Journal of Materials Processing Technology 124 (2002) 238–243

The influence of humidity on the deformation and

fracture behaviour of PMMA
A.M.S. Hamouda*
Faculty of Engineering, Department of Mechanical and Manufacturing Engineering,
Universiti Putra Malaysia, Serdang, 43440 Selangor, Malaysia

Abstract

The effect of relative humidity on the tensile deformation and fracture of poly(methylmethacrylate) (PMMA) has been studied. Seven sets
of samples were conditioned under different environment: as-received specimens; dry specimens (zero moisture content); outdoor exposed
specimens; water saturated specimens; specimens conditioned at three different relative humidities (75.5, 85.1, and 94.6%) using saturated
salt solutions (NaCl, KCl, and KNO3, respectively). The main objective of this study is to investigate the effects of sorbed water on tensile
mechanical properties of PMMA. Properties investigated include development of crazing, tensile fracture stress, strain to fracture and fracture
surface morphology. # 2002 Published by Elsevier Science B.V.

Keywords: Poly(methylmethacrylate); Tensile deformation; Relative humidity

1. Introduction about by polymers. No less dramatic and certainly more

important has been the impact of plastics in agriculture.
Most materials are subject to weathering. The earth’s crust Whatever the application, there is often a natural concern
is degraded by chemical and physical processes as a result of regarding the durability of polymeric materials partly
exposure to the surrounding environment. The rate of dete- because of their relative newness but also because of the
rioration depends on the nature of the material; for the useful lifetime of these materials can be predicted where
hardest rocks the time scale stretches to millions of years their maintenance and replacement can be planned.
whereas for some organic polymers major changes can be The deterioration of a material depends on how and to
induced by exposure of a few days. what extent it interacts with its surroundings. The outdoor
Synthetic polymers offer an impressive range of attractive environment if considered in terms of sunshine, temperature,
properties and in many of their applications they are exposed rainfall and wind, varies widely in duration, intensity and
to the outdoor environment (e.g. sunlight, temperature, sequence. As regards the durability of materials the weight-
moisture). This could lead to the changes of physical and ing, which should be given to severe climatic conditions,
mechanical properties of polymers. For example, polymers depends on the confidence level required in the performance
and composites are widely used externally as well as intern- of the material but in general it is the time-averaged climatic
ally in aircraft because of the substantial weight saving they factors, which should be considered. It is through the pre-
provide. Likewise in boat construction, plastics, composites valence of certain elemental features for prolonged periods
and synthetic fibres are displacing more conventional that the concept of climate has evolved.
materials. A similar development has occurred in the build- In this study, poly(methylmethacrylate) (PMMA) was
ing industry where plastics are competing successfully with selected to investigate the effects of moisture content on
wood, glass, metal and stone as construction materials. The its properties. PMMA is an amorphous glass below 110 8C
modern sports stadium with its transparent canopy, row of and exhibits brittle fracture under normal conditions below
injection-moulded ABS seats and polypropylene ‘grass’ in 80 8C [1]. PMMA have wide range of applications, which
many ways highlights the revolution in materials brought are principally concerned with transparency and light trans-
fer, with good weathering resistance, and with high surface
hardness and durability. They include glazing in aircraft and
other transport, transparent guards and covers, lenses, baths,
*
Tel.: þ603-948-6101-2083; fax: þ603-894-8839. washbasins and sanitary ware, and rear-light assemblies for
E-mail address: [email protected] (A.M.S. Hamouda). road transport. Since most of the PMMA applications are

0924-0136/02/$ – see front matter # 2002 Published by Elsevier Science B.V.

PII: S 0 9 2 4 - 0 1 3 6 ( 0 2 ) 0 0 0 9 6 - 1
 A.M.S. Hamouda / Journal of Materials Processing Technology 124 (2002) 238–243 239

subjected to wet environment, residual stress in the shaping, level of moisture content were tested using Instron 8020
which can add to the moisture absorption and lead to Universal Testing Machine with capacity of 250 kN with
crazing. Further development of crazing will result crack ram velocity of 1.00 mm/min. Tensile stress–strain curves
and rupture. are obtained. The main objectives of the tests are to obtain
The tensile deformation behaviour of PMMA has been the values of tensile strength and strain to fracture, and the
examined by various investigators over a wide range of relation of tensile strength and tensile stress–strain to frac-
temperature and strain rate [2–4]. However, the influence of ture versus water contents for PMMA.
humidity on mechanical properties of PMMA has received
much less attention compared to nylon [5,6]. 2.4. Surface morphology
In this paper, experiments were carried out to identify the
values of tensile stress and strain to fracture for PMMAs Fracture surfaces are examined using JEOL JSM 6400
specimens with different percentage of water concentra- SEM. Prior to SEM examination, the fracture surfaces were
tions. Scanning electron microscope (SEM) analysis was coated with a thin layer of gold–palladium using Polaron
focused on the effects of the moisture on the deformation E5100 SEM Coating Unit to prevent surface charging.
behaviour and fracture mechanics of PMMA at the fracture
point.
3. Result and discussion

