This article was downloaded by: [New York University]

On: 28 November 2013, At: 06:26

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

International Journal of Food

Properties
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/ljfp20

Temperature and water content

influence on thermophysical
properties of coffee extract
a b c
J. Telis‐Romero , A. L. Gabas , M. A. Polizelli & V. R. N. Telis
c

a
Departamento de Engenharia e Tecnologia de Alimentas ,
Universidade Estadual Paulista , São José do Rio Preto, São Paulo,
15054‐000, Brazil Phone: 55‐17‐221 2251 Fax: 55‐17‐221 2251 E-
mail:
b
Departamento de Engenharia de Alimentos , Universidade
Estadual de Campinas , C.P.6121, Campinas, São Paulo,
13083‐970, Brazil
c
Departamento de Engenharia e Tecnologia de Alimentas ,
Universidade Estadual Paulista , São José do Rio Preto, São Paulo,
15054‐000, Brazil
Published online: 02 Sep 2009.

To cite this article: J. Telis‐Romero , A. L. Gabas , M. A. Polizelli & V. R. N. Telis (2000)

Temperature and water content influence on thermophysical properties of coffee extract,
International Journal of Food Properties, 3:3, 375-384, DOI: 10.1080/10942910009524642

To link to this article: http://dx.doi.org/10.1080/10942910009524642

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
 This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
Downloaded by [New York University] at 06:26 28 November 2013
 INTERNATIONAL JOURNAL OF FOOD PROPERTIES, 3(3), 375-384 (2000)

TEMPERATURE AND WATER CONTENT INFLUENCE ON

THERMOPHYSICAL PROPERTIES OF COFFEE EXTRACT

J. Telis-Romero1,*, A. L. Gabas2, M. A. Polizelli1, and V. R. N. Telis1

1
Departamento de Engenharia e Tecnologia de Alimentas, Universidade Estadual
Downloaded by [New York University] at 06:26 28 November 2013

Paulista, São José do Rio Preto, São Paulo, 15054-000, Brazil

2
Departamento de Engenharia de Alimentos, Universidade Estadual de Campinas,
C.P.6121, Campinas, São Paulo, 13083-970, Brazil

*Corresponding author (Tel.: 55-17-221 2251, Fax: 55-17-224 8692, and E-mail:
javier@eta.ibilce.unesp.br)

ABSTRACT

Specific heat, thermal conductivity, thermal diffusivity, and density of coffee extract
were experimentally determined in the range of 0.49 to 0.90 (wet basis) water content
and at temperatures varying from 30 to 82°C. Thermal conductivity and specific heat
were measured by means of the same apparatus- a cell constituted of two concentric
cylinders - operating at steady and unsteady state, respectively. The thermal diffusivity
was measured by the well-known Dickerson's method and density was determined by
picnometry. The results obtained were used to derive mathematical models for
predicting these properties as a function of concentration and temperature.

INTRODUCTION

The world-wide coffee crop production estimate for 1998/99 are of about 107.5 million
bags, and Brazil should take a 31.5% share, making it the major producer and major
exporter. The major producer of soluble coffee in the world is the United States with
30%, followed by Brazil with 20%. Nevertheless, Brazil is the only large producer
directed to the foreign market and is responsible for 60% of the soluble coffee sold
internationally (Cia. Iguacu de Cafe Soliivel, 1999).
Knowledge of thermophysical properties of the coffee extract over a wide range
of concentrations and temperatures is of primary importance to the soluble coffee

375

Copyright © 2000 by Marcel Dekker, Inc. www.dekker.com

 376 TELIS-ROMERO ET AL.

industry, as it makes use of a variety of unit operations such as pumping, heating,

cooling, and evaporation, which must be appropriately designed and controlled.
Some literature are available on determination of specific heat and thermal
conductivity of different fluid food products such as orange juice (Telis-Romero et al.,
1998; Moresi and Spinosi, 1980), reconstituted milk (Reddy and Datta, 1994), apple
juice (Constenla et al., 1989), and tomato juice (Choi and Okos, 1983). Considering
coffee products, the only work that reports thermophysical properties was published by
Singh et al. (1997), but it is related to coffee powders.
Extensive reviews of existing thermophysical properties measurement methods
of foods has been carried out by Reidy and Rippen (1971), Mohsenin (1980), Singh
(1982), and others. Sweat (1995) recommends methods and strategies that can be
employed to measure thermal properties of food products.
Downloaded by [New York University] at 06:26 28 November 2013

