The study of rubber-modified plastics with higher heat resistance and higher toughness and its application

Abstract

For rubber-modified plastics, toughness enhancement is generally at the cost of heat resistance. Of significance, this paper reported a finding that rubber-modified plastics with a special morphology could enhance toughness and heat resistance cooperatively. The special morphology was such that an interface in situ formed between plastic matrix and rubber particle had higher hardness than plastic matrix. As a result, the hard interface not only helped rubber soft component integrate with plastic matrix by covalent bonds to impart plastic matrix high toughness, but also covered rubber nanoparticles as hard shells to protect them from deforming at high temperature. The special morphology had been achieved in rubber-modified epoxies and phenolic molding material. The forming mechanism of the hard interface was studied in detail with AFM, DSC and in situFTIR, by using rubber-modified epoxy resin as an example. The finding could be applied to any rubber-modified plastics as long as the special morphology could be realized.

---

Human beings have entered the Polymer Age.¹ Global plastics production has exceeded 250 million tons per year,² already much more than steel production in volume. Plastics have penetrated into everyone's life. Since the emergence of plastics in 1910, plastics toughening has been a century's challenge in various forms for scientists. Among many a method that has been achieved for the toughening of plastics including thermosets and thermoplastics, rubber-toughening was the most conventional and successful one, due to its low cost and easy processing, especially for plastics such as high impact polystyrene (HIPS), liquid rubber-toughened epoxy resin and EPR-modified polypropylene.^3–5 However, for rubber-modified plastics, toughness enhancement was generally at the cost of heat resistance, and on the other hand, high heat resistance was extremely significant for plastics in many applications. For example, rubber-modified epoxy resin or phenolic molding materials with higher toughness and heat resistance were required in the electronics industry, due to the high temperature required for lead-free solders' application after replacing traditional solders for health reasons. Previously, our group had reported study of a series of nanoscale rubber particles less than 100 nm in diameter, such as carboxylic/acrylonitrile/butadiene rubber nanoparticle (CNBR-NP) and acrylonitrile/butadiene rubber nanoparticle (NBR-NP). Surprisingly, those rubber nanoparticles were able to simultaneously improve heat resistance and toughness of the epoxy resins⁶ and the phenolic molding materials modified by them.⁷ This is different from the conventional mechanism of rubber-modified plastics.^8–10 Many papers^11–15 showed that for epoxy resin modified either by liquid rubber such as carboxyl-terminated acrylonitrile butadiene (CTBN), or conventional solid rubber such as butadiene/acrylonitrile rubber (NBR), the enhancement of toughness was almost at the sacrifice of heat resistance.

In this paper, a mechanism for the simultaneous improvement of heat resistance and toughness of rubber-modified epoxies was studied in detail with AFM, DSC and in situFTIR. A novel morphology was proposed and confirmed by using CNBR-NP modified epoxies as an example. Finally, the mechanism was verified again by NBR-NP modified epoxies with a heat distortion temperature (HDT) increased by about 35%.

Materials and instrumentation, as well as the preparation of rubber nanoparticles, epoxy composites and experimental details were described in the ESI.† The experimental procedures were summarized as follows. CNBR-NP and NBR-NP were prepared by irradiating their corresponding rubber latex with a cobalt-source (dosage: 25 KGy and dose rate: 50 Gy/min) to crosslink the rubber particles before drying with a spray dryer. The epoxy resin modified by CNBR-NP or NBR-NP was prepared as follows. First, a blend of 12 phr CNBR-NP or NBR-NP with 100 phr diglycidyl ether of bisphenol-A (DGEBA) was milled by a three-roll miller three times at room temperature. Then the blend was mechanically stirred for 30 min at 90 °C after the addition of 75 phr curing agent, methyl tetrahydrophenylanhydride (MeTHPA). Subsequently, 1.5 phr initiator, triethanolamine (TEA), was added if necessary, and then the blend was degassed during stirring at 90 °C for 15 min. Finally, the blend was cast into a preheated PTFE mould, and was cured at 130 °C for 15 h. At last the specimens for mechanical measurement were machined in terms of their measuring requirements.

