TEMPO-mediated dispersion polymerization of styrene in the by terrible2


									TEMPO-Mediated Dispersion Polymerization of Styrene
in the Presence of Camphorsulfonic Acid
Department of Chemical Engineering, Inha University, Namgu, Incheon 402-751, Republic of Korea

Received 8 March 2005; accepted 16 June 2005
DOI: 10.1002/pola.20988
Published online in Wiley InterScience (www.interscience.wiley.com).

                     ABSTRACT:   The TEMPO-mediated polymerization of styrene in the presence of cam-
                     phorsulfonic acid (CSA) is carried out using controlled radical dispersion polymeriza-
                     tion. In the absence of TEMPO and CSA, 92% of conversion was achieved within 3 h
                     of polymerization. When TEMPO is solely used, broadening of particle size with nar-
                     row PDI was observed because of the prolonged polymerization time. However, when
                     1:1 molar ratio of CSA/TEMPO was added, the fairly monodisperse PS microspheres
                     having 5.83 lm average size and 3.42% CV (coefficient of variation) were successfully
                     achieved because of the narrow molecular weight of intermediate oligomers and
                     shortening of the polymerization time. This result obviously indicates that the addi-
                     tion of CSA in TEMPO-mediated dispersion polymerization not only shortens the poly-
                     merization time but also greatly improves the uniformity of the microspheres. V 2005

                     Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 44: 62–68, 2006
                     Keywords: camphorsulfonic acid; nitroxide-mediated dispersion polymerization;
                     polystyrene microsphere

INTRODUCTION                                                                   tion and deactivation processes. The most widely
                                                                               studied successful compounds are nitroxides and
Up to date, the design of precise molecular                                    its associated alkylated derivates. Since stable
weights and molecular weight distribution of                                   nitroxide free radical favors to react with the
polymers has been of great importance. Among                                   growing carbon-centered radicals rather than
the various polymerization methods to synthe-                                  with the monomer, the reversible capping reac-
size predetermined molecular structure, con-                                   tion drastically decreases the concentration of
trolled radical polymerizations (CRPs) have                                    the radical chain ends. Because of the conse-
received considerable attention because of its                                 quence of the equilibrium between activation
versatile and simple process; thus, a wide vari-                               and deactivation, the termination reaction is
ety of polymers having various structures has                                  suppressed and the pseudoliving characteristics
been successfully synthesized by several differ-                               are achieved.3
ent CRP mechanisms.                                                               Currently, research interests in CRP are
   The nitroxide-mediated polymerization (NMP)1,2                              being devoted mainly to the synthesis of func-
is one of the generally used strategies, based on                              tional polymers4–7 and in the application of
the common concept of alternating the activa-                                  CRPs to various heterogeneous polymerization
                                                                               techniques, such as (mini)emulsion, suspension,
                                                                               and dispersion polymerization.8–10 Among these
   Correspondence to: Soonja Choe (E-mail: sjchoe@inha.                        polymerizations, dispersion polymerization pro-
Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 44, 62–68 (2006)
                                                                               vides a simple production route of polymeric
V 2005 Wiley Periodicals, Inc.
C                                                                              microspheres in the range of 1–10 lm, which is
                                        TEMPO-MEDIATED DISPERSION POLYMERIZATION OF STYRENE        63

