Effect of the monomer hydrophobicity on the
microemulsion polymerization. A comparison of the
efficiency of water- and oil-soluble photoinitiators.

M.V. ENCINAS, E.A. LISSI and A.M. RUFS

Facultad de Química y Biología, Universidad de Santiago de Chile, Casilla 40,
Correo 33, Santiago, Chile.
(Received: Frebuary 23, 2000 - Accepted: April 31, 2000)

ABSTRACT

Polymerization rates in SDS/monomer/water oil-in-water microemulsions were measured employing a water-soluble (2,2’-azobis(2-amidinopropano), ABAP) and an oil-soluble (2,2’-azobis(isobutyronitrile), AIBN) photoinitiators. Monomers of widely different hydrophobicity, ranging from acrylamide and methacrylic acid to ethylhexyl methacrylate (EHMA) and styrene, were employed in order to evaluate the role of the monomer distribution between the aqueous and microaggregate pseudophases on the photoinitiation efficiency. ABAP was more efficient initiator, irrespective of the monomer distribution. The lower efficiency of AIBN was explained in terms of intracage geminate recombination. The initiation efficiency of AIBN relative to ABAP varied from 0.35 (EHMA) to 0.02 (styrene). The high value for EHMA indicates that the initiation is due to radicals generated in the microaggegates, and that nearly one third of radicals avoids recombination in the original supercage. The low value of styrene is explained in terms of a fast addition to the monomer that reduces the radical exit rate.

KEYWORDS: Microemulsion polymerization, photoinitiation, monomers of different hydrophobicity.

RESUMEN

Las velocidades de polimerización en microemulsiones SDS/monómero/agua se midieron usando un fotoiniciador soluble en agua (2,2’-azobis(2- amidinopropano), ABAP) y un fotoiniciador soluble en la microemulsión (2,2’-azobis(isobutironitrilo), AIBN). Con el objeto de evaluar el rol de la distribución de los monómeros entre las fases acuosa y orgánica en la eficiencia de fotoiniciación se emplearon monómeros de diferente hidrofobicidad. El iniciador más eficiente fue el ABAP, independientemente de la distribución del monómero. La menor eficiencia del AIBN es consecuencia de la recombinación de los radicales en la caja micelar. La eficiencia de iniciación del AIBN relativa a la del ABAP varió desde 0,35 (EHMA) a 0,02 (estireno). El alto valor para el EHMA indica que la iniciación se debe a los radicales generados en el microagregado, solo un tercio de ellos no recombina en la super caja micelar. El bajo valor obtenido para el estireno se explica en términos de la rápida adición al monómero, ésta reduce la velocidad de salida de los radicales.

PALABRAS CLAVES: Polimerización en microemulsión, fotoiniciación, monómeros de diferente hidrofobicidad

INTRODUCTION

Polymerizations in microemulsion (oil-in-water) and/or in swollen micelles of vinyl monomers present kinetic features different from those observed in homogenous solution polymerization¹). Since the micelles are relatively small microcompartments, it can be expected a low average number of free radicals per micelles in the steady state condition, i.e. a fast intramicellar termination rate when two o more radicals are incorporated into the same micelle. This compartmentalization also favors primary intramicellar termination when a pair of radicals is generated in the organic pseudophase²). Hence, lipid-soluble initiators can be expected to be less efficient than water-soluble initiators in the polymerization of vinyl monomers in microemulsion. This effect can be expected to operate independently of the monomer distribution between the aqueous and micellar pseudophases. For hydrophobic monomers, this amount to say that the initiation is more efficient when the radicals are produced outside the main locus of monomer solubilization. In previous works, we have calculated the effect of the initiator distribution employing different carbonyl compounds³) and azocompounds⁴) and a single monomer. The results obtained are compatible with the above discussion, showing that the hydrophobicity of the initiator (and hence the initial radicals) reduces the initiation efficiency. This result can be directly related to a reduced exit rate from the micellar cage, and allows an estimation of the efficiency of this process⁴). In the present work, we try to establish whether the initiator efficiency is also modified by the hydrophobicity of the employed monomer. This work comprises then the study of the initiation efficiency of two initiators of widely different hydrophobicity, ABAP and AIBN in the photopolymerization of several monomers of widely different distribution between the micellar and aqueous pseudophases.

