Study of mechanical and hygrothermal properties of agglomerated cork

A. Mir*, B. Bezzazi**, R. Zitoune***, F. Collombet**** *Unité de Recherche, Matériaux Procédés Environnement, Université Boumerdès, Algérie, E-mail: abdellah_mir@yahoo.fr **Unité de Recherche, Matériaux Procédés Environnement, Université Boumerdès, Algérie. E-mail: bbezzazi@yahoo.fr ***Institut Clément Ader, Université de Toulouse, F-31077 Toulouse, France, E-mail: redouane.zitoune@iut-tlse3.fr ****Institut Clément Ader, Université de Toulouse, F-31077 Toulouse, France, E-mail: francis.collombet@iut-tlse3.fr


Introduction
Cork is a natural material of cellular structure having very interesting properties, which are low density, great dimensional compressibility, good acoustic, vibration absorption, heat insulation, chemical stability and longevity [1].The cork cells are composed mainly of suberin, lignin (40% and 22%, respectively), and of cellulose of 9% [2].Several scientific studies related to the oak cork (growth and production of cork), natural cork (biological and chemical properties and qualities of cork), agglomerated cork (mechanical and hygrothermal characterization) are reported in the literature.With regard to natural cork, the scientific studies are related to the influence of temperature and its environment (wet or dry) on its mechanical and hygrothermal properties.The natural cork immersed in water absorbs in all the three principal directions (axial, radial and tangential).The absorption of water in the walls of cells and noncommunicating cork causes its expansion.The work [3] shows that, the water absorption capacity of cork is twice more significant in water carried at 100°C than in water at ambient temperature.When natural cork is heated at high temperatures, the loss of mass is 15% at 200°C and 62% at 350°C, beyond this temperature, the structure of the walls of cells are damaged considerably [4].Ben Abdallah et al. [5] studied the influence of the heating of cork in a wet medium on variation of mass of Tunisian natural cork.The cork samples are heated at 100°C in the water vapour for about 3 hours then cooled and then these samples are dried in oven at 70°C for 3 hours.The results show a reduction in the density following vaporization of the water contained in the cork.For Tunisian natural cork the thermogravimetric analysis (TGA) carried out at 180°C shows a loss of mass of 4% in the first 10 minutes then 0.3% up to 30 minutes.At 220°C the loss of mass is increasing up to 30 minutes to reach 7.16%, at the end the colour of cork became brown dark.The SEM observations show that, the walls of the cells which are in the beginning corrugated became almost linear (straight) under the effect of heat, these swellings are responsible for the expansion of cork, the mass decreases following the evaporation of the water contained in cork [6].
In the mechanical field, the compression test carried out on natural cork is characterized by three phases: an elastic part corresponding to elastic bending of the walls of cells up to 5% of deformation, a stage (forced stable) corresponding to the compression of the walls of cells between them up to 60%, to 80% of the breaking stress, there is a crushing and a collapse of the cells until the rupture, this behavior is identical in the three principal directions of cork [7].At ambient temperature, the tensile tests on natural cork show a larger tensile module in the radial direction of 60% compared to the tangential direction [8].Strong ones et al, observed in their experiments the Poisson's ratio in the three directions (axial, radial and tangential) up to 30% of deformation compared to the breaking stress, the Poisson's ratios in the axial and radial directions are equivalent [9].
Contrary to natural cork and its limited field of application, the industrial cork covers a wide area up to thermal and acoustic insulations for space vehicles [10,11].Agglomerated cork is obtained from granulated cork compressed in an autoclave and crossed by a water vapour flow at 350°C [12,13].Several scientific studies are made, on black agglomerated cork produced in Portugal.Work of Gil presents the influence of the density on the acoustic properties of the black agglomerated cork panels.The low density cork has maximum of insulating capacity due primarily to the high quantity of air contained in the cells [14].Lastly, several industrial studies on transformed cork are carried out but within the commercial framework.The development of sandwich panels by infusion in "one shot" presents many advantages (simple, inexpensive, allows to produce large panels); on the other hand, it cannot prevent the resin LY 5052 from penetrating inside cork.Within the framework of this study, mechanical characterization and hygrothermal analysis of two types of white agglomerated cork of different density produced in Algeria is presented.Moreover, sandwich panels produced using the cork plates are studied in order to show the influence of the manufacturing process (infusion).The goal of the mechanical characterization is to identify using experiments the shear modulus as well as the failure stress of cork plates and to see the influence of their density as well as the presence of resin in cork material on the intrinsic characteristics.With regard to the hygrothermal study, the influence of the temperature on the variation of mass of the various types of cork is analyzed for large samples using drying oven and small samples using TGA.The capacity of absorption of the various types of cork is also studied.

