (a)Th=300°C (b)Th=500°C
500
C. Initial charge pressure and speed dependences
32
260
480 34
28
0
Fig.5 shows the power contour for two different hot
24
0
0
Hot end temperature (°C)
460
30
temperatures (Th = 300 ° C and 500 ° C) depending on the

0
22
440
0
rotation speed and the load pressure. The evolution of the

420
power with respect to the speed of rotation reaches an optimal
20
26
28
0
400 30
0
24
0 value. Increasing the speed reduces external conduction loss

0
0
22
380 0
18 26
through the motor walls, reduces the heat exchange time and
0 240 0 28
360 20
0
0 increases the frictional loss within each heat exchanger. If N>
340 220 Ncritical, the heat exchange process occurs very rapidly, the
26
320 16
0 18
0 200 0 residence time of the working fluid in the three heat
24
0 exchangers will become very short. The working fluid does
300
3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8
Initial charge pressure (bar) not have time to absorb all the amount of energy delivered by
Fig. 3. Contour of power versus hot temperature and pressure the hot source. Work fluid leaks also increase; which causes
the reduction of the power of the engine SG. The critical
B. Initial charge pressure and cooling flow dependences
value of the rotational speed (Ncritical) depends on both the
The initial charge pressure is imposed to the engine before charging pressure and the hot source temperature. When P i =
it start-up. It represents the amount of working gas flowing 5 bar and Th = 500°C, the optimal power of 260 W is
within the engine. This mass is proportional to the initial obtained for Ncritique = 360 rpm. However, when Pi = 7.3 bar
charge pressure increase. The amount of thermal energy and Th = 500 °C, the optimal power of 340 W is obtained for
exchanged in the engine is closely related to the working gas Ncritique = 380 rpm.
mass. Fig.4 shows the power contour as a function of the
charge pressure and the cooling water flow rate for two (a) T = 300°C
h
8
different hot temperatures (Th = 300 °C and Th = 500 °C).
210 05
2
21
220
The range of flow varies from 0 to 8 l/min and that of the
195
220
200
5
7
pressure varies from 3 to 8 bar. It is clear that increasing the
5
225
21
rg (b r)
cooling rate slightly improves the power produced. For T h = 22

5 230
6
300 °C and Pi= 3.5bar, the power increases from 80 W to 160
205
0
22
235 23
210905
0
210
W when the flow rate varies from 1 l/min to 7.5 l/min. 24

215
5 22
23
0
23
5 23
Increasing the cooling rate decreases the temperature of the
190
5 230
22
0
compression space and therefore improves the temperature
p
4 230 225
22
205
5 0
difference between hot and cold source; which improves the 22
195
200
220
210
21
5
21
power of the Stirling engine.
5
3
310 320 330 340 350 360 370 380
Speed (rpm)
Copyright - 2019
ISSN 1737-9334 Page 76
realization of a perfect seal (to reduce the leaks of working

(b) T = 400°C
h gas)
8 270
-The increase of hot end temperature leads to an increase of
250
26
0 the thermal exchanged energy. Thus, the increase of Stirling
7
engine brake power. However, the temperature of the hot end
230
240
26 0
0 25 should be moderate. It is limited by the melting temperature
220
6
of the material of the hot heat exchanger
250
210
0
0 24
25
20
5
0
230
240
ACKNOWLEDGMENT
0
23
220
4 0
24
This work was supported by the laboratory LAMIH CNRS
0
21
0
UMR 8201 (University of Valenciennes), the laboratory

20
3 230
310 320 330 340 350 360 370 380 390 400 LESTE (ENIM, Monastir, Tunisia) and the European
Speed (rpm)
Commission within the International Research Staff
(c) T = 500°C
8
h
Exchange Scheme (IRSES) in the 7th Framework Programme
360 340
320
FP7/2014-2017/ under REA grant agreement n°612230.
0
0
28
30
7.5 340
These supports are gratefully acknowledged.
280
7
300
0 0
32
260
26
6.5
0
30
6 0
28 REFERENCES
5.5 280
5 260 260
[1] Hachem H., Gheith R., Aloui F., Ben Nasrallah S. Technological
4.5 challenges and optimization efforts of the Stirling machine: A Review.
4 240 Energy Conversion and Management, Vol. 171, pp. 1365–1387, 2018.
240
3.5 [2] Derviş Erol, Hayri Yaman, Battal Doğan. A review development of
240
rhombic drive mechanism used in the Stirling engines. Renewable and
3
320 340 360 380 400 420 Sustainable Energy Reviews, vol. 78, 2017, p.1044–1067.
Speed (rpm) [3] Hachem H., Gheith R., Aloui F., Ben Nasrallah S. Global numerical
Fig. 5. Contour of power versus pressure and velocity (a)Th=300°C characterization of a γ-Stirling engine considering losses and interaction
(b)Th=400°C et (c) Th=500°C between functioning parameters. Energy Conversion and Management.
Vol.96, 2015, p. 532–543.
[4] Hachem H., Creyx M., Gheith R., Delacourt E., Morin C., Aloui F., Ben
V. CONCLUSIONS Nasrallah S. Comparison based on exergetic analyses of two hot air engines:
a Gamma type Stirling engine and an open Joule cycle Ericsson engine.
The experimental study of the Stirling Engine leads us to Entropy; Vol. 17, 2015, p. 7331-7348.
conclude that: [5] Walker G, Fauvel R, Gustafson R, van Bentham J. Stirling engine heat
- The parameters (speed of rotation, initial charge pressure, pumps. Refrigeration 1982, vol. 5(2), p. 91–97.
[6] Gheith R, Aloui F, Tazerout M, Ben Nasrallah S. Experimental
cooling water flow rate and hot temperature) are dependent investigations of a Gamma Stirling engine. Energy Research, 2012, vol. 36, p.
and their interactions significantly affect the power of the 1175–1182.
Stirling engine. [7] Sowale A., Kolios A.J., Fidalgo B., Somorin T., Parker A., Williams L.,
- The optimal rotation speed of the Stirling engine depends on Collins M., McAdam E., Tyrrel S. Thermodynamic analysis of a gamma type
Stirling engine in an energy recovery system. Energy Conversion and
the temperature of the hot end and the pressure of the load. Management, vol. 165, 2018, p. 528–540.
- An optimum value of the rotation speed must always be [8] Karabulut H, Cinar C. Torque and power characteristics of a helium
respected to ensure the proper functioning of the Stirling charged Stirling engine with a lever controlled displacer driving mechanism.
machine (The increase of the rotation speed has double Renewable Energy, 2010, vol. 35, p.138–143.
[9] Cheng C-H, Yang H-S, Jhou B-Y, Chen Y-C, Wang Y-J. Dynamic
effects. On the one hand, it favors the exchanges of heat by simulation of thermal-lag Stirling engines. Applied Energy 2013, vol. 108,
convection and on the other hand, it increases the losses by p.466–476.
viscous friction through the singularities of the machine). [10] Chen W-L, Yang Y-C, Salazar JL. A CFD parametric study on the
-The increase of initial charge pressure leads to an increase of performance of a low-temperature differential γ-type Stirling engine. Energy
Conversion and Management, 2015, vol. 106, p. 635–643.
working fluid mass, which increases the Stirling engine brake [11] Gheith R, Hachem H, Aloui F, Ben Nasrallah S. Experimental and
power. However, the load pressure is limited on the one hand theoretical investigation of Stirling engine heater: Parametrical optimization.
by the capacity of the motor to withstand the high pressure Energy Conversion and Management, 2015, vol. 105, p.285–293.
(resistance of the materials) and on the other by the
Copyright - 2019
ISSN 1737-9334 Page 77
Low capacity diffusion absorption refrigeration:

Experiments and model assessment
Ikram Saâfi#1, Ahmed Taieb#*2, Yosra Ben Salem#*3, Ahmed Bellagi#4
#
Higher Institute of Technological Studies of Kelibia
Road Oued el Khatf 8090, Kelibia, Tunisia
ikram.saafi@gmail.com
ben-salem-yosra@live.fr
*
Thermodynamic and Thermodynamic Research Unit of Industrial Processes
National School of Engineers of Monastir (ENIM)
Avenue Ibn El Jazzar, 5060 Monastir, Tunisia
hmed.taieb@gmail.com
a.bellagi@enim.rnu.tn
Abstract— In this paper, a low capacity commercial absorption- experimental measurements in terms of cooling capacity and
diffusion refrigeration machine using ammonia/water/hydrogen coefficient of performance are less than 1%. Experimental
as working fluids, was tested under different operating tests on a domestic DAR were performed by Ben Jemaa and
conditions in order to master the operation of the refrigerator al [3]. The refrigerator was modelled in dynamic mode for
and to study its performances. A series of thirty experiments
different electric heater powers input. He has used the black-
were realized which consisted of varying the electrical power
supplied to the generator from 15W to 63W in order to identify box technique for the prediction of the transient behavior of
the optimal operating conditions of the machine and to study the the commercial DAR. Starace et al. [4] developed a
various operating modes. All these experiments are all thermodynamic model of the absorption-diffusion cycle
performed in an air-conditioned room at 26°C. It is further without any hypothesis regarding the purity of the refrigerant
noted that the optimal COP found is 0.15 for a cooling capacity leaving the rectifier. Using this model, he compared the
of 7W. performances of the machine with that of another
thermodynamic model proposed by Zohar et al. [5]. Greater
Keywords— Refrigeration machine; absorption-diffusion; accuracy has been shown in predicting the actual state of the
experimental study; performance. machine. In another study, Starace et al. [6] experimentally
validated their model using a prototype of a bubble pump
I. INTRODUCTION coupled to a home magnetron to reduce the start-up transient
The absorption-diffusion refrigeration machine, the subject of the circuit. In order to validate the model, he varied the
of our study, was invented by Platen and Munters in 1928 [1] thermal power supplied to the heat pump, he tested the
and has been recognized as one of the most encouraging operating conditions of the machine in each element of the
sustainable technologies for cold production. The cycle of the machine.
machine (Figure 1) operates at a constant total pressure, and In this paper, an experimental study is carried out on a
uses ammonia as a refrigerant, water as an absorber and low-power commercial refrigerator with a capacity of 25
hydrogen or helium as a non-absorbable auxiliary inert gas. liters, powered by an electrical resistance and which operates
This inert gas is necessary to reduce the refrigerant partial according to the Platen and Munters cycle. This machine
pressure in the evaporator to allow the evaporation process to involves the ammonia/water as refrigerant/absorbent and
take place in the uniform pressure device. hydrogen, pressure equalizer, as an inert gas. It consists
The main feature of this machine is that it has no moving mainly of a generator, a rectifier, a condenser, an evaporator,
parts, hence its good reliability. The circulation of the a gas exchanger, an absorber, a liquid reservoir and a solution
aqueous ammonia solution is driven by a bubble pump and exchanger. All these elements are made of steel.
that of gas flows between the absorber and the evaporator by
gravity. There are many theoretical and experimental studies II. EXPERIMENTAL APPARATUS
on the analysis of the performance of this machine operated A. Description of the cycle
with different energy sources and using various mixtures of
When the machine is energized, the electrical resistance
fluid work in the literature. Mansouri and al [2] has carried
begins to heat the ammonia-rich solution (1), Figure 1, with a
out experimental investigations of a commercial diffusion
thermal energy supply Qg in the lower part of the bubble
absorption refrigerator (DAR) cycle and developed a detailed
pump. This energy supply is the origin of the evaporation of
steady state simulation model of this cycle using Aspen-Plus.
ammonia, the most volatile compound, and small amounts of
He has found deviations between model predictions and
Copyright - 2019
Page 78
ISSN 1737-9334
water. During this evaporation, small bubbles of vapor form in the GHX and the gaseous mixture (10) rich in ammonia
and gather together to form large bubbles occupying the goes to the absorber.
entire section of the tube. Having a lower density than the Inside the absorber, the ammonia vapor in the flow (10) is
liquid mixture, these big bubbles rise and push the liquid absorbed by the lean solution (14), ammonia / water,
carrying with them a quantity of the liquid solution until the returning from the generator. The gaseous mixture thus
upper part of the generator (2). This happens in a vertical tube becomes lighter at point (12) and rises towards the evaporator.
of small diameter. The pumped liquid is discharged by The liquid solution becomes richer in ammonia and colder at
gravity into the annular space of the coaxial tube of the the outlet of the absorber (15). Then, it goes back to the
generator which is in direct contact with the electrical bubble pump through the solution exchanger. And the cycle
resistance. This liquid is then heated further, thereby causing starts again.
the generation of an additional amount of steam.
B. Measuring devices
To follow the operation of the refrigerator, we began by
equipping it with adequate measuring devices as shown in
Figure 2. Eighteen thermocouples type K (Chromel / alumel)
were fixed by clamps on the outer surfaces tubes at the
desired position and are connected in the other side to a data
acquisition system.
Fig 1 Refrigerant circuit of the refrigerating machine.
The resultant poor solution (13) is gravity-fed to the

absorber via a solution exchanger. All of the steam rises to
the condenser through the rectifier to condense the water it Fig 2 Experimental device and different measurement instruments used.
contains and returns to the annulus of the coaxial tube of the
III. RESULTS
generator to rejoin the ammonia-poor solution to the absorber
(14). The ammonia vapor thus purified (4) liquefies in a We started this study by identifying the minimum input
condenser cooled in ambient air by rejecting a thermal power power required to start the machine and ensure the production
Qc. The liquid obtained in point (5) travels through the heat- of cold. By slightly increasing the input power with each test
exchanger gas, thermally insulated, where it is cooled before and by visualizing the evolution of the temperatures of the
accessing the evaporator in point (7). The gas exchanger and various points of the machine, it was possible to determine
the evaporator consist of two coaxial tubes: the hydrogen-rich the minimum power necessary to operate the machine. This
gas leaving the absorber flows into the central tube and the power is of the order of 22W.
binary mixture consisting of the cold inert gas and the
A. Oscillatory operation of the machine
refrigerant leaving the evaporator in the annular space. The
small diameter tube carrying the liquid refrigerant from the This series of tests was started from the minimum starting
condenser (5) to the evaporator (7) is placed in direct contact power of the machine (22W). By slightly amplifying the
with the outer tube of the GHX and the evaporator. The motive power at each test and by visualizing the evolution of
gaseous mixture (12), ammonia / water / hydrogen, rich in the temperatures of the various points of the machine, it was
hydrogen. By passing through the evaporator, the gaseous easily found that all the measured temperatures are
mixture (9), ammonia / water / hydrogen, richer in hydrogen remarkably affected by the oscillatory operation of the
is cooled further. The partial pressure of the ammonia / water generator. In FIG. 3 it can be seen that the evolution of the
system in the flow (8) at the inlet of the evaporator is greatly temperatures of the generator and of the evaporator (for the
reduced. The evaporation of the liquid ammonia is completed heating powers 22, 24 and 30W) present a stationary

ISSN 1737-9334
oscillating regime of the same frequency. The temperature at

the generator oscillates between 205 and 210 ° C, that at the
25
evaporator fluctuates between -12 and -20 ° C. These three
20 63W
powers do not allow to have a constant temperature in the 61W
15
evaporator. 10
58W
55W
5
TEvap(°C)
TGen 30W TGen 24W
50W
TGen 22W 0
TEvap 30W TEvap 24W
47W
TEvap 22W
-5 45W
220 43W
-10
210
-15
200
190 -20
180
0 1 2 3 4 5 6 7 8 9 10 11
170
160
Temps (h)
T (°C)
30
20 Fig 5 Evolution of evaporator temperatures for heating powers: 43, 45, 47, 50,
10 55, 58, 61 and 63W.
0
-10
-20
IV. CHARACTERIZATION OF THE MACHINE
0 1 2 3 4 5 6 7 8 9 10 11 To characterize the thermal exchanges between the machine
Temps (h) and the external environment, we have adopted the notion of
the overall exchange coefficient, one for the cabin-external
Fig 3 Evolution of generator and evaporator temperatures for heating powers:
environment exchange and the other within the cabin between
22, 24 and 30W.
the contents of the cabin and the evaporator. These exchange
B. Stable operation of the machine coefficients can be deduced from measurements in dynamic
In this part, we continued the series of previous tests by or stationary mode.
increasing the input power gradually and by visualizing the The energy balance of the cabin is written as:
evolution of the temperatures of the various points of the (1)
machine. A total disappearance of all the oscillations is noted
in the profiles of the measured temperatures. In particular, the Où is the internal energy of the cabin and respectively
evolution of the temperatures of the generator and the the heat flow exchanged between the cabin and the outside
evaporator (for the heating powers 43, 45, 47, 50, 55, 58 and environment on one side, and the evaporator and the cabin, on
63W) are shown in FIGS. 4 and 5. the other:
(2)
210 (3)
200
190
The energy balance is written then:
180
63W
170 61W (4)
TGen(°C)
160 58W
50
40
55W
50W
Using the result of equation (4), we can write:
47W
30
45W (
20
43W
10
Where is the average mass heat of the cabin, including
0
0 2 4 6 8 10
its contents..
Temps (h)
By dividing the two members by , the average heat
Fig 4 Evolution of generator temperatures for heating powers: 43, 45, 47, 50, mass of the empty cabin, we obtain:
55, 58, 61 and 63W.
(6)
If we put ; ;
(7)
In the particular case of the empty cabin and therefore
. Equation (7) becomes :
(8)
Copyright - 2019
ISSN 1737-9334 Page 80
Based on the temperature dynamic measurements , , TABLE I

et , equation (9) could be solved for 20 experiments Values of parameters , β and γ
(empty cabin or with different volumes of water in the cabin). n° α (h-1) β (h-1) γ (kg-1)
A nonlinear regression procedure is performed using the
MATHEMATICA ® software to determine the model 1 0,954881 0,4022393 0,7888909
parameters α, β and γ and to predict the temperature of the 2 0,916906 0,3917 0,7888909
water and the cabin.
The model of the water temperature is of the form: : 3 0,916906 0,3917 0,981186
4 0,903291 0,3847 0,981186
Where a, b, c, d, e and f are smoothing parameters identified 5 1,11416 0,496746 0,955986
for each temperature (water and cabin) and for each
experiment (cabin empty or not). 6 1,22393 0,568542 0,955986
The experimental values recorded for the cabin temperature 7 1,22393 0,568542 0,955986
as well as the regression model used are shown in Figure 6
for an empty cabin. We note that the model of regeneration 8 1,05858 0,460507
used faithfully represents the experimental data. 9 1,05858 0,460507
Table 1 also shows the found values of the parameters α, β
10 0,945599 0,442812
and γ as well as their average values for the experiments
carried out. 11 0,9099 0,402664
The knowledge of the mean value of γ allows us to determine 12 0,945599 0,442812
. Once this last parameter has been identified, we can
13 0,9099 0,402664
directly deducet et respectively from the
average values of α and β. 14 1,06664 0,486158
15 1,06664 0,486158
16 0,916664 0,355459
1.2. Machine performance study 17 0,886484 0,399174
The performance of the refrigerator studied is shown in 18 0,983791 0,44676
Figures 7 and 8. As previously described, it is noted that a
minimum power required for the start of the refrigerating 19 0,983791 0,44676
machine is in the order of 36W producing a cooling 20 0,902639 0,349796
capacity of 5W.
As the heat supplied to the generator is progressively 21 1,0579 0,410276
increased, the flow rates of the pumped solution and the 22 1,0579 0,410276
steam increase too. Increasing the flow rate of the vapor
23 1,0579 0,410276
and the refrigerant flow allows to have an increasingly
large cooling capacity. The mean 1,0027179 0,4355317 0,9154445
A larger flow rate of the pumped solution allows to absorb

a larger amount of steam. This explains the growth of the
cooling capacity and consequently the coefficient of
performance. It is also recalled that the increase of the motor
temperature leads to a pumped solution poorer in ammonia. A
pumped solution, at the inlet of the absorber, increasingly
poor with a flow more and more increasing leads to an
efficiency of the absorber more and more increasing and a
partial pressure of the refrigerant lower.
Beyond 43W, the cooling capacity stabilizes and the COP
begins to decrease. This is explained by the fact that the heat
exchange at the evaporator is limited by the temperature
along the evaporator and also by the overall internal exchange
Fig 6 Comparison of the cabin temperature profile (Veau = 0l) with the coefficient UA)int between the evaporator and the inside of the
regression model found : T0= 8.40592, a=0.0192634, b= 0.145793, c= - refrigerator.
12968, d= 14.3739, e= 1.06511 et f = 1.06511
Copyright - 2019
ISSN 1737-9334 Page 81
V. CONCLUSIONS
0,16 In this paper, we conducted an experimental study of a
0,15 low-power commercial refrigerator with a capacity of 25
0,14
liters, powered by an electrical power. We have studied the
0,13
performance of the machine for various operating conditions
0,12
which allowed us to completely characterize this machine. In
COP
dynamic mode, the values of (UA)cab and (UA)evap found are

