View PDF VersionPrevious ArticleNext Article
 
DOI: 10.1039/C6RA14396B (Paper) RSC Adv., 2016, 6, 69743-69747

Cycloketyl radical mediated suspension polymerization of styrene

C. R. Yaoab, L. Wang*ab and W. T. Yang*ab
aThe State Key Laboratory of Chemical Resource Engineering, Beijing, 100029, China
bCollege of Materials Science and Engineering, Beijing University of Chemical Technology, Beijing, 100029, China. E-mail: lwang@mail.buct.edu.cn; yangwt@mail.buct.edu.cn; Fax: +86-010-64416338

Received 3rd June 2016 , Accepted 11th July 2016

First published on 12th July 2016

Abstract

Cycloketyl radical mediated living polymerization (CMP) was applied in the suspension polymerization of styrene (St). The polymerization behavior mediated by 9,9′-bixanthydrol (BIXANDL) both in the absence and presence of a traditional thermal initiator benzoyl peroxide (BPO) was investigated thoroughly. The results show that BIXANDL can initiate the suspension polymerization of St effectively and exert moderate control over molecular weights which grew linearly as St conversion increased. When BIXANDL was utilized along with BPO, the polymerization rate was enhanced and larger increases in chain length with increasing conversions were observed.

Introduction

Free radical suspension polymerization, which usually utilizes water as a reaction medium, has many industrial merits such as environment-friendliness, ease of controlling the reaction heat and facility for obtaining directly-employable products.1–4 Therefore, it has long been the main polymerization process in the production of many commercial resins, including polystyrene (PS), poly(vinyl chloride), poly(styrene-acrylonitrile) and poly(methyl methacrylate).

Macromolecular engineering by designing and controlling chain composition, topology and functionality paves a significant route to obtaining tailor-made high performance, advanced functional and even smart materials. It has blossomed into an active and rapidly developing field with the development of various controlled/living radical polymerization (CLRP) methods, such as nitroxide-mediated polymerization (NMP), atom transfer radical polymerization (ATRP) and reversible addition–fragmentation chain transfer polymerization (RAFT).5–11 CLRP not only enables deliberate control over the molecular weight and polydispersity index (PDI) of polymer chains, but also provides an unrivaled tool for synthesizing polymer chains with a rich diversity of compositions, architectures and functionalities.

The controlled/living features of different CLRP methods in suspension polymerization system have been investigated, aiming to enlarge their fields of application to industrial production.12–22 Compared with solution CLRP, it seems that the controlled/living features of suspension CLRP are inferior. This can be ascribed to the presence of water in suspension system, the pseudo-bulk behavior of suspension polymerization and the transformation of liquid droplets to solid beads or powders during polymerization process. Georges et al.12 carried out suspension copolymerization of styrene (St) and butadiene using NMP and yielded polymer with controlled molecular weight and a PDI of 1.36. Taube et al.13 studied TEMPO-mediated suspension polymerization of St and St with acrylonitrile or butyl methacrylate. It was found that the number average molecular weights (Mn) of the polymers increased linearly with monomer conversions and the final PDI lay below 1.5. The influences of different temperatures on the suspension ATRP of methyl methacrylate (MMA) were investigated by Zhu et al.,15 using 1-chloro-1-phenylethane as initiator and copper chloride/bipyridine as catalyst. The curves of ln[M]0/[M] exhibited a linear relationship with reaction time at 90 and 95 °C, while showed a positive deviation after 110 min at lower temperatures (75, 80 and 85 °C) due to the occurrence of autoacceleration. The molecular weight rose with monomer conversions linearly except at 75 °C when marked autoacceleration took place and disrupted the control ability of the polymerization. The PDI ranged between 1.2 and 1.7 at all temperatures. Cao et al.19 conducted suspension ATRP of MMA without the addition of any traditional dispersants and ligands. When acetone was used as solvent, the molecular weights grew linearly with monomer conversions and the PDI ranged between 1.1 and 1.5. Heuts et al.21 investigated suspension RAFT polymerization of MMA and MMA with St or 2-hydroxyethyl methacrylate, using 2-cyanoprop-2-yl dithiobenzoate as RAFT agent. The molecular weight increased linearly and the PDI declined from 1.4 to 1.1 with increasing monomer conversions, demonstrating good controlled/living features.

