低收縮不飽和聚酯樹脂在聚合固化中之微孔洞生成、本質聚合收縮及反應動力之研究

行政院國家科學委員會專題研究計畫 成果報告

低收縮不飽和聚酯樹脂在聚合固化中之微孔洞生成、本質聚 合收縮及反應動力之研究

計畫類別： 個別型計畫

計畫編號： NSC92-2216-E-011-014-

執行期間： 92 年 08 月 01 日至 93 年 07 月 31 日 執行單位： 國立臺灣科技大學化學工程系

計畫主持人： 黃延吉

計畫參與人員： 黃兆緯同學、吳政潔同學、陳紹偉同學、馬鈞奎同學、董志平 同學

報告類型： 精簡報告

處理方式： 本計畫涉及專利或其他智慧財產權，2 年後可公開查詢

中 華 民 國 93 年 11 月 2 日

行政院國家科學委員會補助專題研究計畫成果報告

※※※※※※※※※※※※※※※※※※※※※※※※※

※低收縮不飽和聚酯樹脂在聚合固化中之微孔洞生成、※

※本質聚合收縮及反應動力之研究 ※

※※※※※※※※※※※※※※※※※※※※※※※※※

計畫類別：ˇ個別型計畫 □整合型計畫 計畫編號：NSC 92-2216-E-011-014

執行期間：92 年 8 月 1 日至 93 年 7 月 31 日 計畫主持人：黃延吉教授

共同主持人：

計畫參與人員：黃兆緯同學、吳政潔同學、陳紹偉同學、馬鈞奎 同學、董志平同學

成果報告類型(依經費核定清單規定繳交)：ˇ精簡報告 本成果報告包括以下應繳交之附件：

□赴國外出差或研習心得報告一份

□赴大陸地區出差或研習心得報告一份

□出席國際學術會議心得報告及發表之論文各一份

□國際合作研究計畫國外研究報告書一份

處理方式：除產學合作研究計畫、提升產業技術及人才培育研 究計畫、列管計畫及下列情形者外，得立即公開查詢

□涉及專利或其他智慧財產權，□一年 ˇ二年後 可公開查詢

執行單位：台灣科技大學化工系

中華民國 93 年 10 月 30 日

1

計劃名稱：低收縮不飽和聚酯樹脂在聚合固化中之微孔洞生成、本質聚合收縮及反應動力之 研究

Investigation of Microvoid Formation, Intrinsic Polymerization Shrinkage, and Reaction Kinetics in the Cure of Low-Shrink Unsaturated Polyester Resins

計劃編號：NSC92-2216-E-011-014 執行期限：92/8/1 – 93/7/31

計劃主持人：黃延吉教授

執行機構：台灣科技大學化工系

中文摘要

本計劃探討不同分子量之聚醋酸乙烯 (PVAc)型抗收縮劑(LPA)

對低收縮不飽和聚酯樹脂(UP)其高溫 110^oC 聚合固化過程中樣品之微孔

洞生成、本質聚合收縮及反應動力之影響。實驗結果，吾人以 DSC、

SEM、OM、影像分析、原子力顯微鏡(AFM)、動態機械測定(DMA)、

熱剌激去極化電流分析 (TSC)、及萬能機械測試之整合性測量，包括苯 乙烯(ST)/ UP/ LPA 三成份系在未反應前之相溶性、聚合固化反應動力、

聚合固化反應後之樣品破壞面之微觀型態結構、聚合固化過程中所生成 的微孔洞体積分率及大小分佈、聚合固化後樣品的玻璃轉移溫度、及機 械性質，解釋之。根據 Takayanagi 機械模式，聚合固化後的加工成品可 同時獲致良好的体積收縮控制及可接受的內部可染性色的因素，在文中 亦有探討。