rarely achieved by other single step processes.11    used as a stabilizer and methylene chloride
Such microspheres are being used recently as         (Samchun, Korea) was used to purify the resul-
functional materials in electric, microelectronic,   tant polymer.
information technology, and biorelated applica-
tions.12–14 The particle formation in the disper-
sion polymerization occurs in a complex manner,      Polymerization
since the reaction mechanism is a combination        In the TEMPO-mediated dispersion polymeriza-
of homogeneous and heterogeneous ones. Ini-          tion, 100 g of TPG was first charged to a 250 mL.
tially, all reaction ingredients are dissolved in    four-necked, round glass flask at room tempera-
the reaction medium, and then, the polymeric         ture under nitrogen. After complete dissolv-
particles are nucleated and precipitated upon        ing of styrene (10 g; 96.01 mmol), BPO (0.1 g;
reaching the critical limit of solubility of the     0.421 mmol), PVP (1 g), and CSA (0.206 and
oligomeric species in the medium for further         0.413 mmol) in TPG, nitrogen was continuously
polymerization. The precipitated particles are       purged for 1 h before addition of BPO. The same
spherically stabilized by means of polymeric sta-    molar ratios of [TEMPO]/[BPO] and [CSA]/
bilizers.15,16                                       [TEMPO] were used. The reaction mixture was
   In few literature references, the uniformity of   heated at 125 8C in an oil bath, and the poly-
the polymer microspheres employed by the CRP         merization was conducted under nitrogen atmos-
techniques using dispersion polymerization was       phere at a constant agitation speed of 200 rpm
inevitably poor.8,17,18 The broadening of particle   for a desired time. BPO was added at 70 8C and
size is reported to be caused by the prolonged       aged for 1 h under nitrogen, and then the tem-
polymerization time because of the reversible        perature was raised to 120 8C. It is noted that
termination mechanism of CRPs. In general, a         the polymerization procedure was the same
fast nucleation followed by uniform growth of        as that reported by Gabaston et al.8 During poly-
the primary particles is necessary to obtain         merization, 5 mL of sample was periodically
monodisperse final particles in the dispersion        taken from the reaction vessel in order to char-
polymerization.19 It has been reported that some     acterize the polymerization kinetics and pro-
strong organic acids, such as camphorsulfonic        ducts, including the conversion, molecular we-
acid (CSA) or 2-fluoro-1-methylpyridinium p-tol-      ight, PDI, and particle size. After completion
uenesulfonate, substantially accelerate the rate     of polymerization, the resultant product was
of nitroxide-mediated bulk, solution, and minie-     repeatedly washed with methanol so as to re-
mulsion polymerizations.20–23                        move the remaining PVP, and dried in vacuo
   In this article, we report the multiple effects   for 24 h.
of CSA not only as a rate-accelerating agent but
also as a uniformity-controlling agent in nitro-
xide-mediated dispersion polymerization of styr-     Characterization
ene in the absence and presence of 2,2,6,6-tetra-    The conversion in the dispersion polymerizations
methylpiperidin-1-oxyl (TEMPO).                      was measured gravimetrically. The molecular
                                                     weight and PDI were measured using Waters
                                                     GPC equipped with 510 differential refracto-
                                                     meter and Viscotek T50 differential viscometer.
EXPERIMENTAL                                                           ˚
                                                     105, 103, and 102 A l-Styragel packed high-reso-
                                                     lution columns were used. Universal calibration
                                                     curve was made using the PS standard samples
TEMPO, benzoyl peroxide (BPO), and CSA (rac-         (Polymer laboratories, UK), with molecular we-
emic mixture) were purchased from Aldrich and        ights ranging 580–7,500,000 g/mol. For gel-per-
used as received. Styrene (Junsei Chemicals,         meation chromatography measurement, an ali-
Japan) was purified using an inhibitor-removal        quot of the sample taken from the reaction vessel
column (Aldrich) and stored at À5 8C before use.     was repeatedly washed with excess of methanol
Analytical grade of tripropylene glycol (TPG;        and centrifuged at 13,000 rpm. Then, the poly-
Aldrich) was used after distillation, with calcium   mer was dissolved in methylene chloride and re-
hydride as the medium in the dispersion polym-       precipitated from methanol, and dried in vacuo
erization. Poly (N-vinylpyrrolidone) (PVP; weight-   overnight at 50 8C. The finally obtained PS, dis-
average molecular weight ¼ 40,000; Sigma) was        solved in THF as the mobile phase, was injected
64    OH ET AL.

at a flow rate of 1.0 mL/min. Scanning electron
microscopy (SEM; Hitachi S-4300) was used to
study the morphology of the PS particles. More
than 100 particle diameters were measured from
each SEM microphotograph, using Scion Image1
Analyzer, and were used to calculate the number-
average particle diameter, dn, and coefficient of
variation, CV, defined as follows:

                               ni di
                         dn ¼ P                              ð1Þ
                   P                   P
               ð       ðdi À dn Þ2 =       ni Þ1=2
        CV ¼                                         Â 100   ð2Þ

where ni denotes the number of particles with a
diameter of di.