Izod impact strength was measured with an impact machine (CEAST, AFS/MK3) according to testing standard GB 1843. Heat-distortion temperature was measured with a heat-distortion tester (148-HDPC-6) according to testing standard GB 1634. Glass transition temperature (T_g) was measured by DSC (Perkin-Elmer Pyris DSC-7) according to testing standard GB 19466.2. The sample used for AFM scanning was carefully cut by ultramicrotome with a glass knife at about −100 °C according to the procedures previously described.^16–20AFM images were scanned by Dimension 3100 AFM (Veeco Metrology Group, USA) under tapping mode with tip nominal spring constant of 2.8 N m⁻¹ at room condition. Curing kinetics was monitored by means of DSC and in situFTIR (Nicolet AVATAR 360). DSC was performed from room temperature to 300 °C at a heating rate of 5 °C min⁻¹ under nitrogen, and in situFTIR was carried out at 130 °C for 20 min.

As reported in ref. 6, the toughness of the epoxy resin modified by 12 phr CNBR-NP dramatically was increased by 96% from 11.4 kJ m⁻² to 22.3 kJ m⁻², compared with neat epoxy resin. Significantly, though a soft rubber phase was introduced, its heat resistance showed, instead of a decrease, an increase of 1.1% from 113.2 °C to 114.4 °C and of 3.5% from 136.1 °C to 140.9 °C, in terms of HDT and T_g respectively. Moreover, T_g increased up to 144.5 °C when the content of CNBR-NP was up to 20 phr. It was extremely difficult to explain why a rubber component could increase heat resistance of the epoxy resin since it disagreed with conventional toughening rules that soft modifiers usually lead to deteriorated heat resistance.

Inspired by these puzzling results, we studied the interface between rubber nanoparticles and epoxy matrix using AFM. By carefully adjusting the set-point amplitude ratio to keep a proper tip–sample interacting force, AFM phase image could be used to qualitatively compare the relative hardness of heterogeneous regions of plastics in combination with the information of height images.^21,22 The technique had been also widely applied to image distribution of heterogeneous regions with different hardness in plastics composites.^23–25 Its detailed mechanism was described in the ESI.† Therefore, the relative hardness of epoxy resin, rubber nano-particle and their interfaces could be qualitatively compared by AFM.

In the AFM phase images in Fig. 1a of epoxy resin modified by 12 phr CNBR-NP, the brown spheres corresponding to soft components were rubber nanoparticles here, i.e. CNBR-NP, which was further confirmed by AFM height and TEM images given in the ESI.† Interestingly, around every CNBR-NP always existed a brighter ring. Moreover, the rings were brighter in color than the epoxy resin. This indicated that the ring, i.e. the interface between rubber nanoparticle and epoxy matrix, was harder than the epoxy matrix. A magnified phase image of a CNBR-NP in Fig. 1b showed the brighter interface more clearly, and its cross-section curve through the particle center (Fig. 1c) showed that the phase shift of the interface was about 3° more than that of the epoxy matrix. Therefore, the results indicated that the hardness of the epoxy resin/CNBR-NP interfaces was higher not only than CNBR-NP but also than the epoxy matrix. Additionally, Fig. 1b and 1c showed that the thickness of the interfaces was estimated to be about 7 nm, roughly. Given that CNBR-NP has an average diameter of about 90 nm, a content of 12 phr, and a homogeneous dispersion in epoxy resin as indicated by AFM image in Fig. 1a and TEM image in Fig. S2,† content of the hard interface could be calculated to be about 6.5 phr, assuming that density of the interfaces was equal to that of rubber nanoparticles. In fact, the hard interfaces acted as in situ formed shells covering rubber nanoparticles and the shell was harder than epoxy matrix. Moreover, the hard shells connected the rubber particles and epoxy matrix through covalent bonds, which would be manifested in the following. As a result, about 18.5 phr special core-shells particles, i.e. 6.5 phr hard shells plus 12 phr soft rubber nanoparticles, were uniformly loaded in the epoxy resin. The formation of the hard shells should play the key role in increasing heat resistance of the CNBR-NP modified epoxy resins.²⁶

Fig. 1 AFM phase images of epoxy resin modified by 12 phr CNBR-NP. (a) AFM phase image, (b) magnified AFM phase image of a rubber nanoparticle, and (c) the phase shift from cross-section of the rubber nanoparticle in (b).