0,11
respectively WK-1 and WK-1. These values
0,10
allowed us to evaluate the cooling capacity at 7.5W and the
0,09
COP of the machine in the order of 0.15.
0,08
30 35 40 45 50 55 60 65 70 REFERENCES
QGen (W) [1] B.C. Von Platen, C.G. Munters, Refrigerator, US Patent 1 (1928) 685-
764.
Fig 7 Variation of the experimental COP according to the power supplied to [2] R. Mansouri, M. Bourouis and A. Bellagi “Steady state investigations
the generator. of a commercial diffusion-absorption refrigerator: Experimental study
and numerical simulations,”Applied Thermal Engineering, vol 129, pp
The amount of ammonia that can evaporate inside the 725-734, Jan 2018.
evaporator is then limited. The additional amount produced [3] R. Ben Jemaa, R. Mansouri and A. Bellagi “Dynamic testing and
by the increase in power will not be evaporated and will leave modeling of a diffusion absorption refrigeration,”Applied Thermal
Engineering, vol 67, pp 249-258, July 2016
the evaporator in the liquid state. The cooling capacity [4] A. Zohar, M. Jelinek, A. Levy, I. Borde, The influence of diffusion
produced becomes constant since the amount of ammonia absorption refrigeration cycle configuration on the performance,
evaporated remains the same. Also, an additional heating at Applied Thermal Engineering 27 (2007) 2213–2219.
the generator is not necessary beyond these conditions [5] A. Zohar, M. Jelinek, A. Levy and I. Borde, ‘Performance of
Diffusion Absorption Refrigeration Cycle With Organic Working
because it will only further heat the ammonia vapor which Fluids’, International Journal of Refrigeration, Vol. 32, N°6, pp. 1241
will subsequently be cooled in the rectifier and then in the – 1246, 2009.
condenser, which leads only a loss of energy and a decrease [6] N. Ben Ezzine, R. Garma, A. Bellagi, A numerical investigation
in the COP. of a diffusion–absorption refrigeration cycle based on R124–
DMAC mixture for solar cooling, Energy 35 (2010) 1874–1883.
6
QEvap (W)
2
30 35 40 45 50 55 60 65
QGen (W)
Fig 8 Variation of the cooling capacity according to the power supplied to the
generator.
Copyright - 2019
ISSN 1737-9334 Page 82
Estimated study of the production of electricity from

different types of PV in two Algerian sites
Mohamed SALMI#1, Derradji MEDERREG*2, Djamal FRIDJA#1
#
Department of Physics, University of M’sila , Algeria
1
mohamed.salmi@univ-msila.dz
3
frid.djamal@gmail.com
*
Department of Mechanics, University of M’sila , Algeria
2
Derradji.MEDERREG@univ-msila.dz
Abstract— We present an estimation study on photovoltaic systems physicist Auguste Mouchot had proposed the creation of a lot
in Algeria, for the conversion of solar energy into electrical energy. of thermal machines for industrial and agronomic purposes. A
We used a simulation software "RETS Screen". To make this study, century later in 1951 a famous engineer; Maurice Touchais
we chose two areas taken into consideration by the "RETS Screen" had installed the first solar concentrator on the site Bouzareah
software, namely: Algiers and Tamanrasset. These sites have Algiers.
measured solar global radiation data as well as values of the duration
of sunshine for a period of six years (2000-2005).
To contribute to the photovoltaic system development in
The results of estimation of the power supplied and the yield of Algeria and for the conversion of solar energy into electric
the PV system obtained by simulation with the RETScreen one, we have used the RETscreen software to estimate the
software for Algiers and Tamanrasset are respectively 5884 and electric power generated by the photovoltaic (PV) systems
7565 KWh ; 8 and 7.8%. installed in the regions of Algiers and Tamanrasset.
II. DIFFERENT TYPES OF SOLAR CELLS

Keywords- Solar energy - Photovoltaic - Simulation - RETS II.1. Monocrystalline cells: are the first-generation solar cells,
Screen – Algeria. they are developed from a block of crystallized silicon in a
single crystal. Its manufacturing process is long and
demanding in energy; more expensive, it is however more
I. INTRODUCTION effective than polycrystalline silicon. Silicon in its raw state is
The solar energy is the most powerful and the most melted to create a bar. When the silicon cooling is slow and
promising renewable energy. This clean energy is controlled, a single crystal is obtained. A wafer (silicon wafer)
inexhaustible and is quite available for profitable applications. is then cut in the silicon bar. After various treatments (acid
Indeed, the power of solar radiation on the ground is about surface treatment, doping and creation of the P-N junction,
950W/m2. The total amount of solar energy received at the deposition of antireflection layer, installation of collectors),
ground during a week overtakes the energy produced by the the wafer becomes a cell. The cells are round or nearly square
world’s reserves of petrol, coal, gas and Uranium. But in most and, when viewed closely, have a uniform color. They have a
cases, an electric conversion of the solar radiation energy is yield of 12 to 18%, but the production method is laborious [3].
necessary. The photovoltaic electricity is obtained by the II.2. Polycrystalline cells: are produced from a block of
direct transformation of the solar radiation to electricity by crystallized silicon in the form of multiple crystals. Viewed
photovoltaic cells. The photovoltaic electricity production is in closely, you can see the different orientations of the crystals
a significant growth since the last years, to over 700MW[1]. (different tones). They have a yield of 11 to 15%, but their
The surface area of Algeria is over 2 million km2. It cost of production is lower than monocrystalline cells. These
receives what is the equivalent of 300 billion of Tep a year in cells, thanks to their productivity-enhancing potential, have
the form of solar energy. In term of sunshine duration the today imposed themselves. The advantage of these cells over
daily energy received upon a horizontal surface of 1m2 is of monocrystalline silicon is that they produce little cutting waste
the order of 5kWh on almost the totality of the national and require 2-3 times less energy for their manufacture. The
territory. The annual solar sunshine duration overtakes 2000 wafer is sawn into a silicon bar whose forced cooling has
hours and may achieve 3900 in the high plateaus and the created a poly-crystalline structure. Estimated life: 30 years. A
Sahara, which represents a high average compared with other crystal is a solid with polygonal facades, more or less bright,
countries in the world[2]. with a regular and periodic structure, formed of an ordered
The idea of using solar energy in Algeria is not new, its stack of a large number of atoms, molecules or ions [4].
existence dates back at least to 1870 when the famous
Copyright - 2019
Page 83
ISSN 1737-9334
II.3. Amorphous cells: have a much lower cost of production, the worksheets.
but unfortunately their yield is only 6-8% currently. This - Note: To access the full list of suppliers contained in the
technology makes it possible to use very thin layers of silicon RETScreen Online Product Database, and their contact
which are applied to glass, flexible plastic or metal by a information, the user must select the "All" option from the
vacuum vaporization process. The performance of these "PV Module Type" drop-down list. From the dialog box.
panels is less good than that of polycrystalline or
monocrystalline technologies. However, amorphous silicon III.3. Energy model
makes it possible to produce large-area panels at low cost In this section of the RETScreen Software, there are
using little raw material. Note: In chemistry, an amorphous three parts:
compound is a compound in which the atoms do not respect - Characteristics of the site which contains the name and
any order at medium or long distance, which distinguishes it place of the project and meteorological station.
from crystallized compounds. Glasses are amorphous Parameters of the system that contains the PV field and the
compounds [4]. conditioning of the energy.
Annual energy production which contains the output and
We represent in Figure 1, an example of the I-V the power supplied.
characteristic of a solar cell under the influence of illumination
[5]. III.4. Overall efficiency of the PV system
The model allows the calculation of the overall
FIGURE 1: I-V CHARACTERISTIC OF A SOLAR CELL efficiency of the PV system, in%. This value is the amount
of renewable energy supplied relative to the solar energy
captured by the module. Typically, this value can vary
from 3 to 13%, depending on the type of module, the
performance of the energy conditioning equipment and the
quality of the solar radiation.
III.5.Monthly data
In this section we have inserted in the database of the
software the following data:
Average monthly temperature.
Average monthly global solar radiation.
IV. APPLICATION TO ALGERIAN SITES

In application of the above study for Algerian case, we
III. RETSCREEN SOFTWARE have chosen four cities which represent four different
III.1.Description regions (North, South). They are respectively: Algiers and
The RETScreen software [6] is a program which can be Tamanrasset. These areas are prototypes of the different
easily used to estimate the energy production and the cost of climates of Algeria. The measured values of temperatures
PV project for three basic applications: inside and outside and solar radiation are obtained from the National
electric network, and in water pumping. meteorological Office - Algeria. The geographic
For the evaluation of PV projects, the RETScreen contains coordinates of those sites are given in the table below:
three worksheets:
-Energy model,
-Solar resource evaluation and calculation of the charge, TABLE I. GEOGRAPHICAL COORDINATES OF SITES
-Cost analysis. Location Latitude Altitude (m) Longitude (deg)
(deg)(N)
Algiers 36,43 450 2.83 E
III.2. Product data Tamanrasset 22.47 1378 5.31 E
Some of the model requirements for product data are
The average values of the measured temperature and
presented in the database of these products. These provide
global solar radiation for the different sites are given in the
information on the equipment associated with the project.
figures 2 and 3 below:
- The product database sort routine starts using the "PV
Module Type" selected in the Energy Model worksheet.
FIGURE 2 : GLOBAL SOLAR RADIATION
From the dialog box we select the site chosen for the study,
followed by the supplier, the model and the number of PV
modules. The data can be pasted from the dialog box into
Copyright - 2019
ISSN 1737-9334 Page 84
a)- PV module type: Mono-Si

