Ind. Eng. Chem. Res. 1994,33,1687-1691 1687 Synthesis of Triphenyl Phosphate and Benzyl Benzoate with Phase-Transfer Catalyst in Heterogeneous Liquid-Liquid Reaction System S a t o r u Asai,’ Hidemi Nakamura, Mitsunori Tanabe, and Kenji Sakamoto Department of Chemical Engineering, University of Osaka Prefecture, Sakai, Osaka 593,Japan The synthetic reactions of triphenyl phosphate from diphenylphosphoryl chloride and sodium phenoxide and of benzyl benzoate from benzyl chloride and sodium benzoate with phase-transfer catalyst were studied in a heterogeneous liquid-liquid reaction system using an agitated vessel with a flat interface.
Tetrabutylammonium bromide and 1,2-dichloroethane were used as a catalyst and a solvent, respectively. The behavior of the observed overall reaction rates was explained well by the proposed model. The overall reaction rates were proportional to the organic-phase interfacial concentrations of the actual reactants, tetrabutylammonium phenoxide for the synthesis of triphenyl phosphate and tetrabutylammonium benzoate for that of benzyl benzoate.
Their interfacial concentrations were a unique function of the concentrations of tetrabutylammonium bromide and the respective sodium salts. The intrinsic reaction rate constants, for the synthetic reactions of triphenyl phosphate and benzyl benzoate, which were evaluated by fitting the rate data to the model prediction, were 2. 33 X lo6 and 2. 56 m3/kmol-s at 303 K, respectively. Introduction Phase-transfer catalyst can enhance the rates of heterogeneous liquid-liquid reactions which are unreactive at ordinary temperatures and pressure.
The previous studies have been carried out with a view to obtaining high product yield and high selectivity (Weberand Gokel, 1977; Starks and Liotta, 1978; Dehmlow and Dehmlow, 1983), and the approach from the aspect of chemical engineering allowing for the effect of mass transfer on the overall reaction rate has rarely been achieved. We carried out the alkaline hydrolysis of n-butyl acetate and the oxidation of benzyl alcohol by using phase-transfer catalysts and demonstrated that the overall reaction rates could be explained well by the theoretical solution for phase-transfer catalysis with mass transfer (Asai et al. 1992, 1994). In the present work, we analyzed the synthetic reactions of two esters, that is, triphenyl phosphate from diphenylphosphoryl chloride and sodium phenoxide (system 1) and benzyl benzoate from benzyl chloride and sodium benzoate (system 2). Triphenyl phosphate (system 1)is indispensable for the plastic industry as a plasticizer or flame retarder of polymer (such as polyvinyl chloride, cellulose acetate, etc. ). Until now, triphenyl phosphate has been synthesized by the reaction of phosphoryl chloride with phenol under elevated temperature and pressure by using aluminum chloride or magnesium chloride as a solid catalyst.
However, its severe operating conditions and the corrosion of the reactor by hydrogen chloride formed as a byproduct are the disadvantages of this process. On the contrary, the use of phasetransfer catalyst, with sodium phenoxide instead of phenol as a reactant, makes it possible for the reaction to proceed under mild conditions and to avoid the production of hydrogen chloride. It has been confirmed that the synthetic reaction of triphenyl phosphate between phosphoryl chloride and sodium phenoxide occurs in consecutive steps as follows (Krishnakumar and Sharma, 1985):
Reaction I11 is a rate-determining step. Krishnakumar and Sharma carried out the experiments on the reaction I11 by using Aliquat 336 as a phase-transfer catalyst and chloroform as a solvent. They suggested that the overall reaction rates were explicable qualitatively by the theoretical predictions for the fast pseudo-first-order reaction. However, the evaluation procedure of the interfacial concentration of the actual reactant, that is, the ion pairs consisting of quaternary ammonium cation and phenoxide ion, remained unknown, and the reaction rate constants were not evaluated.
On the other hand, benzyl benzoate (system 2) is used as a plasticizer in surface coating, a solvent or fixative of perfume and fruit flavor, and the solvents of cellulose acetate, nitrocellulose, and insect repellents, etc. Until now, benzyl benzoatehas been manufactured under severe operating conditions, either by dry esterification of benzyl chloride and sodium benzoate in the presence of triethylamine or by condensationof benzaldehyde in the presence of sodium. However, the use of the phase-transfer catalyst will make it possible to operate under mild conditions.