                                                                   Figure 1. The influence of the addition of CSA on
                                                                   the conversion of TEMPO-mediated dispersion poly-
RESULTS AND DISCUSSION                                             merization of styrene in TPG at 125 8C.

Effects of CSA on TEMPO-mediated
Dispersion Polymerization
                                                                   a nonlinear kinetics is observed, while pseudo-
Figure 1 represents the conversion of the                          first-order kinetics is obtained in the presence of
TEMPO-mediated dispersion polymerization                           1.0 molar ratio of [TEMPO]/[BPO]. However,
under various conditions. When neither TEMPO                       when 0.5 and 1 molar ratios of [CSA]/[TEMPO]
nor CSA is employed, that is typically in the dis-                 are incorporated, the living nature of the disper-
persion polymerization, the high conversion of                     sion polymerization is achieved with enhanced
92% is achieved for 3 h because of the high poly-                  conversion and polymerization rate.
merization temperature at 125 8C. However,                            Mayo has reported that the presence of CSA
when 1.0 molar ratio of [TEMPO]/[BPO] is sorely                    in the polymerization suppresses the thermal
employed, low conversion of 63% was obtained for                   radical auto-generation of styrene.24 In the case
48 h. This means that TEMPO retards the poly-                      where modest concentration of CSA ( 0.04 M) is
merization rate unless the molecular weight dis-                   used together with TEMPO in the polymeriza-
tribution is well controlled. To overcome the retar-               tion of styrene, an important aspect is the rapid
dation of polymerization rate in the presence of                   consumption of TEMPO by reacting with CAS
TEMPO, CSA is added in the polymerization sys-                     throughout the TEMPO disproportionation
tem. When 0.5 and 1.0 molar ratios of [CSA]/                       mechanism.22 The reduction of TEMPO leads
[TEMPO] are used, 71.1 and 78.5% conversions                       the equilibrium to favor the formation of active
are obtained for 6 h. As seen in Figure 1, the use                 radicals, thereby increasing the polymerization
of CSA in TEMPO-mediated dispersion polymer-                       rate.22,23 Furthermore, the rate constant for the
ization efficiently accelerates the polymerization                  deactivation between a growing radical and
rate.                                                              TEMPO decreases in the presence of CSA.24,25
   Figure 2 shows the kinetics of the TEMPO-                       Finally, the reduction of TEMPO concentration
mediated dispersion polymerization of styrene,                     results in increasing polydispersity and a loss of
using CSA as the rate-acceleration agent. The                      living character, while the rate of polymerization
linearity in Ln{[M0]/[M]} versus reaction time is                  is accelerated.
a characteristic of living nature of polymeriza-                      Figure 3 shows the molecular evolution dur-
tion, since the number of growing radicals is                      ing the TEMPO-mediated dispersion polymeriza-
maintained constant throughout the dispersion                      tion of styrene without and with the presence of
polymerization. In the absence of TEMPO and CSA,                   CSA. To compare the experimental and theoreti-
                                        TEMPO-MEDIATED DISPERSION POLYMERIZATION OF STYRENE             65

             Figure 2. The influence of the addition of CSA on the kinetics of TEMPO-mediated
             dispersion polymerization of styrene in TPG at 125 8C.

