Documente Academic
Documente Profesional
Documente Cultură
___________________________________________________________________________________________________
9ième Congrès de Mécanique, FS Semlalia, Marrakech 40
(1 + af p ) MT
Esaturé = E0
(1 − f )(1 + af )
p p
V sec
=V sec
1 + bf p − cf p
2
(1 + bf p )(1 + cf p )
p pm
ρ sec = ρ m (1 − f p ) et
kp : coefficient de compression du fluide ρ saturé = ρ m (1 − f p ) + ρ f f p S w
En mettant k p = 0 , on obtient le module de
Vssec ρf fp
compression dans le cas d’un milieu poreux sec k
MT = 1−
sec .
Vssaturé
ρ m (1 − f p )
Le module de cisaillement dans un milieu poreux
saturé est le même dans un milieu poreux sec car le Sw : saturation en eau
fluide ne reprend pas les cisaillements (µ fluide=0) Cas où Sw=100% (saturation totale)
D’après la dernière équation la vitesse Vs (sec) est plus
(1 − f p )(9k m + 8µ m )
µ saturé
MT
= µm grande que la vitesse Vs (saturé)
(9 + 6 f )k + (8 + 12 f )µ
p m p m
Comparaison entre des propriétés effectives :
MT
L’estimation de Module de Young E saturé :
___________________________________________________________________________________________________
9ième Congrès de Mécanique, FS Semlalia, Marrakech 41
50
module de Young, les coefficients de compression et de
45
cisaillement dans le milieu saturé ont des valeurs plus
40 E
MT grandes que ceux dans le milieu sec. Les vitesses de
s a tu r é
35 propagation des ondes de compression Vp et de
30 cisaillement Vs et le rapport Vp/Vs diminuent quand la
E(GPa)
0
les vitesses Vp et Vs sont plus grandes que ceux dans le
0 ,0 0 ,1 0 ,2 0 ,3 0 ,4 0 ,5 milieu sec. Guéguen a montré que la vitesse Vp (sec) est
P o r o s it é f p
plus petite que la vitesse Vp (saturé), la vitesse Vs (saturé)
est plus petite que la vitesse Vs (sec) (confirmation de
Fig 1 : Module d’Young en fonction de la porosité: l’étude théorique). Guéguen a interprété ce résultat, d’une
Grès saturé (Sw=100%): =2.503g/cm3, (eau)=1g/cm3 part le module de compression est plus grand en milieu
k=30.2 GPa, µ= 17.5 GPa, kp (eau)=2.3 GPa [8] saturé alors que le module de cisaillement reste constant,
d’autre part la densité augmente dans tous les cas avec la
saturation. Par suite l’effet de module l’emporte dans le
5
cas des ondes S et Vs diminue. Si la porosité est
importante, l’effet de la densité peut être dominant, du
saturé
VP moins au début de la saturation : Vp et Vs diminuent
4
(Guéguen).
La suite de ce travail sera consacrée à l’étude de
l’influence de la morphologie de pores sur propriétés
VP(km/s)
3
sec macroscopiques effectives, de faire une modélisation de
VP
l’espace poreux pour un milieu de trois phases: matrice
solide, pores occlus (Cop=3KopJ), pores connectés (Ccp=0),
2
et d’étudier d’autres propriétés poromécaniques comme le
tenseur de Biot qui prend en compte la compressibilité de
la matrice solide, l’énergie élastique en fonction de la
1
0,0 0,1 0,2 0,3 0,4 0,5 porosité en utilisant le schéma d’homogénéisation de
Porosité fp Mori-Tanaka.
Bibliographie
Fig 2 : Vitesse Vp en fonction de la porosité dans les [1] H.LEBAIL, Caractérisations microscopiques et
cas sec et saturé. tomographiques de microstructures de matériaux
composites et étude des relations aux propriétés élastiques
associées Thèse de Doctorat, Institut des Sciences
3,0
Vs
saturé Appliquées de Lyon, 2001.
[2] J. L. AURIAULT, heterogeneous medium is equivalent
2,5
macroscopic description possible, Int. J. Engng Sci. Vol.
2,0
29, No. 7, pp. 785-795, 1991.
[3] J.SAROUT, Propriétés Physiques et Anisotropie des
Vs(km/s)
sec
1,5
Vs roches Argileuses : Modélisation Micromécanique et
Expérience Triaxiales Thèse de Doctorat, Université Paris
1,0
XI, 2006.
[4]Q.V.HUYNH, Estimation des propriétés
0,5 poromécaniques effectives des argilites, Thèse de
Doctorat, Institut National de Lorraine, 2006.
0,0 [5] D. BENSAHKOUN, Thèse de Doctorat, Ecole national
0,0 0,1 0,2 0,3 0,4 0,5 supérieur d’arts et métiers de Paris, 2006.
Porosité fp [6] A. Giraud et Al, Effect of pore shape on effective
porothermoelastic properties of isotropic rocks Int. J of
Fig 3 : Vitesse Vs en fonction de la porosité dans les Solids and Structures 45(2008)1–23.
cas sec et saturé. [7] O. COUSSY, et al « Acoustiques des milieux poreux »
Editions TECHNIP
Conclusions et perspectives [8] P. Rasolofosaon, et al, Revu de l’Institut Français du
Nous avons étudié le cas d’un milieu constitue d’une pétrole, Vol. 52, N°6, 1997.
matrice isotrope contenant des inclusions sphériques. [9] Guéguen, V. Palciauskas « Introduction à la physique
Théoriquement, nous concluons que les propriétés des roches » HERMANN EDITEURS DES SCIENCES
macroscopiques effectives (EMT, k MT, µ MT) dans les ET DES ARTS.
deux milieux (sec et saturé) diminuent en fonction de la
porosité fp. Les résultats théoriques montrent que le
___________________________________________________________________________________________________
9ième Congrès de Mécanique, FS Semlalia, Marrakech 42
1er Congres de l'Association Marocaine de Thermique
_____________________________________________________________________________________________________
1* 2
A. Rahmouni , Y. Géraud , Influence de la microstructure et de la
1 1
A.Boulanouar , A.Mellouk ,
1
M. Boukalouch , A. saturation en fluide sur la conductivité
1
Samaouali , I. El Amrani El
3 3
thermique des milieux hétérogènes et poreux
Hassani , M.J.E. Sebbani
1
Laboratoire de Résumé : L’objectif de cet article est d’étudier l’influence de la porosité et
Thermodynamique, de la saturation en fluide sur la conductivité thermique des roches
Département de physique, sédimentaires (calcarénites). La conductivité thermique de ces matériaux
Faculté des Sciences, Avenue poreux est mesurée par la méthode TCS (Thermal Conductivity Scanning).
Ibn Batouta, B.P.1014, Rabat, Les premiers résultats de cette étude indiquent que la conductivité
Maroc. thermique dépend de la porosité, de la structure du réseau poreux et de la
2
Centre de Géochimie de la circulation du fluide dans la roche.
surface UMR-CNRS 7517, 1 rue
Blessing, F 67084 Strasbourg Mots clés : roches sédimentaires, porosité, conductivité thermique, TCS,
Cedex, France. fluide.
3
Laboratoire de Géo-matériaux,
Institut Scientifique, Avenue Ibn
Batouta, B.P.703, Rabat, Maroc.
* a.rahmouni@yahoo.fr
Nomenclature
Nt porosité, %
T température, K
q puissance, W
x distance, m
λ conductivité thermique, W/m.K
1. Introduction
Les matériaux d’origine naturelle (roches, monuments, sols, composites) sont des milieux poreux,
dont la microstructure présente une grande diversité. Ils sont constitués d’une phase solide
généralement hétérogène et une phase poreuse qui peut être totalement ou partiellement saturée par
un ou plusieurs fluides.
Plusieurs auteurs ont étudiés la conductivité thermique des matériaux poreux [1, 2, 3]. La porosité
et la conductivité thermique jouent un rôle important dans les propriétés de transfert des interactions
fluide-roche et pour caractériser les matériaux de construction. La compréhension de la relation entre
ces deux paramètres pourrait être un bon moyen pour comprendre les interactions fluide-roche entre
la microstructure et le réseau poreux et de caractériser la porosité d'un milieu poreux.
2. Matériels et méthodes
Les matériaux utilisés dans cette étude sont des roches sédimentaires calcarénitiques. Des
mesures de la porosité à l’eau et de la conductivité thermique ont été effectués sur des échantillons
3
7×7×7 cm . Ces mesures permettront de mieux comprendre et cerner des caractéristiques du réseau
poreux des matériaux utilisés dans la construction du monument Chella romain Rabat. Le lecteur
2.1. Porosité
La méthode de mesure de la porosité totale à l’eau est celle définie par la norme RILEM [5], qui
prévoit que les échantillons soient imbibés en l’absence d’air, c’est à dire en régime monophasique.
La valeur de la porosité Nt, est ensuite obtenue par la méthode de la triple pesée. Connaissant le
poids sec de l’échantillon W s et son poids imbibé W 1, on obtient le volume poreux occupé en totalité
par l’eau (W 1-W s), puis à partir d’une pesée hydrostatique W 2, on peut calculer le volume total de
l’échantillon (Vt=W 1-W 2). La porosité de l’échantillon s’exprime par :
W − Ws
N t (%) = 1 × 100 (1)
W1 − W2
avec W 1 le poids de l’échantillon saturé en eau, W 2 le poids saturé obtenu par pesée hydrostatique,
W s le poids sec de l’échantillon.
Les deux capteurs de température (T1 et T2) permettent de mesurer la différence de température
due à l’apport d’énergie. La distance entre la source et le capteur thermique mesurant la température
de l’échantillon après échauffement, ainsi que la vitesse de déplacement sont constantes, il est donc
possible de lier l’augmentation de température au point x, θ(x) et la conductivité thermique par une
relation de la forme suivante :
q
θ ( x) = (2)
2π x λ
Figure 2: Disposition des standards et de l’échantillon lors de la mesure de conductivité thermique, schéma
des profils de température.
3. Résultats et discutions
3.1. Porosité
D’après les mesures de différents poids des échantillons qui sont extrait du même bloc (même
minéralogie et pétrographie), on constate que ces échantillons ont une différence de porosité (figure
3). La porosité totate de ces échantillons varie entre 25,69 et 35,83%.