To determine reactions that made the shells harder than epoxy matrix, curing kinetics was carefully studied with DSC and in situFTIR. Fig. 2a gave the curing kinetics curves of different epoxy systems studied by DSC. For the sample composed of DGEBA/MeTHPA without initiator, no peak appeared in the DSC curve as shown by the curve (I) in Fig. 2a. This indicated that no chemical reactions occurred for this sample. However, for the sample composed of DGEBA/CNBR-NP/MeTHPA an exothermic peak appeared due to the introduction of CNBR-NP as indicated by curve (II) in Fig. 2a. The change should be attributed to a reaction which took place between CNBR-NP and epoxy system. Therefore, the function of CNBR-NP in the curing reaction should be similar to that of initiators. That is, CNBR-NP could activate the curing reaction. In contrast to the sample initiated by TEA as shown by curve (III) in Fig. 2a, the catalytic ability of CNBR-NP in the curing reaction was shown to be weak. However, it was key to enhancing heat resistance of the modified epoxy resin. Therefore, it is significant to know why and how CNBR-NP can activate the curing reaction of epoxy resin.

Fig. 2 The curing kinetics curves of epoxy resin studied by DSC and in situFTIR. (a) DSC curing curve of different epoxy systems. Their formulations: (I) 100 phr DGEBA, and 75 phr MeTHPA, (II) 100 phr DGEBA, 75 phr MeTHPA, and 12 phr CBNR-NP, and (III) 100 phr DGEBA, 75 phr MeTHPA, 12 phr CBNR-NP, and 1.5 phr TEA. (b) The evolution of absorbance peak of the nitrile group in CNBR-NP with reacting time measured by in situFTIR at 130 °C. The formulations were 100 phr DGEBA, 12 phr CNBR-NP, and 75 phr MeTHPA.

As is well-known, the nitrile group in CNBR-NP is a Lewis base. Considering its similar structure to acetonitrile (CH₃CN), the pK_a of the nitrile group in CNBR-NP can be approximated to be that of acetonitrile with pK_a = 31.3 in DMSO. The value is higher than that of PhNH₂ in the same solvent which can act as an initiator for epoxy systems.²⁷ Therefore, the nitrile group in CNBR-NP should theoretically speed up the curing reaction as initiators. To confirm the idea, in situFTIR was used to monitor the evolution of the curing reaction at the absence of initiators at 130 °C. The absorbance of the nitrile group in CNBR-NP was at 2236 cm⁻¹ in Fig. 2b, and its intensity decreased quickly from 0.048% to 0.025% in 5 min, and subsequently became plateau. The results again verified the proposed explanation that the nitrile group could accelerate the curing reaction as an initiator, and its initiating reaction could be given by

where R₁ and R₂ were aromatic or alkyl groups, respectively.

To sum up, nitrile group of CNBR-NP could catalyze the curing reaction of epoxy resin. The catalytic reaction made cross-linking density of the CNBR-NP/epoxy resin interfaces higher than epoxy matrix, and subsequently made the formed interfaces harder than epoxy matrix. Therefore, the hard interfaces could act as in situ hard shells covering the rubber particles. As a result, a special core-shell structure was formed, which was illustrated by the schematic diagram in Fig. 3a. The in situ formed hard shells connected soft rubber particles with plastic matrix through covalent bonds. Since the shell was harder than the epoxy matrix, the rubber nanoparticles were protected from deforming at enhanced temperature, and the structure increased heat resistance and toughness of the rubber-modified epoxies cooperatively.