Experimental study
Within the framework of this study two types of cork are considered and are classified based on their density (190 kg/m 3 and 270 kg/m 3 ).The acoustic and thermal tests showed that C190 is the best in the acoustic field and C270 is the best in the thermal field [15].For each type of the cork two batches were studied.The first batch is of specimens made of the cork and the second batch is a sandwich panel made of the cork.The samples of cork plates are referred as C190 and C270; on the other hand sandwich panels made of the cork are referred as C190(R) and C270(R).In the case of the sandwich panels the skin material is made of jute/epoxy [16].

Mechanical tests
Shear tests are carried out in accordance with standard NF EN12090 on both types of corks.The modulus of rigidity is calculated in two orthogonal planes of the agglomerated cork plate.Eq. ( 1) is used for the calculation of Gxy and Gxz.
where A is length L x width B (mm²); tan a is F/e (difference in force by difference in displacement in the elastic zone), d is width (mm) of the plate and G is shear modulus (MPa).
The samples tested are having dimensions 250 x 50 x 30 mm.Fig. 1 shows the assembly used for the shear tests.The cork plates are fixed on two rigid metal supports.These supports are fixed on the jaws of the Zwick tensile testing machine.The rate of travel of the jaw is of 2mm/min.During loading stress versus displacement are recorded (Fig. 1).The analysis of variation of the cork mass versus time at a temperature of 100°C using oven for 24 hours duration is carried out.The samples of cork tested are having the following dimensions 250 x 250 x 30 mm.The samples are weighed before the test and thereafter, samples are weighed for each one hour during 24 hours to quantify the loss of mass.

Thermogravimetric analysis
TGA is carried out in an atmosphere controlled environment on cork samples of dimensions 3 x 3 x 5 mm.
The cycle of temperature imposed on the samples is as follows: rise in the temperature of the cork from 25°C up to 180°C with a speed of 40°C per minute and once 180°C is reached, the cork is held at 180°C for about 30 minutes, then the temperature was reduced to 25°C at the speed of 40°C per minute.In order to see the influence of the maximum temperature on the behavior of the cork samples, a similar cycle is carried out up to 220°C.

Test of water absorption capacity
These tests are carried out in accordance with standard NF EN 1609 in a clean room at controlled temperature and moisture.Table represents the initial mass m 0 of specimens tested before the immersion.Thereafter the specimens partially immersed in water for 24 hours with the help of weight so that the specimen will not float on water.After 24 hours of immersion in water, the samples were drained for 10 minutes (The samples were placed vertically as shown in Fig. 2).Eq. ( 2) gives the capacity of absorption of water Wp (kg/m 2 ).

Table Values of the corks masses before immersion in water
where W p is absorption capacity (kg/m²); m 0 is initial mass (kg); m 24 is mass after 24h (kg); A p is (surface) length x width (m²).