38,00
36,00
34,00
32,00
Porosité (% )
30,00
28,00
26,00
24,00
22,00
20,00
ech1 ech2 ech3 ech4 ech5 ech6
Echantillons
1,20
Conductivité thermique λ (W/m.K)
1,10
1,00
0,90
0,80
0,50
0,40
24,00 26,00 28,00 30,00 32,00 34,00 36,00 38,00
Porosité (%)
1,85
Conductivité thermique λ (W/m.K)
1,7
1,65
1,6
1,55
1,5
24,00 26,00 28,00 30,00 32,00 34,00 36,00 38,00
Porosité (%)
Figure 5 : Variation de la conductivité thermique des échantillons saturés en eau avec la porosité.
Pour la plupart des roches et des minéraux, pour une composition chimique et une structure
données, la conductivité thermique augmente quand la densité augmente [7,8]. Sur la figure 3 on a
représenté la relation entre de la conductivité thermique et la densité des roches sédimentaires. Le
graphe de la figure 3 montre qu’il y a bien une relation entre la densité et la conductivité thermique
des six échantillons. Cette relation apparaît quasi-linéaire (coefficient de régression = 0,9028), ce qui
est également conforme avec les résultats bibliographiques [8].
1,20
1,00
0,90
0,60
0,50
0,40
1580 1600 1620 1640 1660 1680 1700 1720 1740 1760
Densité (kg/m 3)
4. Conclusions
Nous avons utilisé la méthode de la porosité à l’eau pour déterminer la porosité totale des roches
étudiées, et la méthode TCS pour caractériser la conductivité de ces roches, cette technique permet
de définir les relations entre la porosité, la densité et la conductivité thermique.
Les premiers résultats obtenus montrent que conductivité thermique diminuent avec l’augmentation
de la porosité et que le fluide a un effet sur la conductivité. Ces résultats indiquent que la conductivité
thermique dépend de la porosité, de la structure de réseaux poreux et de la circulation du fluide dans
la roche.
Remerciements :
Ce travail a été réalisé dans le cadre du projet d’action intégrée Franco-Marocaine No. MA/07/168,
en collaboration avec nos partenaires de l’Institut Scientifique Rabat et l’Institut de Physique du Globe,
Strasbourg.
Références
[1] Y.A. Popov, et al. Characterization of rock thermal conductivity by high - resolution optical
scanning. Geothermics 28, pp: 253-276, 1999.
[2] C. Gruescu et al. Effective thermal conductivity of partially saturated porous rocks. International
Journal of Solids and Structures 44, pp: 811–833, 2007.
[3] C.Clauser and E. Huenges. Thermal Conductivity of Rocks and Minerals. The American
Geophysical Union, pp: 105-126, 1995.
[4] A. Samaouali et al. Porosity and mineralogy evolution during the decay process involved in the
Chellah monument stones. Environ Earth Sci 59, pp: 1171–1181, 2010.
[5] M. Rosener. Etude pétrophysique et modélisation des effets des transferts thermiques entre roche
et fluide dans le contexte géothermique de Soultz-Sous-Forêts. Thèse de doctorat, Université Luis
Pasteur, Strasbourg, 2007.
[6] F.Surma and Y. Geraud. Porosity and Thermal Conductivity of the Soultz-Sous-Forêts Granite.
Pure and Applied Geophysics 160, pp: 1125–1136, 2003.
[7] K. Besnard. Evolution physico-chimique des matériaux carbonatés en milieu triphasique. DEA
Hydrologie, Hydrogéologie, Géostatistique et Géochimie, Université Paris-Sud, 2000.
[8] A. Hartmann et al. Thermal conductivity from core and well log data. International Journal of Rock
Mechanics & Mining Sciences 42, pp: 1042–1055, 2005.
Résumé:
Dans ce travail des propriétés pétrophysiques de cinq roches différentes ont été mesurées et analysées. Il
s’agit de la conductivité thermique, la vitesse de propagation d’onde P et la porosité. En ce qui concerne
la conductivité thermique, on impose à un échantillon un flux de chaleur unidirectionnel et on mesure la
différence de température aux bornes de l’échantillon. Le test ultrasonique est basé sur la mesure du temps
de propagation d’une onde P dans le sens longitudinal. La porosité est mesurée par la technique de
porosimétrie au mercure, basée sur la mesure de la quantité de mercure insérée dans les pores d’une roche
sous différentes pressions (intrusion de mercure). Les résultats obtenus montrent un effet direct de la
porosité sur La conductivité thermique et la vitesse d’onde P. La relation entre La conductivité thermique
et la vitesse d’onde P est étudiée. Cette relation apparaît quasi-linéaire. Les résultats ont été comparés à
plusieurs modèles théoriques et d’autres résultats expérimentaux présentés dans la littérature. Ces
résultats, conformément à la théorie, montrent la possibilité d’estimer la conductivité thermique à partir de
la vitesse d’onde P et par l’utilisation des méthodes non destructives.
1. INTRODUCTION
Les matériaux d’origine naturelle (roches, monuments, sols, composites) sont des milieux poreux,
dont la microstructure présente une grande diversité. Ils sont constitués d’une phase solide généralement
hétérogène et une phase poreuse qui peut être totalement ou partiellement saturée par un ou plusieurs
fluides. L’étude du rôle de la porosité, de la conductivité thermique et de la vitesse des ondes P dans ces
milieux poreux s’avère fondamentale pour des applications comme la géophysique de la subsurface et des
réservoirs, la sismologie et le domaine de la construction.
De nombreux travaux ont montré que la conductivité thermique d’une roche poreuse dépend
essentiellement de la composition minéralogique, la porosité, la présence des fluides remplissant les
pores; et de la température et la pression [1]. La porosité et la conductivité thermique jouent un rôle
important dans les propriétés de transfert des interactions fluide-roche et pour caractériser les matériaux
de construction. La compréhension de la relation entre ces deux paramètres pourrait être un bon moyen
pour comprendre les interactions fluide-roche entre la microstructure et de caractériser la porosité d'un
matériau. La méthode TCS (Thermal Conductivity Scanning) est une technique utilisée pour des mesures
thermo-physiques. Des études théoriques et expérimentales ont été effectuée sur cette méthode [2, 3]. Elle
est caractérisée par la mesure non destructive des propriétés thermiques d'un grand nombre de minéraux et
de roches.
---------------------------------------
* Auteur correspondant: cocanouar@gmail.com
2. MATERIALS ET METHODES
L’étude a porté sur cinq variétés de pierres de construction extraites du Maroc, volontairement très variés,
se présentaient sous forme 5.5 × 5.5 × 11 cm3. La localisation des échantillons sélectionnés est indiquée sur
le tableau 1. Des mesures de la porosité, de la conductivité thermique, et de la vitesse des ondes P ont été
effectuées sur ces échantillons. Ces mesures permettront de mieux comprendre la relation entre les
propriétés pétrophysiques des matériaux utilisés dans la construction.
3. RESULTATS ET DISCUTIONS
3.1. POROSITE
La porosité contrôle tous les autres paramètres physiques de la roche (densité, perméabilité,
conductivité thermique…). Elle correspond au rapport du volume total des pores sur le volume total de la
roche. Dans la pratique, seule la porosité ouverte d'une roche est mesurable. L'espace aménagé entre les
grains et par les microfissures est souvent comptabilisé avec le volume de la porosité ouverte. La
porosimétrie au mercure est la méthode la plus largement utilisée; elle consiste à mesurer le volume de
mercure susceptible d'être absorbé par une roche (par injection à haute pression sous vide). Dans ce
travail, nous avons utilisé la porosimétrie au mercure.
La roche de Rabat-Salé présente une forte porosité des roches sédimentaires (28 %). la calcarénite de
Taza et le travertin de Meknès sont relativement moins poreux (10.58 et 5.26 %). Les granites d’Agadir et
les marbres (Khénifra) présentent une porosité assez faible (respectivement 0,37 et 0,59%).
3 3
y = -0,0649x + 2,7907
2,5 R2 = 0,9138 2,5
(W /m .K )
(W /m .K )
2 2
0,5 0,5
0 0
0,00 5,00 10,00 15,00 20,00 25,00 30,00 0,00 5,00 10,00 15,00 20,00 25,00 30,00
Porosité (%) Porosité (%)
(a) (b)
Figure 1: Variation de la conductivité thermique avec la porosité des échantillons:(a) secs et (b) saturés.
3.1. VITESSE DES ONDES P- CONDUCTIVITE THREMIQUE
Les vitesses des échantillons sont mesurées à l’état sec et à l’état saturé. Pour les mesures à l’état sec,
les éprouvettes sont séchées à l’étuve à 105 °C pendant 24 heures. Par contre, l’état saturé des éprouvettes
est obtenu en les immergeant graduellement dans l’eau. Les résultats obtenus pour la mesure de la vitesse
des ondes P des échantillons secs varient de 3007 à 5960 m/s et celles des échantillons saturés en eau
varient entre 2900 et 5212 m/s. Ces résultats sont en accord avec les résultats bibliographiques [9]. D’une
manière générale, nous observons que la valeur la plus élevée de la vitesse de l’onde P est obtenue pour
Marbre noir de Khenifra et la valeur la plus faible est obtenue pour biocalcarénite de Rabat-Salé.
Sur les figures 2 (a) et (b), on a représenté la relation entre la vitesse des ondes P et la conductivité des
échantillons saturés et celle des échantillons secs. Cette relation apparaît quasi-linéaire de coefficient de
régression R = 0,814 à l’état sec et R=0.9096 à l’état saturé.
7000 6000
V ite s s e d e s o n d e s P (m /s )
V i te s s e d e s o n d e s P (m / s )
6000 5000
5000
4000
4000
3000
3000
y = 1249x + 1984,7 y = 2415,9x - 1936,6
2000
2000 R2 = 0,814 R2 = 0,9096
1000 1000
0 0
0,8 1,3 1,8 2,3 2,8 3,3 1,9 2,1 2,3 2,5 2,7 2,9 3,1
Conductivité thermique λ (W/m.K) Conductivité thermique λ (W/m.K)
(a) (b)
Figure 2: Vitesse des ondes P en fonction de la conductivité des échantillons:(a) secs et (b) saturés.
Comme résultat, la relation entre les valeurs estimées et les valeurs mesurées de la conductivité thermique
pour chaque échantillon est examinée graphiquement sur les figures 3 (a) et (b). L'erreur de la valeur
estimée est représentée par la distance de chaque point de données des courbes de la ligne diagonale 1:1.
Un point situé sur la ligne indique une estimation précise, comme le montre sur les figures 3 (a) et (b).
3,5
3,0
2,5 2,8
2,0
2,6
1,5
2,4
1,0
2,2
0,5
0,0 2,0
0,0 0,5 1,0 1,5 2,0 2,5 3,0 3,5 2,0 2,2 2,4 2,6 2,8 3,0
(a) (b)
Figure 3: Conductivité estimée, et Conductivité mesurée à partir de vitesse des ondes P des échantillons:(a)
secs et (b) saturés.