cal molecular weights, Mn was calculated using         the presence of CSA is shown in Figure 4. In
the following equation:                                the absence of TEMPO, the PDI continuously
                                                       increases with conversion up to 2.30. When 1.0
                  ½StyreneŠ0                           molar ratio of [TEMPO]/[BPO] is used, the PDI
Calculated Mn ¼              Â conversion
                  ½TEMPOŠ0                             decreases to 1.16 with 63% conversion. Although
                             Â Mw ðstyreneÞ ð3Þ        with the incorporation of CSA the PDI slightly
                                                       increases, the final PDIs of 1.30 and 1.38 are
where Mw (styrene) is the molar mass of styrene.       obtained for 0.5 and 1.0 molar ratios of [CSA]/
   The experimentally measured molecular we-           [TEMPO], respectively. However, the PDI value
ight (Mn) is marginally higher than the theoreti-      is substantially lower than that in the TEMPO-
cally calculated Mn (solid line) in which 1.0          free system. The summary of the TEMPO-medi-
molar ratio of [TEMPO]/[BPO] is used. The              ated dispersion polymerization is listed in Table 1.
higher experimental molecular weight results           These polymerization characteristics indicate that
from the grafting of PVP with PS, since disper-        CSA serves as an efficient rate-enhancing agent
sion polymerization is initiated on the stabilizer     in the TEMPO-mediated dispersion polymeriza-
molecules by abstraction of labile hydrogen.26         tion by minimizing the deterioration of the living
The linear increase in Mn is successfully              nature of the polymerization.
achieved in the TEMPO-mediated dispersion
polymerization with CSA, meaning that the liv-
                                                       CSA as a Particle Size Uniformity-Controlling
ing characteristic is maintained. This is the
same phenomenon as observed in the bulk poly-
merization.20,22                                       The SEM microphotographs of the PS micro-
   The PDI of the PS microspheres prepared by          spheres prepared from various concentrations of
TEMPO-mediated dispersion polymerization in            TEMPO and CSA are displayed in Figure 5 and
66    OH ET AL.

                                                       time, and this causes poor size distribution. Al-
                                                       though the average-molecular weights of the par-
                                                       ticles are relatively low, the PDI is 1.16, indi-
                                                       cating that TEMPO certainly controls the living
                                                          On the other hand, when CSA is involved
                                                       with addition of 1.0 molar ratio of [TEMPO]/
                                                       [BPO] and 0.5 molar ratio of [CSA]/[BPO], the
                                                       dn increases to 3.41 lm, with the poor particle
                                                       size distribution, having the CV of 40.6% as seen
                                                       in Figure 5c. In addition, 71.1% of the conver-
                                                       sion for 6 h and 1.30 of PDI are obtained, which
                                                       means that CSA accelerates the polymerization.
                                                       In addition, compared to Figure 5b, the secon-
                                                       dary small particles are reduced by addition of
                                                       CSA and this seems that CSA supresses the
                                                       birth of the secondary particles. This phenom-
                                                       enon is obvious when more CSA is incorporated,
                                                       as seen in Figure 5d. When 1.0 molar ratios
Figure 3. The influence of the addition of CSA on       of [TEMPO]/[BPO] and [CSA]/[TEMPO] are
the molecular evolution in the TEMPO-mediated dis-     employed, fairly monodisperse PS microspheres
persion polymerization of styrene in TPG at 125 8C.
                                                       with 5.83 lm average size and 3.42% CV are suc-
                                                       cessfully achieved (Fig. 5d). These monodisperse
the results are added in Table 1. Under conven-        microspheres could be produced because of the
tional dispersion polymerization, that is in the       narrow molecular weight of intermediate oligo-
absence of TEMPO and CSA, but in the pres-             mers and shortening of nucleation time, caused
ence of BPO only, the PS microspheres having           by reduction of the concentration of TEMPO by
dn of 1.12 lm and CV of 12.4% are obtained, as         reacting with CSA, which effectively suppres-
seen in Figure 5a. In dispersion polymerization,       ses the generation of secondary particles. This
primary particles are generated from the precip-
itation of relatively large oligomeric species, and
they subsequently grow to microspheres by
absorbing oligomers and monomers from the
medium. In order for the final particles to be
uniform in size, a short nucleation period and
uniform growth of those primary particles are
necessary.16,27 In addition, the coalescence of
growing particles should be prevented. Among
these factors, the particle formation stage, i.e.
nucleation, plays a key role in determining the
final particle size and its distribution. Truly, in
the case of TEMPO-mediated dispersion poly-
merization of styrene, the broadening of particle
size distribution has been observed because of
the prolonged nucleation time.8,18 The small
particles in Figures 5b and 5c are caused by the
birth of secondary particles during polymeriza-
tion. When 1.0 molar ratio of [TEMPO]/[BPO] in
the absence of CSA is used, the dn increases to
2.56 lm and 63% of conversion is obtained dur-
ing 48 h of the reaction, but the particle size dis-   Figure 4. The influence of the addition of CSA on
tribution becomes poor having the CV of 51.2%,         the PDI of PS microspheres prepared by the TEMPO-
as seen in Figure 5b. The formation of secondary       mediated dispersion polymerization of styrene in TPG
particles is obvious because of long nucleation        at 125 8C.
                                         TEMPO-MEDIATED DISPERSION POLYMERIZATION OF STYRENE             67