MATERIALS AND METHODS

Sodium dodecyl sulfate (SDS, Aldrich) was purified by recristallization from boiling ethanol. AIBN (Fluka) was recrystallized from ethanol. ABAP (Wako) was used as supplied. Commercially available methyl methacrylate (MMA), butyl methacrylate (BuMA), iso-butyl methacrylate (iso-BuMA), cyclohexyl methacrylate (CyMA), ethylhexyl methacrylate (EHMA), and styrene monomers were purified by shaking with 10% aqueous NaOH solution, washing with distilled water, and drying with CaCl₂. After this treatment the monomers were purified by vacuum distillation. Methacrylic acid (AcMA), from Aldrich, was vacuum-distilled prior use.

Microemulsions were prepared by adding 0.2 M monomer to an aqueous solution containing 0.52 M SDS. Under these conditions, all monomers are below the saturation limit.

Polymerizations were carried out in a glass dilatometer at 25 ºC. After removing oxygen from the solution by extensive nitrogen bubbling, the solution was directly introduced into the previously desgassed (below 0.01 torr) dilatometer. The samples were irradiated with light from a medium pressure lamp with a glass filter to isolate the 366 nm band. Photoinitiators were used under matched absorption conditions.

RESULTS AND DISCUSSION

After a short lag time, due to the presence of traces of oxygen, the polymerization in homogeneous solution rapidly reaches the steady state condition. This is manifested by a constant rate of volume change, as measured by the decrease of the level of monomer in the dilatometer’s capillary. On the other hand, when the polymerization is carried out in the swollen micelles, a true steady state condition is hardly achieved, with the rate increasing with conversion over a considerably longer time period (Figure 1). This is particularly noticeable when monomers predominantly incorporated into the micellar pseudophase are employed. These data, obtained at conversions below ca. 10 %, are similar to that reported in most polymerizations carried out in microemulsions or in swollen micelles^4-7).

Fig. 1 Conversion-time plots of the microemulsion polymerization of iso-BuMA (0.2 M monomer, 0,52 MSDS) using: (#) ABAP; and () AIBN, as photoinitiators under matched absorption conditions.

The nonlinearity of the contraction vs time plots makes difficult to obtain rate constants that allow a comparison of the initiation efficiencies of both initiators. We choose to define the efficiency (E) as the inverse of the time required to produce a given amount of polymer. For a given monomer, the relative efficiency of both initiators is given by

EAIBN/EABAP = tABAP/tAIBN

(1)

where t_ABAP and t_AIBN are the time required, after starting the polymerization, to produce a given volume diminution when ABAP or AIBN are employed under matched absorption conditions, respectively. This definition of relative efficiency has the advantage that it can be employed irrespectively of the rate law and how it changes with conversion. The values obtained for a series of monomers of widely different hydrophobicity, ranging from acrylamide (mostly solubilized in the aqueous phase) to styrene and ethylhexyl methacrylate (mostly solubilized in the micellar pseudophase) are collected in Table 1. These data show that, for all monomer considered, ABAP is more efficient photoinitiator than AIBN. These initiation efficiency differences are considerably larger than that observed in homogeneous solutions⁸), and must be related to the different distribution of both photoinitiators⁴), showing that, irrespective of the monomer distribution, the production of the geminate radical pair inside the micelle reduces the efficiency of the initiation process. In fact, for all the monomer but styrene (see following discussion), the efficiency of AIBN, relative to that of ABAP, is 0.48 ± 0.09. This implies only a small dependence with the monomer characteristics. However, a closer inspection of the data given in Table 1 shows that the relative efficiency of AIBN seems to be smaller when the monomers are highly soluble in water (acrylamide and methacrylic acid) or is significantly distributed among the micelles and the surrounding aqueous environment (methyl methacrylate⁹)).

Table I. Relative efficiencies of AIBN and ABAP, polymerization rates in benzene solution (R_bz) employng AIBN as photoinitiator, and time required t_AIBN to reach a given volume diminution when AIBN is employed as photoinitiator in microemulsion polymerization.

