Accurate characterization of moisture absorption in polymeric materials

19 January 2017
Youssef K. Hamidi, Gorkem E. Guloglu, and M. Cengiz Altan
The importance of using the exact solution of the hindered diffusion model is demonstrated on experimental data from a nanoclay/epoxy composite.

Moisture absorption is known to detrimentally affect the mechanical integrity and durability of polymeric materials.1–3 For example, a moisture content as low as 0.75wt% can lead to substantial reductions in the tensile strength, elastic modulus, and interfacial shear strength of polymer composites.4 Consequently, it is critically important to accurately characterize moisture absorption in such materials so that their service life behavior (especially for structural load-bearing applications that are designed for long service life) can be predicted.

Although diffusion of a primarily ‘Fickian’ nature is often observed during initial moisture uptake, numerous polymers exhibit non-Fickian behavior over longer timescales.5–19 To describe this behavior, several models have previously been proposed.5–10 Among these models, the Langmuir model—also known as the hindered diffusion model (HDM)—is thought to be particularly effective for describing absorption in polymeric materials10–19 and is capable of capturing both Fickian and non-Fickian behaviors. The affinity of the polymer for water (associated with specific interactions between water and polymer polar groups) and the diffusivity of water into the polymer (related to polymer interstices, or free volume) both play an important role in the HDM framework. For instance, unbound water molecules are free to diffuse and move through the polymer's interstices. However, molecules that interact with the polymer chains stop diffusing and are considered to be bound. The total moisture concentration in a polymeric material is therefore the sum of the bound and unbound concentrations. Furthermore, the bound moisture is added as a sink term to a conventional Fickian diffusion equation. An additional governing equation links both the bound and unbound moisture concentration. The mass intake can be calculated by integrating the total moisture concentration over the material thickness, and an exact analytical solution for the mass intake is available in the literature.9, 10

Despite the prevalence of the HDM in the literature, there is actually little discussion available on exactly how the model should be implemented to recover the set of absorption parameters (i.e., the absorbed moisture content at equilibrium, diffusivity, as well as the rates of unbound molecules becoming bound and vice versa). Indeed, most researchers10–16 use simple approximations of the model's equations. The exact solutions would yield more accurate characterizations, but these are rarely used because of their perceived complexity. Moreover, the last weight-gain data recorded is generally used as a measure of the absorbed moisture content at equilibrium (M) in absorption experiments. During anomalous absorption behavior, however, moisture absorption decelerates after the initial intake and reaches a pseudo-equilibrium, which is often mistaken for equilibrium. The absorption then continues at a significantly lower rate and equilibrium is reached at a much higher level. The ad hoc selection of the last data as M, coupled with the approximations made in the simplified version of HDM, can therefore lead to substantial errors in the calculated absorption parameters.

To accurately characterize moisture absorption in polymeric materials, all absorption parameters (including M) need to be determined from experimental data. In our work,20 we have thus used the HDM in the proper manner to characterize the moisture absorption of an epoxy composite that contained 3wt% nanoclay (I.30E) and that was immersed in water at 25°C. We collected gravimetric data to recover the nanocomposite's absorption parameters, using a search algorithm20, 21 to simultaneously determine the values of all the absorption parameters for which the error between the experimental data and the model is minimized.

The absorption behavior for our nanoclay/epoxy nanocomposite, as predicted by the recovered model parameters, is illustrated in Figure 1. We show the absorption behaviours predicted by both the approximate and exact solutions of the HDM. We find an excellent agreement between the exact solution and our experimental data. Furthermore, we were able to use the data we collected over a four-month period to accurately predict the moisture uptake of the composite after 11 months. The moisture content that we predict with the approximate HDM solution, however, deviates substantially from the experimental values after the initial linear region has ended.

Comparison between the experimental moisture absorption results for a 3wt% nanoclay/epoxy composite, and the predictions obtained from the exact and approximate solutions of the hindered diffusion model (HDM).