Table 1. Properties of Resultant PS Microspheres Prepared by Dispersion Polymerization with Various
Concentrations of TEMPO and CSA

[TEMPO]/[BPO]      [CSA]/[TEMPO]      Time (h)    Conversion (%)    Mn (g/mol)    PDI     dn (lm)     CV (%)

      0                  0                3            92.0           51,250      2.30      1.12      12.4
      1.0                0               48            63.0           16,350      1.16      2.56      51.2
      1.0                0.5              6            71.1           30,126      1.30      3.41      40.6
      1.0                1.0              6            78.5           38,980      1.38      5.83       3.42

result obviously indicates that the addition of         ticle size becomes quite broad because of the
CSA in TEMPO-mediated dispersion polymer-               prolonged polymerization time for 48 h, but the
ization not only shortens the polymerization            PDI narrowed to 1.16, indicating that TEMPO
time needed to obtain substantially high conver-        certainly controls the living nature in the dis-
sion and narrow molecular weight distribution           persion polymerization. When 0.5 and 1.0 molar
but also greatly improves the uniformity of the         ratios of CSA to TEMPO are used, the polymer-
microspheres. However, the living nature of the         ization time is substantially shortened maintain-
polymerization is marginally affected by the            ing the living characteristics. When 1:1 molar
presence of CSA in the dispersion polymeriza-           ratio of CSA to TEMPO is employed, the fairly
tion system.                                            monodisperse PS microspheres having 5.83 lm
                                                        average size and 3.42% CV are successfully
                                                        achieved. Incorporation of CSA reduces the birth
CONCLUSIONS                                             of the secondary particles inducing large aver-
                                                        age particle sizes and small value of CV. These
The TEMPO-mediated controlled radical disper-           monodisperse microspheres could be produced
sion polymerization of styrene in the presence of       because of the narrow molecular weight of inter-
CSA is carried out using BPO as an initiator. In        mediate oligomers and shortening of the nuclea-
the absence of TEMPO and CSA, the conversion            tion time, which effectively suppresses the gen-
of 92% is achieved within 3 h of polymerization.        eration of secondary particles. This result obvi-
When TEMPO is solely used, the resulting par-           ously indicates that the addition of CSA in

             Figure 5. SEM photographs of PS beads prepared under various conditions by dis-
             persion polymerization in TPG at 125 8C. a: [TEMPO]/[BPO] ¼ 0, [CSA] ¼ 0; b:
             [TEMPO]/[BPO] ¼ 1, [CSA] ¼ 0; c: [TEMPO]/[BPO] ¼ 1, [CSA]/[BPO] ¼ 0.5; and d:
             [TEMPO]/[BPO] ¼ 1, [CSA]/[TEMPO] ¼ 1.
68    OH ET AL.

