An alternative method for the determination of siloxane activities toward basic equilibration catalysts

J. Serb. Chem. Soc. 70 (12) 1461 1468 (2005) UDC 678.84:541.121:54 44 JSCS 3384 Original scientific paper An alternative method for the determination of siloxane activities toward basic equilibration catalysts MILUTIN N. GOVEDARICA # Institute of Chemistry, Technology and Metallurgy, Centre of Chemistry, Polymer Department, Njego{eva 12, 11000 Belgrade, Serbia and Montenegro (e-mail: miluting@helix.chem.bg.ac.yu) (Received 27 October 2004, revised 7 April 2005) Abstract: The method used is based on the well-known fact that siloxane equilibrates, once formed, do not change their compositions unless some siloxane compound is added, in which case new equilibrium compositions appear. As these composition changes, as well as their dynamics, are caused solely because of the addition of a particular siloxane compound, they are expected to be specific, and should contain information about the siloxane activities toward the applied equilibration catalyst. It was shown that the viscosities of such systems, measured as a function of reaction times, could be used for the determination of the relative activities of siloxanes. Proceeding from this basic assumption, some commonly used siloxanes were tested in equilibrations catalysed with tetramethylammonium hydroxide, TMAH. The siloxanes were: hexamathylcyclotrisiloxane, D 3, octamethylcyclotetrasiloxane, D 4, tetravinyltetramethylcyclotetrasiloxane, D Vinyl 4, hexamethyldisiloxane, MM, and a linear all-methyl oligosiloxane of number average molecular weight of approximately 800. MD 8.5 M. The following decreasing order of activities toward the TMAH-catalyst was obtained: D 3 >MD 8.5 M>D 4 >D Vinyl 4 >MM. Keywords: siloxane equilibrations, basic equilibration catalyst, activities of siloxanes. INTRODUCTION The siloxane bonds of siloxane compounds easily undergo catalysed scission and reformation, whereby the initial composition, with respect to both molecular type (cyclic, linear) and size, changes and is converted into a new, redistributed one. This process leads to the formation of an equilibrium state, the composition of which is essentially determined by the composition of the starting reaction mixture. The ease of the interaction between the applied catalyst and the siloxane bonds is influenced by many factors, such as, for example, the type of catalyst and the polarity of the siloxane bonds, which is determined by the subtitutents attached to the silicon atoms, steric hindrance, etc. Thus, the over-all behavior of siloxane compounds in equilibrations, regarding their susceptibility and tendency to undergo the # Serbian Chemical Society active member. doi: 10.2298/JSC0512461G 1461

1462 GOVEDARICA reaction, denoted here as the "siloxane activity", is generally expected to be different. The knowledge of these relations, i.e., of different activities, may be very useful when considering equilibration reactions. In the past, the behavior of some siloxane compounds was investigated, 1 and the following decreasing order of activities toward the basic catalysts (tetramethylammonium hydroxide, TMAH), was found: D 3 >D 4 >MD 2 M>MDM>MM, where D = (CH 3 ) 2 SiO and M = (CH 3 ) 2 SiO 1/2. If acidic catalysts were used (sulfuric acid), this order was changed to: D 3 >MM>MDM>MD 2 M>D 4. A number of bromine and chlorine containing disiloxanes was also examined. 2 All these investigations were carried out by following, mostly viscosimetrically, the equilibrations of different cyclic-linear pairs of siloxanes. The obtained results were then used to make conclusions about the activities of the examined siloxanes. In a recently published article, the relative activities of four disiloxanes, in cationic exchange resin catalysed equilibration, were reported. 3 Again, different cyclic-linear pairs (D 4 disiloxane) were equilibrated, this time the molecular weights M n instead of viscosities, as previously were mesured, and used to obtain the desired activities. In the present study, another approach, which will be described in the next section, was used to determine the siloxane activities in base catalysed equilibrations. Its application was demonstrated on a number of siloxane compounds, namely hexamethylcyclotrisiloxane, D 3, octamethylcyclotetrasiloxane, D 4, tetravinyltetramethylcyclotetrasiloxane, D 4 Vinyl, hexamethyldisiloxane, MM, and a linear all-methyl oligosiloxane with a number average molecular weight of approximately 800, MD 8.5 M, using TMAH as the equilibration catalyst. Finally, the results were compared with the previously reported ones. 1 DETERMINATION PROCEDURE Some of the basic elements of the applied determination procedure have already been described in this journal. 