4. CONLUSIONS
Dans ce travail, des propriétés pétrophysiques de 5 échantillons ont été mesurées et analysées L’intérêt de
cette étude est porté sur des échantillons de roches extraites du Maroc pour ramasser des données
individuelles de chaque type de roche qui connaît une rareté des études techniques. Elle apporte des
résultats sur les vitesses d’onde P, la conductivité thermique et la porosité pour bien caractériser ces
roches. Cette étude statistique a montré que la conductivité thermique des roches peut être estimée en
utilisant des méthodes simples, non destructive telle que la vitesse des ondes P.
REMERCIEMENTS
Ce travail a été réalisé dans le cadre du projet d’action intégrée Franco-Marocaine No. MA/07/168, en
collaboration avec nos partenaires de l’Institut Scientifique Rabat et l’Institut de Physique du Globe,
Strasbourg.
REFERENCES
[1] Clauser, C., Huenges, E. 1995.Thermal Conductivity of Rocks and Minerals. The American Geophysical Union,
pp: 105-126.
[2] Popov, Y., Pribnow; D., Sass; J.H., Williams, C.F & Burkhardt, H. 1999. Characterization of rock thermal
conductivity by high resolution optical scanning, Geothermics, 28, 253-276.
[3] Surma, F., Géraud; Y. 2003. Porosity and Thermal Conductivity of the Soultz-Sous-Forêts Granite. Pure and
Applied Geophysics 160, pp: 1125–1136.
[4] Matthieu, A. 2007. Multiscale study of stone decay by salt crystallization in porous networks. Thesis (PhD).
Université de Cergy-Pontoise, France.
[5] Ezzdine, R. 2006. Endommagement des monuments historiques en maçonnerie. Thesis (PhD). Université
Bordeaux 1, France.
[6] Rosener, M. 2007. Etude pétrophysique et modélisation des effets des transferts thermiques entre roche et fluide
dans le contexte géothermique de Soultz-Sous-Forêts. Thesis (PhD). Université Luis Pasteur, Strasbourg, France.
45
11ème congrès de Mécanique 23-26 Avril 2013 - Agadir (Maroc)
1,80
A. Dry
1,2
1,75
Thermal Conductivity (W/m K)
1,1
1
1,70
0,9 y = 1,069x - 2,9597
R2 = 0,9684
0,8
1,65
46
11ème congrès de Mécanique 23-26 Avril 2013 - Agadir (Maroc)
Conclusions
The relations, thermal conductivity against P wave
velocity has the best correlation R=0:9684 (A. Dry) and
R=0.9277 (B. Saturated). This means that the thermal
conductivity of rock can readily be calculated from
laboratory determined P-wave velocity, from the relations:
y = 1.069 x − 2.9597 (A. Dry)
47
CMSS 2013: International Congress on Materials & Structural Stability
Building up Sustainable Materials & Constructions
Abstract:
In this study we focused on seismic attenuation (1/Qc) tomography in Nothern Morocco. For this purpose
two different models are employed : The Single Backscattering model hypothesis of Aki and Chouet (1975)
to calculate values of coda Q (Qc) and the Back Projection technique of Xie and Mitchell (1990) to estimate
lateral variation in Qc via a tomographic inversion . For this investigation, the coda Q method is applied to
a number of 94 local earthquakes with a magnitude between 0.7 and 4. The digital seismograms of these
earthquakes were recorded during the year 2008 by both local temporary and permanent broadband seismic
station network deployed in Northern of Morocco. The Qc quality factor values have been computed at cen-
tral frequencies 0.75, 1.5, 3, 6 and 12 Hz .The lapse time windows is restricted to 30s in order to sample
the earth’s crust only. The Qc results indicate that strong frequency dependence follow a power law for the
entire area. The preliminary results of seismic coda Qc attenuation tomography shows a dependence at each
frequency band, between seismic attenuation and the geology structure units in the study area, especially in the
region of Al Hoceïma and the eastern part of the Rif which are characterized by a high attenuation values due
to active faults area, while a low attenuation values are seen in the west and the south of the Rif in high frequencies.
Keywords: Single Backscattering model, Qc quality factor, Back Projection technique , Attenuation, Tomo-
graphic inversion , Northern Morocco.
C M G A ,M ,
Abstrat :
The attenuation characteristics based on coda waves of Northern Morocco have been
investigated in the present study. A number of 92 local earthquake events recorded digitally by
29 seismic stations, have been selected for estimation of coda wave attenuation with a
magnitude between 0.7 and 3.4. The Qc quality factor values have been computed using the
single backscattering model at central frequencies 0.75, 1.5, 3, 6 and 12 Hz, with four lapse
time windows from 30 sec to 60 seconds duration with a difference of 10 seconds .The
earthquakes have their epicentral distance within 100 km. The mean values of the estimated
Qc vary from 83 (at 0.75Hz) to 1260 (at 12Hz) for 30 seconds. For 40 seconds Qc vary from
127 (at 0.75Hz) to 1406 (at 12 Hz). Similarly for 50 seconds Qc vary from 142 to 1469 and
for 60 second Qc vary from 175 to 1552 .The coda Qc values at four lapse time windows
were found to be Qc = 94.5 f 0.96 (for 30 seconds ) , Qc = 139.5 f 0.86 ( for 40 seconds) ,
Qc = 156.09 f 0.85 ( for 50 seconds ) and Qc = 180.3 f 0.81 (for 60 seconds ) . It is observed that
Qc value for the study area have a strong frequency dependent which could be related to the
size of heterogeneities. The increase in Qc values with lapse time shows the depth dependence
of Qc as longer lapse time windows will sample larger area which agree with many studies
and has been interpreted to be due to increasing Q with depth. The results obtained with this
model are then presented and compared to results from the literature.
Keywords : Attenuation , quality factor, Qc , Northern Morocco, Local earthquake, Single
backscattering
Introduction :
Morocco is a transition zone between the African and European plates. The seismic history
mentions several catastrophic earthquakes in this region. In particular, Northern Morocco has
been most affected by earthquakes in the past .The last relevant one is 24th February 2004
earthquake (Mw = 6.1) which caused great damage in Al Hoceima city and the surrounding
area. One of the important factors in the assessment of the earthquakes in a region is to know
the attenuation characteristics. The attenuation propriety is described by the quality factor Q
.Numerous studies have been done worldwide to understand the attenuation during its
propagation in the earth (Aki, K. (1969), Woodgold, C. R. D. (1994).
Single Backscattering model of Aki and Chouet (1975) has been chosen to be used in the
present study because it facilitates the comparison with other studies made in different zones
within Northern Morocco and elsewhere. The results will be compared with other Q
measurements in other region in the world. To the best of your knowledge, this is the first
study of seismic wave attenuation in the region.
Data
For the estimation of the quality factor Qc a group of 29 stations and 92 events are used, with
magnitudes of between 0.7 and 3.4 recorded in the region during 2008. All instruments are
three components with broadband digital seismic stations .The detailed information regarding
these events in the Fig1.
The Qc value are calculated through the CODAQ subroutine of SEISAN 8.0 [Havskov, J., and
L. Ottemoller (2003)] .And the origin time, epicentre locations and magnitudes of the
earthquakes used were determined by SEISAN Software .
C M G A ,M ,
Fig1 .Locations of the events used in this study .Circles are the events and triangles are the stations selected for coda Q.
Methods of Analysis
In the single backscattering method according to Aki and chouet (1975), the displacement of
π f .t
coda waves is described as : A( f , t ) = S ( f ) t −1 exp( ) (1)
Qc
Where Qc is the quality factor , t −1 is the lapse time and S ( f ) is the coda source factor at
radial frequency .By taking the natural logarithms of equation (1) , the equation can be written
as ln [ A( f , t ).t ] = C − b.t ,Where b and C equal to −π . f / Qc and ln [ S ( f )] respectively.
Qc can be obtained at different frequency f from the slope (=b)
Results
3,5 3,5 y = 0,8657x + 2,1447
y = 0,9614x + 1,9894
3 R2 = 0,9966 3 R2 = 0,9898
2,5 2,5
l o g (Q )
2
lo g (Q )
2
1,5 1,5
1 1
0,5 0,5
0 0
-0,2 0 0,2 0,4 0,6 0,8 1 1,2 -0,2 0 0,2 0,4 0,6 0,8 1 1,2
log(f) log(f)
Fig2 :Plot of mean value of Qc for w=30s Fig3 : Plot of mean value of Qc for w=40s
y = 0,8137x + 2,256
3,5 y = 0,8549x + 2,1934 3,5
R2 = 0,97
R2 = 0,9847
3 3
2,5 2,5
log(Q c)
2 2
Log(f)
1,5 1,5
1 1
0,5 0,5
0 0
-0,2 0 0,2 0,4 0,6 0,8 1 1,2 -0,2 0 0,2 0,4 0,6 0,8 1 1,2
log(f) Log(Qc)
Fig4 : Plot of mean value of Qc for w=50s Fig5 : Plot of mean value of Qc for w=60s
C M G A ,M ,
0.84
Zagros 99f
1200 S.Iberia 100f
0.70
800 0,01
New England 460f
Qc for w=30s dist 100
-1
Qc
0.35
600 NE U.S. 900f
Qc for w=40s) dist 100 Central U.S. 1000f
0.2
400
Qc for w=50s dist 100
200
Qc for w=60s dist 100 1E-3
0
0 2 4 6 8 10 12 14
1 10
Frequency Frequency (Hz)
Fig.6: Plot of average values of Qc with lapse time Fig.7: Comparison of coda-Qc of North of Morocco region
at different frequencies with reported coda-Q of other regions of the world
Conclusion
In this work, coda Qc has been estimated for North Morocco using a single back-scattering
model .The following conclusions have been reached in this study:
The Qc relation for the Northern Morocco, is in the form of: Qc = 157 f 0.87 .
The frequency dependent relationships for lapse time windows were found to be
Qc = 94.5 f 0.96 (for 30 seconds ) , Qc = 139.5 f 0.86 ( for 40 seconds) , Qc = 156.09 f 0.85 ( for
50 seconds ) and Qc = 180.3 f 0.81 (for 60 seconds ) .The observed quality factor is strongly
dependent on frequency and lapse time .
The comparison of estimated Qc with other region indicates that the North of Morocco is
moderately high seismic hazard.
The frequency dependence coefficients (Aki 1980) vary between 0.8 and 0.9, which indicates
moderately high seismic hazard.