In order to relate relative efficiencies to the fraction of radicals initially produced that are able to produce high molecular weight polymer molecule (f), it is necessary to know the relationship between polymerization rate and initiation rate. In homogeneous solution and under steady state conditions, the polymerization rate varies as R_i^0.5 . In microemulsion, the relationship between production of active radicals and polymerization rate can vary from nearly zero to 0.5, even in steady state conditions^1,5,10-12). Under the present conditions, it must be considered that most of the micelles must be devoid of radicals, since the number of micelles is considerably higher than the total number of radicals produced during the reaction time considered. In fact, if it is taken an aggregation number⁴) of ca. 100, the number of micelles is ca. 3.4 x 10¹⁸ micelles/mL. The rate of radical production is, under our experimental conditions, ca. 6 x 10¹⁶ radicals/mL min. Even after 10 minutes photopolymerization, most of the micelles must be devoid of radicals. Two limiting situations can be envisaged under these conditions:

i) The macroradicals remain, after reaching a critical size, in the same micelle. This applies to monomers whose polymer (and hence the macroradicals) are insoluble in water.

ii) The system reaches the steady state condition with a termination that is second order in radicals. This situation applies if the polymerization takes place mainly in the aqueous pseudophase (acrylamide and methacrylic acid) or if the macroradicals (and/or the initial radicals) migrate from micelle to micelle. This situation can lead to steady state conditions and bimolecular termination rates due to the interchange of macroradicals between micelles due to the micelle break up. This process takes place, at least with normal micelles, in the millisecond time scale. Although the presence of hydrophobic polymer could stabilize the microaggregates, migration of the radicals could take place by this mechanism, particularly at low conversions. In this limiting situation it can be expected polymerization rates proportional to Ri^0.5.

In the limiting situation i) the termination process could be disregarded and

Rate = k n

(2)

where n is the total number of chain carrying radicals. In absence of termination,

n = / Ri dt

(3)

and hence

Rate = k / Ri dt = k Ri t

(4)

The produced polymer at a given time is

Polymer = / Rate dt = k Ri / t dt = 0.5 k Ri t2

(5)

Hence, the times required to produce a given amount of polymer are related to the rate of initiation by

RiAIBN /RiABAP = (tABAP/tAIBN)2

(6)

and hence, since the Ri is proportional to the fraction of initial radicals that produce polymer chains,

fAIBN/fABAP = (EAIBN/EABAP)2

(7)

A relationship between the efficiencies and the initiation efficiencies like that given by eqn. (7) also holds if the system achieves a steady state condition by termination in the aqueous pseudophase and/or by radical migration between micelles (limiting condition ii). This analysis allows us to conclude that, under our experimental conditions, it can be expected that eqn. (7) could hold, irrespective of the system considered. This is further supported by previous studies that showed an initiator order of ca. 0.5 in most microemulsion polymerizations4). Values of f_AIBN/f_ABAP obtained with eqn. (7) are also included in Table I.

For AIBN, the low initiation efficiency can be determined by the exit of the original micellar supercage of one of the geminate radicals, and/or by the fraction of radicals formed in the external medium as a consequence of the small fraction of AIBN present in the aqueous pseudophase⁵). However, the fraction of AIBN present in the external medium⁴) is considerably smaller than f_AIBN obtained in this work, discarding so the second possibility. The value of f_AIBN must be then related to the fraction of radicals that avoid recombination in the primary cage. The results of Table 1 show that this fraction is between 0.12 and 0.35 for all the monomers but styrene. For the later monomer, ca. 2 % of the initially formed radicals are able to avoid primary recombination. The reduced rate of primary exit can be attributed to the occurrence of a fast primary radical addition to the monomer, competitive to the radical exit. This process will reduce the hydrophylicity of the radical, reducing its exit rate and promoting its intramicellar combination. A fast addition of the primary radicals to the monomer will be favored by the high reactivity of styrene, particularly with electrophylic radicals13, such as the AIBN derived radicals. This produces the peculiar result that a fast reaction of the primary radicals with the monomer reduces the initiation efficiency.