關鍵詞：聚醋酸乙烯；抗收縮劑；不飽和聚酯樹脂；聚合固化；微孔洞；

反應動力；原子力顯微鏡；体積收縮特性；內部可染色性

ABSTRACT

The effects of poly(vinyl acetate) with different molecular weights as low-profile additives (LPA) on the microvoid formation, intrinsic polymerization shrinkage and reaction kinetics for low-shrink unsaturated polyester resins (UP) during the cure at 110^oC were investigated. The experimental results have been explained by an integrated approach of measurements for the static phase characteristics of the styrene (ST)/UP/LPA system, reaction kinetics, cured sample morphology, volume fraction and size distribution for the microvoid formed during the cure，glass transition temperatures for the cured samples, and mechanical properties by using differential scanning calorimetry (DSC), scanning electron microscopy (SEM), optical microscopy (OM), image analysis, atomic force microscopy (AFM), methods of dynamic mechanical analysis (DMA), the method of thermally stimulated current (TSC), and a universal testing machine. Based on the Takayanagi mechanical model, factors leading to both a good volume shrinkage control and acceptable internal pigmentability for the molded parts have been explored.

Keywords: poly(vinyl acetate), low-profile additive (LPA); unsaturated

INTRODUCTION

Adding specific thermoplastic polymers as low profile additives (LPA) in the unsaturated polyester resins (UP) can lead to a reduction or even elimination of the polymerization shrinkage during the cure¹. However, low-profile polyester molding compounds when formulated with pigments may usually exhibit an unacceptable hazing of the pigment's color. In recent years, unique non-reactive LPAs for high temperature applications have been developed² that give significantly improved deep color pigmentation while maintaining a smooth surface and zero shrinkage, yet the fundamental principle has not been treated. Most recently, Huang et al ³ have studied the effects of volume fraction and average size of microvoid generated during the cure of ST/UP/LPA systems on the volume shrinkage characteristics and internal pigmentability. Three factors in microvoid formation, including the polymerization strain in the crosslinked polyester phases, the interfacial area between the LPA-rich phase and the crosslinked UP phase, and the yield strain for the LPA-rich phase, were proposed.

Since they employed the image analysis method on OM micrographs, only relative volume fraction of microvoids can be measured. Also, the size distribution of microvoids was unable to be measured.

In a study of carbon black filled polyethylene conductive composites, Oakey et al⁴ demonstrated that atomic force microscopy (AFM) can be used to identify void size and absolute volume fraction of microvoids through a comparison of complementary height and phase impages.

The objective of this work is to investigate the effects of poly(vinyl acetate) types of LPA and their molecular weights on the microvoid formation, intrinsic polymerization shrinkage and reaction kinetics for styrene(ST)/UP/LPA ternary systems cured at 110^oC. Above all, the atomic force microscopy (AFM) has been used to identify size distribution and absolute volume fraction of microvoids. Glass transition temperatures by DMA and TSC and mechanical properties of the cured samples have also been studied.

EXPERIMENTAL

Materials. The molecular characteristics of poly(vinyl acetate) types

The UP resin^5-6 was made from maleic anhydride (MA), 1,2-propylene glycol (PG), and phthalic anhydride (PA) with a molar ratio of 0.62:1.28:0.38. The acid number and hydroxyl number were found to be 28.5 and 30.3 by end-group titration, which gives an Mn of 1910 g/mole.

For the sample solution, 10% by wt. of LPA was added, while the molar ratio of styrene to polyester C=C bonds was fixed at MR = 2/1. The reaction was initiated by 1% by wt. of tert-butyl perbenzoate (TBPB). For the sample solution with pigments, 10% by wt. of Bordeaux Red was added as pigments. All the cure reactions were carried out at 110^oC isothermally for 1 hr.

Instrumentation and Procedure. The compatibility of ST/UP/LPA

The absolute quantity of microvoids for the cured ST/UP/LPA systems was measured by AFM method^4,10, where the area method was employed for the determination of volume fraction of microvoids, and the line segment method for the size distribution of microvoids. The details can be found elsewhere¹⁰.

For the measurement of the glass-transition temperatures for the fully cured samples (i.e. cured at 110^oC for 1 h, followed by a postcure at 150^oC for 1 h), dynamic mechanical analysis (DMA)^10-11 and the method of stimulated currents (TSC)^8,12-13 measurements have been carried out.

Impact strength and tensile properties of the fully cured samples have also been measured^10-11.