2. Material and methodology 3.1. Water absorption in PMMA

2.1. Material and conditioning In order to obtain variety of relative humidities (75.5,
85.1, and 94.6%), three types of saturation salt solutions
The material used was commercially available PMMA (NaCl, KCl, and KNO3, respectively) were prepared. Inten-
rod, 16 mm in diameter. Solid cylindrical samples were sities of these saturated salt solutions were calculated as
machined using CNC turning machine from the rods into shown in Table 1. Relative humidities were not monitored as
tensile specimens with dimension f 16 mm  210 mm with this was previously done by Greenspan [11].
gauge diameter 9.70 mm. Two sets of tensile test were The changing in the weights of PMMA samples during the
performed, the first set was carried out on dry specimens conditioning process to different relative humidities and
(zero moisture), and the second set was carried out on liquid water environment were monitored from time to time
specimens with different level of moisture contents. The to get the water content in the samples as a function of time.
second sets of the specimens were conditioned at three It has been found that all the samples exhibited the same
different relative humidities (75.5, 85.1 and 94.6%) using behaviour as time increases water content increases. How-
saturated salt solutions (NaCl, KCl and KNO3, respectively) ever, the rate of water absorption decreases with respect to
at ambient temperature. All the specimens were conditioned time. It was found that samples in liquid water environment
until the water content reach equilibrium. Samples of high leads to the rate of water absorption; followed by 94.6, 85.1,
water content were observed to be somewhat hazy and less and 75.5% relative humidity.
transparent than normal PMMA. It was also found that increasing relative humidities from
To obtain dry samples, the test specimens were heated in a 75.5 to 94.6% had lead to the increasing of water content in
vacuum oven at 50 8C for a period of 70 h [7]. Water PMMA from 0.30 to 0.54% after being conditioned for 45
saturated samples were obtained by storing test specimens days in room temperature. This indicates that the increase of
in liquid water for 45 days at room temperature. water content in PMMA is due to the increases of relative
humidity at surrounding. For water saturated PMMA, it was
2.2. Moisture absorption isothermal found that the water content has the highest value, 0.80%. It
is clear that PMMA will absorb more water in liquid water
The moisture sorption isotherm was constructed using environment rather than in water vapour surrounding. How-
gravimetric method (continuous registration of weight ever these values are relatively low compared to other
changes). Sellable glasses made containers were used to
contain the saturated salt solution and the specimens, which
were placed about 3 cm above the water level and separated Table 1
Intensities of saturated salt solutions
by a filter paper [8].
Salt Volume of still Intensity of saturated
2.3. Tensile test water used (ml) salt solution (g/ml)
Type Weight (g)

Methods of tensile test are based on British Code of NaCl 119.0070 450 0.2646
Practice (BS.2782) [9] and ASTM Standard (D638-91) KCl 122.1578 450 0.2715
KNO3 126.0266 450 0.2801
[10], wherever applicable. All the specimens with different
240 A.M.S. Hamouda / Journal of Materials Processing Technology 124 (2002) 238–243

Fig. 1. The tensile strength response at different humidity levels.

Fig. 2. Average stress versus water content.

polymers such as Nylon 6 that can absorb both liquid water but reduction in hardness and stiffness as described by
and water vapour from surrounding in a proportion app- Hertzberg [13].
roaching 9% by weight [12]. At water contents to 0.80%, the strain to fracture increases
and there is a significant reduction in the modulus. This
3.2. Calculation of mechanical properties behaviour is indicative of water acting as a mild plasticizer
for PMMA, a conclusion similar to that drawn from rheo-
Overall, the values of mechanical properties were optical studies [14]. However, the behaviour of stress to
obtained directly from the Instron machine during the tensile fracture and strain to fracture above 0.80% water concentra-
testing. The tensile stress–strain response for different water tion cannot be obtained since the highest water concentration
contents is shown in Fig. 1. Input parameters just before the in samples is 0.80%. However, as water concentrations reach
testing were carried out were gauge length and gauge 1% or so, it is suggested [15] that some water molecules
diameter. All the calculations for desired output parameters
had been carried out by the computer attached to the
machine. The average and standard deviation values were
also calculated.