We also find that the parameter values we obtain from the approximate and exact HDM solutions are considerably different (see Table 1). For instance, there is an 8.6% difference between the maximum moisture intake values we calculate from the approximate and exact solution (2.11 and 2.17%, respectively). These disparities in the calculated absorption parameters indicate that the diffusion kinetics described by the two versions of the model are fairly different (even though the moisture uptake curves in Figure 1 partially overlap in the early part of the absorption process). In addition, the approximate prediction fails to capture the long-term absorption level. Overall, our results indicate that recovering the absorption parameters with the approximate solution yields an almost six-fold increase in the root-mean-square error per data point compared with the exact solution.

Moisture absorption parameters for the nanoclay/epoxy composite recovered by using the HDM approximate and exact solutions. RMS: Root mean square.

Recovered parametersApproximate SolutionExact Solution
Diffusivity (×10−4mm2/hr)7.617.47
Bound→unbound molecules (×10−3hr−1)7.178.43
Unbound→bound molecules (×10−3hr−1)1.051.33
Maximum moisture (wt%)2.112.17
RMS error/data point (×10−3)13.662.29

In summary, we have conducted a water absorption experiment on a nanoclay/epoxy composite to measure its water-uptake characteristics over time. We have also used both approximate and exact solutions of the popular HDM to predict the water absorption of the polymeric samples. We find that the exact model provides a good match to the experimental data and that it can be used to predict both the short- and long-term absorption of the nanocomposite samples. In addition, our results highlight the importance of using the exact—rather than the approximate—solution of the HDM to characterize moisture absorption in polymeric materials (because the approximate solutions have significantly higher associated errors). Moreover, all absorption parameters (including the equilibrium moisture content) need to be recovered simultaneously from experimental data. In our future work we plan to explore how added inclusions influence the moisture absorption behavior of our polymeric materials.


Youssef K. Hamidi
School of Aerospace and Mechanical Engineering, University of Oklahoma

Youseff Hamidi is an associate professor of engineering at École Nationale Supérieure des Mines de Rabat and a senior research associate at the University of Oklahoma. His research interests are in fundamental and applied topics relevant to fiber-reinforced composites.

Gorkem E. Guloglu
School of Aerospace and Mechanical Engineering, University of Oklahoma

Gorkem Guloglu is a PhD candidate and graduate research assistant. His research interests are continuum mechanics modeling, simulation, and characterization of process-induced damage in polymeric composite materials.

M. Cengiz Altan
School of Aerospace and Mechanical Engineering, University of Oklahoma

Cengiz Altan is the B. H. Perkinson Chair and Director of the School of Aerospace and Mechanical Engineering. His research interests include the mechanics, manufacturing, characterization, and design of polymeric composites. He has developed unique microstructural simulation techniques and experimental methods to describe manufacturing-induced properties of fiber-reinforced composites.


  1. L. Aktas, Y. K. Hamidi and M. C. Altan, Effect of moisture on the mechanical properties of resin transfer molded composites—part II: desorption, J. Mater. Process. Manufac. Sci. 10, pp. 255-267, 2002.

  2. L. Aktas, Y. K. Hamidi and M. C. Altan, Effect of moisture on the mechanical properties of resin transfer molded composites—part I: absorption, J. Mater. Process. Manufac. Sci. 10, pp. 239-254, 2002.

  3. H. J. Barazza, L. Aktas, Y. K. Hamidi, J. Long, E. A. O'Rear and M. C. Altan, Moisture absorption and wet-adhesion properties of resin transfer molded (RTM) composites containing elastomer-coated glass fibers, J. Adhes. Sci. Technol. 17, pp. 217-242, 2003.

  4. E. Pérez-Pacheco, J. I. Cauich-Cupul, A. Valadez-González and P. J. Herrera-Franco, Effect of moisture on the mechanical behavior of carbon fiber/epoxy matrix composites, J. Mater. Sci. 48, pp. 1873-1882, 2013.