In this work, in the first place, the effects of various kinds of phase-transfer catalysts and solvents on the overall reaction rates for the synthesis of triphenyl phosphate were investigated to find out the appropriate catalyst and solvent. Next, the overall reaction rates for both synthetic reactions were measured by varying the concentrations of the chosen catalyst and reactants, as well as the ionic strength of the aqueous phase, and were analyzed by using the theoretical predictions.
Experimental Section The apparatus used in this study was the same type of agitated vessel with a flat interface as that used in a 0 1994 American Chemical Society * To whom correspondence should be addressed. 0888-5885/94/2633-1687$04. 50/0 1688 Ind. Eng. Chem. Res. , Vol. 33, No. 7, 1994 Table 1. Effects of Solvent and Phase-Transfer Catalyst (PTC) on Overall Reaction Rates for Synthesis of Triphenyl Phosphate overall reaction rate (kmol/m%) solvent PTC 0. 67 X l t 7 benzene TBAB toluene TBAB 1. 02 x 10-7 3. 02 X 10-‘ chloroform TBAB 5. 05 X 1,2-dichloroethane TBAB TBAC 3. 9 X le7 1,2-dichloroethane TBAI 1. 77 X le7 1,2-dichloroethane 2. 92 X lk7 1,2-dichloroethane TOMAC noncatalyst 8. 00 X 1,2-dichloroethane Organic phase, 0. 05 kmol/mg PTC and 0. 048 kmoVmS DPPC; aqueous phase, 0. 1 kmol/ma C;ONa and 0. 9 kmol/mS NaOH temperature, 303 K. 0 nC aqueous phase WEr- 5 Q’ t Br- (b) t WOH- e C Y C,H,O(c) Q W K 4 t OH- (e) (d) O’C,H,O- 5 0′ t CI- 4 interface 0 organic phase Q’Br- N, Q*C,H,Ot (C,H,O),POCI F – N 2 E – – . . I- – . – Q T X t (C,H,O),PO (a) Figure 1. Reaction model for synthesis of triphenyl phosphate. benzoate (system 2) in the organic phase.
The experiments were carried out a t agitation speeds of 2. 5 l / s for each phase, a temperature of 303 K, and an ionic strength of 0. 2-3. 0 kmol/m3. Reaction Model and Analytical Procedure The reaction model for the synthetic reaction of triphenyl phosphate in the present heterogeneous reaction system is shown in Figure 1. TBAB (Q+Br), existing initially in the organic phase, transfers to the aqueous phase and dissociates into two ions, Q+and B r , according to reaction b. The dissociated Q+ reacts with C6H50existing in the aqueous phase to form another ion pair Q+C;o- (reaction c).
The lipophilic ion pair ;+C;omovesto the organic phase and reacts with (C,;,O)$’OCl (DPPC) existing there originally (reaction a). The formed ion pair Q+Cl- transfers to the aqueous phase and dissociates partially into Q+ and C1- (reaction d). The resulting ion Q produces Q+C&O- again, and this ‘ catalytic reaction cycle is repeated. The main reaction a was assumed to be first order with respect to Q+C;Oand DPPC, respectively. However, this reaction may be regarded as a pseudo-first-order reaction with respect to Q+C;@-, because [DPPCI ;gt;;gt; [;+C6H50-].
The dissociations ;e of tetrabutylammonium salts Q+X- (X- = C6H50-, C1-, B r , and OH-) may be regarded as instantaneous reversible reactions, as well as that of Aliquat 336 (Asaiet al. , 1991). It is thought that most of ;+OH-,which is formed by reaction e between Q+and OH-, substantially exists in the aqueous phase from the relation of distribution and dissociation equilibria (Asai et al. , 1993). The differential equations and boundary conditions for the present situation may be set up in parallel to those in the previous studies (Asai et al. , 1992, 1994).
The distribution and dissociation equilibria of Q’X- ion pairs (X- CsH50-, C1-, B r , and OH-) are given by the following = expressions: previous work (Asai et al. , 1983). First of all, the effects of various kinds of catalysts and solvents on the formation rates of triphenyl phosphate were examined to search for the appropriate catalyst and solvent. Phase-transfer catalysts used were tetrabutylammonium bromide (TBAB), tetrabutylammonium chloride (TBAC), tetrabutylammonium iodide (TBAI), and trioctylmethylammonium chloride (TOMAC), and organic solvents were 1,Bdichloroethane, benzene, toluene, and chloroform.