TEMPO-mediated dispersion polymerization not               10. Cao, J.; He, J.; Li, C.; Yang, Y. Polym J 2001, 33,
only shortens the nucleation time to obtain sub-               75.
stantially high conversion and narrow molecular            11. Barrett, K. E. J. Dispersion Polymerization in
weight distribution but also greatly improves                  Organic Media; Wiley: London, 1975.
                                                           12. Fudouz, H.; Xia, Y. Adv Mater 2003, 15, 892.
the uniformity of the microspheres. However,
                                                           13. Ugelstad, J.; Stenstad, P.; Kilaas, L.; Prestvik,
the living nature of the polymerization is mar-
                                                               W. S.; Rian, A.; Nustad, K.; Herje, R.; Berge, A.
ginally affected by the presence of CSA in this                Macromol Symp 1996, 101, 491.
polymerization system.                                     14. Covolan, V. L.; Mei, L. H. I.; Rossi, C. L. Polym
                                                               Advan Technol 1997, 8, 44.
                                                           15. Tseng, C. M.; Lu, Y. Y.; El-Aasser, M. S.; Vanderh-
This work is performed by the financial support of the          off, J. W. J Polym Sci Part A: Polym Chem 1986,
Korea Research Foundation (Grant KRF-2004-042-                 24, 2995.
D00055) during 2004–2005.                                  16. Paine, A. J. Macromolecules 1990, 23, 3109.
                                                           17. Holderle, M.; Baumert, M.; Mulhaupt, R. Macro-
                                                               molecules 1997, 30, 3420.
                                                           18. Shim, S. E.; Oh, S.; Chang, Y. H.; Jin, M.-J.;
REFERENCES AND NOTES                                           Choe, S. Polymer 2004, 45, 4771.
                                                           19. Shen, S.; Sudol, E. D.; El-Aasser, M. S. J Polym
 1. Rodlert, M.; Harth, E.; Rees, I.; Hawker, C. J. J          Sci Part A: Polym Chem 1994, 32, 1087.
    Polym Sci Part A: Polym Chem 2000, 38, 4749.           20. Georges, M. K.; Veregin, R. P. N.; Kazmaier,
 2. Grubbs, R. B.; Dean, J. M.; Broz, M. E.; Bates, F.         P. M.; Hamer, G. K.; Saban, M. Macromolecules
    S. Macromolecules 2000, 33, 9522.                          1994, 27, 7228.
 3. Hawker, C. J. Trends Polym Sci 1996, 4, 183.           21. Odell, P. G.; Veregin, R. P. N.; Michalak, L. M.;
 4. Matyjaszewski, K.; Shigemoto, T.; Frechet, J. M. J.;       Brousmiche, D.; Georges, M. K. Macromolecules
    Leduc, M. Macromolecules 1999, 29, 4167.                   1995, 28, 8453.
 5. Baek, K. Y.; Kamigaito, M.; Sawamoto, M. J             22. Veregin, R. P. N.; Odell, P. G.; Michalak, L. M.;
    Polym Sci Part A: Polym Chem 2002, 40, 1972.               Georges, M. K. Macromolecules 1996, 29, 4161.
 6. Narrainen, A. P.; Pascual, S.; Haddleton, D. M. J      23. Cunningham, M. F.; Tortosa, K.; Lin, M.; Keosh-
    Polym Sci Part A: Polym Chem 2002, 40, 439.                kerian, B.; Georges, M. K. J Polym Sci Part A:
 7. Chen, M.; Ghiggino, K. P.; Mau, A. W. H.; Riz-             Polym Chem 2002, 40, 2828.
    zardo, E.; Thang, S. H.; Wilson, G. J. Chem            24. Mayo, F. R. J Am Chem Soc 1968, 90, 1289.
    Comm 2002, 19, 2276.                                   25. Baldavi, M. V.; Mohat, N.; Scaiano, J. C. Macro-
 8. Gabaston, L. I.; Jackson, R. A.; Arms, S. P.; Mac-         molecules 1996, 29, 5497.
    romolecules 1998, 31, 2883.                            26. Lok, K. P.; Ober, C. K. Can J Chem 1985, 63, 209.
 9. Georges, M. K; Veregin, R. P. N.; Kazmaier, P. M.;     27. Shen, S.; Sudol, E. D.; El-Aasser, M. S. J Polym
    Hamer, G. K. Macromolecules 1993, 26, 2987.                Sci Part A: Polym Chem 1993, 31, 1393.

To top