References
Aki, K. (1969). Analysis of the seismic coda of local earthquakes as scattered waves, J.
Geophys. Res. 74, 615–631.
Aki, K.(1973) .Scattering of P-waves under Montana LASA, J.Geophys. Res. 78, 1334–1346.
Aki, K. , and B.Chouet (1975) . Origin of the coda waves: source attenuation and scattering
effects, J. Geophys. Res.80 , 3322–3342.
Havskov, J., and L. Ottemoller (2003). SEISAN: the earthquake analysis software for
windows, Solaris, Linux, and Mac OSX Version 8.0,244 pp.
Woodgold, C. R. D. (1994). Coda Q in the Charlevoix, Quebec, region: lapse-time
dependence and spatial and temporal comparisons, Bull. Seism. Soc. Am. 84, 1123-1131.
11ème congrès de Mécanique 23-26 Avril 2013 - Agadir (Maroc)
PREDICTION DES PROPRIETES ELASTIQUES ET ACOUSTIQUES EFFECTIVES
DES MILIEUX POREUX
Hashin et Strickman (Hashin et al; 1963) développent une Dans un milieu homogène isotrope et élastique, la
approche variationnelle qui permet d’identifier, pour les vitesse des ondes de propagation plane de type compressif
milieux isotopes, un domaine plus étroit pour les modules (onde P) ou de cisaillant (onde S) est donnée par :
effectifs que les moyennes de Voigt et Reuss. Ces modèles 4
k+ µ
conduisent à un encadrement assez large des modules Vp = 3 ,V =
µ (5)
élastiques de milieux composites, essentiellement lorsque ρ s
ρ
les phases en présence ont des propriétés élastiques très
distinctes. 2.1.1. Propriétés effectives d’un milieu poreux sec
Certains auteurs ont appliqués des modèles pour
décrire le comportement des milieux poreux. Ainsi En mettant K inc = 0 et µ inc = 0 dans les équations (3) et
Guéguen (1997) a utilisé le modèle différentiel auto (4), on obtient les propriétés élastiques ( ksec *
, µ sec
* *
, E sec )
cohérent pour déterminer les modules élastiques des
milieux poreux. Sarout (2006) a utilisé le modèle Mori- dans le cas du milieu poreux contenant des inclusions vides.
Tanaka pour décrire la propagation des ondes élastiques
*
k sec = k m (1 − f p )(1 + af p ) −1 (6)
dans des roches argileuses sèches.
= µ m (1 − f p )(1 + bf p )
−1
µ sec
* (7)
Cette étude concerne la modélisation du comportement
= E0 (1 − f p )(1 + cf p )
mécanique et acoustique des milieux poreux en utilisant * −1 (8)
Esec
une technique d’homogénéisation basée sur le modèle de
Mori-Tanaka, dont le but est de relier les propriétés Les propriétés acoustiques effectives (VP, VS) dans le cas du
mécaniques et acoustiques effectives à la microstructure de milieu poreux sec contenant des inclusions sphériques sont:
ces matériaux. Une technique expérimentale ultrasonique a
été utilisée pour déterminer les propriétés élastiques à partir (1 + af p ) , 1
V p = V pm Vs = Vsm
(1 + bf p )(1 + cf p ) (1 + cf p )
de mesure des vitesses des ondes P.
2. Méthode d’homogénéisation 2.2.2. Propriétés effectives d’un milieu poreux saturé
La méthode d’homogénéisation consiste généralement Dans le cas d’un milieu poreux saturé par un fluide, le
à remplacer un milieu réel hétérogène par un milieu fictif tenseur élastique s’écrit :
homogène équivalent (Xu, 2004). Nous considérons un
MT
C sat = C f − (1 − f p )(C f − C m )[ (1 − f p ) I +
milieu biphasique élastique linéaire isotrope constitue d’une [
f p I + S E : C m−1 : (C f − C m ) ] ]
−1 −1
15
11ème congrès de Mécanique 23-26 Avril 2013 - Agadir (Maroc)
1 (1 − af p ) 4 (1 − f p )
V psat = km + µm
ρ m (1 − f p ) + ρ f f p (1 + bf p ) 3 (1 + cf p ) 50
Eth
µm (1 − f p ) 45
Vssat = Eexp
ρ m (1 − f p ) + ρ f f p (1 + c f p ) 40
E (G Pa)
30
4,3
25
th
4,2 VP
exp
4,1 VP 20
4,0 15
3,9
VP (km/s)
10
3,8 0 10 20 30 40 50
3,7 fp (%)
3,6
3,5
Figure 4: Comparaison des mesures du Module de Young E
3,4
en fonction de la porosité pour des calcarenites saturées
3,3
0 10 20 30 40 50
avec les prédictions du modèle Mori-Tanaka.
fp (%)
Les matériaux utilisés dans cette étude sont des roches
sédimentaires calcarenitiques prélevées dans une carrière à
Figure 1: Comparaison des données de vitesse d'ondes P en proximité de la ville de Rabat. Nous avons utilisé la
fonction de la porosité pour des calcarenites secs avec les méthode de la porosité à l’eau pour déterminer la porosité
prédictions du modèle Mori-Tanaka (Km=63.3 GPa, µ m= totale des échantillons. Une technique ultrasonique a été
17.1 GPa (Lion, 2004)). utilisée pour déterminer les vitesses des ondes P des
échantillons.
50
45
Eth 3. Résultats et discussion
Eexp
40
D’après les résultats de mesures, on constate que ces
35 échantillons ont une différence de porosité. La porosité
E (GPa)
16
11ème congrès de Mécanique 23-26 Avril 2013 - Agadir (Maroc)
Le module de compression K* est plus grand en milieu méthodes d’homogénéisation. Thèse de Doctorat, Institut
saturé, le module de cisaillement µ* reste inchangé et la National Polytechnique de Lorraine.
vitesse VP est plus importante en milieu sec, ce dernier Guéguen, Y., Chelidze, T., Le Ravalec, M., 1997.
résultat est interprété par le fait que la densité est plus grand Microstructures, percolation thresholds, and rock physical
en milieu saturé. Contrairement des auteurs (Guéguen et al, properties. Tectonophysics 279, pp. 23-35.
1992) ont comparé les vitesses VP et VS en milieu sec (S=0)
Le Ravalec, M ., 1995. Vitesses et perméabilité des roches :
et saturé en eau (S=1), ils ont observé en général VP (sec) <
modélisation du rôle des fluides et des fissures. Thèse de
VP (saturé) et VS (saturé) < VS (sec) (confirmation de
doctorat, Université de Rennes.
l’étude expérimentale). Beck (2006) a observé que la
vitesse de compression VP est plus importante dans l’état Rahmouni. A et al., 2010. Influence de la microstructure et
saturé pour le tuffeau blanc et la pierre de Sébastopol. Il est de la saturation en fluide sur la conductivité thermique des
interprété ce résultat par le fait que le module de milieux hétérogènes et poreux. 1er Congrès de l'Association
compression est plus grand en milieu saturé et que la Marocaine de Thermique, 6-7 Mai Settat (Maroc).
vitesse du son est plus grande dans l’eau que dans l’air. Sarout, J., 2006. Propriétés physiques et anisotropie des
roches argileuses : modélisation micromécanique et
4. Conclusions
expérience triaxiales. Thèse de Doctorat, Université Paris
XI.
Nous avons étudié un milieu constitué d’une matrice
solide contenant des inclusions sphériques vides ou saturées Xu, Y., 2004. Approches multi-échelle pour l’étude du
en utilisant le schéma de Mori-Tanaka. Les résultats comportement des systèmes polyphasiques - application
obtenus montrent que les propriétés élastiques et aux milieux poreux non saturés. Thèse de Doctorat, Ecole
acoustiques effectives diminuent avec l’augmentation de la National des Ponts et chausses.
porosité et que le fluide n’a pas un effet sur les vitesses Vs.
Le modèle Mori-Tanaka présente d’une part l’avantage de
la détermination de modules élastiques effectifs même pour
un milieu dont espace poreux est vide, et d’autre par ce
modèle offre l’avantage de prendre en compte les effets
d’interaction entre les pores, ce qui légitime le calcul des
propriétés élastiques et acoustiques effectives jusque des
valeurs élevées de la porosité.
Les vitesses acoustiques dans un milieu dépendent de
ses propriétés élastiques et de sa densité volumique qui sont
générées par la matrice solide (composition minéralogiques,
taille des grains) et l’espace poreux (porosité, géométrie des
inclusions) associé qui est saturé ou non saturé par un ou
plusieurs fluides. L’étude des propriétés effectives des états
secs et saturés par l’utilisation de ce modèle est simple, car
ils représentent un milieu biphasique constitue d’une phase
solide (matrice) et d’une phase fluide (eau ou air), mais les
états partiellement saturés sont plus complexes, car ils
représentent un milieu constitue de trois phases : la matrice
solide et deux phases fluides (eau et air) c’est l’objet d’une
autre étude.
Remerciements
Ce travail a été réalisé dans le cadre du projet d’action
intégrée Franco-Marocaine No. MA/07/168, en
collaboration avec nos partenaires de l’Institut Scientifique
Rabat et l’Institut de Physique du Globe, Strasbourg.
Références
17
Sixième Colloque Maghrébin de Géophysique Appliquée, Meknès les 3, 4 et 5 mai 2013
Abstract
The determination of P-wave velocity is an easy and simple task and it can be determined
in field as well as in laboratory. Determination of P-wave velocity is non-destructive and easy
to apply, that is why it is increasingly being used in geotechnical engineering. The present
work deals with the use of simple and non destructive technique, ultrasonic velocity to predict
the porosity of calcarenite that are characteristic in historical monument. Good correlations
between P-wave velocity and porosity were found, which indicate them as appropriate
technique for estimating the porosity.
Keywords: Calcarenite, Ultrasonic velocity, Porosity; P-wave velocity.
Introduction
Ultrasonic velocity measurements have been used for many years in geotechnical
practice and mining science. They are widely used to determine properties and states of
materials, in the case of engineering solids measurements of ultrasonic wave propagation
velocities are routinely used to determine elastic constants (Vasconcelos et al. 2007). There
has been an Increasing use of ultrasonic velocity measurements for nondestructive
characterization of material microstructures and mechanical properties (Kahraman et al.
2008). Therefore, it is Important to have appropriate practical methods for making velocity
measurements on a variety of material samples.