Our group has developed a new CLRP method called cycloketyl radical mediated living polymerization (CMP).23,24 It involves the formation of cycloketyl xanthone radicals (CX) by the homolytic cleavage of 9,9′-bixanthene-9,9′-diol (BIXANDL) under thermal or photo stimuli. The CX radicals can initiate and control the solution polymerization of common monomers (MMA, St and butyl acrylate), achieving good controlled/living features. Furthermore, CMP is a simple, environment-friendly and metal/odor/color free method, therefore, affords a promising approach to industrial production.

In the present work, CMP was applied in the suspension polymerization of St for the first time to evaluate the initiation and control ability of BIXANDL. It was found that BIXANDL could initiate the suspension polymerization of St effectively and achieve moderate control over molecular weights. When BIXANDL was utilized in combination with a traditional thermal initiator benzoyl peroxide (BPO), the polymerization proceeded with more rapid kinetics and larger increments of molecular weights with increasing St conversions.

Experimental

Materials

St (99.0%) was purchased from Beijing Sinopharm Reagent Co., Ltd and purified by passing through a column filled with inhibitor removal resin. Xanthone (99.0%), zinc powder (97.5%) and BPO (97.0%) were supplied by Alfa Aesar Chemical Co., Ltd. Polyvinyl alcohol (degree of hydrolysis: 88%) was provided by Thermo Fisher Scientific Co., Ltd. Ammonium chloride (97.5%), tetrahydrofuran (THF, AR), ethanol (AR) and petroleum ether (boiling range: 60–90 °C) were obtained commercially from Beijing Modern Oriental Technology Development Co. Ltd and used as received. BIXANDL was synthesized according to the literature method23 and purified repeatedly prior to use.

Suspension polymerization of St by CMP

When only BIXANDL was used, 100 mL of deionized water and 1 g of aqueous PVA solution (1 wt%) were added to a four-necked flask equipped with a thermometer, a gas supply and a mechanical stirrer set at 300 rpm. Subsequently, desired amounts of BIXANDL were dissolved in 20 g of St and added into the flask which was stirred in an oil bath and purged with nitrogen for 30 min. During the polymerization, samples were taken out from the flask periodically and weighed accurately (m1). The samples were dissolved in THF, followed by being precipitated in ethanol and dried until constant weights (m2) were reached. The conversion of St was determined by gravimetry using the equation below:
image file: c6ra14396b-t1.tif
where A is the weight fraction of St in the initial reaction mixture.

When both BIXANDL and BPO were used, the concentration of BIXANDL was kept at 1 wt% with respect to St and the amounts of BPO were varied. The experimental procedure was similar to the case when only BIXANDL was used.

Characterization

The number average molecular weights (Mn) and polydispersity index (PDI) were determined by gel permeation chromatography (GPC, Waters, USA). The GPC was equipped with a Waters 1515 HPLC pump, packing PSS columns (Styragel HT-3-4-5) and a Waters 2414 differential refractive index detector. THF was used as eluent at a flow rate of 1 mL min−1 at 30 °C. The column system was calibrated with linear PS standards provided by Polyscience.

Results and discussion

According to the formulation and technical process of the suspension polymerization of St in industry,25 polyvinyl alcohol (PVA) was chosen as dispersing agent and the polymerization temperature ranged between 80 and 95 °C in the present study. The suspension polymerization behavior of St initiated by BIXANDL both in the absence and presence of BPO was investigated separately.