Specific heat measurements are often made by means of a calorimeter (Riedel,

1951, Hwang and Hayakawa, 1979), which is a simple technique although requiring a
careful calibration as a result of the heat capacity of the apparatus. The differential
scanning calorimeter is the best alternative for experimentally determining the specific
heat, but has the disadvantage of being expensive (Constenla et al., 1989; Sweat, 1995).
Most of works on thermal conductivity measurements of food products are
concerned with solid materials (Donsi et al, 1996; Lopez-Ramos et al., 1993; Pham and
Willix, 1989). Many measurement techniques have been described, such as the guarded
hot plate (ASTM C177 American National Standard Institute, 1970) or the line heat
source probe (Sweat and Haugh, 1974; Choi and Okos, 1983). In liquids, the main
source of experimental errors is convection during measurements. Sweat (1995)
recommends the addition of 0.5% agar to water when measuring its thermal
conductivity with a line heat source probe. For oils and water at high temperatures,
about 1% by weight of fiberglass "wool" can be added to suppress convection. In order
to minimize uncertainties due to convection, Bellet et al. (1975) developed an apparatus
based on a cell constituted of two coaxial cylinders, separated by a narrow space that is
filled with the fluid sample. According to these authors, convection can be avoided if
the space between the cylinders is sufficiently small, and the difference between wall
temperatures is not very large. The thermal conductivity is obtained from the equations
describing heat transfer in steady state conditions. Mathematical modeling of unsteady
state operation allows for evaluation of the specific heat of the fluid employing the
same device, which constitutes the main advantage of this method.
Thermal diffusivity can be estimated indirectly from the thermal conductivity
divided by the product of specific heat and mass density. This method of evaluation has
the inconvenient of adding up the experimental errors involved in each one of these
quantities. Alternatively, thermal diffusivity can be measured directly using a transient
heating technique developed by Dickerson (1965).
The purpose of the present work was to measure density, specific heat, thermal
conductivity and thermal diffusivity of coffee extract as a function of temperature and
water content and to develop correlations for predicting these properties under different
conditions.

MATERIALS

All the experimental measurements were made with samples prepared from the same
batch of concentrated coffee extract (51°Brix) supplied by COCAM (Catanduva, Sao
 THERMOPHYSICAL PROPERTIES OF COFFEE EXTRACT 377

Paulo, Brazil) and stored at -18°C prior to its use. The moisture content of the initial
batch was determined in a vacuum oven (48 hrs, 60°C), resulting in 47.6% of moisture
(wet basis). Samples with different water contents were obtained by diluting the
concentrated coffee extract with distilled water.

METHODS

Measurement of Density

Density of coffee extract at different temperatures and concentrations was determined

in triplicate by weighing, in an analytical balance, the coffee extract contained in a
Downloaded by [New York University] at 06:26 28 November 2013

standard volumetric pycnometer (Constenla et al., 1989). Sample temperature was

varied by equilibration on a thermostatic bath. The pycnometer of 25 ml was previously
calibrated with distilled water at each temperature.

Determination of Thermal Conductivity and Specific Heat

Thermal conductivity and specific heat at various temperatures and water contents were
measured using the method described by Bellet et al. (1975), based on a cylindrical cell,
where the liquid whose properties are being determined fills the annular space between
two concentric cylinders. The equipment, shown in Figure 1, presented the following
physical characteristics:

- two coaxial copper cylinders (A and B), 220 mm length, separated by a 2 mm

annular space, which was filled with the sample;
- 50 mm thick covers (C) made of a low thermal conductivity material (0.225
W/m.°C), to prevent axial heat transfer;
- inner cylinder (A) containing a heater (D) made with a constantan wire (resistance
15 Q), electrically insulated by a varnish and coiled around a copper stick;
- two thermocouples type T (E) to measure temperature differences between the two
cylinders located at half-length of the cell. The wires were placed inside 0.5 mm
gaps, parallel to the cell axis.

To keep the external temperature constant, the cell was immersed in a thermostatic bath
(model MA-184, Marconi, Sao Paulo, Brazil) containing water. The power input to the
heater resistance was made by means of a laboratory DC power supply (model MPS-
3006D, Minipa, Sao Paulo, Brazil), which permitted to adjust the current with a
stability of 0.05%. A HP data logger model 75.000-B, an interface HP-IB and a HP PC
running a data acquisition program written in IBASIC monitored temperatures with an
accuracy of 0.6°C.
In the steady state, the solution of the Fourier equation in cylindrical
coordinates, with boundary conditions corresponding to heat transfer between two
concentric cylindrical surfaces kept at constant temperatures, permits to determine
thermal conductivity of the sample. The specific heat can be measured with the same
apparatus operating in the unsteady state. Details of the cell operation and calculations
involved are described elsewhere (Telis-Romero et al., 1998; Bellet et al., 1975).
 378 TELIS-ROMERO ET AL.