Most researchers (Assefa et al. 2003; Soroush, H. et al. 2011) demonstrated that the
prediction of petrophysical properties in rocks is often difficult due to possession of complex
textures and petrophysical properties- porosity and permeability, which are dependable and
influenced by the diagenetic processes they may have undergone right from the time of
deposition to the late diagenesis set in. Certain studies that combined seismic, petrophysical
and petrological data to established useful relationship between velocity and petrophysical of
rocks were carried out by Yasar et al. (2004) and Kahraman et al. (2008)
The study of compressional and shear wave propagation a material to dry and saturated
evaluates the physical properties of materials such as porosity, cracking state and its elastic
properties in modeling of the wave propagation. For rocks, it is often interested in the
compressional waves propagation (P-wave) and shear waves (S-wave). For a P-wave, the
vibration occurs in the direction of propagation and affects the volume of the rock. For S-
wave, the vibration propagates in a plane normal to the direction of propagation; it is slower
than the P-wave and does not propagate in water. The P wave and S-wave are characterized
by their velocities VP and VS respectively and their attenuation.
The paper further investigates the result of the laboratory measurement of acoustic
velocity in calcarenite samples. This study explores velocity data- compresional wave and
porosity. Correlations between the P-wave velocity value and the porosity value obtained
from laboratory measurements were investigated.
Materials and experimental procedure
The materials used in this study are sedimentary rocks calcarenitiques taken from a
quarry near the town of Rabat. Calcarenite monuments of Rabat is a rock of Plio-Quaternary
1
Sixième Colloque Maghrébin de Géophysique Appliquée, Meknès les 3, 4 et 5 mai 2013
age, is characterized by high porosity and variable (18-47%) (Zaouia et al.2005). Its chemical
composition is very rich in calcium carbonate. Measurements of P-wave velocity and porosity
of were carried out on samples 7×7×7 cm3.
Porosity
The method of measuring the total porosity to water is that defined by (Rosner 2007),
which provides that the samples are soaked in the absence of air, ie in monophasic system.
The porosity value Nt, is then obtained by the method of three weighing. Knowing the dry
weight of the sample Ws and its weight soaked W1, we obtain the total pore volume occupied
by water (W1-Ws), then from a hydrostatic weighing W2, we can calculate the total volume of
the sample (Vt = W1-W2). The porosity of the sample is given by:
W − Ws (1)
N t (%) = 1 × 100
W1 − W2
With W1 the weight of the sample saturated with water, W2 weight saturated obtained by
hydrostatic weighing, Ws is the dry weight of the sample.
P-wave velocity
Measuring propagation velocity of ultrasonic waves in a material such as stones is ways
to further investigation of the mechanical tests have the advantage of being non-destructive
testing (Yasar et al. 2004). Wave propagation in solids is based on the theories of elasticity
that link efforts and low strains. These elastic waves are polarized in two perpendicular
directions in the two plane waves: longitudinal compression wave called the primary wave (P-
wave) representing the vibration of particles parallel to the direction of wave propagation, and
transverse wave called shear secondary wave (S-wave), representing the vibration of particles
perpendicular to the direction of propagation.
The measuring principle is based on the production of a pulse train of vibration using a
transmitter (transducer) applied on one side of the sample to be tested. After crossing a path
length (L) in the rock, the impulse of vibration is converted into an electrical signal with a
receiver (transducer) and the time base to display the electronic delay ultrasonic waves (t). As
soon as the measured value is stable for 3 seconds, a beep sounds and the speed of sound
appears. The P-wave velocity is given by the formula:
L
V= (2)
t
For there to be a perfect contact between the rocks and the transducers, it is necessary to
sand and level the part of the surface where the transducer is fixed, and it is recommended to
use a paste made of coupling kaolin and glycerol at both ends of rocks and taking care to
ensure that the transducers are pressed against the surface to be tested. The standard 54 kHz
transducers are calibrated on the display unit must be controlled by a calibration test. The
calibration factor is marked on the standard toolbar (20.6 m).
Results and discussions
Based on measurements of different sample weights that are extracted from the same
block (same mineralogy and petrography) shows that these samples have a difference of
porosity. The total porosity of the samples varies between 25.69 and 35.83%. The average
porosity is 31.83%.
The results obtained for the measurement of P-wave velocities of dry samples range from
3.57 to 3.8 km/s and those of water saturated samples vary between 3.6 and 3.9 km/s. These
results are consistent with literature results (Boulanouar et al 2012).
The results of P-wave velocity and porosity of the rocks were analyzed using the method
of least square regression method. The equation of best fit line and coefficient of
determination (R2) were determined for each regression. The graphs between P-wave velocity
and porosity of saturated samples and dry samples are shown in Figs. 1 and 2.
2
Sixième Colloque Maghrébin de Géophysique Appliquée, Meknès les 3, 4 et 5 mai 2013
3,85
3,85
3,8 3,8
y = -0,021x + 4,3239 y = -0,0224x + 4,4008
R2 = 0,8899 R2 = 0,9356
3,75
3,75
3,7
3,7
3,65
3,65
3,6
3,55 3,6
3,5 3,55
24,00 26,00 28,00 30,00 32,00 34,00 36,00 38,00 24,00 26,00 28,00 30,00 32,00 34,00 36,00 38,00
Porosity (%) Porosity (%)
Fig.1. P-wave velocity versus porosity Fig.2. P-wave velocity versus porosity
for dry samples. for saturated samples.
38 38
36 36
Estimated porosity (%)
32 32
30 30
28 28
26 26
24 24
24 26 28 30 32 34 36 38 24 26 28 30 32 34 36 38
Measured porosity (%) Measured porosity (%)
Fig.3. Estimated porosity and measured porosity Fig.4. Estimated porosity and measured porosity
from P-wave velocity for dry samples. from P-wave velocity for saturated samples
Data from each test were used in the developed empirical equations to estimate the
porosity of the calcarenite rocks. The estimated values of porosity of the calcarenite rocks
were plotted against the measured values for each test (Figs.3 and 4). The error in the
estimated value is represented by the distance of each data point from the 1:1 slope line. A
point lying on the 1:1 slope line indicates an exact estimation, whereas, away from the line
shows the error, as shown in Figs.3 and 4.
Conclusions
The graphs of P-wave velocity against porosity have been presented. From the graphs it
can be seen that, porosities of the rock specimens are directly proportional to P-wave velocity.
The correlation between the two parameters used to estimate the porosity of the calcarenites
that are characteristic in historical monument.
References
Assefa, S et al. 2003. Velocity of Compressional and Shear waves in limestone. Geophy. Prospecting, 51:1-13.
Boulanouar, A., Rahmouni, A., Boukalouch, M., and Sebbani, M. 2012. Corrélation entre la vitesse d’onde P et
la conductivité thermique des matériaux hétérogènes et poreux. MATEC Web of Conferences 2, 05004.
Kahraman, S. et al.2008. Determination of physical properties of carbonate rocks from P-wave velocity. Bull
Eng Geol Environ 67:277–281.
Rosener, M.2007. Etude pétrophysique et modélisation des effets des transferts thermiques entre roche et fluide
dans le contexte géothermique de Soultz-Sous-Forêts. Thèse de doctorat, Université Luis Pasteur, Strasbourg.
Soroush, H. et al. 2011. Evaluation of rock properties using ultrasonic pulse technique and correlating static to
dynamic elastic constants The 2nd South Asain Geoscience Conference, GEOIndia.
Vasconcelos, G. et al.2007. Prediction of the mechanical properties of granites by ultrasonic pulse velocity and
schmidt hammer hardness. Masonry Confer, USA.
Yasar, E., Erdogan, Y. 2004. Correlating sound velocity with the density, compressive strength and Young’s
modulus of carbonate rocks. International Journal of Rock & Mining Sciences, vol. 41, p. 871-875.
Zaouia, N et al. 2005. Superficial alteration and soluble salts in the calcarenite weathering. case study of
almohade monuments in Rabat. Environ Geol 48: 742–747.
3
11ème congrès de Mécanique 23-26 Avril 2013 - Agadir (Maroc)
system.
1. Introduction
The porosity value Nt, is then obtained by the method of
Porosity and thermal conductivity play an important three weighing. Knowing the dry weight of the sample Ws
role in transfer properties of fluid-rock interactions. The and its weight soaked W1, we obtain the total pore volume
understanding of the relationship between these two occupied by water (W1-Ws), then from a hydrostatic
parameters could be a good way to understand the fluid- weighing W2, we can calculate the total volume of the
rock interactions between the porous and the fracture sample (Vt = W1-W2). The porosity of the sample is given
network and to predict the porosity of a porous media. by:
Thermal conductivity of rocks is a critical parameter W − Ws
N t (%) = 1 × 100 (1)
used in geothermal utilization design. Generally, the W1 − W2
thermal conductivity ranges given in the compilations of
rock properties (Clauser and Huenges, 1995) are too broad with W1 the weight of the sample saturated with water, W2
to be useful in the application of a specific site because weight saturated obtained by hydrostatic weighing, Ws is
thermal properties are highly dependent on site specific the dry weight of the sample.
characteristics. The distribution and spatial variability of
2.1. Thermal Conductivity
the thermal conductivity are needed as input parameters in
the design process. Knowledge of the thermal conductivity The technique presented here eliminates the effect of
of rocks is essential for solving many geothermal utilization the thermal contact resistance. It is based on the change in
design problems (Schilling, 1999). surface temperature of the sample after a known heat input
The thermal conductivity of rocks depends primarily and controlled (Popov et al.1999). The experimental design
on: is identical to that used by (Rosener 2007). The meter
• Its mineral composition. comes in the form of a movable block, consisting of two
• Its structural and textural features. These include mineral temperature sensors sensitive to electromagnetic radiation
size fine grained or coarse grained, mineral shape and the and a heat source, aligned parallel to the axis of movement,
presence of pores. Also the presence of micro cracks. so it is possible to measure the temperature of the sample
• The amount of pore water present. before and after heating. The block moves at constant speed
• The condition it is in, when tested (e.g., temperature, (V=4.99 mm/s) near the surface of the sample, along a
water content) (Clauser and Huenges, 1995). measurement line, and thus allows to obtain a profile of
In this study, correlations between the thermal thermal conductivity.
conductivity value and the porosity values. The effect of The two temperature sensors (T1 and T2) to measure
water saturation on the thermal conductivity is studied. We the temperature difference due to the energy input. The
used the method of water porosity to determine the total distance between the source and temperature sensor
porosity of the samples and the method optical scanning to measuring the temperature of the sample after heating and
characterize the thermal conductivity of these samples. the speed are constant, it is possible to link the increase in
temperature at x, θ (x) and conductivity heat by a relation of
2. Materials and methods the form:
q
The materials used in this study are sedimentary rocks θ ( x) = (2)
calcarenitiques taken from a quarry near the town of Rabat. 2π x λ
Calcarenite monuments of Rabat is a rock of Plio- With, q the source strength, x the distance between the
Quaternary age, is characterized by high porosity and source and the thermal sensor and λ the thermal
variable (18-47%) (Zaouia et al.2005). Its chemical conductivity at the measurement point (Popov et al.1999).