TABLE 2: OBTAINED RESULTS USING RETSCREEN SOFTWARE (Mono-si)
Energy model Alg. Tam.
Nominal efficiency of the PV module (%) 10.3 11,7
Nominal temperature of the operating cells(C°) 45 45
FIGURE 3 : TEMPERATURE OF SITES Annual solar radiation (Mwh/m2) 6,01 8,65
Annual average temperature (°C) 19,6 22,8
Average efficiency of the inverter (%) 90 90
Various losses in PV field (%) 10 10
Absorption rate of energy (%) 95 95,0
Global efficiency of PV system (%) 8,0 7,8
30 Tamanrasset Captured energy (MWh) 2 ,932 3,891
28
Algiers
26 Energy supplied(MWh) 2,506 3,327

24
22
20
G (MJ m-2 )
18 b)- PV module type: Poly-Si

16
14 TABLE 3: OBTAINED RESULTS USING RETSCREEN SOFTWARE (Poly-si)
12
10 Energy model Alg. Tam.
8
6 Nominal efficiency of the PV module (%) 10 11
0 2 4 6 8 10 12
Mois Nominal temperature of the operating cells(C°) 45 45

Annual solar radiation (Mwh/m2) 5,99 8,65
Various losses in PV field (%) 10 10
Absorption rate of energy (%) 95 95
Global efficiency of PV system (%) 7,5 7.5
Tamanrasset Captured energy (MWh) 6,882 8,848
30
Algiers
Energy supplied(MWh) 5,884 7,565
25
Temperature (C°)
20
b)- PV module type: a-si

15
10
TABLE 4: OBTAINED RESULTS USING RETSCREEN SOFTWARE (a-si )
5
Energy model Alg. Tam.
0
Nominal efficiency of the PV module (%) 9.2 10
1 2 3 4 5 6 7 8 9 10 11 12
Mois Nominal temperature of the operating cells(C°) 45 45

Annual solar radiation (Mwh/m2) 5,99 8,65
V. RESULTS Various losses in PV field (%) 10 10
5.1.Production of energy Absorption rate of energy (%) 95 95
Tables 2, 3 and 4 show the results obtained using the Global efficiency of PV system (%) 7,4 3,7
RETScreen software from the sites selected in our study using
PV Module Manufacturer - ABC Inc. Captured energy (MWh) 1,943 2,647
Energy supplied(MWh) 1,661 2,263
The most important of the result for us is the annual value

of the energy supplied, the maximum value of the electric

ISSN 1737-9334
energy supplied (fig.5): (7,565MWh) was produced by the applications photovoltaïques. Thèse de doctorat, L’institut
mono module if at Tamanrasset and the minimum value national des sciences appliquées de Lyon .
(1,661MWh) was produced by the a-si module for the Algiers [4] Zhao .J, Wang .A., Green. M. A. (1999). High Efficiency
site. PERL Silicon Solar Cells on FZ, MCZ and CZ Substrates,
FIGURE 4: PRODUCED ANNUAL AVERAGE ENERGY GIVEN BY Tech. Digest 11th Internat. Photovolt. Science and
Engineering Conf. Sapporo, Japan. 557 .
THE THREE PV TYPES
[5] Benmoussa. W.C., Amara. S, Zerga. (2007). Etude
comparative des modèles de la caractéristique
courant-tension d’une cellule solaire au silicium
monocristallin. Revue des Energies Renouvelables ICRESD-
07 Tlemcen 301 – 306.
[6] Leng G. J, Ziegier U. T, Meioche. N, Parakh . D. S,
Sheriff. F, Bourque .K, Poirier. J, T.Anderson, Richard.A.
(2006). Logiciel RETScreen, Ministre de Ressources
naturelles Canada .
VI. CONCLUSION
We have presented the fundamentals of the principle of
the photovoltaic effect and the various basic constituents of a
conventional photovoltaic cell, which make it possible to
calculate the efficiency of a solar cell from the current-voltage
characteristic and the different constituents of a conventional
photovoltaic cell.
A brief description has been made of the RETScreen
software used in estimating the efficiency of a photovoltaic
system. We chose two typical cities representing the north and
the south of the Algerian territory. These cities are
respectively: Algiers and Tamanrasset.
The results of estimation of the power supplied and the
yield of the PV system obtained by simulation with the
software RETScreen for Tamanrasset and Algiers are
respectively: 7565, 5884 KWh by the type PV module (Mono-
si) and the best efficiency is 8 and 7.8%.
The performance of photovoltaic systems is strongly affected
by the incident radiation and temperature.
REFERENCES
[1] Kaldellis .C. J. K.( 2004). Optimum techno-economic
energy autonomous photovoltaic solution for remote
consumers throughout Greece, ” Energ. Convers. Mgmt,
vol45, N°17, pp. 2745-2760.
[2] Salmi. M, Chegaar. M, Mialhe. P. ( 2011). A Collection of
Models for the Estimation of Global
Solar Radiation in Algeria, ”.Energy Sources, Part B:
Economics, Planning, and Policy, vol. 6, pp.187–191.
[3] Abdo. F. (2007). Croissance de couches minces de silicium
par epitaxie en phase liquide à basse temperature pour
Copyright - 2019
ISSN 1737-9334 Page 86
Impact of the secondary phase ZnS on CZTS

performance solar cells
Naïma.Touafek#*1, Ramdane. Mahamdi*2, Chahrazed. Dridi§3
#
Higher National School of Biotechnology “Toufik Khaznadar” (ENSB), Constantine, Algeria
ntouafek@yahoo.fr
*
LEA, department of electronics, University of Batna2, Algeria
ra_mahamdi@yahoo.fr
§
LPMS ,Mentouri brothers,Constantine1 University ,Algeria
chazed2011@gmail.com
Abstract— In the present study, ultra-thin layer ZnS is inserted interface properties, while ZnS phase can block current flow
in the structure of standard solar cell between CdS buffer layer and introduce dead areas [10]. According to the recent
and CZTS absorber layer to represent the Second Phase (SP) studies, ZnS is the main secondary phase in the Cu2ZnSnS4
often forms on the top of CZTS. The impact of this layer on the layer obtained from a sulphurised Zn/CuSn metallic stack.
performance of CZTS solar cells is illustrated by the diverse
The excess Zn trends to diffuse from back contact region to
results obtained by simulation using SCAPS-1D. The formation
of ZnS on the CZTS surface has harmful effects on the solar top surface of Cu2ZnSnS4 layer accumulating in the form of
cells parameters where the conversion efficiency (η) decreases ZnS [11]. The solar cell with a higher Zn concentration shows
by 2%. When varying thickness of ZnS SP of 2% to 20% a large quantity of isolated ZnS grains at Cu2ZnSnS4 top
corresponding of the total absorber layer the efficiency surface which is close to CdS/Cu2ZnSnS4 hetero-junction
decreases by about 0.65%. The ZnS SP can deteriorate the interface [12], the ZnS phase in CZTS is predicted to be
photovoltaic power conversion efficiency from 10.2% to levels of resistive barriers for carriers, which is generally deleterious
8.8%, depending on the material buffer layer and the thickness to solar cell performance [13,14].
of ZnS SP. Using the ZnS as buffer layer mitigates harmful
effects of the ZnS secondary phase forms on the top of CZTS
The purpose of this work is to study, using SCAPS-1D [15]
absorber layer.
simulation package, the impact of the secondary phase ZnS
Keywords— CZTS; Solar cell; Secondary Phases; CdS; ZnS;
on the photovoltaic output parameters of CZTS solar cells.
SCAPS-1D.
We examine the influence of ZnS layer thickness, the doping
concentration as well as the effect of ZnS as buffer layer in
I. INTRODUCTION the presence of the ZnS second phase.
Absorber kesterite material CuZnSnS4 (CZTS) has been II. DEVICE STRUCTURE AND SIMULATION DETAILS
recently attracted for fabrication of thin film solar cells due to
earth-abundant, low-cost, non-toxic elements, tunable band The standard devices based CZTS consist of a thin-film
gap energy (from 1.48 to 1.63 eV) closer to the optimum multilayer deposited on soda lime glass (SLG),
band-gap value of 1.45 eV for a single junction solar cell [1], Mo/MoS2/CZTS/ CdS/i-ZnO/ZnO:Al. A back metal electrode
economic process ability like spray deposition [2,3], sol-gel Molybdenum (Mo), MoS2 formed between absorber and back
and spin coating [4,5],and exceptionally high absorption contact; a p-CZTS absorber layer, an n-type buffer layer,
coefficient (over 104 cm-1) [6]. But, despite the advantages typically CdS to create the junction of the solar cell, a doped
owned by this material, the solar cells based on the CZTS ZnO layer (TCO), and ZnO:Al transparent front contact.
exhibit relatively weak conversion efficiencies, CZTSSe solar Metallic Ni/Al contact grids complete the cell.
cell with efficiency of 12.6% [6] and only 9.1% of the CZTS In this study, we insert a thin layer between the buffer
solar cell [7] due to our understanding of CZTS based solar layer CdS and the absorber CZTS, Fig.1, to represent
cell which is relatively limited compared to other cells, like secondary phase ZnS generally formed on the top of absorber
CIGS, CdTe and Silicon thin film. The CZTS cells are still layer [16].
suffering from low electronic properties mainly due to the Numerical simulation is an effective way to test and
high carrier recombination both in the bulk and in the predict the thin film solar cell performance [15]. In this work,
interface between absorber (CZTS) and the buffer layers we have used a simulation program called Solar Cell
where atomic arrangement is disturbed severely, add to the Capacitance Simulation in 1 Dimension (SCAPS-1D) to
secondary phases of ZnS, SnS and Cu2SnS3 appear in CZTS predict the changes to CZTS based solar cell performance
films, at both the front surface of the CZTS films and the that are introduced by the presence of the secondary phase
back contact interface [8, 9]. These secondary phases can be ZnS at the top of CZTS absorber.
detrimental to the device performance, as the SnS phase has
lower bandgap than CZTS phase, and could modify the
Copyright - 2019
ISSN 1737-9334 Page 87
Fig.1 Schematic structure of CZTS based thin-film solar cells (layer

