Ind. Eng. Chem. Res. 2003, 42, 5437-5439
5437
Cationic Reactivity of Olefins Present in the C5 Fraction Giuseppe Gozzelino,* Aldo Priola, and Marco Sangermano Dipartimento di Scienza dei Materiali e Ingegneria Chimica, Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Torino, Italy
A study of the cationic reactivity of olefins present in the C5 fraction has been performed under conditions similar to those used for synthesis of hydrocarbon resins. The most important olefins present in the C5 fraction, viz. 2-methyl-1-butene, 2-methyl-2-butene, 1-pentene, cyclopentene, cis-2-pentene, trans-2-pentene, and 3-methyl-1-butene, were considered. Their relative reactivities were evaluated by using cyclopentene as a standard reference monomer in a series of copolymerizations at 40 °C in cyclohexane solution in the presence of AlCl3-HCl-xylene complex as catalyst. Owing to the very low molecular weight values of the resulting oligomers, the systems were considered as copolymerizations in which transfer to monomer is operative. A copolymerization equation including monomer transfer was proposed from which the R parameters, related to the propagation and chain-transfer reaction, were obtained. The equation, by using the KelenTu¨do¨s treatment, fits very well the experimental results, and the R values were obtained for the various pair of monomers. The system show almost ideal behavior, and a reactivity order for the olefins was obtained. Introduction The use of the C5 hydrocarbon fraction for producing hydrocarbon resins is an important industrial application of cationic catalysts.1 The C5 fraction produced in the steam cracking of naphthas contains about 50% of olefins and dienes which can react, to different extents, in the polymerization process and allow the production of hydrocarbon resins having different properties. Despite the existence of many patents on the cationic polymerization of the C5 fraction, no data are available on the relative reactivities of the monomers in the reacting system. As a part of research on the utilization of the C5 fraction in cationic reactions, a systematic study on the relative reactivities of the C5 monomers has been undertaken. We first considered the most important olefins present in the C5 fraction, and then the dienes and the other unsaturated hydrocarbons were taken into account. In this work, a comparison of the cationic reactivities of the C5 olefins with cyclopentene, as a reference comonomer, is reported using experimental conditions similar to those adopted for synthesis of hydrocarbon resins. Experimental Section Reagents. The monomers, 3-methyl-1-butene, 2-methyl-1-butene, 2-methyl-2-butene, cis-2-pentene, trans-2pentene, 1-pentene, and cyclopentene, were Fluka products, pure grade (GC purity >99%). Cyclohexane, n-hexane, aluminum chloride, hydrochloric acid, and xylene were RPE Carlo Erba products, purified as previously reported.2 The catalyst was obtained by bubbling anhydrous hydrochloric acid at atmospheric pressure into a stoichiometric slurry of AlCl3-xylene until a homogeneous solution was obtained. The analysis performed on the * To whom correspondence should be addressed. Tel./fax: +39-011-5644652. E-mail:
[email protected].
obtained catalyst showed that the molar ratio between the components HCl, AlCl3, and xylene was 0.5:1.09: 1.15. Copolymerization. The copolymerizations were carried out in glass microreactors, volume about 15 mL, equipped with a magnetic stirrer and a sealed port for introduction of the reagents or withdrawal of the products by means of a syringe. The monomer mixture was maintained at the selected temperature (40 °C) using a thermostatic bath. For each run, 4 mmol of reagents was used per 10 mL of reaction volume. Copolymerization was started by introducing the catalyst (about 0.05 µmol of AlCl3), and after 5 min water containing 5% NH4OH was added. Each pair of monomers was tested by a series of runs with initial monomer molar ratio in the range between 3:1 and 1:3. To consider the microreactor as a differential one, only results of experiments in which the conversion for both monomers was lower than 10% has been taken into account. All manipulations of monomers and catalyst were carried out under dry N2. Analyses. The compositions of the copolymers were determined by GC analysis of the residual monomers in the sample withdrawn from the reaction mixture; a programmed temperature increase (3 min at 25 °C, 10 °C/min up to 75 °C) on a Varian 3700 instrument equipped with a DB-1 J&W megabore column, a FID detector, and a Varian 4270 integrator was used. The Mn values of the samples were determined by means of a Knauer vapor pressure osmometer equipped with Digital Meter G06. GPC measurements were carried out with a Waters Instruments HPLC by using THF as solvent, Styragel columns, PL50, 100, 500 and 10 000 A, and a refractive index detector. Results and Discussion The reactivities of the olefins were tested in a series of copolymerizations carried to low conversion using cyclopentene as a reference monomer. Cyclopentene was selected because of its abundance and reliable GC determination.