composition is very rich in calcium carbonate. To obtain an absolute value of the thermal conductivity of
Measurements of thermal conductivity and porosity of the material analyzed, it is sufficient to align the sample and
were carried out on samples 7×7×7 cm3. These measures a standard of known thermal conductivity λstd (in this case
will help to better understand and identify the λstd=1.35 W/m.K). The thermal conductivity of the material
characteristics of the porous network of materials used in will be determined from λstd and report temperature
the construction of the monument Roman Chella Rabat. differences:
θ std
2.1. Porosity λ ( x ) = λ std (3)
θ (x)
The method of measuring the total porosity to water is
that defined by (Rosner 2007), which provides that the Sample preparation is simple. To avoid a change of
samples are soaked in the absence of air, ie in monophasic heat absorption due to color variations in the sample
18
11ème congrès de Mécanique 23-26 Avril 2013 - Agadir (Maroc)
surface, a layer of black paint is applied. In this way, the Fig.2. Estimated porosity and measured porosity from
coefficient of heat absorption is maximum and thermal conductivity for dry samples.
homogeneous. The range of measurable conductivity is
between 0.2 and 70 W/m.K, and the measurement error is 1,85
1,55
Based on measurements of different sample weights
that are extracted from the same block (same mineralogy 1,5
and petrography) shows that these samples have a 24,00 26,00 28,00 30,00 32,00 34,00 36,00 38,00
1,10 24
1,00 22
22 24 26 28 30 32 34 36 38
0,90
0,80
Measured porosity (%)
0,70 y = -0,0216x + 1,6303
R2 = 0,7798
0,60 Fig.4. Estimated porosity and measured porosity from
0,50 thermal conductivity for saturated samples
0,40 The results show that the thermal conductivity
24,00 26,00 28,00 30,00 32,00 34,00 36,00 38,00 decreases with increasing porosity. Linear relationships
Porosity (% ) between conductivity and porosity appear on figures 1 and
3. The correlation between the two parameters used to
Fig.1. Thermal conductivity versus porosity for dry
estimate the porosity of a porous media.
samples.
As a result, the relationship between the estimated
38
values and the measured values of porosity for each sample
is examined graphically in Figs. 2 and 4. The error in the
36 estimated value is represented by the distance that each data
point plots from diagonal line curves 1:1. A point lying on
34
the line indicates an exact estimation, as shown in Figs.2
Estimated porosity (%)
32 and 4.
30 3.2. Influence of water saturation on the thermal
conductivity
28
The effect of partial saturation in water varies
26
depending whether the rock is porous or fractured. Porosity
24
in porous rocks consists of “bottlenecks” formed at the
contact between individual grains and the bulk pore space.
22 Dry bottlenecks act as thermal contact resistances between
22 24 26 28 30 32 34 36 38
grains, while the bulk pore volume contributes according to
Measured porosity (%)
its size to the effective thermal conductivity (Clauser and
Huenges 1995). In Fig.5 shows how the thermal conductivity
varies in a rock calcarenite porosity of 31.07% with the
19
11ème congrès de Mécanique 23-26 Avril 2013 - Agadir (Maroc)
degree of water saturation. It is found that the thermal Schilling, F.R.: 1999. A transient technique to measure
conductivity increases as a function of water saturation, and thermal diffusivity at elevated temperatures,11, 1115- 1124.
that the thermal conductivity does not vary linearly with the Surma, F.and Geraud, Y. 2003. Porosity and Thermal
change of water saturation. This has been observed in other Conductivity of the Soultz-Sous-Forêts Granite. Pure and
materials such as sandstone and granite (Popov et al. 1999). Applied Geophysics 160, pp: 1125–1136.
For low water saturations, the transfer of heat takes place
Zaouia, N et al. 2005. Superficial alteration and soluble
mainly at the contact points between the particles forming
salts in the calcarenite weathering. case study of almohade
the material, where the small increase in the value of the
monuments in Rabat: Morocco. Environ Geol 48: 742–747.
thermal conductivity. The increase in water saturation,
resulting in the formation of thermal bridges of heat within
the material, hence the increase in conductivity of the
material.
2,10
T h e rm a l c o n d u c tiv ity (W /m /K )
1,80
1,50
1,20
0,90
EX
0,60 EY
0,30 EZ
0,00
0 20 40 60 80 100
20
CMSS 2013: International Congress on Materials & Structural Stability
Building up Sustainable Materials & Constructions
Strasbourg, France
The prediction of elastic and acoustic properties effective of porous materials has
been an important development in many areas, such as physics rock, construction
and the study of the behavior of historical monuments. This interest has resulted in
the development of models which correspond to variations of fundamental
approaches. The study of compressional and shear wave propagation a material to
dry and saturated evaluates the physical properties of materials such as porosity,
cracking state and its elastic properties in modeling of the wave propagation. For
rocks, it is often interested in the compressional waves propagation (P-wave) and
shear waves (S-wave). The modelling acoustic and elastic properties in dry or
saturated rocks illustrate the influence of saturation. In this work we propose to
model the elastic modulus (Young's modulus E, compression modulus K) and the
propagation velocities of compressional waves VP and shear waves VS in partially
saturated rocks according to the saturation of rock samples which pore space is
occupied by a two-phase fluid (water and gas) by the application schema 2
homogenization Mori-Tanaka. The modelling results for a medium partially
saturated by a two-phase fluid system show the influence of degree of saturation on
the elastic and acoustic properties.
P
Rabat, Morocco
3Institute of Earth Science "JaumeAlmera", CSIC, Barcelona, Spain.
The main objective of this work is to analyze seismic attenuation (1/Q c) using a
single backscattering model hypothesis of Aki and Chouet (1975). For this
purpose, the recordings of 66 local earthquakes (epicentre distance < 100 km)
during 2008 in Northern Morocco have been used with a magnitude (Ml ) less
than 4. The Qc quality factor values have been computed at four central
frequencies 0.75, 1.5, 3, 6 and 12 Hz and analyzed for two horizontal and vertical
components for performing the average values. Four lapse time windows seconds
from 30 to 60 duration with a difference of 10 seconds have been analyzed to
study the lapse time dependence of Qc. We obtained a strong average frequency
dependence follow a power law Q n=Q0fn where Q0 is Qc at 1Hz and n is the power
of frequency dependent .The frequency dependent relationships obtained are
Qc=(143.75±1.09)f(0.864±0.006) for the vertical component ,
Qc=(149.12±1.08)f (0.85±0.05) and for the N component , and
Qc=(140.42±1.81)f (0.902±0.04) for the E component. The values estimated of coda Q
shows independent on the component of wave motion consistent with (Jen-Kuang
Chung 2009 and Priyamvada Singh 2012 ), thus only one component sufficient to
treat the attenuation in this region. The mean values of the estimated Qc of the
vertical component vary from 76 (at 0.75) to 1147.6 (at 12 Hz) for 30 seconds coda
window length, for 40 seconds coda window length Qc vary from 122.48 (at 0.75)
to 1255 (at 12 Hz ) while for 50 seconds coda window length Qc vary from 141.4
(0.75) to 1420.8 (at 12 Hz ) . Similarly for 60 seconds coda window length Q c vary
3
from 173.89 to 1495. The increase in Q c values with lapse time shows the depth
dependence which agrees with many studies. The results obtained with this model
O
are presented and then compared to results from the literature.
France.
3Department of Earth Science, Scientific Institute, Mohamed V University, Rabat,
Morocco.
3
P
Morocco
On 24th February 2004 a significant earthquake (Md = 6.4) occurred in the north
of Morocco causing great damage in the vicinity of Al Hoceima region. This area is
characterized by a complex faulting system as a result of compressional tectonic
forces. Three short period stations are set in this area of interest and recordings
from these stations were used in this study. In order to complete our knowledge of
attenuation, 60 local earthquakes are recorded a few days after the great
earthquake with magnitude Ml 2.6-5.0 to estimate seismic attenuation. For this
purpose, we applied the single backscattering model of Aki & Chouet 1975 in the
frequency range for 1 to 4 Hz. The study of coda waves was limited to a relatively
short lapse time (10 Seconds) in order to sample the earth’s crust only .The values
of Qc estimated for all the three stations show a strong frequency dependent
relationship of the form Q c=Q0fn, where Q0 is Qc at 1Hz , and n represents the
degree of frequency dependence , and reflect the level of crustal heterogeneities to
varying degrees. The average frequency dependent attenuation relationship has
been obtained which indicates that the attenuation is high in this region. Finally to
conclude our work, the values of Q0 suggest that Al Hoceima area is highly
heterogeneous and the n parameter indicates a meaning frequency dependence of
Qc .
3
P
1*
Abdelaali Rahmouni ,
Abderrahim Boulanouar ,
1
CORRELATION BETWEEN THERMAL
1
Abderrahim Samaouali , CONDUCTIVITY, POROSITY AND DENSITY
2
Yves Géraud , Mohamed
1
Boukalouch , Jamal OF CALCARENITE ROCKS USING OPTICAL
1
Sebbani SCANNING METHOD
1
Laboratory of Thermodynamics,
Department of Physics, Faculty Abstract: A non-contact and non-destructive optical scanning method
of Science, Mohammed V provided in principle a approach to the study of the relationship between the
University-Agdal, P.O. Box thermal conductivity and other physical properties of rocks. The present
1014, Rabat, Morocco. work deals with the use of a non-destructive technique, optical scanning, to
2
University of Lorraine, ENSG, predict the porosity and density of calcarenite rocks that are characteristic in
UMR 7359-GeoRessources, historical monument. The influence of anisotropy and water saturation on
Nancy Cedex, France. the thermal conductivity of calcarenite rocks was studied. A correlation
* Corresponding author between thermal conductivity, porosity and density of calcarenite rocks was
(a.rahmouni@yahoo.fr) found, which indicated them as an appropriate technique for estimating the
physical properties.
Nomenclature
ϕ porosity, %
λ thermal conductivity, W/m.K
3
ρ density, g/cm
T temperature, °C
1. Introduction
Porosity and thermal conductivity play an important role in transfer properties of fluid-rock
interactions. The understanding of the relationship between these two parameters could be a good
way to comprehend the fluid-rock interactions between the porous and the fracture network and to
predict the porosity of a porous media.