Inner cylinder (A) External cylinder (B)

Cover (C)

Sample
Power supply

SSSSSi^SKiSSSi^^

SKS^SKSKSSKKSK^^

Thermocouple (E)
Downloaded by [New York University] at 06:26 28 November 2013

Sample supply tube Cover (C)

Thermocouple (E)

Figure 1. Cross section of the cell used for thermal conductivity and specific heat
measurements

Calibration of the cell was performed with distilled water and glycerin. This
permitted to calculate density (p') and specific heat (C p ') of the cell material (copper),
which must be known in order to determine sample properties. The copper properties
obtained by using water were 8890 kg/m3 and 403.4 J/kg °C for density and specific
heat, respectively. With glycerin, respective values were 8886 kg/m3 and 406.6 J/kg °C.
Experiments were made in order to check if these values were affected by temperature
and results showed differences within a ± 2% range. Considering that differences were
small, constant average values of p ' and C p ' were adopted.

Measurement of Thermal Diffusivity

Thermal diffusivity was determined using the method proposed by Dickerson (1965).
The experimental apparatus consisted of a cylindrical cell (24.75 x 10"3 m internal
radius and 248.5 x 10"3 m length) made of chromium plated brass with two nylon
covers with thermal diffusivity of 1.09 x 10'7 m2/s, which is similar to the most of
liquid food products. Two thermocouples type T were fixed at the center and on the
external surface of the cell. The cell was immersed in an well-agitated thermostatic bath
(model MA-184, Marconi, Sao Paulo, Brazil) heated at a constant rate, and the
evolution of temperatures at the wall and at the center of the cell was monitored.
Temperatures were monitored employing the same data acquisition system used in
thermal conductivity and specific heat measurements.
The calculations were based on the solution of the equation of energy
conservation, considering unsteady state and constant unidimensional (radial) heat flux,
described in details in previous work (Telis-Romero et al., 1998; Dickerson, 1965).
 THERMOPHYSICAL PROPERTIES OF COFFEE EXTRACT 379

1 " 1 ' I 1
I ' 1
X
1200- -

•X
1150- X • \ x
-

x \<x •

1100- -

\ w X

1050- Xv X
Downloaded by [New York University] at 06:26 28 November 2013

1000- •

0.5 0.6 0.7 0.S 0.9

Vteter content (mass of water/total mass)

Figure 2. Experimental density of coffee extract as function of water content and

temperature. ( • ) 82°C; (O) 65°C; (*) 48°C; (X) 30°C; ( ) Predictions of
Equation 2

Data Analysis

All statistical analysis were performed using the GLM procedure while fitted functions
were obtained by using the REG procedure from the SAS statistical package (SAS,
1985). The suitability of the fitted functions was evaluated by the coefficient of
determination (r2), the level of significance (p), and residual analysis.

RESULTS AND DISCUSSION

Specific heat, thermal conductivity, thermal diffusivity, and density of coffee extract at
ten selected water content levels varying from 0.49 to 0.90 (wet basis) were determined
at 30, 48, 65, and 82°C, adding up to 40 experimental values of each thermal property.
The obtained data are presented in Figures 2 to 5.
Experimental values of density shown on Figure 2 presented a very strong
dependence on water content, decreasing as the solvent fraction in the solution was
increased. The temperature dependence, however, showed to be less important.
Polynomial functions, simultaneously dependent on temperature and water content
were investigated in order to correlate experimental data. The best fit, resulting in a r2
value of 0.981, is given by Equation 1.
 380 TELIS-ROMERO ET AL.

4000
Downloaded by [New York University] at 06:26 28 November 2013

0.5 0.6 0.7 0.8 0.9

V\feter content (mass of water/total mass)

Figure 3. Experimental specific heat of coffee extract as function of water content

and temperature. (•) 82°C; (O) 65°C; (*) 48°C; (X) 30°C; ( ) Predictions of
Equation 3

1 ' 1 * 1

M y
0.50-

0.45-

•
i •

•B
0.40- -
"TO

0.35- Y -

0.5 0.6 0.7 0.8 0.9

Wkter cxxitent (mass of water/ total mass)

Figure 4. Experimental thermal conductivity of coffee extract as function of water

content and temperature. (•) 82°C; (O) 65°C; (*) 48°C; (X) 30°C; ( )
Predictions of Equation 5
 THERMOPHYSICAL PROPERTIES OF COFFEE EXTRACT 381