3.3. Influence of water on mechanical behaviour of PMMA

The tensile stress–strain response is shown in Fig. 1 for

individual test samples having 0, 0.40, and 0.80% of sorbed
water. The average stress at fracture sf as a function of water
concentration is shown in Fig. 2 while the strain to fracture ef
is shown as a function of water concentration in Fig. 3.
With increase of water content to 0.40%, the average
stress to fracture falls gradually from 75.519 to 62.566 MPa
and then increase to 67.080 MPa at 0.44% water content
before falls more rapidly to 43.250 MPa at 0.80% water
content. The mechanism by which water is accommodated
in the PMMA microstructure appears to be changing with
increase in water concentration. The sorbed water, probably
by hydrogen bonding to the carbonyl groups in the PMMA
side chain, lowers inter-chain interaction and thereby
reduces glass transition temperature. This leads to the imp-
rovement of flexibility, ductility and toughness of polymers Fig. 3. Average strain at fracture versus water content.
 A.M.S. Hamouda / Journal of Materials Processing Technology 124 (2002) 238–243 241

Fig. 5. Fracture of dry.

in Fig. 6(a). Figs. 7–10 show similar scan for sample with
water content 0.30, 0.40, 0.44, and 0.46%, respectively.
Several broad, ribbed bands surround the smooth, slow-
growth portion of the fracture surface and the fast growth
region beyond the banded area is rough. The smooth pre-
crack source area, the ribbed bands and the remaining rough
fracture surface are characteristic features of tensile fracture
Fig. 4. Average tensile strength versus water content. in the PMMA. The ribbed bands are indicative of instability
in the crack propagation process. Their presence has been
discussed by Doyle [16] and attributed by him to a periodic
begin to cluster and act more as filler particles than as a slowing of the crack front movement due to energy expen-
plasticizer which leads to stress concentration effects, accent- diture in the development of craze bundles ahead of the fast
uates craze development and causes early fracture. moving crack front.
The influence of sorbed water on tensile strength for the The fracture surface for samples with 0.54% water con-
PMMA is shown in Fig. 4. All values cited are average tent is different from samples with lower water content.
values obtained from tests on three specimens. Dry samples Necking appeared prior to fracture as shown in Fig. 6(b)
have the highest average tensile strength, 73.397 MPa. With indicates that new deformation modes have developed and
the increase of water content, the tensile strength falls
gradually to 63.194 MPa at 0.40% water content and then
increases to 74.414 MPa at 0.54% water content before
reduces to 59.843 MPa at 0.80% water content. The drop
in tensile strength at 0.80% (water saturated) is about 18.5%
compared to the dry specimens.
As water content increases from 0 to 0.80%, there was a
significant change in deformation behaviour for PMMA
from brittle mode to ductile mode as shown in Fig. 1.
The area under the stress–strain curve up to fracture for
PMMA, which indicates the toughness, increases with
increasing of water content in PMMA from 0 to 0.80%.
This means that the increase of water in PMMA has led to Fig. 6. Diagrams of necking: (a) dry; (b) wet.
the increment of toughness. With higher water content,
ability of PMMA to absorb energy up to fracture will be
improved.

3.4. Surface morphology

The fracture surface morphology of PMMA did not vary

appreciably with the water content of the samples range
from 0 to 0.46%. The reason for this is that all the samples
fractured without shear yielding by craze breakdown and
subsequent rapid crack propagation. Fig. 5 shows a portion
of the fracture surface near to the fracture origin for a dry
sample. No necking was found for the dry sample as shown Fig. 7. Fracture at 0.30% water content.
242 A.M.S. Hamouda / Journal of Materials Processing Technology 124 (2002) 238–243

Fig. 8. Fracture at 0.40% water content. Fig. 11. Fracture at 0.54% water content.

Fig. 9. Fracture at 0.44% water content.

Fig. 12. Fracture at 0.80% water content.

the fracture surface morphology changes. Both crazes and Fig. 11 shows a high magnification scan of a portion of the
shear bands are present. Instead of continued craze growth, fracture surface of 0.54% water content sample near to the
shear deformation modes develop which lead to necking and source of the fracture. Fracture has developed from lateral
more extensive plastic deformation. The surface crazes are extension and tearing of a small surface diamond cavity and
relatively small, and the growth is impeded by shear bands. features of equivalent geometry are present on the opposite
In many of the surface crazes the SEM scan shows that a fracture surface. The semi-circular, fan-shaped region show-
central cavity has started to form. With continued deforma- ing radial tear lines indicates the extent of the relatively slow
tion, some of these small cavities open up into relatively crack growth prior to rapid crack extension. Characteristics
large diamond shaped, or truncated diamond shaped cav- such as those discussed above can be explained by; the
ities. While many of the small cavities have formed at the presence of small crazes and diamond shaped cavities on the
centre of surface crazes, others have apparently formed in external surfaces, and a fracture that develops by further
regions where there is no visible evidence of crazes. These extension and growth of one of the diamond cavities, are also
may have initiated from local stress concentration sites noted in water saturated samples with 0.80% water content
arising from the presence of contaminants. Similar cavities whose fracture surface is shown in Fig. 12.
are reported to occur in cellulose by Cornes et al. [17], which
does not craze. 4. Conclusions