thicknesses not to scale)
This software tool is a one dimensional solar cell device

simulator, developed at the University of Gent [17], allows
the definition of thin-film solar cell devices stacks of layers
with a large set of parameters and solves for each point the
fundamental solar cell equations: the Poisson equation and
the continuity equations for electrons and holes. Definable Fig.2 Comparison between the (J-V) curves for the simulated and the
parameters include the thickness, doping, defect and reported experimental data [21]
interface–state densities and cross-sections, optical absorption
coefficient, band-gap and electron affinity. The Shockley- III. RESULTS AND DISCUSSION
Read-Hall (SRH) model is used to describe the recombination The current-voltage (J-V) results from simulation using the
currents in deep bulk levels. All the bulk defects layers are parameters given in table I, with and without ZnS Secondary
positioned at the mid band gap [17]. For the simulation under Phase (SP) are compared with measurement data from [18] in
illumination the AM1.5 standard spectrum is used and the cell Fig.2. The results show that the measured JV curve is very
operating temperature is set at 300 K. The series and shunt well reproduced by the parameters model, with SP, which
resistances are taken in to account in this study with values validates our set of parameters as a baseline for simulating the
respectively fixed at 0.96 and 670 Ω.cm-2 [18]. impact of the SP parameters on solar cell performance. The
JV parameters from simulations and measurements are
TABLE I
Input Parameter Values For The Simulation Of CZTS Solar Cells With
displayed in table II.
SCAPS-1D.
Layer properties
CZTS CdS ZnS MoS2 i:ZnO
W(nm) 700 var Var 100 200
Eg (eV) 1.50 2.4 3.7 1.6 3.3
χ (eV) 4.1 4.215 4.5 4.2 4.4
ε/ε0 7 10 10 13.6 9
Nc (cm-3) 2.2*1018 2.2*1018 1.8*1018 2.2*1018 2.2*1
018
Nv (cm-3) 1.8*1019 9.1*1018 1.8*1019 1.8*1019 1.8*1
019
νn (cm/s) 1*107 1*107 1*107 1*107 1*107
νp (cm/s) 1*107 1*107 1*107 1*107 1*107
µn 60 100 100 100 100
(cm2/Vs)
µp 20 25 25 25 25
(cm2/Vs)
Doping Var (A) 1*1017(D) Var (D) 1*1016 (A) 1*1018
(cm-3)
TABLE II
Measured and Simulated Solar Cell J-V Parameters.
Jsc
Voc (mV) (mA/cm2) FF (%) η (%)
Simulation With ZnS 0.7265 20.2895 60.07 8.85
Simulation W/o ZnS 0.7322 22.2277 61.18 9.96
Fig.3 Dependence of FF and η, Jsc and Voc on the thickness of the secondary
Experimental [18 ] 0.683 20.7 62.5 8.8
phase ZnS.
Copyright - 2019
Page 88
ISSN 1737-9334
A. Modeling with various ZnS thickness C. Material buffer layer effect

One important parameters having a great impact on the Generally, CZTS solar cell is used with CdS buffer, which
solar cells performance, is the thickness of layers constituting has many advantages such as large band gap and the carrier
the device. In this part, we have investigated the effects of density. Otherwise, the presence of Cadmium, a toxic
the ZnS SP layer thickness on the device electrical parameters material, is a disadvantage. To remedy the toxic CdS problem,
as termes of efficiency (ɳ), open-circuit voltage (Voc), short- the researchers try to replace the CdS by other buffer layers
circuit current density (Jsc), and fill factor (FF). This is shown such as ZnS. In this section, as the second phase on the top of
in Figure 3. The other layers properties are kept constant and CZTS layer is ZnS we proposed to use the same type of
the doping of ZnS layer is fixed at 1016 cm-3, while varying material as a buffer layer. We try to compare the calculated
the ratio between ZnS thickness and absorber layer CZTS results with those obtained with conventional CdS buffer
layer.
one’s (WZnS/WCZTS), the total thickness of absorber and ZnS
SP is kept constant to 700 nm. In this simulation the thickness of buffer layer has been
We remark at figure 3 that the persence of utra-thin ZnS changed from 20nm to 100 nm and simulation result is
SP layer between CdS and CZTS affect severly the illustrated in figure 5. It can be seen that, regardless of the
performance of the solar cell where we noted a drop of Voc, material constituting the buffer layer, all cell parameters,
Jsc and FF which leads to the drop of efficiency by about 2%. except for fill factor, decrease with increasing thickness buffer
Increase further the rotio WZnS/WCZTS , the Jsc varies from layer. For both CdS and ZnS buffer layer, the drop of
22.185 mA/cm3 for WZnS/WCZTS= 2% to 19.8 mA/cm3 for parameters is more pronounced for CdS at the presence of
WZnS/WCZTS= 20% . That is a loss of 2.4 mA/cm3, which is secondary phase.
not negligible, corresponding to the loss in the absorption of It shows that the thickness increase of both CdS and ZnS
the photons in the space charge region occurs in the ZnS layer as buffer layer, from 20 nm to 100 nm decreases the efficiency
due to the high band gap of this material (E g=3.7 eV). of the cell in CdS as buffer layer by about 0.7 %, however for
Although increase of FF and Voc, according to the thickness, ZnS as buffer layer the efficiency drop by about 0.22 %,
the conversion efficiency drops drastically from 9.23 % to which is due to reduced transmission of light to the absorber
levels of 8.58%. caused by the lower band gap of CdS compared to the ZnS
ones. We noted a diminution of about 1.6 mA/cm2 for CdS
B. Modeling with various ZnS doping versus only 0.8 mA/cm2 for ZnS ones.
Figure 4 shows open circuit voltage (Voc), short circuit- However, the ZnS as buffer layer has shown the best
current density (Jsc), fill factor (FF) and conversion efficiency performance than the CdS ones, we note an augmentation of
(ɳ) as a function of donor concentration ZnS secondary phase. efficiency from 0.86% to 1.16 % when we pass the CdS buffer
All parameters increase slightly with doping ZnS layer to the ZnS one. So ZnS could be an alternative buffer to
concentration. This mainly can be explained by the reduction the CdS in the production of CZTS solar cells.
of the width space charge region in the ZnS second phase due
to the augmentation of the carrier concentration. Which
increases the number photons absorbed in the absorber layer. Fig.5 Dependence of FF, η, Jsc and Voc on the thickness of buffer layer
IV. CONCLUSIONS
In the present work, the effect of the secondary phase ZnS,
present on the top of CZTS layer, on the performance of solar
cell is investigated numerically by using one dimensional
SCAPS-1D computer software. The results demonstrate that
ZnS second phase decreases the Voc, Jsc, FF and η of solar
cells. The high carriers’ concentration ZnS secondary phase,
enhance slightly the all performance solar cells, indicating a
very little dependence on the doping concentration.
We have shown that the impact of the secondary phase
ZnS deteriorate the device efficiency from 11% to levels of
8.58% depending on the thickness of ZnS SP. Independently
to the material constituting the buffer layer, the efficiency
decreases sharply, especially for CdS buffer layer. In the
presence of de secondary phase ZnS on the top of CZTS the
ZnS as buffer layer has shown the best performance than the
CdS ones. ZnS could be an alternative buffer to the CdS in
the production of CZTS solar cells.
Fig.4 Dependence of (a) FF and η , (b) Jsc and Voc on the doping
concentration of the second phase ZnS.

ISSN 1737-9334
[3] G. Grinciene, and all,“Spray pyrolysis approach to CZTSSe thin

films. Influence of solvents on film characteristics” IOP
Publishing Ltd. 2018; vol .33, pp
[4] B. Patel and V. Gohe, “Optimization of sol–gel spin-coated
Cu2ZnSnS4 (CZTS)thin-film control parameters by RSM
method to enhance the solar cell performance”. Electronic
materials. 2018.
[5] S. Akhanda, R. Matin, M. S. Bashar, M. Sultana, A.
Kowsar, M. Rahaman and Z. H. Mahmood, “Effect of
annealing atmosphere on structural and optical properties of
CZTS thin films prepared by Spin-coating”. BaVngladesh
Journal of Scientific and Industrial Research. 2018; 53(1):13-
20.
[6] Wang, W., Winkler, M. T., Gunawan, O., Gokmen, T., Todorov,
T. K., Zhu, Y., Mitzi, D.B. Adv.Energy Mater., 2014; 4:
1301465
[7] Green M.A, Emery K Hishikawa Y, Warta W and Dunlop E D
2015 Prog.Photovoltaics 23805
[8] A.Kumara, P. Ranjana, A. D. Thakura “Secondary Phases in
CZTS Thin Films Grown Using Direct Liquid Coating
Approach Materials Today: Proceedings. 2018 vol.5, pp.99–103.
[9] PROGRESS IN PHOTOVOLTAICS: RESEARCH &
APPLICATIONS Prog. Photovolt: Res. Appl. (2011)Published
online in Wiley Online Library (wileyonlinelibrary.com). DOI:
10.1002/pip.1174
[10] Yi Ren, Jonathan J. S. Scragg, Christopher Frisk, Jes K. Larsen,
Shu-Yi Li, and Charlotte Platzer-Bj€orkman “Influence of the
Cu2ZnSnS4 absorber thickness on thin film solar cells Phys.
Status Solidi A 212, No. 12, 2889–2896 (2015) / DOI
10.1002/pssa.201532311
[11] L. Wei, C. Jian, Y. Chang, L. Fangyang, H. Xiaojing,
Nanotechnology 25 (2014) 195701.
[12] Wei Li, Jian Chen, Chang Yan, Xiaojing Hao . “The effect of
ZnS segregation on Zn-rich CZTS thin film solar cells”. Journal
of Alloys and Compounds 632 ,2015, pp.178–184.
[13] Wujisiguleng Bao and Masaya Ichimura, “Influence of
Secondary Phases in Kesterite-Cu2ZnSnS4 Absorber Material
Based on the First Principles Calculation”, Hindawi Publishing
Corporation International Journal of Photoenergy. 2015, Article
ID 592079, 6 pages http://dx.doi.org/10.1155/2015/592079
[14] Katy Hartman1, Bonna K. Newman1*, J.L. Johnson2, Hui Du3,
P.A. Fernandes4,5, Vardaan Chawla6, Trudy Bolin7, Bruce M.
Clemens6, Antonio F. da Cunha4, Glenn Teeter3, Michael A.
Scarpulla2, and Tonio Buonassisi1, “ Detection of ZnS phases
ACKNOWLEDGMENT in CZTS thin-films by exafs”, 978-1-4244-9965-6/11/$26.00 ©
[15] J. Petterson, C. Platzer-bjorkman, U. Zimmermann, and M.
We acknowledge the use of SCAPS-1D program Edoff. “Baseline model of graded-absorber Cu(In,Ga)Se2 solar
developed by Marc Burgelman and colleagues at the cells applied to cells with Zn1-xMgxO buffer layers”. Thin Solid
University of Gent in all the simulation reported in the paper. Films. vol. 519, pp.7476-7480, 2011.
. [16] Woo K, Kim Y,Yang W,Kim K, Kim I, Oh Y, Kim J.Y and
Moon J. “Band-gap-graded Cu2ZnSn(S1-x,Sex)4 Solar Cells
Fabricated by an Ethznol-based, Particulate Precursor Ink
REFERENCES Route”.Scintific Reports. 2013.
[17] Metzger, Wyatt K. “The potential and device physics of
[1] D.B. Mitzi, O. Gunawan, TK. Todorov, K. Wang, S. Guha. interdigitated thin-film solar cells”. Journal of Applied Physics.
“The path towards a high-performance solution processed 2008,vol. 103, No.9, pp.094515,.
kesterite solar cell”. Solar Energy Materials and Solar Cells. [18] C.yan, K. Sun, J. Huang, S. Johnston, F. Liu, B. P. Veettil, K.
2011; 95: 1421–1436. Sun, A. Pu, F. Zhou, J.A. Stride, M.A. Green, and X. Hao,
[2] U. Chalapathi, B. Poornaprakash, S. H. Park, “Effect of “Beyond 11% Efficient Sulfide KesteriteCu2ZnxCd1-xSnS4 Solar
Substrate Surface on the Growth of Spray Deposited Cell: Effect of Cadmium Alloying ”Energy Lett. 2017; 2, 930-
Cu2ZnSnS4 thin films”. Chalcogenide Letters.2018; Vol. 15, No. 936.
10 , p. 475 -481