4,5 For the sake of clarity and completeness, a full description of the procedure follows. The starting point in the discussion about the procedure is the fact that the composition of a siloxane equilibrium mixture, once formed, cannot change. However, if any siloxane compound or a mixture of siloxane compounds, the composition of which differs from that of the equilibrium mixture, is added, the situation alters. Such a mixture starts to change and keeps on changing its composition, until an equilibrium is reestablished. The formation of the new equilibrium state, i.e., the transition from one to another equilibrium, is obviously caused by the added siloxane species and is, therefore, expected to be specific, i.e., some kind of a "finger print", for a particular siloxane. Since this transition is nothing else than a redistribution process of the siloxanes, which involves the change of the molecular type and size, and, consequently, of the molecular weight distribution and average molecular weight, this means that all variables which reflect such changes might be used for the determina-

SILOXANE ACTIVITIES DETERMINATION 1463 tion of activity relations of siloxanes. A variable which can be expected to fulfil the above requirements, is the viscosity. It is sensitive enough toward molecular weight changes, and simultaneously, its experimental determination is easy. In other words, the viscosimetrical results obtained during the attainment of new equilibria of some arbitrarily chosen equilibrate and a series of siloxane compounds, should give appropriate information about the relative activities of these particular siloxane compounds. Following this idea, the experimental results obtained by applying the common dilute-solution viscosimetry method were expressed in terms of inherent viscosities, inh,cm 3 /g: inh =ln( s / 0 )/c =ln( s / 0 )/c where: is the flow time, the indices s and 0 refer to solution and pure solvent, respectively, and c is the solution concentration, g/cm 3. Thereafter, these inherent viscosities were used to calculate x values, defined as the fraction of the viscosity change at an arbitrarily chosen reaction time, with respect to the total viscosity change: x =( inh(actual) inh(initial) /( inh(final) inh(initial) )= inh(actual) / inh(total) The calculated x values for different siloxanes, presented graphically as a function of the corresponding reaction time, gave a set of curves, each individual curve representing the corresponding siloxane compound. A simple comparison of these curves demonstrates immediately the desired order of relative activites. Of course, in order to obtain consistent and comparable results, the amounts of the starting equilibrate are to be held constant, and those of the added silxoanes have to be equimolar, i.e., the mole ratio "equilibrate"/"siloxane compound" needs to be constant throughout the entire experimental series. EXPERIMENTAL Materials All investigated siloxanes, namely D 3,D 4,D Vinyl 4 and MM, were obtained from ABCR & Co.KG, Germany, and used as received. The only exception was MD 8.5 M, a linear all-methyl oligosiloxane, which was synthesised by equilibrating the corresponding amounts of D 4 with MM. The equilibrate used as the reactant, i.e., as the reaction milieu in all experiments, was prepared by equilibrating D 4 with 1 wt. % of MM, an arbitarily chosen composition; actually, any other equilibrate could have been used. In this equilibration, a macroporous cation-excange resin Duolite C 26 (Diamond Shamrock, USA) was used as the acidic catalyst, mainly because of its easy-to-handle form and simple separation from the reaction products. 4,5 Tetramethylammonium hydroxide, THAM (Merck Schuchardt, Germany) was used as the equilibration catalyst in the equilibrations aimed at determining the relative activities. The original 25 % aqueous solution was treated in the manner previously described in the literature, 1 i.e.,itwas evaporated using a vacuum pump. Equilibrations All equilibrations were performed in a three-necked, round-bottomed flask equipped with a mechanical stirrer, reflux condenser and a thermometer. The flask was placed into an oil bath, and the temperature was held constant at 70 o C. The concentration of the TMAH-catalyst was 0.5 wt.%.