Shear tests
The Figs. 3, a and b shows the behaviour in shear in planes X0Y and X0Z for the cork C190 and C270 respectively.For C190 cork, the linear elastic phase in planes X0Y and X0Z is observed for a stress of 0.12 MPa.The modulus of rigidity calculated in these two planes is about 2.6 MPa.Beyond 0.12 MPa one notes a nonlinear behaviour until the rupture.The maximum stress of rupture is  = 0.3 MPa in plane XOY, which is lesser (24%) when compared to plan X0Z.With regard to cork C270, one can note an identical linear behaviour for a stress of 0.12 MPa.The modulus of rigidity calculated in this zone GX0Y ≈ GX0 = 2.62 MPa.Beyond this stress nonlinear behaviour is observed until the total rupture.The maximum stress of rupture is 0.33 MPa in plan X0Y, which is 20% larger when compared to plane X0Z.In the Fig. 5, a, the maximum breaking stress of 2.9 MPa is identical for both C190 and C190(R) corks.The modulus of rigidity is of 1.54 MPa for C190 and 2.75 MPa for C190(R), which is an increase of 78% when compared to C190, C190 has a linear behavior up to 6.5 mm of displacement then a nonlinear behavior until rupture.Cork C190(R) has a linear behavior up to 5 mm of displacement then a nonlinear behavior until the rupture, this behavior can be related to the presence of resin inside the cork because of its larger granularity (3 to 5 mm), indeed, the presence of large pellets supports spaces between grains what allows the resin inside the cork and makes it harder but fragile.
We notice in the Fig. 5, b that the modulus of rigidity calculated in the linear elastic zone is 2.25 MPa for C270 and 3.08 MPa for C290(R), which shows an increase of 37% when compared to C270.This increase can be again related to the presence of resin in C270.The modulus of rigidity of the epoxy resin is about 3 GPa.It is also noted, that the maximum stress of rupture of about 2.9 MPa which is identical for C270 and C270(R) cork material.The Fig. 7, a shows the variation of the mass of C190 cork with time of drying at a temperature of 100°C in oven.C190 cork loses in the first 10 hours around 5% of its mass then it is stabilized, C190(R) loses 3.3% in the first 10 hours then 0.7% up to 24 hours.This difference in loss of mass can be explained by the fact, that the presence of resin in cork C190(R) plays a role in moisture absorption.
It is noted that C270 cork loses more mass than cork C270(R).In the first 4 hours C270 loses 4% of its initial mass, where as after 10 hours the variation of its total mass compared to its initial mass is only 4.5%.Beyond 10 hours of drying in oven, no variation of mass is observed.With regard to material C270(R), it loses a mass greater than C270 material in the first hour.In this situation, one can note 2.3% of mass lost for C270(R) and 1.2% of mass lost for C270.At the end of 6 hours, cork C270(R) loses 3.5% of its initial mass, thereafter no variation of b a mass was observed.The difference in loss of the mass is slightly less than that of C190 and C190(R), this can be related to the structure of these two types of cork.It is found that if the cork material is dense, it is less sensible to moisture.Fig. 7 Variation of the mass of cork C190, C190(R), C270 and C270 (R) according to the warming time at 100°C

Analysis of variation of mass of cork during TGA test
The Fig. 8, a shows the variation of mass expressed in percentage for agglomerated cork (C270) carried at 180°C during 35 minutes in TGA.A significant amount of loss of mass in the first 5 minutes (3.84%) is observed.Between 5 to 35 minutes the mass varies slightly (0.84%).The total loss of mass at the end of 35 minutes is 4.68%.Similar work of Ben Abdallah et al. on Tunisian untreated cork shows a loss of mass of 5% in the first 8 minutes and then a stable behaviour up to 30 minutes [5].During the first 5 minutes at 220°C, the variation of the mass of C270 cork is (4.13%).After the first 5 minutes the loss of mass of the cork continues and at the end of 35 minutes, the cork lost a total mass of 7.16% (Fig. 8, a).The Fig. 8, b represents the variation of the mass expressed in percentage of the C270 and C270 (R) carried at 220°C.Loss of mass is different for the two types of the cork tested.Although the loss of mass is high during the first 5 minutes, the amount of loss is not the same for C270 (R) and C270; it is 5.39% for C270 (R) and 6.78% for C270.From 5 minutes up to 35 minutes the loss of mass is continuous and steady.This variation of mass can be explained by the fact that C270 cork presents a higher percentage of presence of air between the grains, in cork C270 (R) the presence of resin inside cork prevents the mass from decreasing considerably.At 220°C the cork changes to maroon dark colour, which is similar to the results of Ben Abdallah et al. [5].