ISSN 1737-9334
Optical absorption of new n-doped GaN0.38yAs1-

1.38ySby/GaAs QWs suitable for the design of 0.8 -
1 eV solar cells
C.Bilel#1, A. Rebey*
#
Physics Department, College of Science, Jouf University, P.O. Box: 2014, Sakaka, Saudi Arabia
chakroun@ju.edu.sa
*
Qassim University, College of Science, Kingdom of Saudi Arabie
Abstract— We investigated the optical absorption of n-doped

GaN0.38yAs1-1.38ySby/GaAs quantum wells (QWs) using a self-
) [16] and ( , ) [17].
consistent calculation combined with the 16-band anti-crossing The energy 0.95 eV was obtained for the couples
model. GaNAsSb material seems to be a potential alternative for ( , ) [5] and ( ,
GaAs-based photodetectors operating at 0.8 - 1 eV energy range.
The increase of the well width from 4 to 10 nm in GaN0.38yAs1- ) [18]. Moreover, the subbands and inter-band
1.38ySby/GaAs QWs caused a red-shift of the fundamental transition of un-doped GaAs/GaNxAs1-x-ySby/GaAs QWs were
transition energy accompanied with a significant decrease of calculated using the envelope function approximation [19].
the absorption coefficient. The wavelength emission of the The study of n doping and Stark effects on the band structure
studied QWs achieved the values 0.8 and 0.95 eV for specific and inter-band transition of GaAs/GaNxAs1-x-ySby/GaAs single
values of Sb composition. A rise of doping density induced an quantum well (SQW) can be useful to enhance the optical
increase of . The Stark effect on energy and was also
absorption process at the energies 0.8 and 0.95 eV.
discussed. Additionally, was adjusted in order to obtain the
energies 0.8 and 0.95 eV of the electrically polarized The aim of this work is to investigate the n doping effect
GaNAsSb/GaAs QWs. on the electronic band structure and optical absorption of
Keywords— GaNAsSb/GaAs QWs; Absorption coefficient; n GaAs/GaNxAs1-x-ySby/GaAs SQW. The increase of the well
doping; Stark effect; energy range 0.8 – 1 eV. width from 4 to 10 nm causes a red-shift of the fundamental
transition energy as well as a decrease of the absorption
I. INTRODUCTION coefficient. Specific values of Sb composition give rise to the
Over the past few years, GaNAsSb material has considered energies emissions 0.8 and 0.95 eV of un-doped SQW.
to be an alternative candidate for GaAs-based 0.8 - 1 eV Owing to doping effect, the absorption process is enhanced.
photodetectors [1,2] and photovoltaic cells [3,4]. These The Stark effect on the optical transition and absorption was
materials show a rapid band gap reduction covering a large also examined. Sb composition was adjusted in order to
infrared range [5,6]. The dependence of the band gap energy obtain electrically polarized GaAs/GaNxAs1-x-ySby/GaAs
of GaNxAs1-x alloys on the N composition can be theoretically SQW operating exactly at 0.8 and 0.95eV.
exhibited by using the conduction band anticrossing model
(C-BAC). The rate of band gap reduction for this material The (16×16) Hamiltonian can describe the electronic band
was found to be around 180 meV/%N [7-9]. For GaSbyAs1-y, structure of bulk GaNxAs1-x-ySby [20,21]:
the Sb composition dependence of the band gap energy can
be described by the valence band anticrossing model (V-
BAC). This alloy displays 16 meV/%Sb as a rate of the band
gap reduction [10,11]. The combination of V-BAC and C-
BAC models offers the possibility of examining the electronic (1a)
band structure of GaNxAs1-x-ySby quaternary alloy [12,13]. where is the modified k.p matrix by adding
The x and y-dependence of the band gap energy of GaN xAs1-x- ∆ECBM, ∆EVBM and ∆Eso (bands edges offsets between
ySby can be modeled in a large composition range
GaAs and GaSb) to the diagonal elements [21]. )
( and ) [14]. The matrix is formed by the energies levels of N atoms (EN) and
simultaneous incorporation of N and Sb atoms affects the Sb atoms (ESb and spin-orbit splitting level ESb-so):
conduction and valence bands leading to a rapid reduction of
band gap energy which reached a minimum value of 0.5 eV
(i.e. 2.48 µm). Experimentally, the optical measurements of
the elaborated GaNxAs1-x-ySby/GaAs structures show that the
value of band gap energy is around 0.80 eV (i.e. 1.55 µm) for
the couples ( , ) [15], ( ,

ISSN 1737-9334
CSb
Parameters
(eV) (eV) (eV) (eV)
GaNAsSb 1.05 0.5 -0.1 0.09

[10] [10] [10] [10]
For GaNxAs1-x-ySby/GaAs QWs, subbands and confined

(1b) potential can be computed using the envelope function
approximation [23,24]. The total Hamiltonian is given by the
matrix is formed by the term
following expression: , where
describing the coupling between Sb
isoelectronic localized states and extended valence band (VB):
is the Hamiltonian of system in a direction
parallel to the GaNAsSb-GaAs interface. ( for
electron and h for heavy hole) is the carrier effective mass in
-direction. is the
Hamiltonian of the system in quantizing z-direction where
is the total potential energy. For un-doped
(1c)
CSb is a fitting parameter taken equal to 1.05 eV [10]. GaNAsSb/GaAs QWs, is restricted to the band
Similar to Sb levels interactions with VB states, the matrix discontinuity between the GaAs barrier and
is composed of the term and GaNAsSb well. For n doped GaNAsSb/GaAs QWs, the
reveals the coupling between the N localized states and the coupled Schrödinger-Poisson equations are written as [25,26]:
extended conduction band (CB) states. The matrix
can be written as:
(2a)
(1d)
CN is a fitting parameter equal to 2.7 eV [22]. The BAC
parameters used in this work are reported in TABLE I.
(2b)
GaNxAs1-x-ySby material can be lattice-matched to GaAs
with a concentration ratio [22]. and are, respectively, the eigenvalues and
TABLE I eigenvectors of . The envelope wavefunctions
satisfy the boundary condition at the interfaces and
A Summary of BAC parameters used in this work. EN represents the N
localized level. ESb and ESb-so are Sb localized levels. CN and CSb are . and are the Hartree and
fitting parameters. , and are conduction, exchange-correlation potentials respectively. is
valence and spin-orbit splitting band edges offsets, respectively. obtained by solving the Poisson equation discretized using the
finite differences method (FDM) taking into account the
boundary conditions. The exchange-correlation potential
ESb- is induced by the many-body effects [27]. is
EN CN ESb
Parameters so
(eV) (eV) (eV) the potential induced by the applied electric field. and
(eV)
are the donor and confined electron density
GaNAsSb 1.65 2.7 -1.0 -1.6 respectively. e is the absolute value of the electron charge and
is the local dielectric constant of the medium. The
[10,20] [10,20] [10] [10]
equations (2a) and (2b) were discretized along the
confinement z-direction using the FDM.
Copyright - 2019
Page 92
ISSN 1737-9334
To solve the coupled Schrödinger-Poisson equations, an