Fig. 3 (a) A schematic illustration of a hollow nanoparticle harder than epoxies covering a rubber particle. (b) Normalized DSC curing curves of different epoxy resins. Their formulations: (I) 100 phr DGEBA, and 75 phr MeTHPA, (II) 100 phr DGEBA, 75 phr MeTHPA, and 12 phr NBR-NP, and (III) 100 phr DGEBA, 75 phr MeTHPA, 12 phr BNR-NP, and 1.5 phr TEA.

As shown in Fig. 2a, a stronger peak at lower temperature occurred in DSC curing curve when 1.5 phr TEA was added. The results showed that the catalytic ability of CNBR-NP was much weaker than that of TEA. If the catalytic ability of the rubber nanoparticles used was improved the shell would become harder. Since the catalytic ability of CNBR-NP originated from the basic properties of the nitrile group, the acidity of the carboxyl group in CNBR-NP had the possibility to deactivate the catalytic efficiency of the nitrile group. As a result, the curing reaction became difficult. Therefore, when NBR-NP replaced CNBR-NP, it was expected that the catalytic ability of nitrile group became stronger and the shell would become harder. Fig. 3b showed the DSC curing kinetic curves of the DGEBA/MeTHPA/NBR-NP system. When NBR-NP replaced CNBR-NP, the exothermic peak shown by the curve (II) in Fig. 3b was stronger, sharper and occurred at lower temperature, compared with the curve (II) of the DGEBA/MeTHPA/CNBR-NP system in Fig. 2a. It was attributed to the enhanced catalytic ability of NBR-NP over CNBR-NP. Introducing TEA to the blend, a novel peak at even lower temperature occurred as shown by the curve (III) in Fig. 3b, which was derived from the TEA-initiated curing reaction. Interestingly, the peak intensity for the curing reaction initiated by the nitrile group was stronger than that initiated by TEA. It was mainly due to both the high content of nitrile groups and its high catalytic efficiency. The results again confirmed that it was the nitrile group that effectively initiated the curing reaction.

Table 1 shows the mechanical and physical properties of epoxies modified by NBR-NP. For the epoxy resin initiated just by NBR-NP without any other initiators, the curing reaction could be also completed because the catalytic ability of NBR-NP was stronger than that of CNBR-NP as demonstrated above. Moreover, its HDT was raised by 35% from 98 °C to 132 °C, compared with the neat epoxy resin. For the epoxies modified by NBR-NP, the curing reaction started only from rubber nanoparticles surfaces, and hence the cross-linking density near the rubber particles should be much higher than that of the bulk. Therefore, the formed interface was much harder than the epoxy matrix. And the formed hard interfaces could act as hard shells to impart higher heat resistance to epoxy resins. As a result, epoxy resin with higher heat resistance was prepared. On the contrary, the curing reaction started not only from the interface but also from epoxy matrix when TEA was also added. Therefore, it limited the diffusion of epoxy group to the interface, which affected the interface curing reaction.

Table 1 Physical properties of NBR-NP-modified epoxies^a

Samples	Izod impact strength /standard deviation (kJ m^−2)	Flexural strenth (MPa)	T g (°C)	HDT (°C)
a (A) neat epoxy system with formulations of 100 phr DGEBA, 75 phr MeTHPA, and 1.5 phr TEA; (B) epoxy system modified by NBR-NP cured without TEA with formulations of 100 phr DGEBA, 75 phr MeTHPA, and 12 phr NBR-NP; (C) epoxy system modified by NBR-NP cured with TEA with formulations of 100 phr DGEBA, 75 phr MeTHPA, 12 phr BNR-NP, 1.5 phr and TEA.
(A)	9.0/2.4	105	102	98
(B)	15.6/2.4	96.51	133	132
(C)	16.3/3.3	84.8	114	111

---

In conclusion, toughness and heat resistance could be cooperatively enhanced for rubber-modified plastics with a special morphology. The morphology was such that rubber nanoparticles were covered with an in situ formed shell. And the shell must have greater hardness than the plastic matrix. The hard shells not only integrated the soft component of rubber with plastic matrix by covalent bonds to impart high toughness to the plastic matrix, but also covered rubber nanoparticles as hard shells to protect them from deforming at higher temperature. More significantly, the results would help material scientists design rubber-modified plastics with high performance in both toughness and heat resistance.