BIXANDL as initiator

First, we carried out the suspension polymerization of St with BIXANDL as initiator. The polymerization was conducted with different concentrations of BIXANDL (0.5, 1 and 2 wt% with respect to St) at 80 °C and with 1 wt% of BIXANDL at different temperatures (80, 90 and 95 °C), in order to gain an insight into the effects of the loading of BIXANDL and temperature. The conversion-time plots are given in Fig. 1. It can be seen from Fig. 1A that St conversion grew with polymerization time, reaching 18.4% at 5.0 h and then achieving 49.5% at 12 h with 0.5 wt% of BIXANDL. The polymerization accelerated with increasing loading of BIXANDL. St conversions at 11 h with 1 and 2 wt% of BIXANDL were 60.0% and 81.4%, respectively. Fig. 1B shows that increasing temperature enhanced the polymerization rate remarkably at a fixed loading of BIXANDL. It took 11 h for St conversion to reach 48% at 80 °C, while the time for 80% St conversion at 90 and 95 °C was about 10 and 6 h, respectively. Such behavior has also been reported by Zhu et al.15 and Cao et al.19 Hence, BIXANDL can act as efficient initiator for the suspension polymerization of St and the polymerization was speeded up markedly when elevating the concentration of BIXANDL and temperature.
image file: c6ra14396b-f1.tif
Fig. 1 Plots of conversion vs. time for the suspension polymerization of St with different concentrations of BIXANDL at 85 °C (A) and with 1 wt% of BIXANDL at different temperatures (B).

The evolutions of Mn and PDI with conversion when using BIXANDL as initiator are illustrated in Fig. 2. At 85 °C, Mn increased slightly when the concentrations of BIXANDL were 0.5 and 1 wt%, while obvious linear growth of Mn from 60k to 100k was observed with 2 wt% of BIXANDL. The high initial Mn at the early stage was in accordance with the results obtained in our previous research.23 However, a large deviation of Mn from theoretical Mn was observed in the present work, indicating a rather low initiating efficiency of BIXANDL in the suspension polymerization of St. The PDI of PS at various concentrations of BIXANDL ranged between 1.7 and 2.0 when St conversions were below 60.0%. This shows that the controlled ability of CMP in suspension polymerization is much less impressive than that in solution (THF) polymerization when a final PDI as low as 1.29 was recorded.23 In the literature, Zhu et al.15 reported that the PDI of suspension ATRP of MMA was around 1.5 in most cases, which was noticeably higher than that in typical ATRP. The PDI rose further at the late stage of polymerization especially when 2 wt% of BIXANDL was incorporated, which is indicative of a further broadening of molecular weights. It can be seen from Fig. 2B that temperature also has a substantial effect on the growth intervals of Mn during the polymerization. The growth intervals of Mn for the polymerization at 80, 90 and 95 °C were from 80k to 120k, 70k to 100k and 54k to 74k, respectively. Therefore, elevated temperature can bring about a drop in both the initial Mn and its increment during the polymerization. This correlates well with the results of Zhu et al.,15 except that a dramatic negative deviation of Mn-conversion curve from linearity was observed in their case when monomer conversion was over 25%. Thus, moderate control over Mn can be achieved when using BIXANDL as initiator and a high loading of BIXANDL and low temperature are beneficial for a larger increase in Mn. Nevertheless, the results of PDI are not satisfying.


image file: c6ra14396b-f2.tif
Fig. 2 Plots of Mn & PDI vs. conversion for the suspension polymerization of St with different concentrations of BIXANDL at 85 °C (A) and with 1 wt% of BIXANDL at different temperatures (B).