1.5-

1.4-
s
o
X 1.3-
diffusivit

1.2-
Downloaded by [New York University] at 06:26 28 November 2013

5 1.1-

1.0
0.5 0.6 0.7 0.8 0.9
Water content

Figure 5. Thermal diffusivity of coffee extract as function of water content and

temperature. (•,<>) 82°C; (•,()) 65°C; (A,A) 48°C; (B,D) 30°C. Open symbols
correspond to calculated values using Equation 8, ( ) Predictions of Equation 7

= 1422.57-451.98X w -0.16T (1)

where p is the density of coffee extract (kg/m3), Xw the water content (mass of
water/total mass), and T is the temperature (°C). A theoretical approach as the one
proposed by Constenla et al. (1989) and based on a thermodynamic expression for the
specific volume of a multicomponent solution, was also considered. This resulted in
Equation 2, which gave a slight smaller value of r^ (0.967). In this case, the thermal
effect on density is included in the value of pure water density, p w .

= p w (l.44-0.46X w ) (2)

The specific heat of the coffee extract (Figure 3) increased as water content was
elevated from 0.49 to 0.90. Temperature rising was also responsible for higher values
of specific heat. Experimental data could be well correlated by a linear function of
temperature and water content given by Equation 3, which resulted in a coefficient of
determination (r2) of 0.993.

C p = 1439.65 + 2633.72XV + 1.99T (3)

 382 TELIS-ROMERO ET AL.

where Cp is the specific heat (J/kg °C). The correlation obtained was similar to that
presented by Telis-Romero et al. (1998) for orange juice. Thermal conductivity
increased with increasing water content, but the effect of temperature on the
experimental values was not so well defined (Figure 4). Nevertheless, a linear
dependence on concentration and temperature, represented by Equation 4, could well
adjust the data with a r2 value of 0.986.

k = 0.154 + 0.391X w +1.48xl0" 4 T (4)

where k is the thermal conductivity (W/m °C). Also an additive model, based on the
apparent thermal conductivity and volume fraction of each component in the solution,
presented by Constenla et al. (1989), was tested in adjusting thermal conductivity of
Downloaded by [New York University] at 06:26 28 November 2013

coffee extract. In this model, as in Equation 2, the effect of temperature is incorporated

through the conductivity and density of pure water. The final correlation is given by
Equation 5 and resulted in a coefficient of determination (r2) of 0.923.

k = (kwp/pj(0.0556 + 0.794Xw) (5)

where kw is the thermal conductivity of pure water. The thermal diffusivity of the
coffee extract increased at higher water contents and at higher temperatures, as shown
in Figure 5. An empirical model dependent on these two variables is presented as
Equation 6, with a r2 value of 0.944.

cc = 7.92xl0~ 8 +5.93xl0~ 8 X w +0.0212xl0" 10 T (6)

where a is the thermal diffusivity (m2/s). An alternative model for predicting thermal
diffusivity is based on the expression developed by Riedel (1969) cited by Singh
(1982). The correlation considers the temperature effect by the inclusion of the thermal
diffusivity of pure water, ctw. Equation 7 presents the best adjustment obtained (r2 =
0.931).

( ) (7)

The constants in Equation 7 are a little different of the ones proposed by Riedel (1969),
but this could be explained by considering that it was originally developed based on
experimental data of a miscellaneous of food products, including solids and semi-
solids.
Thermal diffusivity can also be calculated according to its definition (Equation
8), what was done in this work by using the measured values of density, specific heat,
and thermal conductivity.

a =— . (8)
pc p
The relative error between calculated and experimental thermal diffusivity, defined
according Equation 9, showed that the calculated values were always lower than
 THERMOPHYSICAL PROPERTIES OF COFFEE EXTRACT 383

experimental ones (Figure 5). The mean value of error was -11.35%, with a standard
deviation, s, of 3.87%. The comparison of calculated and experimental data showed
that greater errors occur at higher temperatures. This lead to the conclusion that as
temperature increases convection inside the Dickerson's cell could cause super-
estimation of experimental thermal diffusivity.

A% error = (otcai - a exp ). 100/aexp (9)

CONCLUSIONS

The experimental results of this work showed that temperature and mainly water
Downloaded by [New York University] at 06:26 28 November 2013

content have a strong influence on thermophysical properties of coffee extract.

Theoretical and empirical models could be successfully adjusted to density, specific
heat, thermal conductivity, and thermal diffusivity data. Equations 2, 3, 5, and 7 are
recommended to estimate these properties for coffee extract in the range of 30 to 82°C
and 49 to 90% of moisture content (wet basis).

