Viscoelastic and Mechanical Properties of Polyimide-Clay Nanocomposites

Mohamed O. Abdalla, Derrick Dean*, Sandi Campbell

Tuskegee-Center for Advanced Materials, 101 Chappie James Center, Tuskegee

University, Tuskegee, Alabama 36088,

NASA Glenn Research Center, Cleveland, Ohio,

44135.

Abstract

Polyimide-clay nanocomposites were prepared by blending 2.5 and 5 wt.% of PGV (Na

montmorillonite) and two organically modified PGV (PGVC10COOH, PGVC12) with a

methanol solution of PMR-15 precursor. DMA results showed a significant increase in

the thermomechanical properties (E

′

and E

′′

) of 2.5 wt.% clay loaded nanocomposites in

comparison with the neat polyimide. Increasing the clay loading to 5 wt.% degraded

these properties. Higher T

s were observed for 2.5 wt.% nanocomposites compared to

the neat polyimide. A lower CTE was observed only for the PGV/PMR-15

nanocomposite. Flexural properties measurements for the 2.5 wt.% nanocomposites

showed an improvement in the modulus, strength and elongation. This trend in the

tensile properties was not observed for the 5 wt.% nanocomposites.

1. Introduction

Only a few studies [1,2,3] have addressed the high performance thermosets

nanocomposites such as PMR-15 (Fig 1.). Polymerizable reactive mixtures (PMR),

polyimides are a group of a variety of thermosetting polymers used in high temperature

environments (T

~ 300-430ºC). PMR-15 is a state of the art thermoset polyimide mainly

used as a matrix resin for carbon fiber-reinforced composites for aerospace applications.

It was perfected by Serafini et. al. [4] at NASA Lewis Research Center in 1972. The

general advantages of PMR-15 composites are their thermal stability up to 300°C, good

mechanical properties, relatively easy processing and low cost. In this study, we will

report the synthesis and morphological characterization of layered silicates (LS) and

organically-modified layered silicates (OLS)/PMR-15 nanocomposites. We will also

report the viscoelastic and mechanical properties of the consolidated nanocomposites

versus those of the neat polymer.

2. Experimental

2.1. Preparation of organically- modified layered silicates

The OLS were synthesized by a cation-exchange reaction between the PGV (Na

montmorillonite, Nanocor, Inc.) and the ammonium salt of modifiers´ (dodecylamine and

11-aminoundecanoic acid). PGV was dispersed in water at 70-80ºC. Excess modifier

(twice the cation exchange capacity of the clay) was dissolved in water at 70-80ºC and an

equivalent amount of concentrated HCl acid was added to the solution. The dispersion of

PGV was added to the solution of the modifier and this mixture was stirred vigorously for

1 h. A white precipitate was isolated by suction-filtration, placed in a 600 ml beaker with

400 ml of hot water, and stirred for 1 h. This process was repeated two times to ensure

Q6.30.1

https://doi.org/10.1557/PROC-726-Q6.30

Downloaded from https://www.cambridge.org/core. University of Arizona, on 27 Jul 2018 at 06:48:46, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.

the removal of the excess ammonium salt. The filter cake was then freeze-dried over-

night.

Fig 1. Synthetic scheme for PMR-15

2.2. Preparation of polyimide-clay nanocomposites.

A desired amount (10-15 g) of PMR-15 precursor solution and varying amounts (2.5

and 5 wt.%) of PGV or OLS were added. Methanol (100-150 ml) was added for dilution.

The mixture was stirred at room temperature for 1.5 hours with a mechanical stirrer to

perform complete clay dispersion into the polymer matrix. The methanol was evaporated

off at 50-60°C. A viscous solution of PMR-15/clay nanocomposite was obtained. The

viscous nanocomposites solutions were imidized (or b-staged) at 204°C (5.1°C/minute)

for 1 h and 232°C (1.4°C/minute) for 30 minutes. The imidized products were ground to

form a golden powder.

2.3. Consolidation of the nanocomposites

The b-staged powders were placed in a mold [4 × 4 × 0.039 in] and a thermocouple of

a digital thermometer was connected to the mold. The mold was placed in the carver

laboratory hydraulic press and heating started. Pressure (2000 psi) at 270°C was applied

and the mold was held at these conditions for 2 h. The heat was turned off and the mold

was left to cool to room temperature before the molded samples were removed. The

molded samples were post-cured in an air-circulating oven at 315ºC for 5 h.