The experiments were carried out at 303 K and agitation speeds of 2. 5 l/s for each phase, by using an organic phase containing 0. 05 kmol/m3 catalyst and 0. 048 kmol/m3 diphenylphosphoryl chloride (DPPC) and an aqueous phase containing 0. 1 kmol/m3 sodium phenoxide (C&ONa) and 0. 9 kmoll m3 sodium hydroxide. The overall reaction rates were obtained from the decrease in the aqueous concentration of C6H5ONa with time. The concentration of CeHsONa in the aqueous phase was measured by an ultraviolet spectrophotometer. The experimental results are listed in Table 1.
The reaction did not substantially occur for the noncatalyst system, but the overall reaction rate was accelerated several thousand times by the addition of any phase-transfer catalyst. From these results, the best solvent and catalyst were proven to be 1,Zdichloroethane and TBAB, respectively. Therefore, 1,2-dichloroethane and TBAB were used in the synthetic experiments of two esters (systems 1and 2). The lower organic phase was 0. 048-0. 24 kmol/m3 DPPC (system 1)or 0. 441-1. 76 kmol/m3 benzyl chloride (system 2) diluted with 1,Zdichloroethane. The upper aqueous phase consisted of the mixed solutions of 0. -0. 5 kmol/m3 sodium phenoxide (system 1)or 0. 05-0. 5 kmol/m3sodium benzoate (system 2), with 0. 1-2. 9 kmol/m3 sodium hydroxide. TBAB (Q+Br) of 0. 005-0. 08 kmol/m3 was dissolved in the organic or aqueous phase. Sodium hydroxide was used to adjust the ionic strength and to prevent the production of phenol or benzoic acid. Samples of 1. 0 X 1O-e-2. 0 X 1O-e m3were taken from the upper and lower phases a t intervals of 15 min. The concentrations of triphenyl phosphate (TPP) or benzyl benzoate formed in the organic phase were measured by high-performance liquid chromatography.
The concentration of tetrabutylammonium cation Q+ in the aqueous phase was determined by the orange II-chloroform method with an ultraviolet spectrophotometer (Scott, 1968). The aqueous concentrations of bromide ion B r , chloride ion C1-, and benzoate ion C&&O2- were measured by ion chromatography. The aqueous concentration of phenoxide ion C6HsO- was taken as being approximately equal to the concentration of C,&ONa. The concentration of OHwas determined by titration with aqueous HC1 solutions. The overall reaction rates were evaluated from the time course of the concentrations of TPP (system 1)or benzyl
From the observed values of the concentration of CsHsO-, C1-, B r , OH-, and Q+ in the aqueous phase: [X-l,, = [&+x-lb + LX-1, (3) [Q+Ioh = [&+C6H,0-lb+ [Q+C1-lb+ [&+Br-lb + [Q’OH-Ib -I[;+lb (4) M s transfer of ;+OH-from the aqueous phase to the as organic phase may be considered to be negligible as Ind. Eng. Chem. Res. , Vol. 33, No. 7,1994 1689 mentioned above. Moreover, the aqueous-phase concentration of OH- is considerably higher than that of Q+. Thus, the following relations may hold: [;+OH-], = [OH-], [OH-], (5) (6) Dj/Djw= 1- (0. 41[C6H,COzNal + 0. 081[NaCl] + 0. 593[NaBrl + 0. 138[NaOHl) for system 2 (11) (j = Q’Cl-, Q’Br-, and Q+C6H,CO;) Solving the relevant differential equations with eqs 1-3, 5, and 6, one obtains the following expressions for the overall reaction rate, that is, the mass-transfer rate N1 of Q+CsHsO- across the interface and for the interfacial concentration of Q+C;O- in the organic phase: 4 (7) The diffusivities of C1-, B r , OH-, C,;0-, and CSHSCOZions in the aqueous electrolyte solutions were estimated from the Vinograd-McBain equation (1941).