1464 GOVEDARICA The samples used for viscosity determination were withdrawn from the reaction mixtures at arbitrarily selected time intervals. The reaction was stopped by neutralizing the catalyst with aqueous HCl solution, then washed with water and finally dried over anhydrous sodium carbonate. Viscosity measurements The viscosities were measured using an Ubbelohde-type viscosimeter in toluene at 30 o C. The initial solution concentrations were 0.15 g/cm 3. The measured flow times were then used to calculate the inherent viscosities. RESULTS AND DISCUSSION All determinations of relative activities of siloxanes toward equilibration catalysts were performed by means of equilibrations between an arbitrarily chosen equilibrate and a siloxane compound, the activity of which was to be determined. The equilibrate used as a reaction milieu was obtained by equilibrating a mixture of 99 wt.% of D 4 with 1 wt.% of MM. In order to check the obtained equilibrate, it was subjected to a stability-test, the results of which are given in Table I. It can be seen, that the equilibrate did not change its viscosity, which means its composition, also over a sufficiently long period of time (ten hours), either in contact with different equilibration catalysts or on standing alone. TABLE I. Viscosities measured during the equilibrate stability-test Time/h Inherent viscosities * /cm 3 g -1 Pure equilibrate Equilibrate + TMAH catalyst Equilibrate + Duolite C 26 catalyst 0 11.82 11.82 11.82 (0.16) (0.16) (0.16) 2 11.76 12.24 11.84 (0.04) (0.14) (0.04) 4 11.43 12.24 11.68 (0.22) (0.14) (0.17) 6 11.48 12.19 11.83 (0.10) (0.07) (0.11) 8 11.56 12.07 11.99 (0.04) (0.15) (0.10) 10 11.66 12.07 11.67 (0.09) (0.04) (0.06) *Averages obtained from 3 separate experiments and (in parenthesis) the corresponding standard deviations The results of determinations of activies toward an acidic equilibration catalyst, Duolite C 26 cation-exchange resin, were previously reported in this journal. 4,5 In this work, the method was applied to the determination of the activities toward a basic equilibration catalyst. Tetramethylammonium hydroxide, TMAH, was selected to represent this type of catalyst. Siloxanes, the activity of which was

SILOXANE ACTIVITIES DETERMINATION 1465 determined, were: D 3,D 4,D 4 Vinyl,MMandMD 8.5 M. Using the procedure described above, the inherent viscosities of equilibrating mixtures were measured. The results are presented in Table II, from which it can be immediately seen that D 3 TABLE II. Inherent viscosities of equilibrating mixtures Time/min Inherent viscosities * /cm 3 g -1 D 3 D 4 D Vinyl 4 MM MD <8.5> M 0 9.43(0.13) 8.05(0.04) 7.68(0.13) 9.46(0.06) 6.13(0.07) 10 10.73(0.11) 20 11.20(0.21) 30 12.35(0.08) 45 13.51(0.32) 60 13.88(0.06) 9.35(0.04) 9.25(0.07) 9.89(0.11) 5.15(0.07) 120 14.34(0.15) 11.19(0.10) 11.29(0.10) 8.36(0.10) 3.53(0.05) 180 14.02(0.23) 12.02(0.10) 13.14(0.11) 7.24(0.03) 2.51(0.04) 240 13.89(0.29) 13.03(0.12) 13.70(0.08) 6.21(0.06) 2.27(0.05) 300 13.84(0.14) 13.74(0.12) 14.31(0.11) 5.72(0.06) 2.10(0.07) 360 14.04(0.19) 15.09(0.38) 5.21(0.05) 2.12(0.03) 480 14.28(0.30) 13.96(0.31) 15.60(0.23) 4.69(0.10) 2.10(0.06) 3000 14.16(0.34) 13.98(0.25) 16.94(0.40) 3.25(0.07) 2.19(0.04) * These values are averages obtained from 5 7 individual runs and (in parenthesis) the corresponding standard deviations reacted very fast and the equilibration was completed after two hours. Hence, in the equilibrations with D 3, the viscosities were also measured at times shorter than one hour. Furthermore, with the cyclic siloxanes, D 3,D 4 and D Vinyl 4, the viscosities increased as a function of reaction time, because the cyclics acted as chain-extenders. On the other hand, the addition of linear siloxanes, MM and MD 8.5 M, increased the concentration of terminal units, shortened the chains and provoked a decrease of the measured viscosities. An anomaly was observed in the results obtained for MM. The first measurement after one hour, showed an unexpected viscosity increase. In order to ascertain whether this increase was geniuine, a supplementary series of equilibrations was caried out in which the viscosities were measured at shorter equilibration times. The results are presented in Table III, from which it can be seen that there was a real initial viscosity increase. This phenomenon can be explained as follows. Since MM enters relatively slowly into the reaction, it acts at the very beginning as an inert solvent. It is known that the addition of inert solvents shifts the equilibrium cyclosiloxanes linear siloxanes to the left, increasing the concentration of cyclosiloxanes and, simultaneously, the molecular weight of the linear polymer. 6 These two effects, act in opposite directions with respect to the viscosity of the reacting mixtures, making any prediction of the final viscosity val-