3. Results and discussions

3.1. Preparation of organically modified layered silicates (OLS)

XRD curves of the LS and OLS are shown in Fig. 2. The peak observed for the

LS (PGV) at 2

= 6.61° (d

001

= 13.34 Å), which corresponds to the basal spacing between

the layers of the silicate layers, has shifted to lower angles. PGVC10COOH, modified

MeO

O, CH

(

)

MDA

BTDE

Heat, Pressure

(

)

Imidized Prepolymer,Formulated molecular weight,FMW = 1500

Thermally Cross-linked Polyimide

Q6.30.2

https://doi.org/10.1557/PROC-726-Q6.30

Downloaded from https://www.cambridge.org/core. University of Arizona, on 27 Jul 2018 at 06:48:46, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.

11-aminoundecanoic acid, exhibited a sharp intense peak at 4.97° corresponding to a d-

spacing of 17.78 Å. While the PGVC12, modified using dodecylamine, exhibited a sharp

intense peak at 2

= 4.36° corresponding to a d-spacing of 20.54 Å. Thus, in both

modified PGV clays, the interlayer spacing has expanded signifying the incorporation of

the modifiers between the silicate layers.

2345678910

100

200

300

400

500

600

700

800

(Degrees)

6.61

001

= 13.34 Å)

4.97

001

= 17.78 Å)

4.36

001

= 20.54 Å)

PGV

PGVC10COOH

PGVC12

Relative Intensity

2θ

Fig. 2. XRD curves of PGV and organically modified PGV

3.2. X-ray diffraction analyses of PMR-15/LS and OLS nanocomposites

XRD curves of the (a) B-staged and (b) consolidated (2.5% clay loading)

nanocomposites are shown in Fig. 3. Table 1 summarizes the analyses of these XRD

curves. The B-staged prepolymers exhibited weaker peaks that were shifted to lower

angles compared to the organically modified PGV, suggesting the PMR-15 oligomers

diffused and intercalated between the clay layers leading to partial disruption of the clay

layers. The broadening of the peaks for the consolidated PGV and

PGVC10COOH/PMR-15 nanocomposites indicated that further delamination in the

silicate clay layers has occurred without further loss in their regularity. For the

consolidated (fully cured) PGVC12 nanocomposite, disappearance of the peak at 2

4.43º suggested complete loss of regularity of the silicate clay layers indicative of an

exfoliated structure.

2345678910

100

200

300

400

500

600

700

800

PGVC12

PGVC10COOH

(Degrees)

PGV

Relative Intensity

2θ

Fig. 3. (a) XRD curves of B-staged

2.5% clay loaded LS and OLS nanocomposites

2 345678910

100

200

300

400

500

600

700

800

(Degrees)

PGVC12

PGVC10COOH

PGV

Relative Intensity

2θ

(b) XRD curves of consolidated

2.5% clay loaded LS and OLS nanocomposites

Q6.30.3

https://doi.org/10.1557/PROC-726-Q6.30

Downloaded from https://www.cambridge.org/core. University of Arizona, on 27 Jul 2018 at 06:48:46, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.

In Fig. 4, the XRD curves of the (a) B-staged and (b) consolidated (5% clay loading)

nanocomposites are shown, respectively. A similar trend to that observed in the

structural development of the 2.5% clay loaded nanocomposites was observed, however,

delamination of the clay layers for the PGVC12/PMR-15 nanocomposite begins at the B-

stage and is maintained after consolidation.

2345678910

100

200

300

400

500

600

700

800

(Degrees)

PGVC12

PGVC10COOH

PGV

Relative Intensity

2θ

Fig. 4. (a) XRD curves of B-staged

5% clay loaded LS and OLS nanocomposites.

2345678910

100

200

300

400

500

600

700

800

(Degrees)

PGVC12

PGVC10COOH

PGV

Relative Intensity

2θ

(b) XRD curves of consolidated 5% clay loaded

LS and OLS nanocomposites.