RESULTS AND DISCUSSION

Interrelationship Between the Polarity Difference of UP and LPA and the Compatibility of Styrene/UP/LPA Systems. The molecular

µ

Compatibility of Styrene/UP/LPA Systems During the Cure. During

cause a less compatible ST/UP/LPA system, with the precipitation of microgel particles being more clearly identified. Adding a higher molecular weight of LPA can also lead to the increase in viscosity for the ST/UP/LPA system, and the phase separation rate during the cure may then be slowed down. The reason why the PVAc3S and PVAc4S systems were more compatible than the PVAc2S system is due to the effect of increasing viscosity (increasing phase separation time) predominating over the effect of decreasing compatibility (increasing the degree of phase separation) for the former systems with higher LPA molecular weights.

Relationship Between Morphologies and Mechanical Properties – the Takayanagi Models. For the cured LPA-containing UP resin systems

Between the LPA-covered microgel particles there are some lightly St-crosslinked polyesters chains and polystyrene chains (taken together as the P2 phase), which have different compositions of St and UP than those in the P1 phase, that are dispersed in the LPA phase (R phase). Hence, the characteristic globule microstructure may be represented by the parallel-parallel-series (P-P-S) model as shown in Fig. 1(a). In contrast, for the ST/UP/LPA systems with PMMA as an LPA, a two-phase microstructure would arise¹⁴, which can be represented by a P-(P-P-S) model (Fig. 1(b)).

Cure Kinetics. For the ST/UP/PVAc systems, except for the PVAc4S

Effects of LPA on Volume Shrinkage After the Cure. The volume

Effects of LPA on Internal Pigmentability After the Cure. Adding

≅ -2.2%),

≅

Effects of Microvoid Formation and Intrinsic Polymerization on Volume Shrinkage and Internal Pigmentability. Pattison et al.

This strain can increase to the point that stress cracking propagates through the weak LPA phase, relieving this strain, forming microcracks and/or microvoids, and compensating for the overall volume shrinkage by the microcrack or microvoid space. For the ST/UP/LPA system containing PVAc as an LPA, the relative volume fraction of microvoid data, vm’, in Table 4, as measured by the image analysis method on OM micrographs, show that the higher the vm’, the lower the fractional volume shrinkage, which has supported the volume shrinkage compensation through strain relief mentioned above.

Besides microvoid formation, the intrinsic polymerization shrinkage⁹ is also important in determining the volume shrinkage control during the cure of ST/UP/LPA system. This is especially true of the ST/UP/LPA system with a reactive LPA, which will not be discussed in this report.

The microvoid formation is intimately connected with the internal pigmentability². As incident light enters an internally pigmented part of the cured ST/UP/LPA system, the intensity of reflective light to the pigment can be greatly reduced due to the severe light scattering which may occur at the solid/air (microvoid) interface inside the parts, leading to the hazing of a pigment’s color. In theory, the higher the volume fraction of microvoids, the worse the internal pigmentability. For the ST/UP/LPA system

containing PVAc as an LPA, except the PVAc1S and PVAc2S systems, the relative volume fraction of microvoid data shown in Table 4 agreed with the experimental results, where the system with a larger volume fraction of microvoids, such as the PVAc4S system, would lead to a worse internal pigmentability (i.e. a higher L^* value).

In addition to the volume fraction of microvoids, the average size of microvoids is also influential for the internal pigmentability. It should be noted³ that when the average size of the microvoids becomes larger than 0.05-0.1

µm, which is about 1/10 to 1/5 of the wavelength of visible light

Size Distribution, Average Size, and Volume Fraction of Microvoids by AFM. For the cured ST/UP/LPA systems, the size distribution and the

Data in Table 5 (AFM pictures not shown) show that the average size of microvoids, sm, for the four ST/UP/LPA systems ranged from 0.19 to 0.27

µm. The s

decreasing L^* value). (Table 4) Besides, the higher percentage of microvoids in the bracket of 0.25-0.45 µm for the PVAc2S system (Table 5) would greatly enhance the light scattering between the air/solid interface of the cured samples, and hence can lead to the poor performance of internal pigmentability.

