Contribution to the Study of the Water Behavior of Starch-Based Composite Films

: In our previous studies, we developed a composite based on cassava starch. However, it is observed, like all starchy materials, that it is very hydrophilic and its behavior depends on its water content. This is explained by the attachment of water molecules to the hydroxyl groups of starch via hydrogen bonds. The barrier to water absorption has been reinforced by the incorporation of natural hydraulic lime of the NHL-3.5 type. This study aims to contribute to the knowledge of the water behavior of starch-based composite films. Here, it is the study of the diffusion and absorption of water vapor in a biofilm made of cassava starch reinforced with coconut fibers. To do this, we carried out tests on the sample which made it possible to obtain measurements concerning the hydric behavior of the material. Two approaches were used to evaluate these phenomena under well-defined conditions. The two methods contribute to the same result for the hydric behavior of the composite. The value of the diffusion coefficient obtained is 0.723.10-3 mm2/d. The results obtained are convergent and consistent with those obtained in the literature


Introduction
The use of biodegradable packaging materials as an alternative to conventional petrochemical-based polymers is based on the environmental issues associated with conventional materials [1][2][3][4].One of the functions of films is to slow down the transfer of water vapor.There is a large amount of data on vapor transfer depending on the nature of the matrix and the reinforcements [5][6][7].The water vapor permeability values are difficult to compare due to the diversity of temperature conditions and especially the relative humidity (RH) values at which the measurements are made [3,6,8].The wide chemical variety of starches is reflected in the water vapor permeability of these films.However, for a given composition, it depends on the level of plasticizer, and the measurement conditions.Nevertheless, the values remain very high for starch-based films compared to those of synthetic films such as low density polyethylene which is 0.0036.10-10g/m.s.Pa [9].While those of amylose and amylopectin are respectively equal to 11.95.10-10g/m.s.Pa and 14.44.10-10g/m.s.Pa.[9].The objective of this work is to study the water behavior of a film based on cassava starch reinforced with coconut fibers [10] by analyzing the diffusion and absorption of water vapor.From experience, we obtained results that allowed us to study these phenomena.The diffusion coefficient was analyzed and determined by two different approaches.The results are convergent and the value obtained is 0.723.10 - mm 2 /d.

Raw Materials [3]
To produce the sample, we used cassava starch powder (Figure 1. A), distilled water available in the laboratory, glycerol supplied by Polychimie Abidjan S.A., hydraulic lime of the NHL-3.5 as a barrier to water absorption, the anti-foaming agent EROL ADT obtained from PMC OUVRIE (France) and the fibers of coconut mesocarps as reinforcements (Figure 1. B).
The purity of the starch used was evaluated by proton nuclear magnetic resonance called 1H-NMR-TF.Its density was determined equal to 1.8 g/cm 3 with a dispersion of 1.1%.

Formulation and Elaboration of the Film
It is recognized that the phases of biofilms are sensitive to hygrometry [8,11].Indeed, depending on the degree of the latter, these materials absorb water; which locally modifies the phases, inducing a heterogeneity which weakens them.In addition, the reinforcement was carried out with coconut fibers after preliminary tests relating to the determination of the proportions of the mass of starch, the quantity of water and glycerol.Table 1 summarizes the different proportions retained for the final formulation.The sample is shown in Figure 2.

Experimental Phase
The 10 mm side and 1 mm thick films are cut and each is exposed to a temperature between 20 and 25 o C under different relative humidity RH = 72 and 98%.Each sample, of initial mass M i, is then removed at time intervals t to be weighed; we obtain for each time, a mass M t .The water absorption rate has the expression: The mass of water absorbed at time t, , can be calculated as follows [12,13]: where M∞ is the mass at equilibrium in g, M t is the mass at equilibrium at time t in g, L half the thickness of the polymer film in mm, π is the angle made by the diffusion hemisphere, D the diffusion coefficient and n the number of samples having undergone the test.The representation of the absorption rate (AR) as a function of time makes it possible to determine the absorption coefficient.On the other hand, from equation ( 2), we obtain the expression for the diffusion coefficient: The absorption process can be materialized by a spherical cap represented in Figure 3.

Results
The experimental measurements are listed in Table 2.   First, the maximum amount of liquid that can be absorbed by the sample without its mechanical properties being degraded by contact with the actual exposure atmosphere of use is determined.Figure 3. shows that the highest absorption rates are observed on the first and second day which are respectively equal to 2% and 6% for the respective relative humidity of 98% and 72%.It is also noted that after five days, the absorption rate is identically equal to 16% under the two atmospheres of exposure.The absorption phenomenon seems to reach a ceiling after five days.From the fourth day, the two samples show the same absorption rate; to reach equilibrium on the fifth day at 14%.The water diffusivity or diffusion coefficient D in the starch-based material was estimated using equation (2).We draw the graph of the quantity # !/ % as a function of / & ⁄ (Figure 5).From this representation, we will determine the diffusion coefficient.

Determination of the Diffusion Coefficient
We used two different methods that achieved the same result.