Thermal conductivity of rocks is a critical parameter used in geothermal utilization design.
Generally, the thermal conductivity ranges given in the compilations of rock properties [1] are too
broad to be useful in the application of a specific site because thermal properties are highly dependent
on site specific characteristics. The distribution and spatial variability of the thermal conductivity are
needed as input parameters in the design process. Knowledge of the thermal conductivity of rocks is
essential for solving many geothermal utilization design problems [2].
The thermal conductivity of rocks depends primarily on:
• Its mineral composition.
• Its structural and textural features. These include mineral size fine grained or coarse grained,
mineral shape and the presence of pores. Also the presence of micro cracks.
• The amount of pore water present.
• The condition it is in, when tested (e.g., temperature, water content) [1].
Physical properties, such as porosity and density, are important parameters for the quality of
building stones. Good correlations are established between thermal conductivity and the physical
properties of calcarenite rocks. The effect of anisotropy and water saturation on the thermal
conductivity was studied. We used the method of water porosity to determine the total porosity of the
samples and the method optical scanning to characterize the thermal conductivity of these samples.
2. Materials
The materials used in this study are sedimentary rocks calcarenite, collected in a quarry near the
town of Rabat, Morocco. The monuments of Rabat are all constructed by the plioquaternary
calcarenite that constitutes the basement of the whole region. A few kilometers to the north of the city,
quarries still provide this ornamental rock widely used by stone craftsmen. It is characterized by
variable and high porosity (18–47%) and thus an elevated permeability [3]. Its chemical composition is
3. Characterization techniques
3.1. Porosity
The method of measuring the water total porosity is that defined by the standard RILEM [6], which
provides that the samples are soaked in the absence of air, ie in monophasic regime.
The porosity value ϕ, is then obtained by the method of the triple weighing. Knowing the dry weight
of the sample W s and its weight soaked W 1, we obtain the total pore volume occupied by water (W 1-
W s), then from a hydrostatic weighing W 2, we can calculate the total volume of the sample (Vt = W 1-
W 2). The porosity of the sample is given by:
W −Ws
φ = 1 × 100 (1)
W1 − W 2
with W 1 the weight of the sample saturated with water, W 2 weight saturated obtained by hydrostatic
weighing, W s is the dry weight of the sample.
3.2. Density
Bulk density is determined either by measuring precisely sample dimensions and sample weight,
after drying in an oven at 105°C during 24 h, or by hydrostatic weighing method which is based on the
Archimedes principle on a sample saturated and submerged in a wetting fluid: water. In this test,
samples are dried in an oven at 105°C until constan t weight (W s). Then, the samples are saturated by
imbibition in a cell under vacuum during 24 h, by complete immersion during a day in outgassed
water. Then a hydrostatic weighing of the saturated samples immersed in water (W 2) and a weighing
in air of the saturated samples wiped with a wet rag (W 1) are carried out. Dry (ρdry) and saturated (ρsat)
bulk density can then be calculated from:
Ws
ρ dry = (2)
W1 − W 2
W1
ρ sat = (3)
W1 − W 2
1,1 E3Y
E3Y
1,8 E3Z
1 E3Z
1,7
0,9
1,6
0,8
1,5
0,7
1,4
0,6
1,3
0,5
1,2
0,4
0 10 20 30 40 50 60
0 10 20 30 40 50 60
Position (mm) Position (mm)
Figure 1: Profile of thermal conductivity of the sample (29.82%) to dry and saturated state in directions (ox, oy
and oz).
The measurements of thermal conductivity carried out on porous materials depend on the physical
properties of the solid (mineralogical properties and thermal properties, if necessary, especially in
case of anisotropy), the pores (air, vacuum, fluid ...), and the proportion of each of them. On the basis
of two measurements of thermal conductivity, one of the dry material, the other of saturated material,
Schärli and Ribach (1984) provided an estimation of the porosity using the following formula:
ln( λ sat ) − ln( λ dry )
φ = (5)
ln( λ water ) − ln( λ air )
with ϕ is the calculated porosity, λdry and λsat are respectively the thermal conductivities measured for
the dry and saturated sample, λwater and λair are respectively the thermal conductivities values for the
water and air.
0,30
0,25
Porosity
0,20
0,15
E3X
E3Y
0,10 E3Z
0,05
0 10 20 30 40 50 60
Position (mm)
Figure 2: Variation of porosity of the sample (29.82%) as a function of the position in directions (ox, oy and oz).
Figure 2 shows the variation of the calculated porosity from equation (5) to the sample (29.82%) in
the directions (ox) (oy) and (oz). The porosity of this sample is more important in the perpendicular
direction (E3Z), which shows a low thermal conductivity in the dry state and a high conductivity in the
saturated state.
A.Dry
B.Saturated
1,20
Thermal Conductivity (W/m.K)
1,85
Figure 3: Thermal conductivity versus porosity for dry (A) and saturated (B) samples.
A.Dry B.Saturated
1,200 1,85
Thermal Conductivity (W/m.K)
Therm al C onductivity (W /m .K )
1,100 1,8
1,000 1,75
0,900 1,7
y = 1,5849x - 1,3939
1,65
0,800 R2 = 0,7839
y = 1,4162x - 1,3853
R2 = 0,9028 1,6
0,700
1,55
0,600
1,5
0,500 1,9 1,92 1,94 1,96 1,98 2 2,02
1,55 1,6 1,65 1,7 1,75 1,8
3
3
Density (g/cm )
Density (g/cm )
Figure 4: Relation between the thermal conductivity and density of the dry (A) and saturated (B) samples.
1,80
1,40
1,20
1,00 E3X
E3Y
0,80
E3Z
0,60
0,40
0 20 40 60 80 100
Figure 5: Effect of water saturation on the thermal conductivity of a calcarenite rock (29.82%), in directions (ox,
oy and oz).
5. Conclusion
In this study, basic physical properties of 6 rocks samples have been measured and analyzed in an
integrated manner. Laboratory measurements have been carried out on the following physical
parameters: porosity, density and thermal conductivity.
A strong coefficient of determination was found between thermal conductivity with porosity and
density of the tested calcarenite rocks. This correlation is used to estimate the porosity and density of
a porous media.
The results show that the thermal conductivity decreases with increase of the porosity. Anisotropy
and water saturation have an effect on the thermal conductivity of calcarenites rocks.
References
[1] C. Clauser and E. Huenges, Thermal Conductivity of Rocks and Minerals, The American
Geophysical Union, pp. 105-126, 1995.
[2] F.R. Schilling, A transient technique to measure thermal diffusivity at elevated temperatures, Vol.
11, pp.1115-1124, 1999.
[3] N. Zaouia, M. ELwartiti, B. Baghdad, Superficial Alteration and Soluble Salts in the Calcarenite
Weathering Case Study of Almohade Monuments in Rabat: Morocco, Environmental Geology,
Vol. 48, No. 6, pp. 742–747, 2005.
[4] AFPC-AFREM, Détermination de la Masse Volumique Apparente et de la Porosité Accessible à
l’Eau. Méthodes Recommandées pour la Mesure des Grandeurs Associées à la Durabilité,
Compte-rendu des Journées Techniques, Toulouse, France, pp. 121–124, 11–12 Décembre
1997.
[5] ISRM, Suggested Method for Rock Characterization, Testing and Monitoring, International Journal
of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, Vol. 18, No. 6, p. 109, 1981.
[6] M. Rosener, Etude pétrophysique et modélisation des effets des transferts thermiques entre roche
et fluide dans le contexte géothermique de Soultz-Sous-Forêts, Thèse de doctorat, Université Luis
Pasteur, Strasbourg, 2007.
[7] Y. A. Popov et al, Complex Detailed Investigations of the thermal properties of rocks on the basis
of a Moving Point Source, Physics of the solid earth, Vol. 21, No. 1, pp: 64-70, 1983.
[8] F. Surma and Y. Géraud, Porosity and Thermal Conductivity of the Soultz-Sous-Forêts Granite,
Pure and Applied Geophysics, Vol. 160, pp. 1125–1136, 2003.
[9] Y.A. Popov et al. Characterization of rock thermal conductivity by high - resolution optical scanning.
Geothermics, Vol. 28, pp. 253-276, 1999.
[10] K. Besnard, Evolution physico-chimique des matériaux carbonatés en milieu triphasique, DEA
Hydrologie, Hydrogéologie, Géostatistique et Géochimie, Université Paris-Sud, 2000.
[11] A. Hartmann et al. Thermal conductivity from core and well log data, International Journal of Rock
Mechanics & Mining Sciences, Vol. 42, pp. 1042–1055, 2005.
Effect of porosity and water saturation on the mechanical properties and P-wave
velocity of calcarenite rocks used in the construction of historical monuments of
Rabat (Morocco)
A. RAHMOUNI1, Y. EL RHAFFARI1, A. BOULANOUAR1, A. SAMAOUALI1,
M. BOUKALOUCH1, Y. GÉRAUD2 AND M.J.E. SEBBANI1
1. Laboratory of Thermodynamics, Department of Physics, Faculty of Science, Mohammed V University, P.O. Box 1014,
Rabat, Morocco. Email: a.rahmouni@yahoo.fr
2. University of Lorraine, ENSG, UMR 7359-GeoRessources, Nancy Cedex, France.
Abstract where kw and kg are the bulk modulus for the liquid and
The objective of this paper is to study the effect of gas. S is the degree of water saturation.
porosity and water saturation on the mechanical and In a homogeneous isotropic elastic medium, the velocity
acoustical behavior of isotropic porous materials of a of wave propagation plane compressive-type (P-wave) is
solid matrix containing dry, saturated and partially given by:
k
saturated spherical pores, using a homogenization 4
Vp
µ
technique based on the Mori-Tanaka model. The 3 (2)
mechanical properties of calcarenite rocks are
determined from P-wave velocity measurements. The 2. Materials and methods
comparisons between the predictions of Mori-Tanaka
2.1. Materials
model and experimental results are discussed and
analyzed. The rock used in this work is the calcarenite. It is mainly
Keywords: Calcarenite rock, porous media, used in various sectors: construction, renovation of
historical monuments, stones, sculptures. This
homogenization, Mori-Tanaka, mechanical properties,
sedimentary rock colored yellow ocher is found mainly
P-wave velocity, porosity, water saturation. in the Rabat- Salé region, it was used for the construction
1. Introduction and restoration of various historical monuments of
The study of the role of porosity and fluid flow in the Rabat. It is characterized by high porosity (18–47%) and
mechanical and acoustical behavior of porous media is elevated permeability [7, 8]. Its chemical composition is
fundamental to understanding phenomena in different very rich in calcium carbonates [9].
fields (geophysical subsurface, seismology, engineering, 2.2. Mechanical properties and P-wave
monuments, and construction). Indeed, the presence of velocity of a dry and saturated rocks
porosity and water saturation results in a modification of
mechanical and acoustic properties which can be The estimate of the macroscopic elastic tensor of a dry
modelled by homogenization techniques. and saturated porous media corresponding to the Mori-
C f (1 p ) (C f Cm ) (1 p ) I
MT
homogenization to describe the behavior of porous Cdry
p I S E : Cm1 : C f Cm
media. Thus Guéguen et al. [1] used the differential self-
1
MT
C sat
consistent model to determine the elastic moduli of
1
porous media. Xu [2] applied changing scale methods for
modeling the behavior of unsaturated porous media
whose solid phase consists of a linear elastic material. The elastic properties and P-wave velocity in the dry
kdry km (1 ) (1 a )1
Miled et al. [3] studied the prediction of elastic moduli of porous media containing spherical pores are respectively:
dry m 1 1 b 1
isotropic porous materials constituted of a solid matrix (3)
based on well-known Mean-Field Eshelby-based
Edry E0 1 1 c
(4)
1
homogenization schemes.