For the ST/UP/LPA systems, the volume fraction of microvoids, vm, measured by AFM ranged from 18 to 27% (Table 5), which would be way too high. It is anticipated that the method of small angle X-ray scattering (SAXS) can allow us to measure the size and volume fraction of microvoids accurately in the future. (A SAXS system will be installed in our university at the end of 2004.) Nevertheless, adding a higher molecular weight of LPA can lead to a higher vm (by AFM), which shows the same trend as that in Table 4 (by the image analysis method on OM micrographs).

Glass Transition Temperatures by DMA. Table 6 shows DMA results for

β

The Tg values displayed in Table 6 reveal that, as the MR increases, the Tg1α exhibits an increase followed by a decrease; it reaches a maximum at a 2/1 MR (Tg1α = 162.6^oC), which is a trend similar to those reported in the literature³.

Table 6 also shows the DMA results for the cured UP resins containing 10 wt% PVAc as an LPA. We propose a P-P-S model (Fig. 1(a)) for all of the four compatible ST/UP/LPA systems. For the Tg listed in Table 6 for the ST/UP/LPA systems, Tg1α is the glass-transition temperature for the major continuous phase of the St-crosslinked polyester (i.e., phase P1, Fig. 1), Tg1β (^oC) is the β relaxation temperature for the polyester segments between the crosslink in mainly the densely St-crosslinked polyester phase [i.e., phase P1, Fig. 1(a)], Tg2 is the glass-transition temperature for the LPA concontinuous phase [i.e., phase P2, Fig. 1(a)], and TgR is the glass-transition temperature of the R phase. For the PVAc systems represented by a P-P-S model, the value of Tg2 cannot be clearly identified by DMA, which may be superposed with Tg1β.

It is found that the Tg1α was higher for the ST/UP/LPA systems than that of the neat UP system at MR = 2/1. (162.8-173.3 ^oC vs. 162.6^oC) It is ascribed to the fact⁷ that adding PVAc as an LPA may enhance the formation of microgel particles during the cure and the conversion of polyester C=C bonds may be increased, leading to a higher crosslinking density for the St-crosslinked network and hence a higher Tg1α.

Glass Transition Temperatures by TSC. Table 7 shows TSC results for the

The variation of Tg1α with MR shows the same trend as that of DMA in Table 6.

3

Based on the TSC results, Tg1α was also higher for the ST/UP/LPA systems than for the neat UP system at MR = 2/1 (i.e. 157.2-157.6^oC vs 141.9^oC) However, Tg1α was unable to be identified for the PVAc3S and the PVAc4S systems. It can be due to the relatively large volume fraction of microvoid generated in the cured sample (see Tables 4 and 5), which may greatly reduce the intensity of depolarization current in the TSC test below the sensitivity range of the TSC detector.

Mechanical Properties. Table 8 shows the impact strength and tensile

CONCLUSIONS

For ST/UP/LPA systems, the microvoid formation can affect both the volume shrinkage control and internal pigmentability for the cured sample.

In this work, by employing the PVAc4S type of LPA (Mn = 106000 g/mol; Tg

= 27.2^oC) and the MA-PA-PG type of UP resin (with a molar ratio of 0.62:1.28:0.38 for MA, PG, and PA), a good volume shrinkage control (∆V/V0 = -2.2%), and acceptable pigmentability (L^* = 28-29) for the molded parts have been achieved.

The atomic force microscopy (AFM) can be used to measure the size distribution, average size, and volume fraction of microvoids generated during the cure of ST/UP/LPA system. A higher LPA molecular weight can lead to a higher volume fraction of microvoids and hence a better volume shrinkage control. (PVAc4S > PVAc3S > PVAc2S > PVAc1S) Although the volume fraction of microvoids measured by AFM ranges from 18 to 27%, which is way too high, yet the volume fraction of microvoids measured by the indirect method of image analysis and the direct method of AFM shows the same trend. In the future work, it is suggested that the small angle X-ray scattering technique be applied to measure the volume fraction and the size distribution of microvoids for the ST/UP/LPA systems.

The average size of microvoids, sm, for the four ST/UP/LPA systems as measured by AFM ranged from 0.19 to 0.27 µm. The sm was the largest for the PVAc2S system, followed by the PVAc1S, the PVAc3S, and the PVAc4S systems, which was generally in agreement with the trend of increasing internal pigmentability (i.e. decreasing L^* value). Besides, the higher percentage of microvoids in the bracket of 0.25-0.45 µm for the PVAc2S system would greatly enhance the light scattering between the air/solid interface of the cured samples, and hence can lead to the poor performance of internal pigmentability.