GaAs/GaN0.38yAs1-1.38ySby/GaAs SQW
iterative procedure was used. More calculation details of
subbands, confined electron density, Fermi level, absorption
1.6 e1
coefficient and quantum confinement Stark effect are 10 nm
Transition Energy, T e1-h1 (eV)

reported in references [26,28]. Te1-h1 8 nm
1.4 h1 6 nm
II. RESULTS AND DISCUSSION 4 nm
We have theoretically examined the electronic band 1.2

structure of un-doped GaNxAs1-x-ySby/GaAs SQW. In fact,
quantum well GaNxAs1-x-ySby layer is lattice-matched to 1.0 1.3 µm
barrier GaAs using the ratio [22]. Sb
1.55 µm
introduces two localized levels and situated at 0.8
1.0 and 1.6 eV below the GaAs VB edge, respectively [10]. N
localized level is positioned at 0.23 eV above the GaAs 0.6
CB edge [22]. According to the QWs confinement conditions,
the well width domain was optimized basing on the 0.04 0.08 0.12 0.16 0.20
following expressions: Sb composition, y
a.
(3a) Fig. 1. Dependence of the transition energy of GaAs/GaN0.38yAs1-
1.38ySb y/GaAs SQW on Sb composition for different values of the well

width Lw. The band diagram of subbands is presented in Inset figure.
(3b)
The variation of the absorption coefficient of
GaAs/GaN0.38yAs1-1.38 ySby/GaAs SQW as a function of
where and are the electron effective mass of well
GaNAsSb and barrier GaAs, respectively. the well width is shown in Fig. (2a and 2b). For
and are the De , can be equal to 1.3 µm and 1.55 µm
Broglie thermal wavelength and the thermal energy, for and , respectively. The
respectively. For GaNAsSb/GaAs QWs QW, and absorption coefficient magnitude is around
are equal to 3.6 and 11.3 nm, respectively. As shown and , respectively. Then, the
in the inset of Fig. 1, the fundamental inter-band transition
increase of from 4 to 10 nm induces a significant red-shift
is between the electron subband e1 and the heavy
hole subband h1. Fig. 1 shows the dependence of the of energy. To correct this shift, we adjusted the
transition energy on the Sb composition (i.e. N for each used value of . The adjusted is equal to 10.8
composition ) for varying from 4 to 10 %, 9.7 % and 9.1 % for , 8 nm and 10 nm,
nm. The increase of induces a significant decrease of respectively. Likewise, the rise of causes a decrease of
. Due to the band anticrossing effect, the coupling . For , reduces to
interaction between Sb localized levels and the valence band
when reaches 10 nm. For
maximum VBM of GaAs leads to a rise of the VBM of
GaAs1-ySby [12-14]. The coupling interaction between N , decreases to
level and the conduction band minimum CBM of GaAs
makes the CBM of GaNxAs1-x move downward [7-9]. Along for . We noted the presence
the z-quantizing direction of the studied GaAs/GaN0.38yAs1- of two allowed transition and in the
1.38ySby/GaAs SQW, the confined electrons level e1 and
absorption spectra for , as shown in the insets
heavy holes level h1 move downward and upward with the
zoom figures. Experimentally, Ohtani et al. [29] investigated
rise of Sb composition, respectively. The wavelength of the
the well width dependence of optical absorption in n-doped
fundamental transition achieves 1.3 and 1.55 µm InAs/AlSb quantum structures. They signaled a red-shift of
(i.e. 0.95 and 0.80 eV, respectively) for specific values of the absorption peak with increasing the well width.
(as displayed in Fig. 1).

ISSN 1737-9334
to blue when the in-plane carrier density increases to 5 1012

8 (a) 10 nm (x5) cm-2. It should be noted that the difference of the transition
8 nm Te1-h3
energy shift value of GaN0.38yAs1-1.38 ySby/GaAs QWs
6 nm Te1-h3
4 nm compared to other systems may be due to the difference in
6
abs (10 cm )
their physical properties such as the total confining potential,

-1
1.04 1.08 1.12

range of doping density and carrier effective masses. On the
Te1-h1 = 1.3 µm
4
4 other hand, we suggested adjusting the Sb composition

Te2-h2
for each value of in order to keep the fundamental
2 transition wavelength equal to 1.3 and 1.55 µm.
0
970 (a) ySb = 13.0 %
8 (b) 10 nm Te1-h3
(x5) ySb = 18.1 %
8 nm 965
6 nm
Te1-h1 (meV)
4 nm Te1-h3 960
6
abs (10 cm )
-1
0.85 0.90 0.95 1.00

955
815
4 Te1-h1 =
4
1.55 µm Te2-h2
810
2
805
0
0.6 0.8 1.0 1.2 1.4 (b) Te1-h1 = 1.3 µm
E (eV)
Te1-h1 = 1.55 µm
(10 cm )
-1
10
Fig. 2. Well width dependence of the absorption coefficient of
5
GaAs/GaN0.38yAs1-1.38ySby/GaAs SQW for equal to (a) 1.3 µm and

(b) 1.55 µm.
Te1-h1
abs
In order to enhance the optical absorption process, we

have studied the n doping effect on the electronic and optical
properties of GaAs/GaN0.38yAs1-1.38 ySby/GaAs SQW. 1
The dependence of the fundamental transition energy
on the in-plane donor density is presented in Fig. 3a. 1 10
The well width is . The Sb compositions give N2D 11 -2
d (10 cm )
rise to and 1.55 µm are 13.0 % and 18.1 %,
respectively. The used range is between and Fig. 3. Variation of (a) the transition energy and (b) the
. Owing to n doping effect, we have found

absorption coefficient magnitude of GaAs/GaN0.38yAs1-
that energy shifts of 10 meV to higher energies. The
obtained behavior is mainly due to the band-bending effect at 1.38ySb y/GaAs SQW as a function of the in-plane donor density .
the interface GaNAsSb (well) –GaAs (barrier). The rise of In Fig. 3b, we plotted the variation of the absorption
doping density causes the change of subbands positions and
the enhancement of the electrons confinement. Likewise, we coefficient magnitude at the fundamental transition
signaled in our previous work [25] a blue-shift (10 meV) in and µm as a function of the in-plane
n-doped GaAs/GaN0.58yAs1-1.58yBiy/GaAs QWs when the
donor density . In fact, values are in the ranges 13.1
in-plane doping density reaches . In another
study, Kim et al. [30] studied the doping effect on the optical – 13.4 % and 18.3 – 18.6 % for and
properties of GaAs.76Sb.24/In.26Ga.74N.06As.94/GaAs QWs.
, respectively. For , we found
They stated that the fundamental transition shifts of 20 meV

ISSN 1737-9334
that increases from to (a) N2D 12 -2 ySb = 0.134

d = 3 10 cm
960 ySb = 0.186
Te1-h1 (meV)
with increasing from to . In the
same donor density range and for , 940
rises from to . This 800

behavior was reported in the study of InGaN/GaN LEDs
structures [31] and n-doped GaNAsBi/GaAs QWs and 780
[25].On the other hand, Tan et al. [32] experimentally
investigated the absorption in GaNAsSb/GaAs p-i-n
photodetectors at the wavelength 1.3 µm. They stated that the 19.0 (b) Te1-h1 = 1.3 µm
absorption coefficient is about 1.3 104 cm-1. Te1-h1 = 1.55 µm
18.5
ySb (%)
We have investigated the effect of the applied electric field 18.0
on the optical absorption and inter-band transition in n-doped
GaAs/GaN0.38yAs1-1.38 ySby/GaAs SQW. The in-plane 13.5
donor density is equal to . Fig. 4a 13.0
exhibits that without an applied electric field 12.5
, the use of Sb compositions 13.4 % and 18.6
(c) Te1-h1 = 1.3 µm
% gives rise to a fundamental transition equal to 0.95
(10 cm )
-1
(i.e. 1.3 µm) and 0.80 eV (i.e. 1.55 µm), respectively. Under
10
Te1-h1 = 1.55 µm
an external electric field of up to , the
5
fundamental transition energy shifts (20 meV) to

Te1-h1
lower energies due to the quantum confined Stark effect. This

abs
red shift can be attributed to the modification of the electron 1

and hole subbands under the effect of the applied electric
field [33]. The Stark effect is more pronounced for holes as a
consequence of their heavier effective masses and the lower 0 20 40 60 80 100 120 140
valence band edge offset [34]. In order to maintain the F (kV/cm)
wavelength at 1.3 µm and 1.55 µm for each used
value of the applied electric field, we adjusted the Sb Fig. 4. Electric field dependence of (a) the transition energy , (b)
composition as shown in Fig. 4b. The dependence of the the adjusted value of Sb composition and (c) the absorption coefficient
absorption coefficient magnitude at the fundamental
magnitude of GaAs/GaN0.38yAs1-1.38ySby/GaAs SQW.
transition on the electric field F is displayed in Fig.
III. CONCLUSION
4c. For , decreases from
to when F increases from In summary, we have theoretically investigated the n
doping effect on the electronic and optical absorption of
zero up to 140 kV/cm. For , lattice-matched GaAs/GaN0.38yAs1-1.38 ySby/GaAs SQW
operating at 1.3 and 1.55 µm. We have performed a self-
decreases from to with
consistent calculation combined with (16×16) BAC model.
increasing F to the value of 140 kV/cm. In another study, Kuo
For each value well width ranging from 4 to 10 nm, the use
et al. [35] investigated the Stark effect the on absorption
of specific Sb composition gives rise to the wavelengths
coefficient spectra of Ge/SiGe QWs operating at 1.55 µm.
emissions and . The increase of causes a significant
They signaled that the absorption coefficient magnitude
decrease of absorption coefficient magnitude . This latter is
decreases from 104 cm-1 to 2 103 cm-1 when the bias voltage
enhanced to when the studied SQW is n doped with of up
rises from zero to 2 V. The theoretical optimization of the
damental transition
Stark effect on the optoelectronic properties of such QWs
energy is stated when SQW is under an applied electric field
improves the absorption process at the telecomm wavelengths
F of up to 140 kV/cm. Sb composition is adjusted for each
1.3 and 1.55 µm. The present investigation could be useful
value of F to keep at 1.3 and 1.55 µm. The increase of the
for the elaboration of highly efficient photodetectors based on
external electric field induces a slight decrease of the
n-doped GaNAsSb/GaAs QWs.
absorption coefficient.
Copyright - 2019
Page 95
ISSN 1737-9334
ACKNOWLEDGMENT [17] J.C. Harmand, A. Caliman, E.V.K. Rao, L. Largeau, J. Ramos, R.