ACKNOWLEDGMENTS

The authors would like to thank the Sao Paulo State Research Fund Agency, FAPESP,
for financial support (Proc. 98/08738-7) and COCAM - Cia. de Café Solúvel e
Derivados for helpful suggestions and sample material donation.

REFERENCES
American National Standard Institute. 1970. Standard method of test for thermal
conductivity of materials by means of the guarded hot plate, 1970 Annual Book of
ASTM Standard, C177-63, Philadelphia.
Bellet, D., Sengelin, M., and Thirriot, C. 1975. Determination des proprietes
thermophysiques de liquides non-newtoniens a l'áide d'une cellule a cylindres
coaxiaux. International journal of Heat and Mass Transfer. 18(10): 1177-1187.
Choi,.Y., and Okos, M. R. 1983. The thermal properties of tomato juice concentrates.
Transactions of the ASAE. 26: 305-311.
Cia. Iguaçu de Café Solúvel 1999. Homepage service: http://www.iguacu.com.br
(09/20/99).
Constenla, D. T., Lozano, J. E., and Crapiste, G. H. 1989. Thermophysical properties of
clarified apple juice as a function of concentration and temperature. Journal of Food
Science. 54(3): 663-668.
Dickerson, R. W. Jr. 1965. An apparatus for measurement of thermal diffusivity of
foods. Food Technology. 19(5): 198-204.
Donsi, G., Ferrari, G., and Nigro, R. 1996. Experimental determination of thermal
conductivity of apple and potato at different moisture contents. Journal of Food
Engineering. 30: 263-268.
Hwang, M. P., and Hayakawa, K. I. 1979. A specific heat calorimeter for foods.
Journal of Food Science. 44: 435-438, 448.
 384 TELIS-ROMERO ET AL.

López-Ramos, A., Palmisano, E., Dombey, A., Pimentel, J. A., Fayés, D., and
González-Mendizábal, D. 1993. Thermal properties of tropical fruits and vegetables.
Revista Española de Ciencia y Tecnología de Alimentos. 33: 271 -283.
Mohsenin, N. N. 1980. Thermal Properties of Foods and Agricultural Materials.
Gordon and Breach, New York.
Moresi, M., and Spinosi, M. 1980. Engineering factors in the production of
concentrated fruit juices. 1. Fluid physical properties of orange juices. Journal of
Food Technology. 15: 265-276.
Pham, Q. T., and Willix, J. 1989. Thermal conductivity of fresh lamb meat, offals and
fat in the range -40 to +30°C: measurements and correlations. Journal of Food
Science. 54: 508-515.
Reddy, Ch. S., and Datta, A. K. 1994. Thermophysical properties of concentrated
Downloaded by [New York University] at 06:26 28 November 2013

reconstituted milk during processing. Journal of Food Enginnering. 21: 31-40.

Reidy, G. A., and Rippen, A. L. 1971. Methods for determining thermal conductivity in
foods. Transactions of the ASAE. 14: 248-254.
Riedel, L. 1951. The refrigerating effect required to freeze fruits and vegetables.
Refrigerating Engineering. 59: 670-673.
Riedel, L. 1969. Measurements of thermal diffusivity of foodstuffs rich in water.
Kaltetechnik Klimatisierung. 21(11): 315.
SAS Institute Inc. 1985. SAS® User's guide: Statistics, Version 5 edition. Cary: SAS
Institute Inc.
Singh, P. C., Singh, R. K., Bhamidipati, S., Singh, S. N., and Barone, P. 1997.
Thermophysical properties of fresh and roasted coffee powders. Journal of Food
Process Engineering. 20: 31-50.
Singh, R. P. 1982. Thermal diffusivity in food processing. Food Technology. 36(2): 87-
91.
Sweat, V. E. 1995. Thermal properties of foods. In: Engineering Properties of Foods,
ed. M. A. Rao and S. S. H. Rizvi, 2nd edition, Marcel Dekker, Inc., New York, pp.
99-138.
Sweat, V. E. and Haugh, C. G. 1974. A thermal conductivity probe for small food
samples. Transactions of the ASAE. 17: 56-58.
Telis-Romero, J., Telis, V. R: N., Gabas, A. L., and Yamashita, F. 1998.
Thermophysical properties of Brazilain orange juice as affected by temperature and
water content. Journal of Food Engineering. 38(1): 27-40.

(Received November 30, 1999; revised March 16, 2000; accepted April 5, 2000)