The following conclusions can be drawn:

1. The strain to fracture for PMMA increases with the
increasing of water content.
2. The tensile strength for PMMA reduces with the
increasing of water content.
3. It is suggested that the water acts as mild plasticizer for
PMMA at water contents to about 0.80%.
4. The fracture toughness of PMMA increase as the water
content increase from 0 to 0.80%, which indicates that
the ability of energy absorbing up to fracture will be
Fig. 10. Fracture at 0.46% water content. increased.
 A.M.S. Hamouda / Journal of Materials Processing Technology 124 (2002) 238–243 243

5. PMMA samples with higher water content have [4] M. Kawagoe, A.M. Morit, Fracture mechanisms of polymethyl
tendency to deform plastically before fracture where methacrylate under static torsion in alcohol environment, J. Mater.
Sci. 29 (1994) 6041.
crazes, shear bands, and diamond shaped surface [5] A.M.S. Hamouda, M.S.J. Hashmi, The influence of water sorbed on
cavities were observed. the mechanical behaviour of nylon under static and dynamic loading,
6. Fracture has occurred from growth, and then breakdown, Mech. Solids Mater. Eng. c (1992) 753–758.
of a surface craze and the smooth region surrounding the [6] A.M.S. Hamouda, M.S.J. Hashmi, Mechanical behaviour of nylon at
high strain rate and room temperature, J. Key Eng. Mater. 86 (1993)
source is indicative of the extent of craze growth prior
109–114.
to the length of the bar, which is the direction of the [7] W.M. Cheng, G.A. Miller, J.A. Manson, R.W. Hertzberg, L.H. Sperling,
principal stress. Mechanical behaviour of polymethyl methacrylate: Part 3. Activation
processes for fracture mechanism, J. Mater. Sci. 25 (1990) 1931.
[8] J. Shen, C.C. Chen, J.A. Saver, Effect of sorbed water on properties
of low and high molecular weight PMMA: 1. Deformation and
Acknowledgements fracture behaviour, J. Polym. 26 (1985) 511.
[9] W.M. Cheng, G.A. Miller, J.A. Manson, R.W. Hertzberg, L.H.
The financial support provided by the Department of Sperling, Mechanical behaviour of polymethyl methacrylate: Part 1.
Tensile strength and fracture toughness, J. Mater. Sci. 25 (1990) 1917.
Mechanical and Manufacturing Engineering, UPM, is grate- [10] Standard Test Method for Tensile Properties of Plastics, ASTM,
fully acknowledged. The author would thank and express his 1993.
sincere appreciation to Mr. Te Chuan Keat, for his help with [11] L. Greenspan, Humidity fixed points of binary saturated aqueous
the experimental work and Dr. Elsadig Mahdi and Ms. solutions, J. Res. Nat. Bur. Stand. A 81A (1) (1977) 51.
Rohani Yusof, for their help in preparing the manuscript. [12] M.W.A. Paterson, J.R. White, Effect of absorption on residual
stresses in injection-moulded Nylon 6,6, J. Mater. Sci. 27 (1992)
6229.
[13] R.W. Hertzberg, Deformation and Fracture Mechanics of Engineer-
References ing Materials, 3rd Edition, Wiley, New York, 1989, p. 206, 387.
[14] J.A. Johnson, D.W. Jones, The mechanical properties of PMMA and
[1] I. Yasufumi, N. Brown, Environmental crazing and intrinsic tensile its copolymers with ethyl methacrylate and butyl methacrylate, J.
deformation in polymethyl methacrylate: Part 2. Morphology, J. Mater. Sci. 29 (1994) 870.
Mater. Sci. 11 (1976) 425. [15] M. Kawagoe, M. Kitagawa, Craze initiation in polymethyl
[2] I. Yasufumi, N. Brown, Environmental crazing and intrinsic tensile methacrylate exposed to n-alkanes, J. Mater. Sci. 22 (1987) 3000.
deformation in polymethyl methacrylate: Part 1. Mechanical [16] M.W. Doyle, A mechanisms of crack branching in polymethyl
behaviour, J. Mater. Sci. 11 (1976) 417. methacrylate and the origin of the bands on the surfaces of fracture, J.
[3] P.J. Burchill, Effect of absorbed water, temperature and strain rate on Mater. Sci. 18 (1983) 687.
the yield strength of two methylmethacrylate-based plastics, J. Mater. [17] P.L. Cornes, K. Smith, R.N. Haward, J. Polym. Sci. Polym. Phys.
Sci. 24 (1989) 1936. Edn. 15 (1997) 995.