(5)
(1 d )
Many studies have been devoted to the modeling of
V p2dry V p2m
(1 a ) (1 b )
effective mechanical and acoustic properties of partially (6)
saturated porous media [4, 5]. Gassmann's equations [6]
provide the effective elastic moduli of saturated rock for
low frequency, provided you know in advance the where km and µ m are the elastic bulk and shear modulus
effective elastic moduli of dry rock. The bulk modulus of matrix, E0 is Young's modulus of the matrix, ϕ is the
1 S
of fluid Kf is then written: porosity.
1 S (1) The elastic properties and P-wave velocity in the
Kf Kw Kg saturated porous media containing spherical pores are
respectively:
13ème Congrès de Mécanique 11 - 14 Avril 2017 (Meknès, MAROC)
ksat km (1 a ) (1 e ) 1
sat dry m 1 1 d
(7) predictions of Mori-Tanaka model: (A) dry state, (B)
1 saturated state.
1 1 a
(8)
The Young's modulus E in the saturated porous media is
E sat E0
1 b c 2
(9) more important than those in dry porous media. This
1 f g 2
result is consistent with experimental results obtained on
1 a kf
a porous media: in saturated porous media.
kunsat kdry
The comparison between Eq. (6) and Eq. (10) and
1 b kf
(11)
experimental results is plotted in Fig. 2. This figure
showed a good agreement between Mori-Tanaka model
Eunsat
Edry (12) and experimental results for dry and saturated calcarenite
1
kf
1 k f
sedimentary rocks, which can be used to predict the
acoustic behavior of calcarenite rocks.
The variation of P-wave velocity of a porous media
5,0
according to the degree of water saturation S is given by: (A) Dry state th
kunsat µ
VP
4,8
4 exp
VP
Vp
4,6
unsat
unsat 3 (13) 4,4
VP (km/s)
4,2
3,6
From measurements of different sample weights, the
3,4
water total porosity of all samples varies between 25.69
3,2
and 35.83 % [8]. The results of measurement of P-wave 0 10 20 30 40 50
velocities of dry samples range from 3.56 to 3.8 km/s Porosity (%)
and those of water saturated samples vary between 3.59
and 3.9 km/s [11].
5,0
(B) Saturated state th
4,8 VP
50 exp
(A) Dry state 4,6 VP
Eth
45
Eexp 4,4
40 4,2
VP (km/s)
35 4,0
E (GPa)
3,8
30
3,6
25
3,4
20 3,2
15 3,0
0 10 20 30 40 50
0 10 20 30 40 50 Porosity (%)
Porosity (%)
Fig.2. Comparison of P-wave velocity data as a function
of porosity for saturated calcarenite rocks with the
50
(B) Saturated state
Eth predictions of Mori-Tanaka model: (A) dry state, (B)
45
Eexp saturated state.
40 3.2. Effect of degree of water saturation
35 The results of the modeling of a partially saturated
E (GPa)
30
porous media using the Mori-Tanaka model, show that
the bulk modulus k and Young’s modulus E increases
25
with the degree of water saturation S (Figs. 3 and 4).
20 From the curves, we see a slow increase of elastic
15 moduli up to 80 % of saturation, then for strong
0 10 20 30 40 50
saturation, the bulk modulus k and Young’s modulus E
Porosity (%) are increasing rapidly, which is in the same sense as the
work of Gregory [4].
Fig.1. Comparison of measurements of Young's modulus
Theoretically, using the Mori-Tanaka model for a
E as a function of porosity for calcarenite rocks with the
partially saturated porous media constituted by a solid
13ème Congrès de Mécanique 11 - 14 Avril 2017 (Meknès, MAROC)
matrix containing spherical pores, and when the porosity corresponding to degrees of saturation situated between
is important, which is the case, and for low or medium 80% and 90%.
saturation, the effect of density is dominant: the P-wave 4. Conclusion
velocity decreases as the density increases. By This work is to predict the elastic and acoustic behavior
approaching a strong saturation, the effect of bulk of calcarenite sedimentary rocks used in the construction
modulus becomes preponderant: P- wave increases with of historical monuments of Rabat, Morocco.
density. The result is that the curve of variation of P- The elastic properties in saturated state are more
wave velocity with the degree of saturation S presents a important than those in dry state. A good agreement is
minimum (Fig. 5). shown between the Mori-Tanaka model and
experimental results for dry and saturated calcarenite
25,0 kth 27 sedimentary rocks, which can be used to predict the
24,5
kexp elastic behavior of calcarenite rocks.
24
24,0
Furthermore, we have established two new equations that
21 describe the variation of bulk modulus k and Young's
kexp (GPa)
kth (GPa)
23,5
modulus E as a function of degree of water saturation by
18
23,0 using the Mori-Tanaka model.
22,5
15 Références
22,0 [1] Guéguen Y, Chelidze T, Le Ravalec M.
12
21,5 Microstructures, percolation thresholds, and rock
0,0 0,2 0,4 0,6 0,8 1,0
physical properties. Tectonophysics, 279:23-35. 1997
Degree of water saturation S
[2] Xu Y. Approches multi-échelle pour l’étude du
Fig.3. Bulk modulus k versus degree of water saturation comportement des systèmes polyphasiques -
S for a calcarenite sample of porosity 33.5%. application aux milieux poreux non saturés, Thèse
Ecole National des Ponts et chausses 2004.
24,00
Eth
33 [3] Miled K, Sab K, Le Roy, R. Effective elastic
23,92 Eexp 30
properties of porous materials: Homogenization
23,84
schemes vs experimental data. Mechanics Research
27
Communications; 38: 131–135. 2011
Eexp (GPa)
Eth (GPa)
3,0
Materials and Environmental Science; 5: 931-936,
VP
VP
3,90
2,8
2014.
3,85 2,6 [9] Samaouali A, Laanab L, Boukalouch M, Géraud Y.
3,80 2,4 Porosity and mineralogy evolution during the decay
0,0 0,2 0,4 0,6 0,8 1,0
Degree of water saturation S
process involved in the Chellah monument stones.
Environ Earth Sci; 59: 1171–1181, 2010.
Fig.5. P-wave velocity versus degree of water saturation [10] Huynh Q V. Estimation des propriétés
S for a calcarenite sample of porosity 33.5%. poromécaniques effectives des argilites: apport des
This minimum, degree of saturation limit, often méthodes d’homogénéisation. Thèse de Doctorat,
corresponds to a fairly high degree of saturation, greater Institut National Polytechnique de Lorraine , 2006.
than 80%. [11] Rahmouni A, Boulanouar A, Boukalouch M,
We carried out measurements of P-wave velocity to Géraud Y, Samaouali A, Harnafi M, Sebbani J.
different degrees of saturation on two samples of Prediction of Porosity and Density of Calcarenite
different porosities. The curves shown in Fig. 5, shows Rocks from P-wave Velocity Measurements,
the variation of P-wave velocities measured with the International Journal of Geosciences, 4, 1292-1299,
degree of saturation for a calcarenite sample. We find we 2013.
find the general shape of the curve found in the literature
[1] and the theoretical model, with minima
2è me C ongrè s International
Matériaux et Stabilité Structurelle
R abat, Maroc, 22-25 Novembre 2017
X -ray computed tomography (CT) is a non-destructive technique with wide applications in various domains,
medical, geological, rock mechanics, engineering and construction. The application of X -ray computed
tomography (CT) imaging to rocks has been used for many years to study and understand the microstructures
proprieties, such as, porosity, density and permeability. It reveals the internal structure of objects, determined
by variations in density and atomic composition. In this work, we present X -ray computed tomography (CT)
measurements of calcarenite rocks. X -ray computed tomography has used to characterize and determine
measures of porosity and density for calcarenite. The profiles of vertical and horizontal variations of porosity
and density were performed and analyzed. The reconstruction of two and three-dimensional imaging of these
rocks are presented to estimate their porosity and density.
K eywords: Calcarenite, monument, porosity, density, X -ray Computed Tomography, Hounsfield Units.
A uteur correspondant :
125
Spatial Distribution of attenuation in northern
Morocco and their interpretation
Abstract
In the present study, we investigated the spatial distribution of attenuation based on the
Coda wave of northern part Morocco. For this purpose, we analyzed sixty-six broadband
events from local earthquakes which recorded by twenty-four stations deployed by the Sci-
entific Institute of Rabat during 2008. Also, we selected five frequencies (0.75, 1.5,3,6 and
12Hz) and four lapse time windows (30s, 40, 50 and 60s). Then, the attenuation values
are extracting by applying the back-scattering model proposed by Aki and Chouet (1975).
Additionally, the spatial distribution maps of the attenuation of Coda wave for this region
are generated. The maps describe the attenuation (inverse of quality factors) distribution in
this area. We conclude that the region known several anomalies of Coda attenuation for all
frequencies and all windows used, like Al Hoceim and Nador regions with high attenuation.
This is due to due of the high seismic activity and the presence of many faults. They are also
compared with the heat flow map for this study zone. Moreover, the attenuation of Coda
wave decreases with increasing depth into the crust of this area. This last result is explained
by the fact that the earth becomes homogeneous more one penetrates into the crust of this
area. Finally, these results can be used not only for scientific research but also for practical
field.
∗
Speaker
sciencesconf.org:afsc2018:187715