The distribution coefficients Q+X- and the dissociation constants KQ+x- (X-= C;O-, C1-, B r , OH-, and CeH;02- were taken from the previous paper (Asai et al. , 1993). The mass-transfer coefficients ku and ku of relevant species in the aqueous and organic phases, respectively, were predicted from the empirical correlation of Asai et al. (1983). where Results and Discussion When the main reaction a in the organic phase for the synthesis of triphenylphosphate occurs in the fast pseudofirst-order regime, that is,? ;gt; 3, eqs 7 and 9 can be written as: ml and K1 refer to mQ+m50- [email protected], respectively. nd The value of the dimensionless number y can be evaluated by fitting the observed mass-transfer rates N1 with eqs 7 and 8, using [;+]I, calculated from e s 2-4. Thus, one can determine the reaction rate constant from the evaluated value of ? by means of eq 9. For the synthesis of benzyl benzoate, on the other hand, the overall reaction rate and the interfacial concentration of Q+CeH;O2- in the organic phase can be estimated by the replacement of C;0-, subscripts 1and 5, and DPPC in eqs 7-9 with C;€SCOZ-, subscripts 6 and 7, and C;CHzC1, respectively.
Physical Properties The relevant physical properties were measured, or predicted, as follows. The densities, the viscosities, and the interfacial tentions of the solutions used in all the experiments were measured in a similar way to those in a previous paper (Asai et al. , 1992). The diffusivities Djw of Q+Cl-, Q+Br, Q+C;O-, and Q+C,J-I;Oz- in water were predicted in a similar manner to the previous work (Asaiet al. , 1991)by using the NernstPlanck equation and taken as 8. 28 X 10-10, 8. 32 X 10-10, 7. 52 X lO-lO, and 6. 51 X W0 m2/s,respectively, at 303 K.
The diffusivities of Q+C;0- and Q + C ~ H S C ~ Z the in and 2. 06 organic phase were predicted to be 2. 63 X X mz/sat 303 K from the observed diffusivity of Q+Br = 3. 07 X 10-9 m2/s (303 K), by the same procedure as in the previous paper (Asai et al. , 1991). The diffusivitiesof tetrabutylammonium salts in the mixed electrolyte solutions for the reactive systems were predicted from the following expression based on the correlation procedure proposed by Hikita et al. (1979): Dj/Djw= 1- (0. 239[C6H,ONa1 + O. OSl[NaClI + for system 1 (10) 0. 593[NaBrl + 0. 138[NaOHI)
Figure 2 represents the log-log plot of the observed overall reaction rates, that is, the mass-transfer rates N 1 of Q+C;,O- against the bulk concentration of DPPC in the organic phase. The data points can be correlated by the straight line with slope equal to 0. 5, indicating that this reaction system is of first order with respect to DPPC. According to eq 12, Nl/(~l[DPPClb)1/2 should be proportional to the interfacial concentration [Q+C,H,O-liof Q+C;0- in the organic phase, and the reaction rate constant K can be determined from the slope of the straight line of a plot of N/(Dl[DPPC]b)1/2 vs [Q+C,H,O-],.
A l experimental data for the synthesis of l triphenyl phosphate (system 1) are plotted in Figure 3. Symbols in Figure 3 are listed in Table 2. The experimental values corresponding to the ordinate are proportional to the interfacial concentration [;+C,H,O-], of Q+C;O- in the organic phase, as expected. From the slope of the straight line, the second-order reaction rate constant K was determined to be 2. 33 X 108 m3/kmol-sat 303 K, independent of the physical properties in the aqueous solution. The increase in the overall reaction rates N1 is attributable to an increase of the interfacial concentration of Q+C;0-.
The increase in the initial concentration of Q + B r in the organic phase increases the concentration of Q+C;o- in the aqueous phase, which is formed by ion exchange between B r and C6HSO-9 thus increasing the concentration of the organic phase interfacial concentration of Q+C6H;-. Similarly, the increase in the initial concentration of C6HsONa in the aqueous phase results in an increase in the interfacial concentration of ;+c;oin the organic phase. The interfacial concentration [Q+C6H60-li increases with the concentration of NaOH. This may be considered as follows.