Although relative f values are similar for all monomers (but styrene) considered, there seems to be a tendency to lower relative values for the more hydrophylic monomers. This tendency could be due to two different effects. In the first place, the efficiency of ABAP could be lower when most of the monomer is incorporated into the micelles. The highly hydrophylic ABAP derived radicals most probably initiates the chain reaction in the aqueous phase, even with monomers of low water solubility. When the concentration of monomer in the aqueous solution is very low, the ABAP derived radicals can terminate in the aqueous phase, prior to its reaction with the micelle incorporated monomer. In agreement with these results, Capek et al.12) reported a decrease of the polymerization rate with the monomer hydrophobicity in the microemulsion polymerization of alkyl acrylates initiated by a water-soluble (ammonium peroxodisulphate) initiator. Also, it can be argue that, when the monomer resides in the aqueous phase (acrylamide and methacrylic acid) initiation of the polymerization chain must take place in the external medium. Due to the rather high hydrophobicity of the AIBN derived radicals, re-entry to the micelles could be competitive to the addition process. In this case, the initiation efficiency could be lower than the fraction of radicals that leaves the original cage. On the other hand, if the monomer mostly resides in the micelles, re-entry of the radical to an empty micelle will lead to initiation and hence the rate of initiation could be equated to the rate of radicals exit from the original cage. The present data do not allow differentiating these two possibilities. This could be tested employing a totally excluded radical trapp, such as negatively charged nitroxide.

Data of Table 1 show that the value of t_AIBN for all the alkyl methacrylates is very similar. If it is considered that the polymerization rate in homogeneous solutions (measured also in mm/min) are very similar (see values of polymerization rates in benzene, Table 1), the similarities of t_AIBN values would indicate that the efficiencies of radical initiation in the microemulsion is nearly independent of the monomer distribution between the micelles and aqueous pseudophases.

In conclusion, the present data indicate that, for monomers of widely different hydrophobicity, the water-soluble photoinitiator is ca. twice as efficient as the micelle incorporated compound. This is related to the super-cage intramicellar termination of the geminate pair produced inside the micelles.

ACKNOWLEDMENTS

The financial support of FONDECYT (# 1970414) and DICYT.

1. J.E. Puig. Polymeric Materials Encyclopedia; J.C. Salamone Ed., CRC Press: Boca Raton, FL; Vol. 6, p. 4333.        [ Links ]

2. P.L. Kuo, N.J. Turro, C.M. Tseng, M.S. El-Aasser and J.W. Vanderhoff. Macromolecules, 20, 1216 (1987).        [ Links ]

3. M.V. Encinas, E.A. Lissi and A.E. Olea. J. Polym. Sci.: Polym. Chem. Ed., 21, 2157 (1983).        [ Links ]

4. M.V. Encinas, E.A. Lissi, A.M. Rufs and J. Alvarez. Langmuir, 14, 5691 (1998).        [ Links ]

5. L.M. Gan, C.H. Chew and I. Lye. Macromol. Chem., 193, 1249 (1992).        [ Links ]

6. J.E. Puig, V.H. Pérez-Luna, M. Pérez-González, E.R. Macías, B.E. Rodríguez and E.W. Kaler. Colloid Polym. Sci., 271, 114 (1993).        [ Links ]

7. I. Capek and J.P. Fouassier. Eur. Polym. J., 33, 173 (1997).        [ Links ]

8. J. Alvarez, M.V. Encinas and E.A. Lissi. Macromol. Chem. Phys., 200, 2411 (1999).        [ Links ]

9. M.V. Encinas, E. Guzmán and E.A. Lissi. J. Phys. Chem., 87, 4770 (1983).        [ Links ]

10. J.S. Guo, M.S. El-Aasser and J.W. Vanderhoff. J. Polym. Sci.: Part A: Polym. Chem. 27, 691 (1989).        [ Links ]

11. L.M. Gan, C.H. Chew, S.C. Ng and S.E. Loh. Langmuir, 9, 2799 (1993).        [ Links ]

12. I. Capek, V. Juranicova, J. Barton, J.M. Asua and K. Ito. Polymer Inter., 43, 1 (1997).        [ Links ]

13. M.V. Encinas, M. Rivera and E.A. Lissi. J. Polym. Sci., Polym. Chem. Ed., 16, 1709 (1978).        [ Links ]