10.1021/ie030026+ CCC: $25.00 © 2003 American Chemical Society Published on Web 09/13/2003
5438 Ind. Eng. Chem. Res., Vol. 42, No. 22, 2003
The composition data could be treated with the classical two-parameter Lewis-Mayo copolymerization, equation but this treatment implies that high-molecular-weight copolymers are formed. In our system, the reaction products have a complex structure due to isomerization and chain-transfer reaction, as has been reported in a previous paper dealing with the cationic polymerization of 3-methyl-1-butene.2 As a product of the copolymerization, only very low molecular weight copolymers were obtained: Mn determinations revealed that the oligomers contain 4-8 structural units, independent of the monomer conversion. If we consider that the concentration of the catalytic system was very low (see Experimental Section), we must conclude that chain-transfer reactions are important. Assuming that chain transfer to monomer is the dominant chain-transfer process, as found for many cationic systems,3 we can treat our system according to these simplified equations which are related to the propagation (k) and chain-transfer reactions (ktr):
Figure 1. η versus ξ plot for the copolymerization of 2-methyl1-butene/cyclopentene (b) and 2-methyl-2-butene/cyclopentene (O).
dM1 + + ) k11M+ 1 M1 + k21M2 M1 + ktr11M1 M1 + dt ktr21M+ 2 M1 dM2 + + ) k12 M+ 1 M2 + k22M2 M2 + ktr12M1 M2 + dt ktr22M+ 2 M2 + + + dM1 M1 k11M+ 1 + k21M2 + ktr11M1 + ktr21M2 ) (1) dM2 M2 k M+ + k M+ + k M+ + k M+ 12
1
22
2
tr12
1
tr22
2
Figure 2. η versus ξ plot for the copolymerization of 1-pentene/ cyclopentene (b) and trans 2-pentene/cyclopentene (O).
By introducing steady-state conditions related to the propagating species and to the chain-transfer process, we obtain an equation similar to the classical copolymerization equation:
dM1 M1 R1(M1/M2) + 1 ) dM2 M2 R2 + (M1/M2)
(2)
in which we define R1 and R2 as the ratios between the rate constants related to the propagation and chaintransfer processes:
R1 )
k11 + ktr11 k12 + ktr12
R2 )
k22 + ktr22 k21 + ktr21
For the evaluation of the parameters of eq 2, despite the better performance of nonlinear fitting methods, the Kelen-Tu¨do¨s data treatment,4 a linear fitting method, has been applied. We consider this simple method sufficient both to obtain accurate values of Ri and to verify the applicability of eq 2 to our system.5 The following equation was used:
(
η ) R1 +
)
R2 R2 ξR R
in which
η)
G R+F
ξ) G)
F R+F
x(y - 1) y
R ) xFminFmax F)
x2 y
Figure 3. η versus ξ plot for the copolymerization of cis-2-pentene/ cyclopentene (b) and 3-methyl-1-butene/cyclopentene (O).
where x and y are the composition of the feed and the copolymer, respectively. Figures 1-3 show the η versus ξ plots for the C5 olefins copolymerized with cyclopentene. In all cases, the experimental data yield straight lines from which the values of R1 and R2 can be obtained (see Table 1). The high correlation coefficients reported in the last column of Table 1 indicate that the two-parameter model is a satisfactory approximation to describe these systems quantitatively. Some GPC measurements, performed on the products, indicate the presence of some badly resolved peaks, probably due to formation of co-oligomers. The R1R2 values are also reported in Table 1: in the case of pentene isomers these values are very near unity, indicating almost ideal copolymerization behavior related to the propagation and monomer chain-transfer reactions.