In addition to the DSC rate and conversion profiles for cure kinetics, the DMA and TSC results, which allow us to identify Tg in each phase region as a result of phase separation during the cure, have been useful for the in-depth analysis of the cure behavior of ST/UP/LPA systems.

REFERENCES

1. Atkins, K.E. In Sheet Molding Compounds: Science and Technology;

Kia, H.G., Ed.; Hanser Publishers: New York, 1993; Chapter 4.

2. Atkins, K.E.; Rex, G.C.; Reid, C.G.; Seats, R.L.; Candy, R.C. 47^th Annu. Conf., Cincinnati, OH, Feb. 1992, 7-D.

3. Huang, Y.J.; Chen, T.S.; Huang, J.G.; Lee, F.H. J. Appl. Polym. Sci.

2003, 89, 3336.

4. Oakey, J.; Marr, D.W.M.; Schwartz, K.B.; Wartenberg, M.

Macromolecules 2000, 33, 5198.

5. Huang, J.W. M.S. Thesis, National Taiwan University of Science and Technology, 2004.

6. Huang, Y.J.; Jiang, W.C. Polymer 1998, 39, 6631.

7. Huang, Y.J.; Su, C.C. J. Appl. Polym. Sci. 1995, 55, 305.

8. Ma, J.K. M.S. Thesis, National Taiwan University of Science and Technology, 2004.

9. Huang, Y.J.; Liang, C.M. Polymer 1996, 37, 401.

10. Chen, S.W. M.S. Thesis, National Taiwan University of Science and Technology, 2004.

11. Huang, Y.J.;Horng, J.C. Polymer 1998, 39, 3683.

12. Ibar, J.P. Fundamentals of Thermal Stimulated Current and Relaxation Map Analysis; SLP Press: New Canaan, CT, 1993.

13. Huang, Y.J.; Lee, S.C.; Dong, J.P. J. Appl. Polym. Sci. 2000, 78, 558.

14. Dong, J.P.; Huang, J.G.; Lee, F.H.; Roan, J.W.; Huang, Y.J. J. Appl.

Polym. Sci. 2004, 91, 3388.

15. Pattison, V.A.; Hindersinn, R.R.; Schwartz, W.T. J. Appl. Polym. Sci.

1974, 18, 2763.

16. Cook, W.D.; Delatycki, O. J. Polym. Sci. Polym. Phys. Ed. 1974, 12, 2111.

17. Huang, Y.J.; Chen, T.S; Huang, J.G.; Lee, F.H. J. Appl. Polym. Sci.

2003, 89, 3347.

Table 1. LPAs Used in This Study

LPA codes monomer M

M

PD

T

(C)

PVAc1S VAc 23000 67000 2.9 29.7

PVAc2S VAc 65000 134000 2.1 28.9

PVAc3S VAc 88000 167000 1.9 29.5

PVAc4S VAc 106000 231000 2.2 27.2

a:by GPC (g/mole); b:by DSC.

Table 2. Molar Volumes and Dipole Moments for UP and LPA UP or LPA dipole

moment µ (debye/mole)

molar volume V (cm

/mol)

dipole moment per unit volume µ’

( =(µ

/V)

)

MA-PG-PA 2.79 1063.1 0.086

(b) Non-Reactive

LPAs

PVAc1S 10.39 17040 0.080

PVAc2S 17.86 50340 0.080

PVAc3S 20.78 68150 0.080

PVAc4S 22.81 81970 0.080

Table 3. Phase Separation Time, t

, for ST/UP/LPA Uncured Systems at 25

C, and Final Conversions of Total C=C Bonds, α,as Measured by

DSC for ST/UP/LPA Systems Cured at 110

C

LPA added t

(min) Upper layer (wt.%) α(%)

Neat UP 89.2

LPAs

PVAc1S

∞

PVAc2S

∞

PVAc3S

∞

PVAc4S

∞

a: one phase (after 1440 min)