  5. S. Roy, W. X. Xu, S. J. Park and K. M. Liechti, Anomalous moisture diffusion in viscoelastic polymers: modeling and testing, J. Appl. Mechan. 67, pp. 391-396, 1999.

  6. C. Maggana and P. Pissis, Water sorption and diffusion studies in an epoxy resin system, J. Polym. Sci. Part B Polym. Phys. 37, pp. 1165-1182, 1999.

  7. P. M. Jacobs and F. R. Jones, Diffusion of moisture into two-phase polymers part 1: the development of an analytical model and its application to styrene-ethylenelbutylene-styrene block copolymer, J. Mater. Sci. 24, pp. 2331-2336, 1989.

  8. A. R. Berens and H. B. Hopfenberg, Diffusion and relaxation in glassy polymer powders: 2. separation of diffusion and relaxation, Polymer 19, pp. 489-496, 1978.

  9. H. G. Carter and K. G. Kibler, Langmuir-type model for anomalous moisture diffusion in composite resins, J. Compos. Mater. 12, pp. 118-131, 1978.

  10. L. R. Grace and M. C. Altan, Three-dimensional anisotropic moisture absorption in quartz-reinforced bismaleimide laminates, Polym. Eng. Sci. 54, pp. 137-146, 2014.

  11. W. Liu, S. V. Hoa and M. Pugh, Water uptake of epoxy-clay nanocomposites: experiments and model validation, Compos. Sci. Technol. 68, pp. 2066-2072, 2008.

  12. T. Glaskova and A. Aniskevich, Moisture absorption by epoxy/montmorillonite nanocomposite, Compos. Sci. Technol. 69, pp. 2711-2715, 2009.

  13. Y. Joliff, L. Belec and J. F. Chailan, Modified water diffusion kinetics in an unidirectional glass/fibre composite due to the interphase area: experimental, analytical and numerical approach, Compos. Struct. 97, pp. 296-303, 2013.

  14. P. Scott and J. M. Lees, Water, salt water, and alkaline solution uptake in epoxy thin films, J. Appl. Polym. Sci. 130, pp. 1898-1908, 2013.

  15. F. B. Boukhoulda, L. Guillaumat, J. L. Lataillade, E. Adda-Bedia and A. Lousdad, Aging-impact coupling based analysis upon glass/polyester composite material in hygrothermal environment, Mater. Design 32, pp. 4080-4087, 2011.

  16. R. Sridhar, H. N. N. Murthy, B. Karthik, K. R. V. Mahesh, M. Krishna and P. Ratna, Moisture diffusion through nanoclay/vinylester processed using twin-screw extrusion, J. Vinyl Addit. Technol. 20, pp. 152-159, 2014.

  17. T. I. Glaskova, R. M. Guedes, J. J. Morais and A. N. Aniskevich, A comparative analysis of moisture transport models as applied to an epoxy binder, Mechan. Compos. Mater. 43, pp. 377-388, 2007.

  18. G. LaPlante, A. V. Ouriadov, P. Lee-Sullivan and B. J. Balcom, Anomalous moisture diffusion in an epoxy adhesive detected by magnetic resonance imaging, J. Appl. Polym. Sci 109, pp. 1350-1359, 2008.

  19. S. Popineau, C. Rondeau-Mouro, C. Sulpice-Gaillet and M. E. R. Shanahan, Free/bound water absorption in an epoxy adhesive, Polymer 46, pp. 10733-10740, 2005.

  20. G. E. Guloglu, Y. K. Hamidi and M. C. Altan, Fast recovery of non-fickian moisture absorption parameters for polymers and polymer composites, Polym. Eng. Sci., 2016.

  21. L. Aktas, Y. K. Hamidi and M. C. Altan, Combined edge and anistropy effects on fickian mass diffusion in polymer composites, J. Eng. Mater. Technol. 126, pp. 427-435, 2004.

DOI:  10.2417/spepro.006846