1466 GOVEDARICA ues very uncertain. The experimental results an initial viscosity increase allow the assumption that the second effect an increase of the molecular weight prevails. If this assumption were correct, then the addition of a true inert solvent should cause the same effect as the addition of MM did. An experimental series, performed with the corresponding amounts of xylene as a true inert solvent, the results of which are also shown in Table III, fully confirmed the above argumentation; namely, an initial viscosity increase was observed. However, this phenomenon, because of being limited to short reaction times, and moreover, being of relatively small magnitude, could by no means alter the final results, i.e., the determination of activities. TABLE III. Inherent viscosities obtained in the supplementary experimental series Time/min Inherent viscosities * /cm 3 g -1 Equlibrate + MM Equilibrate + xylene 0 9.55 8.95 (0.12) (0.05) 15 10.05 9.45 (0.06) (0.12) 30 10.17 9.36 (0.04) (0.10) 45 10.35 9.38 (0.11) (0.04) 60 10.13 9.29 (0.04) (0.05) 120 8.83 9.28 (0.10) (0.06) * Averages obtained drom 3 separate experiments and (in parenthesis) the corresponding standard deviations It should be noted that this effect applies to the other siloxanes, too. However, it was not registered, because the cyclics D 3,D 4 and D 4 Vinyl act on viscosities in the same direction as inert solvents, and the activity of MD 8.5 M was obviously high enough to cover and to suppress the observed effect. Returning to Table II, the x =( inh(actual) )/( inh(total) ) values were calculated and are graphically presented as a function of reaction times in Fig. 1. In this way, the following order was obtained: D 3 >MD 8.5 M>D 4 >D 4 Vinyl >MM. As expected, D 3 was found to be significantly the most active siloxane, due to the instability of its six-membered planar ring, 6 just as it was also with the acidic catalysts. 5 Thus it takes a special position among the siloxanes. The lowest activity was demonstrated by MM and D 4 Vinyl, as a consequence of the reduced positive charge on their silicon atoms, caused by the electron-donating substituents. CH 3 and CH=CH 2,re-

SILOXANE ACTIVITIES DETERMINATION 1467 Fig. 1. x as a function of reaction time. 1 D 3 ( );2 MD 8.5 M( );3 D 4 ( ); 4 D 4 vinyl ( );5 MM( ) spectively. The two remaining siloxanes, D 4 and MD 8.5 M, were positioned in-between. This finding was, more or less, in compliance with the general expectation that the activities of all the higher siloxane homologues, both cyclic and linear, should become more and more similar. 5 CONCLUSION It can be concluded that the determination procedure described in this and previous articles, 4,5 could be used for the determination of the order of activities toward the basic tetramethylammonium hydroxide catalyst, giving the following result: D 3 >MD 8.5 M>D 4 >D Vinyl 4 >MM. Moreover, the obtained results coincided rather well with the previously reported ones: D 3 >D 4 >MD 2 M>MDM>MM. 1 A comparison of the orders of the activities obtained for an acidic 5 and a basic type of catalyst presented here shows that in both cases, as expected, D 3 was the most active siloxane monomer. All the other siloxane compounds which appeared in both series changed very regularly their positions in the corresponding orders of activities, when the catalyst type was changed: from MM>D 4 >MD 8.5 Mtowardan acidic, to MD 8.5 M>D 4 >MM toward a basic catalyst. These results fit well into the general idea concerning the siloxane equilibration polymerization. 7 Finally, the presented alternative determination method, the main advantages of which can be seen in its general applicability to all siloxane compounds, and in its simplicity, was shown to be correct in its use not only in reactions with a considerable number of siloxane compounds, but also with respect to both types of catalysts, acidic and basic. Acknowledgement: I wish to thank Mr. S. Petrovi} for his technical assistance.