BIXANDL and BPO as initiators

When only BIXANDL was utilized, it played dual roles of initiator and mediating agent. As shown in the ESI part of our previous paper,23 the low concentration of CX radicals at the beginning of polymerization and their poor initiating activity cause a gradual increase in the concentration of chain radicals. This exerts multiple negative effects including reduced polymerization rate, large initial Mn and worsened control ability of CMP. Attempting to improve its control ability, we introduced the commonly-used thermal initiator BPO into the present system. The suspension polymerization behavior of St with different molar ratios of BPO to BIXANDL (1[thin space (1/6-em)]:[thin space (1/6-em)]2, 1[thin space (1/6-em)]:[thin space (1/6-em)]1 and 2[thin space (1/6-em)]:[thin space (1/6-em)]1) at 80 °C and with the molar ratio of BPO to BIXANDL being 1[thin space (1/6-em)]:[thin space (1/6-em)]1 at different temperatures was studied, while maintaining the concentration of BIXANDL at 1 wt%. The conversion-time plots are presented in Fig. 3. At a fixed amount of BIXANDL, the polymerization was expedited with increasing ratio of BPO to BIXANDL. The St conversion at 11 h was only 48% without the addition of BPO (Fig. 1B), while the corresponding St conversion with the molar ratio of BPO to BIXANDL being 1[thin space (1/6-em)]:[thin space (1/6-em)]2 was 70%. When one or two molar equivalent of BPO with respect to one molar equivalent of BIXANDL was incorporated, the polymerization almost completed within 10 h. Similarly, Georges et al.12 have compared the suspension polymerization of St at different ratios of BPO to TEMPO and found that the polymerization slowed down when the ratio was lower. Fig. 3B shows a similar trend of conversion versus time to the case without BPO, that is, the polymerization exhibited more rapid kinetics with increasing temperature. The St conversion at 8 h reached approximately 62% at 80 °C, while the conversions at the same time exceeded 75% and 90% at 85 and 90 °C, respectively.
image file: c6ra14396b-f3.tif
Fig. 3 Plots of conversion vs. time for the suspension polymerization of St with different molar ratios of BPO to BIXANDL at 80 °C (A) and with the molar ratio of BPO to BIXANDL being 1[thin space (1/6-em)]:[thin space (1/6-em)]1 at different temperatures (B).

GPC analysis (Fig. 4A) demonstrated a drastic influence of BPO on the initial Mn and its increments during the course of polymerization. When the molar ratio of BPO to BIXANDL was 1[thin space (1/6-em)]:[thin space (1/6-em)]2, Mn grew from 65k to 130k. In comparison with the corresponding case in the absence of BPO, it is conceivable that the control over Mn in suspension CMP was improved. When the molar ratios of BPO to BIXANDL increased to 1[thin space (1/6-em)]:[thin space (1/6-em)]1 and then to 2[thin space (1/6-em)]:[thin space (1/6-em)]1, the increments of Mn reduced to 32k (56k to 88k) and then to 16k (40k to 56k). Thus, the initial Mn dropped and the growth intervals of Mn narrowed when the amount of BPO was raised. The PDI of PS prepared in the presence of BPO resembled the case without BPO. A shoulder peak emerged in the GPC traces at the late stage when the molar ratio of BPO to BIXANDL equaled 1[thin space (1/6-em)]:[thin space (1/6-em)]2, with an associated rising of PDI. The evolutions of Mn and PDI versus conversion for the polymerization with the molar ratio of BPO to BIXANDL equaling 1[thin space (1/6-em)]:[thin space (1/6-em)]1 at different temperatures are compared in Fig. 4B. Mn increased from 56k to 88k at 80 °C, whereas the growth interval was from 33k to 64k at 90 °C. The PDI rose sharply at the late stage at 90 °C, due to the appearance of a shoulder peak in GPC trace. Consequently, the suspension polymerization of St conducted at 80 °C with the molar ratio of BPO to BIXANDL being 1[thin space (1/6-em)]:[thin space (1/6-em)]1 demonstrated a balance of appropriate polymerization rate and moderate control over Mn.


image file: c6ra14396b-f4.tif
Fig. 4 Plots of Mn & PDI vs. conversion for the suspension polymerization of St with different molar ratios of BPO to BIXANDL at 80 °C (A) and with the molar ratio of BPO to BIXANDL being 1[thin space (1/6-em)]:[thin space (1/6-em)]1 at different temperatures (B).

Fig. 5 presents the GPC traces of PS prepared by suspension polymerization performed at 80 °C with the molar ratio of BPO to BIXANDL equaling 1[thin space (1/6-em)]:[thin space (1/6-em)]1. The GPC peaks were without a shoulder peak and shifted to lower elution time, pointing to continuous growth of Mn during the course of polymerization. It seems that small tailing exists in GPC traces, which led to the broadening of molecular weights, reflected in the increase in PDI during the polymerization.


image file: c6ra14396b-f5.tif
Fig. 5 GPC traces of PS prepared by suspension polymerization carried out with the molar ratio of BPO to BIXANDL being 1[thin space (1/6-em)]:[thin space (1/6-em)]1 at 80 °C.