Teissier, L. Travers, G. Ungaro, B. Theys, I.F.L. Dias, Semicond. Sci.
We gratefully acknowledge the financial support from the Technol. 17 (2002) 778.
“Direction Générale de la Recherche Scientifique et de la [18] M. Gladysiewicz, R. Kudrawiec, M.S. Wartak, IEEE J. Quant.
Technologie” (DGRST), Tunisia. Electron. 50 (2014) 996.
[19] S.A. Lourenço, I.F.L. Dias, L.C. Poças, J.L. Duarte, J.B.B. de Oliveira,
J.C. Harmand, J. Appl. Phys. 93 (2003) 4475.
REFERENCES [20] F. Bousbih, S. Ben Bouzid, R. Chtourou, F.F. Charfi, J.C. Harmand, G.
[1] S. Fedderwitz, A. Stöhr, S. F. Yoon, K. H. Tan, M. Weiß, W. K. Loke, Ungaro, Mater. Sci. Eng. C 21 (2002) 251.
A. Poloczek, S. Wicaksono, D. Jäger, Appl. Phys. Lett. 93 (2008) [21] M.M. Habchi, A. Ben Nasr, A. Rebey, B. El Jani, Infrared Phys.
033509. Technol. 61 (2013) 88.
[2] H. Luo, J.A. Gupta, H.C. Liu, Appl. Phys. Lett. 96 (2005) 211121. [22] D.P. Samajdar, U. Das, A.S. Sharma, S. Das, S. Dhar, Curr. Appl.
[3] K.H. Tan, W.K. Loke, S. Wicaksono, D. Li, Y.R. Leong, S.F. Yoon, P. Phys. 16 (2016) 1687.
Sharma, T. Milakovich, M.T. Bulsara, E. A. Fitzgerald, Appl. Phys. [23] I. Vurgaftman, J.R. Meyer, J. Appl. Phys. 94 (2003) 3675.
Lett. 104 (2014) 103906. [24] G. Bastard, Wave Mechanics Applied to Semiconductor
[4] A. Marcos, N. Faleev, R.R. King, C.B. Honsberg, J. Vac. Technol. B Heterostructures, Les Editions de Physique, Les Ulis Cedex, France,
34 (2016) 02L106-1. 1990.
[5] Y.T. Lin, T.C. Ma, T.Y. Chen, H.H. Lin, Appl. Phys. Lett. 93 (2008) [25] A. Ben Nasr, M.M. Habchi, C. Bilel, A. Rebey, B. El Jani, Journal of
171914. Semiconductors 49 (2015) 593.
[6] S.A. Lourence, M.A.T. da Silva, I.F.L. Dias, J.L.Duarte, J.C. Harmand, [26] C. Bilel, M.M. Habchi, A. Rebey, B. El Jani, Physica E Low Dimens.
J.Phys.:Condens. Matter. 23 (2011) 325801. Syst. Nanostruct. 69 (2015) 232.
[7] K. Uesugi, N. Morooka, I. Suemune, Appl. Phys. Lett. 74 (1999) 1254 [27] C. Bilel, M.M. Habchi, A. Rebey, B. El Jani, Thin Solid Films 581
[8] J. Wu, W. Shan, W. Walukiewicz, Semicond. Sci. Technol. 17 (2002) (2015) 70.
860 [28] A. Ben Jazia, H. Merjri, H. Maaref, K. Souissi, Semicond. Sci.
[9] W. Shan, W. Walukiewicz, K.M. Yu, J.W. Ager III, E.E. Haller, J.F. Technol. 12 (1997) 1388 and references therein.
Geisz, D.J. Friedman, J.M. Olson, S.R. Kurtz, H.P. Xin, C.W. Tu, [29] M.M. Habchi, C. Bilel, A. Ben Nasr, A. Rebey, B. El Jani, Mater. Sci.
Phys. Status Solidi B 223 (2001) 75. Semicond. Process. 28 (2014) 108.
[10] K. Alberi, J. Wu, W. Walukiewicz, K.M. Yu, O.D. Dubon, S.P. [30] K. Ohtani, N. Matsumoto, H. Sakuma, H. Ohno, Appl. Phys. Lett. 82
Watkins, C.X. Wang, X. (2003) 37.
[11] Liu, Y.-J. Cho, J. Furdyna, Phys. Rev. B 75 (2007) 045203. [31] H.Y. Ryu, K.S. Jeon, M.G. Kang, Y. Choi, J.S. Lee, Opt. Express. 21
[12] R. Teissier, D. Sicault, J.C. Harmand, G. Ungaro, G. Le Roux, L. (2013) A190.
Largeau, J. Appl. Phys. 89 (2001) 5473. [32] K.H. Tan, S.F. Yoon, W.K. Loke, S. Wicaksono, T.K. Ng, K.L. Lew,
[13] K.I. Lin, K.L. Lin, B.W. Wang, H.H. Lin, J.S. Hwang, Appl. Phys. A. Stöhr, S. Fedderwitz, M. Weiβ, D. Jäger, N. Saadsaoud, E.
Express 6 (2013) 121202. Dogheche, D. Decoster, J. Chazelas, Opt. Express. 16 (2008) 7720.
[14] N. Ben Sedrine, C. Bouhafs, J.C. Harmand, R. Chtourou, V. [33] A. Dimoulas, K.P. Giapis, J. Leng , G. Halkias, K. Zekentes, A.
Darakchieva, Appl. Phys. Lett. 97 (2010) 201903. Christouc, J. Appl. Phys. 72 (1992) 1912.
[15] C.Z. Zhao, H.F. Guo, T. Wei, S.S. Wang, K.Q. Lu, Physica B: Phys. [34] L. Vina, R.T. Collins, E.E. Mendez, W.I. Wang, L.L. Chang, L. Esaki,
Cond. Matter. 485 (2016) 35. Superlattices and Microstructures 3 (1987) 9.
[16] J.C. Harmand, G. Ungaro, J. Ramos, E.V.K. Rao, G. Saint-Girons, R. [35] Y. Kuo, Y. K. Lee, Y. Ge, S. Ren, J. E. Roth, T. I. Kamins, D. A. B.
Teissier, G. Le Roux, L. Largeau, G. Patriarche, J. Cryst. Growth 227 Miller, J. S. Harris, IEEE J. Sel. Topics Quantum Electron. 11 (2005)
(2001) 553. 338.
Copyright - 2019
ISSN 1737-9334 Page 96
Editor biographie:
Dr. Ahmed Rhif (Tunisia) is a Researcher & Engineer (PhD,

Eng). He has more than 11 years of experience on Scientific
Research, Teaching and industrial projects. He is actually the
Dean of the International Centre for Innovation &
Development (ICID). Ahmed Rhif has worked as a Technical
Responsible Chief in LEONI (International Leader of Wiring
Fibers Companies) and has occuped also the task of Project
Manager and Method Engineer in both SMSI (electronic
development industry) and CABLITEC (Engineering
automobile company). Then he was a Lecturer at both the Private University of
Sousah (UPS) and the High Institute of Applied Sciences and Technologies of
Sousah (ISSATso) and now he is working as Lecturer in the High Institute of
Applied Sciences and Technologies of Al Qayrawan (ISSATk). His research interests
include Modelling, Control Systems and Engineering as well as the implantation of
the international standard of quality (ISO-TS 16949). Dr. Ahmed Rhif is also a
Trainer of Trainers certified from the Canadian Centre of Training in Montreal and
Coach certified NLP from the American Board of Neuro-linguistic programming.

Renewable Energy &amp; Applications Anglophone Issue

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Renewable Energy &amp; Applications Anglophone Issue

Încărcat de

Drepturi de autor:

Formate disponibile

ISSN: 1737-9334

Proceedings of Engineering & Technology

Renewable Energy & Applications

International Centre for Innovation & Development

Editorial Borad Samira Chader Kerdjou (ALG)

Tounzi Abdelmounaïm (FRA)

Fabien Delaleux (FRA)

Georges Descombes (FRA)

Hana Boudjedir (ALG)

Fawaz Massouh (FRA)

Nachida K. Merzouk (ALG)

Khenfer Nabil (ALG)

Mohamed Elamine Djeghlal (ALG)

Manar Slimane (MAR)

Abdelfettah Barhdadi (ALG)

Mohamed Gherbi (ALG)

Mohamed Benbouzid (FRA)

Sara Zatir (ALG)

Philippe Guibert (FRA)

Mustapha Hatti (ALG)

Saoussen Hammami (TUN)

Rahmani Lazhar (ALG)

Tahar Bahi (ALG)

Fathi Mohammed (ALG)

Diagnosis of Defects by Principal Component Analysis a Gas Turbine. Page 23

Group contribution method for the estimation of flammability proprieties. Page 30

Numerical investigation of an inclined offset jet. Page 33

Performance evaluation of Gamma type Stirling engine. Page 74

Meeting Bioclimatic Knowledge Needs of Algerian

Copyright - 2019 Page 1

Practically, the implementation of such method is the result

Copyright - 2019 Page 4

Extraction and valorization of the active ingredients of Saccocalyx

Number phone : 00 213 5 51 15 81 45

Copyright - 2019 Page 5

Algeria is considered among the countries known Drying:

We introduce into a 2-liter pyrex glass flask, 40 g of

Without forgetting to note its weight with a scale

Figure 02: Saccocalyx saturated

It is an agar diffusion method that aims to

Figure 03: Installation of a modified clevenger type

Gas chromatography coupled with mass

GC / MS is a separation analysis method that

Note that for Ain chouhada we did not test the

Copyright - 2019 Page 7

Figure 07: Content of the chemical classes

Figure 08: Content of three major constituents of

The results of the chemical analysis of the

Copyright - 2019 Page 8

HE with a light yellow color and a strong and

The difference between the yields of the two sites is

This difference in yield between the two sites can

The class Monoterpenes Hydrocarbons follows

Copyright - 2019 Page 9

For some authors, chemotypes constitute, within the

Copyright - 2019 Page 10

We note that Saccocalyx sematioides HE from both Determination of hydrodistillation yields of

Copyright - 2019 Page 12

Fuel Cells and Hydrogen Storage: Challenges Facing Vehicle Manufacturers

Keywords— Fuel cell; Hydrogen storage; vehicle

by 2031 [4]. In other words, oil consumption for the transport

In the longer term, fuel cell technology seems to be very

III. FUEL CELLS AND THEIR APPLICATIONS

Figure 4: The Sir William Grove experience 1839

Renewable Energy & Applications Anglophone Issue

Renewable Energy & Applications Anglophone Issue