1468 GOVEDARICA IZVOD ALTERNATIVNI METOD ZA ODRE\IVAWE AKTIVNOSTI SILOKSANA PREMA BAZNOM KATALIZATORU EKVILIBRACIJE MILUTIN N. GOVEDARICA IHTM-Centar za hemiju, Odeqewe za polimerne materijale, Wego{eva 12, 11000 Beograd Kori{}ena metoda se zasniva na poznatoj ~iwenici, da jednom formirani siloksanski ekvilibrati vi{e ne mewaju svoj sastav, osim ako im se doda neko siloksansko jediwewe, u kom slu~aju dolazi do stvarawa novog ravnote`nog sastava. Treba o~ekivati da ova promena sastava, kao i dinamika promene s obzirom da su izazvane iskqu~ivo dodatkom siloksanskog jediwewa budu specifi~ne i da sadr`e informacije o aktivnosti dodatog siloksanskog jediwewa prema primewenom katalizatoru ekvilibracije. Pokazano je da viskoziteti reakcionih sme{a mereni u zavisnosti od vremena trajawa reakcija mogu da se iskoriste za odre ivawe relativnih aktivnosti siloksana. Polaze}i od ove osnovne pretpostavke, ovom metodom odre ene su aktivnosti nekoliko siloksana, i to u bazno katalizovanim ekvilibracijama. Katalizator je bio tetrametilamonijum hidroksid, TMAH, a siloksani: heksametilciklotrisiloksan, D 3, oktametilciklotetrasiloksan, D 4, tetraviniltetrametilciklotetrasiloksan, D 4 Vinyl, heksametildisiloksan, MM, i jedan linearni metil supstituisani oligosiloksan sredwe brojne molarne mase od oko 800, MD 8.5 M. Dobijen je slede}i opadaju}i redosled aktivnosti: D 3 >MD 8.5 M>D 4 >D 4 Vinyl >MM. (Primqeno 27. oktobra 2004, revidirano 7. aprila 2005) REFERENCES 1.S. W. Kantor, W. T. Grubb, R. C. Osthoff, J. Am. Chem. Soc. 76 (1954) 5190 2. W. Simmler, Makromol. Chem. 57 (1962) 12 3. M. Cazacu, M. Marcu, A. Vlad, D. Caraiman, C. Racles, Eur. Polym. J. 35 (2000) 1629 4. M. N. Govedarica, J. Serb. Chem. Soc. 65 (2000) 639 5. M. N. Govedarica, J. Serb. Chem. Soc. 66 (2001) 429 6. M. G. Voronkov, V. P. Mileshkevich, Yu. A. Yuzhelevskii, The Siloxane Bond, Consultants Bureau, New York, 1978, pp. 160, 163 7. T. C. Kendrick, B. M. Parbhoo, J. W. White, in Comprehensive Polymer Science,G.Allen,J.C. Bevington, Eds. Pergamon Press, London, 1989, pp. 460, 479.