Analysis of absorption of water
Histogram in Fig. 9 shows the variation of the capacity of water absorption Wp (kg/m²) of the C270, C190 corks at partial immersion condition in a vessel during 24 hours.We notice that the capacity of absorption of C270 cork is much higher when compared to C190 cork.C270 has a capacity of absorption of about 4.86 kg/m², where as C190 is only about 0.38 kg/m².This difference is certainly related to there granulometry of the two types of corks tested.Fig. 10 shows the air voids of C190, which is larger than that of C270.More the proportion of air voids, water absorption capacity of the cork is less.It is also noted that the presence of resin in material C270 (R) led to a capacity of absorption 26 % larger than that of C270.This increase can be explained by the fact that low porosities can be filled by a small quantity of resin (Fig. 11, a) which permits a good retention of water.However, a great quantity of resin fills the bigger sizes of porosity of C190 (Fig. 11, b) which results in a weaker capacity of absorption for materials C190 (R).After 24 hours of immersion of water, C270 had a significant deformation; however C190 cork did not change its form.Fig. 12 shows the agglomerated cork C190, C270 tested for water absorption.

Conclusion
In this paper two of types of agglomerated corks (medium and low density) and sandwich panels are studied for its shear strength and hygrothermal properties.This study took into account the influence of presence of resin inside the cork of sandwich panel of cork.During shear tests, presence of resin in the C220(R) cork showed an increased modulus of rigidity of 1.8 times compared to the C220 cork and of 2.8 times that of C190.In hygrothermal analysis, the influence of the temperature on the variation of the mass of cork with and without resin was observed.At 100°C, 180°C and 220°C, cork loses a significant mass in the first phase; this loss is not identical for the various types of the cork with or without presence of resin.The water absorption capacity showed that the various types of the cork do not behave same manner in wet medium, further it is noted that the variation of the density of cork and the presence of resin influence the water absorption capacity.These results highlight the influence of manufacturing process on the cork panels made of infusion process, which shall help the designers to simulate cork panels during design of structures.The mode of development of sandwich panels with cork core by the method infusion cannot prevent the resin from penetrating inside cork.This study characterizes white cork agglomerated before and after the development of sandwich panels to see the influence of the resin on the mechanical and hygrothermic characteristics of cork.Two types of cork are considered, they are characterized by their density of 270 kg/m 3 (C270) and of 190 kg/m 3 (C190).One notes that the modulus of rigidity of C270 cork is 20% larger than that of C190.The presence of resin increases also the module of shearing 47% for C270, 64% for C190.The hygrothermic analysis out of drying oven at 100°C shows a fall of significant mass in the first 7 hours.The analysis thermogravimetry at 220°C, the fall of significant mass is recorded in the first 5 minutes.

Fig. 2
Fig. 2 Draining of samples for 10 minutes

Fig. 3
Fig. 3 Stress versus displacement for the shear test in the planes X0Y and X0Z, (a) C190, (b) C27 Fig. 4 shows the first (elastic zone) and final stage (rupture of cork) of the cork tested.The rupture of the cork is identical to the work of Reis et al. [17].The modulus of rigidity of agglomerated cork of 11 mm thickness and 270 kg/m 3 of density was found by Reis et al. is 4.86 MPa for pellets of 1 mm to 4 mm.

Fig. 4
Fig. 4 Typical mechanical behaviour of agglomerated cork during shear test (a): elastic domain, (b): final failure Fig. 5 represents the curves of the stress versus displacement during the shear tests in plan (X0Y) of the cork C190, C190(R), C270 and C270(R).In the Fig.5, a, the maximum breaking stress of 2.9 MPa is identical for both C190 and C190(R) corks.The modulus of rigidity is of 1.54 MPa for C190 and 2.75 MPa for C190(R), which is an increase of 78% when compared to C190, C190 has a linear behavior up to 6.5 mm of displacement then a nonlinear behavior until rupture.Cork C190(R) has a linear behavior up to 5 mm of displacement then a nonlinear behavior until the rupture, this behavior can be related to the presence of resin inside the cork because of its larger granularity (3 to 5 mm), indeed, the presence of large pellets supports spaces between grains what allows the resin inside the cork and makes it harder but fragile.

Fig. 5 Fig. 6 SEM
Fig. 5 Stress versus displacement during shear test in (XOY) plan of specimens, (a) C190 and 190(R), (b) C270 and C270 (R)The observations using SEM show the presence of resin in the cork.The quantity of resin in cork depends on its nature.Figs.6, a and bshows the space in the cork and resin penetration, which explains the rigid behavior of the cork with resin in shearing test compared to the agglomerated cork without resin.