Acknowledgements

We primarily acknowledge financial support from National Basic Research Program of China (2005CB623800).

Notes and References

1. M. Rubinstein and R. H. Colby, Polymer Physics, Oxford University Press, London, 1st edn, 2003, pp. 1.
2. L. Pilato, Phenolic Resins: A Century of Progress, Springer,New York, 2010, p. V.
3. H. G. Karian, Handbook of Polypropylene and Polypropylene Composites, Marcel Dekker Inc., New York, 2nd edn, 2003.
4. A. J. Peacock, Handbook of Polyethylene Structure, Properties, and Applications, Marcel Dekker Inc., New York, 2000.
5. E. Martuscelli, P. Musto, G. Ragosta, Advanced Routes for Polymer Toughening, Elsevier, Amsterdam, 1996.
6. F. Huang, Y. Liu, X. Zhang, G. Wei, J. Gao, Z. Song, M. Zhang and J. Qiao, Macromol. Rapid Commun., 2002, 23, 786.
7. H. Ma, G. Wei, Y. Liu, X. Zhang, J. Gao, F. Huang, B. Tan, Z. Song and J. Qiao, Polymer, 2005, 46, 10568.
8. B. Wetzel, P. Rosso, F. Haupert and K. Friedrich, Eng. Fract. Mech., 2006, 73, 2375.
9. A. C. Garg and Y. W. Mai, Compos. Sci. Technol., 1988, 31, 179.
10. S. H. Wu, Polymer, 1985, 26, 1855.
11. P. Punchaipetch, V. Ambrogi, M. Giaberini, W. Brostow, C. Carfagna and N. A. D'Souza, Polymer, 2002, 43, 839.
12. D. Ratna, Polymer, 2001, 42, 4209.
13. S. Balakrishnan, P. R. Start, D. Raghavan and S. D. Hudson, Polymer, 2005, 46, 11255.
14. R. Thomas, S. Durix, C. Sinturel, T. Omonov, S. Goossens, G. Groeninckx, P. Moldenaers and S. Thomas, Polymer, 2007, 48, 1695.
15. C. W. Wise, W. D. Cook and A. A. Goodwin, Polymer, 2000, 41, 4625.
16. M. Zhang, Y. Liu, X. Zhang, J. Gao, F. Huang, Z. Song, G. Wei and J. Qiao, Polymer, 2002, 43, 5133.
17. W. Fan, L. Wang and S. Zheng, Macromolecules, 2009, 42, 327.
18. J. Marquardt, R. Thomann, Y. Thomann, J. Heinemann and R. Mülhaupt, Macromolecules, 2001, 34, 8669.
19. X. Zhang, Y. Liu, J. Gao, F. Huang, Z. Song, G. Wei and J. Qiao, Polymer, 2004, 45, 6959.
20. L. Zheng, A. F. Xie and J. T. Lean, Macromolecules, 2004, 37, 9954.
21. G. Bar, Y. Thomann, R. Brandsch and H. J. Cantow, Langmuir, 1997, 13, 3807.
22. D. Raghavan, M. Vanlandingham, X. Gu and T. Nyuyen, Langmuir, 2000, 16, 9448.
23. T. Karino, Y. Ikeda, Y. Yasuda, S. Kohjiya and M. Shibayama, Biomacromolecules, 2007, 8, 693.
24. B. D. Mather, M. B. Baker, F. L. Beyer, M. A. G. Berg, M. D. Green and T. E. Long, Macromolecules, 2007, 40, 6834.
25. C. Schellenberg, S. Akari, M. Regenbrecht, K. Tauer, F. M. Petrat and M. Antonietti, Langmuir, 1999, 15, 1283.
26. M. Jose, P. Rodriguez and P. Virgil, Macromolecules, 1986, 19, 55.
27. F. G. Bordwell, Acc. Chem. Res., 1988, 21, 456.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c0py00413h

---