The increase in the concentration of NaOH brings out the decrease in the distribution of Q + B r into the aqueous phase owing to enhancement of the salting-out effect, reducing the bulk concentration of Q+ in the aqueous phase. This inhibits the formation of Q+CsH;- in the aqueous phase, while 1690 Ind. Eng. Chem. Res. , Vol. 33, No. 7, 1994 7 t i v , , I , I , , , , , , , , , i 5 [QtC;C02-]i X 0. 01 0. 1 [ m ] b 1 5 (kmol/m3) 5 0 10 I O4 ( k r n o l ~ n – ~ ) 15 Figure 2. Effect of bulk concentra-of DPPC in organic phase on overall reaction rates at 303 K, [Q+Br-lo= 0. 02 kmol/m3, [CeHSONalo = 0. 1 kmol/m3, and [NaOHIo = 0. 9 kmol/m3.
Figure 4. Determination of second-order reaction rate constant k for synthesis of benzyl benzoate. For definitions of symbols, see Table 3. Table 3. Symbols in Figure 4 symbol A A A A 0 0 0 0 0 0 ” R 10 E IC,H,CH,ClI (kmol/m3) 0. 441 0. 882 1. 323 1. 764 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 0. 882 CQ+Brlo [CeHsCOzNalo [NaOHlo (kmol/ms) (kmol/m3) (kmoUm3) 0. 05 0. 05 0. 05 0. 05 0. 01 0. 02 0. 03 0. 06 0. 08 0. 02 0. 02 0. 02 0. 02 0. 02 0. 02 0. 02 0. 02 0. 02 0. 03 0. 03 0. 03 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 05 0. 1 0. 15 0. 2 0. 3 0. 4 0. 5 0. 1 0. 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 95 0. 9 0. 85 0. 8 0. 7 0. 6 0. 5 0. 4 1. 9 0. 4 0. 9 1. 9 zrO 1 8 a 2 4 6 a l a [QfC&150-]i x 1O5 (kmol. m-3) Figure 3. Determination of second-order reaction rate constant k for synthesis of triphenyl phosphate. For definitions of symbols, see Table 2. Table 2. Symbols in Figure 3 symbol 0 0 0 0 0 0 0 8 0 V V V (kmol/mq 0. 048 0. 048 0. 048 0. 048 0. 048 0. 048 0. 048 0. 048 0. 048 0. 048 0. 048 0. 048 0. 048 0. 048 0. 048 [m] (kmol/mi) [Q’Br-I 0. 005 0. 01 0. 02 0. 03 0. 05 0. 06 0. 02 0. 02 0. 02 0. 02 0. 01 0. 01 0. 02 0. 02 0. 02 – [C~HSON~IO [NaOHlo (kmol/m3) (kmol/m3) . 1 0. 1 0. 1 0. 1 0. 1 0. 1 0. 2 0. 3 0. 4 0. 5 0. 2 0. 5 0. 1 0. 1 0. 1 0. 1 0. 5 0. 5 0. 1 0. 1 0. 1 0. 9 0. 9 0. 9 0. 9 0. 9 0. 9 0. 8 0. 7 0. 6 0. 5 0. 8 0. 5 0. 1 0. 4 1. 9 2. 9 1. 5 2. 5 0. 9 0. 9 0. 9 v A A A A A A 0 Similar tendencies were also observed for the synthesis of benzyl benzoate (system 2). The plot of Nl/(D,. [C6H5CH,Cl]b)1/2 [Q+CBH5CO;li for all the experivs mental data are shown in Figure 4, and the symbols are shown in Table 3. From the slope of the straight line, the second-order reaction rate constant h was determined to be 2. 56 m3/kmol. s, a t 303 K, independent of the physical properties in the aqueous solution.
This value is remarkably smaller than that for the synthesis of triphenyl phosphate (system 1). a m 8 Conclusion The synthetic reactions of triphenyl phosphate from diphenylphosphoryl chloride and sodium phenoxide and of benzyl benzoate from benzyl chloride and sodium benzoate were accelerated remarkably by the addition of the phase-transfer catalyst, tetrabutylammonium bromide. The overall reaction rates for the synthesis of triphenyl phosphate were quantitatively interpreted in terms of the theoretical solution for the irreversible fast pseudo-firstorder reaction, eqs 8,9, and 12.