Table 4. Volume Shrinkage, ∆V/V

, L

for Both Sides of Molded Parts, and Relative Volume Fraction of Microvoids, v

, for ST/UP/LPA Cured

Systems

LPA added ∆V/V

(%) L

(upper) L

(botto m)

v

(%)

a

22.45(0.1 8)

- PVAc1S -6.54(0.08) 25.97(0.08) 29.95(0.3

0)

0.72 PVAc2S -5.46(0.05) 38.44(0.08) 56.00(0.0

6)

0.80 PVAc3S -3.76(0.11) 23.31(0.02) 24.39(0.0

9)

1.75 PVAc4S -2.24(0.20) 27.62(0.03) 29.46(0.0

5)

3.35 a: The values in parentheses represent the estimated standard errors.

Table 5. Absolute Volume Fraction of Microvoids, v

, Average Size of Microvoids, s

, and Size Distribution of Microvoids for ST/UP/LPA

Cured Systems by AFM Void Size

(µm) PVAc1S PVAc2S PVAc3S PVAc4S

0 – 0.15 45.6% 25.4% 47.2% 50.8%

0.15 – 0.25 17.1% 17.2% 22.5% 22.6%

0.25 – 0.35 13.2% 33.0% 13.2% 13.3%

0.35 – 0.45 11.2% 17.2% 11.2% 13.3%

0.45 – 0.55 5.6% 5.9% 5.6% -

0.55 – 0.65 5.6% - - -

0.65 – 0.75 1.8% 2.0% - -

Total 100% 100% 100% 100%

Avg. Void Size, sm

0.24 µm 0.27

µm 0.20 µm 0.19 µm

Distribution of Voids

Unimodal Bimodal Unimodal Unimodal

Volume Fraction of

Voids, vm

17.9% 18.6% 25.5% 27.1%

Table 6. Glass-Transition Temperatures of Fully Cured ST/UP/LPA Systems Based on tanδ by DMA

LPA MR Model T

T

T

T

Neat UP 1/1 157.6 89.9

2/1 162.6 92.4

3/1 157.8 101.4

6/1 140.4 100.0

LPA

PVAc1S 2/1 P-P-S 167.7 121.5 60.6 PVAc2S 2/1 P-P-S 173.3 114.0 57.4 PVAc3S 2/1 P-P-S 164.1 113.1 52.4 PVAc4S 2/1 P-P-S 162.8 112.4 52.7

Table 7. Test Conditions and Glass-Transition Temperatures of Fully

Cured ST/UP/LPA Systems by TSC (a)

LPA MR Polar.

Temp (

C)

Polar.

Time (min)

Elec.

Field (V/mm)

Sample Thickness

(mm)

2/1 150 0.01 120 1.601

3/1 150 0.01 120 1.324

6/1 150 0.01 120 1.547

LPA

PVAc1S 2/1 150 60 500 0.952

PVAc2S 2/1 150 60 500 0.964

PVAc3S 2/1 150 60 500 0.957

PVAc4S 2/1 150 60 500 0.948

(b)

LPA MR Model T

T

T

T

Neat UP 1/1 117.7 105.2

2/1 141.9 118.3

3/1 141.5 -

6/1 138.8 116.2

LPA

PVAc1S 2/1 P-P-S 157.6 88.7 24.8 PVAc2S 2/1 P-P-S 157.2 77.4 23.9

PVAc3S 2/1 P-P-S - 78.3 24.1

PVAc4S 2/1 P-P-S - 69.2 23.4

Table 8. Impact Strength, Tensile Strength, Ultimate Tensile Strain, and Young’s Modulus for ST/UP/LPA Cured Systems after Isothermal

Cure at 110

C for 1 h and Postcure at 150

C for 1 h LPA

added

Impact strength

(J/m)

Tensile strength (MPa)

Ultimate tensile strain (%)

Young’s modulus (MPa)

(a)

(b)

Fig. 1 The Takayanagi models for mechanical behavior of cured

LPA-containing UP resin systems. (a)parallel-parallel-series (P-P-S)

model, and (b)parallel-parallel-parallel-series (P-(P-P-S)) model. The

area of each diagram is proportional to a volume fraction of the phase.