Method 1
Notes: The time is, effectively, recorded in hours before being converted into days to build the point cloud.We choose here to represent the variation of the mass # !/ % according to the square root of time ' (t in days)).For example, for ' /L=8 (with l=0.5mm), we will have, here, t=16 hours=0.6666days (Table 3).By observation of Figure 5, we observe, objectively, that the curve has, among other things, the appearance of a power function.The phenomenon described could therefore be modeled by a function of the form: 8 9 : ; ; =, 9, ?@A (4) For 0, B # .The curve passes through the coordinate point (0,0).It is therefore written in the form: The linearization of equation ( 5) gives: If we have: The equation ( 5) becomes: The coefficients a, b and B are determined in the form using the data in table 4: I H O =J O (10) Table 4. Values for determining parameters a and b.We have: = 0,28 and 9 0,060 with I 2,8010 The equation ( 5) becomes: It is represented in Figure 6. and shows the evolution of the mass of the composite under the conditions defined by the experiment.From equations 2 and 12, we get: D 0,723. 10 g mm /d.

Method 2
The representation of is give by the folowing expression [14]: ( ) The diagram below (Figure 7) shows the results obtained for orders n=2, 3 and 4. The highest coefficient of determination is obtained for n=4 (0.583).For n=1, r 2 = 0.358.
We can therefore conclude that for a good convergence of equation ( 13), it will be necessary to choose a very high order, which requires a more robust calculation program.

It is therefore judicious to opt for the simplified form of equation (1) obtained for very low values of t (equation 2):
The problem here to be solved is to determine the time interval in which equation ( 2) is valid.By making the change of variable 1/ 2 t x = , we obtain the equation below: The graph of should lead to a straight line passing through the origin with slope: as a function of 1/2 t gives the following graph (Figure 8): The determination of D must be done in the linear part, that is to say in the time interval [0 -81] h either [0 - 3,375] days.The optimization carried out using the SigmaPlot software leads to the following result: D = 7,293.10 - mm 2 /j with a standard deviation of 1,179.10 - mm 2 /j (r 2 = 0,9544).

Discussion
The difference between the absorption rates on the first and second day is due to the greater absorption of water vapor under 98%RH (AR = 2%) than under 72%RH (AR = 6%) by the movies.This high absorption rate is due to the voids created by the plasticizer between the polymer chains which will be filled by water vapor over time [15].The gaits show that there are absorption kinetics.This is well proportional to the hygrometry.
The equality of the absorption rates (16%) after five days under the two exposure atmospheres (72% RH, 98% RH) is essentially due to the botanical nature of the matrix and the presence of reinforcements in the composite material.This is confirmed in the work of Davidovic A. S. [16] where a value of approximately 15% is obtained with expanded starch successively reinforced with natural fibers (wheat straw, hemp, cellulose and cotton linter).This is the saturation state of the composite.The voids were filled with steam.
Angles M. N. et Al.[12], working on virgin films based on cassava starch, found a higher value equal to 62% due to the absence of reinforcement.Indeed, the presence of fibers create tortuous paths between the polymer chains preventing the penetration of water molecules into the composite.The value of the diffusion coefficient, which is 0.723.10 - mm 2 /d, is influenced by the presence of lime, which completely opposes the penetration of water vapour.It is somewhat lower than that found by Anglès M. N. and Al.[12] which was 15.206.10 -3 mm 2 /d.This is favored by the nature of the matrix and especially by the fibers used in the case of our work.On the other hand, this is explained by the presence of hydraulic lime contained in the clay [17].

Conclusion
The study of the behavior of new materials requires particular importance with regard to the use made of them.This is the case of biopolymer composites for the analysis of their hydrophilic behavior.The aim of this work was to make a contribution to the study of the water behavior of starch-based composite films.To do this, we determined the diffusion and analyzed the water vapor absorption of a biofilm made of starch reinforced with coconut fibers.Tests are carried out to obtain measurements.These have made it possible to study these phenomena.The diffusion coefficient, equal to 0.723.10 - mm 2 /d, was determined by two different approaches, the results of which converge.The latter, compared to those of the literature found by other authors, show a very high agreement with the behavior of such composites.The value of the diffusion coefficient obtained shows that the composite film has a convincing water behavior in the field of packaging.The results of this study will be of great importance for our future materials.We will indeed use, simultaneously for validation.the two methods of this study to analyze the water behavior of future composites developed.

Figure 3 .
Figure 3. Representation of the diffusion cap.

Figure 4
Figure 4 shows the evolution of the absorption rate as a function of time and the humidity rate of films based on cassava starch plasticized with glycerol and reinforced with coconut mesocarp fibres.

Figure 4 .
Figure 4. Evolution of the absorption rate as a function of time and humidity rate.

Figure 7 .
Figure 7. Evolution of the mass according to equation (11) for for the orders n=2, 3 et 4.

Figure 8 .
Figure 8. Graph of experimental values of t 0 M M M ∞ − as a function of 1/2 t .

Table 2 .
Experimental values for the determination of the rate of absorption and the coefficient of diffusion.