3.3. Viscoelastic properties of PMR-15/LS and OLS nanocomposites

In Fig. 5 (a) the storage modulus (E

′

) and (b) loss modulus (E

′′

) of the 2.5% clay

loaded nanocomposite are shown. Addition of the clays increases the level of the E´ in

the glassy region, with the highest value observed for the PGVC12, which has an

exfoliated morphology by XRD. The effect of the clays on the relaxation behavior and T

can be seen by observing both the E

′

and E

′′

curves. Both curves indicate an increase in

the T

, accompanied by a broadening of this relaxation indicative of restriction of

segmental relaxation. The values for T

s (which were measured from the E

′′

curves) are

summarized in Table 3. The average T

increase was 9ºC, with the PGVC12/PMR-15

exhibiting a 16ºC increase. The E

′′

curves also show a broad

relaxation centered near

100

C, which has been attributed to crosslink motion in PMR-15. This relaxation

becomes broader upon addition of clay, with the largest effect again observed for the

PGVC12 clay. These results (T

and

relaxation changes) indicate polymer-clay

interaction at least at the segment level [5]. This effect is not observed for the samples

containing 5% clay, however (Fig. 6). In fact, both of the samples containing organically

modified clays have glassy E

′

values lower than that for PMR-15, while the PGV/PMR

sample has an E

′

value equal to that of the PMR-15. The T

was increased by an average

of 14ºC for these systems. This finding is corroborated by the E

′′

curve, which also

shows little or no effect of the clays on the

relaxation.

Q6.30.4

https://doi.org/10.1557/PROC-726-Q6.30

Downloaded from https://www.cambridge.org/core. University of Arizona, on 27 Jul 2018 at 06:48:46, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.

100

1000

10000

Storage Modulus (MPa)

0 100 200 300 400

Temperature (°C)

––––––

Neat

"–––––– PGV

#–––––– PGVC10COOH

––––

–– PGVC12

Fig. 5. (a) Storage modulus of PMR-15/LS

and OLS nanocomposites (2.5% clay loading).

100

1000

Loss Modulus (MPa)

0100200300 400

Temperature (°C)

–

––––– PMR-15

"–––––– PGV

#–––––– PGVC10COOH

$–––––– PGVC12

(b). Loss modulus of PMR-15/LS and OLS

nanocomposites (2.5% clay loading).

100

1000

10000

Storage Modulus (MPa)

0100200300 400

Temperature (°C)

!–––––– PMR-15

–––––– PGV

#–––––– PGVC10COOH

$–––––– PGVC12

Fig. 6. (a) Storage modulus of PMR-15/LS

and OLS nanocomposites (5% clay loading).

# # #

$ $

100

1000

Loss Modulus (MPa)

0100200300 400

Temperature (°C)

!–––––– Neat

"–––––– PGV

#–––––– PGVC10COOH

$–––––– PGVC12

(b). Loss modulus of PMR-15/LS

and OLS nanocomposites (5% clay loading).

Table 3. T

s (ºC) of neat PMR-15, LS and OLS/PMR-15 nanocomposites (2.5 and 5% clay loading).

% Clay

loading PGV PGVC12 PGVC10COOH

0 351 351 351

2.5 357 367 358

5 361 369 366

3.4. Flexural properties of PMR-15/LS and OLS nanocomposites

Table 4 summarize the flexural properties of the 2.5% and 5% clay loaded

nanocomposites, respectively. For the neat PMR-15 polyimide, the flexural modulus was

3.5 GPa. With the incorporation of 2.5% PGV, the flexural modulus increased to 4.3

GPa, representing a 23% increase. The nanocomposites loaded with 2.5% of the OLS,

PGVC10COOH and PGVC12, showed a 63 and 31%, respectively, increase in the

modulus compared to the neat PMR-15. The increase in the flexural modulus and

strength of the nanocomposites is accompanied by a small increase in the elongation at

break (except for the PGVC12/PMR-15). As shown in Table 4, increasing the clay

loading to 5% did not show a significant improvement in the flexural modulus, strength

and elongation of the nanocomposites (except for the PGVC10COOH/PMR-15). In fact,

a decrease was observed. This variation in the trend may be caused by variations in the

clay-polymer interface caused by an unknown degree of modifier degradation, since these

modifiers are known to exhibit degradation onsets well below the PMR-15 crosslinking

Q6.30.5

https://doi.org/10.1557/PROC-726-Q6.30

Downloaded from https://www.cambridge.org/core. University of Arizona, on 27 Jul 2018 at 06:48:46, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.

temperature of 316

C [6,7]. The morphological heterogeneity of the intercalated

morphology is also primary contributor. It has been suggested that a more homogeneous,

exfoliated morphology would exhibit even better properties [8]. Examination of the

flexural strength at clay loading of 2.5% reveals a 14 and 49% increase for the PGV and

PGVC10COOH, respectively, while no change is observed for the PGVC12 sample. The

enhanced properties for the PGVC10COOH relative to the PGVC12 sample are assumed

to be due to the more favorable interaction between the carboxylic acid functional groups

of the modifier and the polymer [9].