The interfacial concentration of &[email protected] was a unique function of the concentrations of Q+Br-, CsH&Na, and NaOH. The interpretation was the same also for the synthesis of benzyl benzoate. a El V V V 0. 048 0. 048 0. 048 0. 096 0. 144 0. 240 0. 02 0. 02 0. 02 0. 02 0. 02 0. 02 the equilibrium distributions of Q’CsHsO- become more favorable to the organic phase with an increase in NaOH concentration, due to the salting-out effect. Therefore, the increase in [Q+C6H,0-] with NaOH concentration, i. e. , the enhanced overall reaction rates N I ,suggests that the latter effect is large compared with the former effect.
Ind. Eng. Chem. Res. , Vol. 33, No. 7,1994 1691 The reaction rate constant for the synthetic reactions of triphenyl phosphate and benzyl benzoate were determined to be 2. 33 X 108 and 2. 56 m3/kmol-s,respectively, at 303 K, independent of the physical properties in the both phases. Nomenclature D = diffusivity, m2/s I = ionic strength of solution, km0Vm3 K = dissociation constant in aqueous phase, kmol/m3 5 = second-order reaction rate constant, m3/kmolms k L = mass-transfer coefficent, m/s m = distribution coefficient between organic and aqueous phases = Q+C&C027 = cfl5co2- Literature Cited Asai, S. ; Hatanaka, J. ; Uekawa, Y. Liquid-LiquidMass Transfer in an Agitated Vessel with a Flat Interface. J. Chem. Eng. Jpn. 1983,16,463-469. Asai,S. ; Nakamura, H. ; Furuichi, Y. The Distribution and Dissociation Equilibria of PhmTransfer Catalyst Tricaprylmethylammonium Chloride and Ita Aqueous-PhaseMasa Transfer. J. Chem. Eng. Jpn. 1991,24,653-658. A d ,S. ; Nakamura, H ; . Furuichi, Y. Alkaline Hydrolysis of n-Butyl Acetate with Phase T r a d e r CatalystAliquat 336. AIChEJ. 1992, 38,3974. N = mass-transfer rate, kmol/mZ. Q+ = [CH&Hzh14N+ X-= CaaO-, C1-, B r , OH-, and CBHSCOZ[ I = concentration, kmol/m3 Greek Letters 7 = dimensionless number defined by eq 9 6 = film thickness, m Superscript – = organic phase Subscripts Asai,S. ; Nakamura, H. ; Tanabe, M. ; Sakamoto, K. Distribution and Dissociation Equilibria of Phase-Transfer Catalyst Tetrabutylammonium Salta. Znd. Eng. Chem. Res. 1993,32,143&1441. A d ,S. ; Nakamura, H. ; Sumita,T. Oxidation of Benzyl Alcohol Using Hypochlorite Ion via Phase-Transfer Catalysis. AIChE J. 1994, in press. Dehmlow, E. V. ; Dehmlow, S. 5.
Phase Transfer Catalysis; Verlag Chemie: Weinheim, 1983. Hikita, H. ; Asai,S. ; Ishikawa, H. ; Seko, M. ; Kitajima,H. Diffusivities of CarbonDioxide in Aqueous Mixed Electrolyte Solutions. Chem. Eng. J. 1979,17,77-80. Krishnakumar, V. K. ; Sharma, M. M. Kinetics of Formation of Triphenyl Phosphate: Phase-Transfer Catalysis in a LiquidLiquid System. Znd. Eng. Chem. Process Des. Dev. 198~424,12931297. b = bulk i = interface j = species J’ obs = observed w = water 0 = initial 1 = Q+CsHaO2 = Q+C1- Scott,G. V. SpectrophotometricDeterminationof Traces of Cationic 1 Surfactants with Orange 1 .
A n d . Chem. 1968,40,768-773. Starks,C. M. ; Liotta, C. Phase Transfer Catalysis; Academic Press: New York, 1978. Vinograd, J. €2. ; McBain, J. W. Diffusion of Electrolytes and of the Ions in Their Mixtures. J. Am. Chem. SOC. 1941,63,200&2015. Weber, W. P. ; Gokel, G. W. Phase Transfer Catalysis in Organic Synthesis; Springer-Verlag: Berlin, 1977. Receiued for reuiew June 11, 1993 Revised manuscript received December 27, 1993 Accepted January 31, 1994. Abstract published in Advance ACS Abstracts, March 15, 1994. 3 = Q+Br 4 = Q+OH5 = C&50-