Ind. Eng. Chem. Res., Vol. 42, No. 22, 2003 5439 Table 1. R1 and R2 Values Obtained in the Copolymerization of C5 Isomers (M1) with Cyclopentene (M2) as Comonomer monomer (M1)
R1a
R2a
R1R2a
correl coeff
2-methyl-1-butene 2-methyl-2-butene 1-pentene cis-2-pentene trans-2-pentene 3-methyl-1-butene
1.34 ( 0.13 0.74 ( 0.05 0.78 ( 0.14 0.89 ( 0.16 0.53 ( 0.10 0.70 ( 0.09
0.27 ( 0.05 0.38 ( 0.06 1.08 ( 0.17 1.20 ( 0.11 1.69 ( 0.13 2.14 ( 0.22
0.36 ( 0.11 0.28 ( 0.06 0.84 ( 0.29 1.07 ( 0.29 0.90 ( 0.24 1.50 ( 0.34
0.993 0.998 0.986 0.996 0.994 0.994
a
The standard error of estimate was evaluated by a linear fitting of the data based on the Kelen-Tu¨do¨s treatment. Table 2. Values of 1/R2 Obtained with the One-Parameter Equation from the Data of Table 1 monomer
1/R2
2-methyl-1-butene 2-methyl-2-butene 1-pentene cis-2-pentene trans-2-pentene 3-methyl-1-butene
2.89 1.79 0.92 0.83 0.57 0.47
The 1/R2 values give an indication of the reactivities of the different olefins toward the reference monomer cation in propagation and chain-transfer reactions. On the basis of the 1/R2 values (Table 2), the following order of cationic reactivity for the C5 olefins investigated was found: 2-methyl-1-butene > 2-methyl-2-butene > cyclopentene > 1-pentene > cis-2-pentene > trans-2pentene > 3-methyl-1-butene. The reactivity order of propagating cationic species should take into account different factors, i.e.: -double bond electron density, i.e., the presence of methyl groups linked to the unsaturated carbon atoms; -steric hindrance, i.e., the presence of substituents on both of the double bond carbon atoms (internal olefins, cis and trans isomers); -ring strain, in the case of cyclopentene; and -monomer energy level, in the case of cis and trans isomers. All these factors can, to different extents, influence the reactivity of an olefin either in propagation or in chain-transfer reactions; a comprehensive review on this subject was reported6 calling attention to the great complexity of this problem.
In our system, all these factors reflect their relative importance influencing the reactivity order toward the cyclopentene monomer. Anyway, at this stage of investigation they cannot be discriminated. The experimental reactivity order found is clearly valid only in the specific set of conditions adopted (counteranion, temperature, reference monomer); further work will consider the influence of changing these conditions in order to obtain more general conclusions. Conclusions Cationic copolymerizations of olefins with cyclopentene were carried out assuming the propagation and chain transfer to monomer as dominant processes in the monomer consumption. A simplified Lewis-Mayo equation was used to treat the experimental data, and a Kelen-Tu¨do¨s statistical treatment gave accurate reactivity ratio values (R). An almost ideal copolymerization is observed with R1R2 values close to unity. Although the order of cationic reactivity toward the cyclopentene reference monomer could be easily evaluated by the proposed copolymerization method, the factors that influence the observed order cannot be properly established on the basis of the molecular structure of the monomers. Literature Cited (1) Kredenburgh, W.; Faley, K. F.; Scarlatti, A. N. Encyclopedia of Polymer Science and Engineering, 2nd ed.; Wiley Publ.: New York, 1987; Vol. 7, p 761. (2) Priola, A.; Gozzelino, G.; Ferrero, F. Cationic Oligomerization of 3-Methyl-1-Butene Catalyzed by BF3-Protonic Donor Complexes. Polym. Bull. 1985, 13, 245. (3) Kennedy, J. P.; Marechal, E. Carbocationic Polymerization; Interscience Publ.: New York, 1982; p 206. (4) Kelen, T.; Tu¨do¨s, F. Analysis of the Linear Methods for Determining Copolymerization Reactivity Ratios. J. Macromol. Sci., Chem. 1975, A9, 1. (5) Kennedy, J. P.; Kelen, T.; Tu¨do¨s F. Critical Reexamination of Cationic Monomer Reactivity Ratios. J. Polym. Sci. 1975, 13, 2277. (6) Reference 3, p 374.
Received for review January 13, 2003 Revised manuscript received June 4, 2003 Accepted July 23, 2003 IE030026+