Mechanism of the suspension CMP of St in the absence and presence of BPO

The present results can be interpreted by the plausible mechanism outlined in Scheme 1. In the suspension polymerization of St using BIXANDL, BIXANDL decomposes under thermo stimuli, producing two CX radicals. Though CX radicals have low activity to add to the double bonds of monomer due to the conjugation effect of the two benzene rings, chain radicals possessing sufficient initiating activity are formed after the addition of the first monomer. Then following a persistent radical effect, an activation–deactivation equilibrium between chain radicals and CX radicals is established, which is crucial to the controlled/living feature of CMP. Clearly, a higher concentration of BIXANDL led to a larger amount of chain radicals produced. And by elevating temperature, the decomposition rate of BIXANDL, the addition rates of monomer to CX and chain radicals as well as the activation rate of dormant species were all enhanced. Thus, the polymerization demonstrated more rapid kinetics when the concentration of BIXANDL and temperature were raised. The mechanism described above also suggests that maintaining an adequate concentration of CX radicals is crucial to the control ability of CMP. This accounts for the small increments of Mn with 0.5 and 1 wt% of BIXANDL and the high Mn at the initial stage in all cases. At higher temperature, the initiating steps accelerated, resulting in a decrease in the initial Mn. However, side reactions such as termination and chain transfer events also occurred more frequently, leading to worsened control over Mn.
image file: c6ra14396b-s1.tif
Scheme 1 Schematic mechanism of the suspension polymerization of styrene mediated by cycloketyl radicals.

The situation in the suspension polymerization of St using BIXANDL along with BPO is somewhat different. The chain initiation rate of BPO (ki2) is higher than that of BIXANDL (ki1), resulting in a dominant role of BPO in chain initiation, while BIXANDL mainly played the role of deactivating chain radicals. When the molar ratio of BPO to BIXANDL was 1[thin space (1/6-em)]:[thin space (1/6-em)]2, the concentration of CX radicals was higher than that of chain radicals, suggesting that CX radicals could deactivate the chain radicals and transform the active species into dormant state effectively, according to the equilibrium in Scheme 1. Therefore, the increment of Mn was the largest under such circumstances. When one or two molar equivalent of BPO with respect to one molar equivalent of BIXANDL was added, the concentration of CX radicals was below that of chain radicals. Hence, CX radicals could not deactivate chain radicals efficiently and the occurrence of termination or chain transfer reactions brought about faded control over Mn.

In addition, the large deviations of Mn from theoretical values and high PDI are associated with the combined effects of the complexity in the mechanism of CMP as well as the complicated suspension polymerization process. Typical solution CMP exhibits a transformation period after which all BIXANDL has been transformed into the end groups of polymer chains and CX-terminated polymer chains are formed,23 which causes a small deviation of Mn from theoretical value and a little higher PDI. In suspension polymerization which possesses pseudo-bulk polymerization feature, however, gel effect may overlap with the transformation period. This impeded the effective deactivation of chain radicals by CX radicals from the beginning, leading to large deviation of Mn and large PDI. In addition, the CX radicals taking effect in the activation–deactivation equilibrium might still initiate new chains, causing a loss in the control ability of CMP gradually. It has also been found that CX radicals can undergo unwanted hydrogen transfer reactions, which leads to the formation of xanthone. Both effects led to a decrease in the concentration of CX radicals, which caused a gradual lost in the controllability of CMP. The sharp increase in PDI at the late stage of polymerization in some cases might be due to the high viscosity of monomer droplets when St conversions were high, which promoted the occurrence of the autoacceleration effect.15,21 The restriction of chain radicals in mobility due to high viscosity made it hardly possible for them to be deactivated by the CX radicals, which brought about a further broadening in molecular weights.

Conclusions

In summary, BIXANDL can initiate the suspension polymerization of St effectively and exert moderate control over molecular weights which increase linearly as St conversion increases. When utilizing BIXANDL along with BPO, the polymerization demonstrates more rapid kinetics and larger increases in Mn with increasing St conversions. Further research will be focused on improving the control ability of CMP in dispersed systems in order to be applied in industrial production.