Table 5. Flexural properties of neat PMR-15, LS, and OLS/PMR-15 nanocomposites.

Clay

(%)

Modulus

(GPa)

Strength

(MPa)

Elongation

at Break

(%)

Neat 0 3.5 ± 3% 96 ± 2% 2.6 ± 6%

PGV 2.5

4.3 ± 2%

3.4 ± 3%

109 ± 3%

76 ± 4%

2.8 ± 5%

2.3 ± 3%

PGVC10COOH 2.5

5.7 ± 4%

3.4 ± 1%

143 ± 4%

110 ± 3%

2.9 ± 6%

1.6 ± 5%

PGVC12 2.5

4.6 ± 2%

2.9 ± 3%

97 ± 3%

57 ± 5%

1.5 ± 5%

2.0 ± 6%

3.4. Linear coefficient of thermal expansion of PMR-15/LS and OLS nanocomposites

Table 6 shows the coefficients of thermal expansion (CTE) of the neat PMR-15 and

the LS and OLS nanocomposites. It was found that the CTE decreased only for the

PGV/PMR-15 nanocomposite while it increased in the case of the OLS/PMR-15

nanocomposites. A similar trend has been reported for thermoplastic polyimide

nanocomposites [10]. As stated in the previous section, potential variations in the

interface and the heterogeneous morphology could possibly lead to the increases in the

CTEs of the OLS/PMR-15 nanocomposites that are observed.

Table 6. CTEs (µm/ºC) of the neat PMR-15 and the LS and OLS/PMR-15 nanocomposites (2.5 and 5%

clay loading).

% Clay

loading PGV PGVC12 PGVC10COOH

0 46.3 46.3 46.3

2.5 34.4 54.1 53.0

5 39.0 58.5 56.2

Q6.30.6

https://doi.org/10.1557/PROC-726-Q6.30

Downloaded from https://www.cambridge.org/core. University of Arizona, on 27 Jul 2018 at 06:48:46, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.

4. Conclusion

High temperature PMR-15/LS and OLS nanocomposites have been prepared using

convention organic ammonium modifiers. Structural evolution as a function of curing

has been studies using XRD. Incorporation of 2.5% clay showed an improvement in the

flexural modulus, strength and no reduction in the elongation. Doubling the clay loading

percentage resulted in degradation of the nanocomposites flexural properties. Higher T

were measured for all the nanocomposite compared to the neat PMR-15, with the highest

values obtained for the 5% clay loaded samples. The effect of the silicates on the

relaxation behavior (i.e. Tg enhancements and broadening of relaxation) indicated

polymer-clay interactions at the segmental level. An improvement in the CTE was

observed for the PGV/PMR-15 nanocomposites.

5. References

[1] Islam M, Dean D, Campbell S American Chemical Society Polymeric, Materials

Science and Engineering 2001;84.

[2] Ganguli S, Dean D, Jordan K, Vaia R. Intercalated Cyanate Ester/Silicate

Nanocomposites, submitted to polymer.

[3] Campbell S, Scheiman D, Faile M, Papadopoulos D to appear in J. of High

Performance Polymers.

[4] Serafini TT, Delvigs P, Lightsey GR. J. Appl. Polym. Sci., 1972;16;905.

[5] Utracki LA. Polymer alloys and blends: thermodynamics and rheology, New York: Hanser

Publishers, 1990.

[6] Islam M, Dean D, Campbell S American Chemical Society Polymeric Materials

Science and Engineering 2001;84.

[7] Campbell S, Scheiman D, Faile M, Papadopoulos D. to appear in J. of High Perform.

Polym.

[8] Le Baron P, Wang Z, Pinnavaia T Appl Clay Sci 1999;15;11.

[9] Agag T, Koga T, Takeichi T Polymer 2001;42;3399-3408.

[10] Hsiao S, Liou, G., Chang, L, J Applied. Polym Sci 2001,80;11;2067-2072.

Q6.30.7

https://doi.org/10.1557/PROC-726-Q6.30

Downloaded from https://www.cambridge.org/core. University of Arizona, on 27 Jul 2018 at 06:48:46, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.