Acknowledgements

This research is financially supported by the Natural Science Foundation of China (no. 21404004) and Fundamental Research Funds for the Central Universities of China (no. ZY1409).

Notes and references

  1. B. W. Brooks, Chem. Eng. Technol., 2010, 33, 1737 CrossRef CAS.
  2. J. Qiu, B. Charleux and K. Matyjaszewski, Prog. Polym. Sci., 2001, 26, 2083 CrossRef CAS.
  3. M. Antonietti and K. Tauer, Macromol. Chem. Phys., 2003, 204, 207 CrossRef CAS.
  4. P. B. Zetterlund, Y. Kagawa and M. Okubo, Chem. Rev., 2008, 108, 3747 CrossRef CAS PubMed.
  5. H. Fischer, Chem. Rev., 2001, 101, 3581 CrossRef CAS PubMed.
  6. Y. Yagci and M. A. Tasdelen, Prog. Polym. Sci., 2006, 31, 1133 CrossRef CAS.
  7. W. A. Braunecker and K. Matyjaszewski, Prog. Polym. Sci., 2007, 32, 93 CrossRef CAS.
  8. G. Moad, E. Rizzardo and S. H. Thang, Acc. Chem. Res., 2008, 41, 1133 CrossRef CAS PubMed.
  9. C. J. Hawker, A. W. Bosman and E. Harth, Chem. Rev., 2001, 101, 3661 CrossRef CAS PubMed.
  10. K. Matyjaszewski and N. V. Tsarevsky, J. Am. Chem. Soc., 2014, 136, 6513 CrossRef CAS PubMed.
  11. M. R. Hill, R. N. Carmean and B. S. Sumerlin, Macromolecules, 2015, 48, 5459 CrossRef CAS.
  12. M. K. Georges, R. P. N. Veregin, P. M. Kazmaier and G. K. Hamer, Macromolecules, 1993, 26, 2987 CrossRef CAS.
  13. C. Taube and G. Schmidt-Naake, Chem. Eng. Technol., 2001, 24, 1013 CrossRef CAS.
  14. M. A. D. Gonçalves, V. D. Pinto, R. C. S. Dias, M. R. P. F. N. Costa, L. G. Aguiar and R. Giudici, Macromol. React. Eng., 2013, 7, 155 CrossRef.
  15. C. Y. Zhu, F. Sun, M. Zhang and J. Jin, Polymer, 2004, 45, 1141 CrossRef CAS.
  16. Y. Fuji, T. Ando, M. Kamigaito and M. Sawamoto, Macromolecules, 2002, 35, 2949 CrossRef CAS.
  17. T. Nishikawa, M. Kamigaito and M. Sawamoto, Macromolecules, 1999, 32, 2204 CrossRef CAS.
  18. C. Granel, P. Dubois, R. Jérôme and P. Teyssié, Macromolecules, 1996, 29, 8576 CrossRef CAS.
  19. J. Cao, L. F. Zhang, X. W. Jiang, C. Tian, X. N. Zhao, Q. Ke, X. Q. Pan, Z. P. Cheng and X. L. Zhu, Macromol. Rapid Commun., 2013, 34, 1747 CrossRef CAS PubMed.
  20. A. Limer, A. Heming, I. Shirley and D. Haddleton, Eur. Polym. J., 2005, 41, 805 CrossRef CAS.
  21. J. D. Biasutti, T. P. Davis, F. P. Lucien and J. P. A. Heuts, J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 2001 CrossRef CAS.
  22. N. Rocha, J. F. J. Coelho, B. Barros, P. M. L. Cardoso, P. M. Gonçalves, M. H. Gil and J. T. Guthrie, J. Appl. Polym. Sci., 2013, 127, 3407 CrossRef CAS.
  23. X. F. Zheng, M. Yue, P. Yang, Q. Li and W. T. Yang, Polym. Chem., 2012, 3, 1982 RSC.
  24. X. G. Huang, L. Wang and W. T. Yang, Polym. Chem., 2015, 6, 6664 RSC.
  25. P. L. Shanta, US Pat., 2694700A, 1954.
This journal is © The Royal Society of Chemistry 2016
Click here to see how this site uses Cookies. View our privacy policy here.