Penultimate effect in the initiation of the cationic polymerization of isobutylene: Kinetic study of initiation with (3-chloro-1,1,3-trimethylbutyl)benzene

Article abstract of DOI:10.1021/ma020189u

The rate constant of ionization of Cum-IB₁-Cl was measured and compared with that of 2-chloro-2,4,4-trimethylpentane (TMPCl). Replacement of a methyl group in TMPCl with a phenyl group in Cum-IB₁-Cl resulted in overall reactivity of

Full text of DOI:10.1021/ma020189u

5320
Macromolecules 2002, 35, 5320-5322
It was the goal of this study to measure the rate
constant of ionization (ki) of Cum-IB1-Cl and compare
it to that of 2-chloro-2,4,4-trimethylpentane (TMPCl)
and to determine the rate constant of deactivation
(reversible termination) (k-i) and the rate constant of
addition (kc) of a reactive π-nucleophile (2-phenylfuran,
P en u ltim a te Effect in th e In itia tion of th e
Ca tion ic P olym er iza tion of Isobu tylen e:
Kin etic Stu d y of In itia tion w ith
(3-Ch lor o-1,1,3-tr im eth ylbu tyl)ben zen e†
Moon Su k Kim a n d Ru d olf F a u st*
2-PhFu) for the corresponding Cum-IB1 and TMP+
+
Polymer Science Program, Department of Chemistry,
University of Massachusetts Lowell, One University Avenue,
Lowell, Massachusetts 01854
cations. The results were expected to identify the
underlying reason(s) for the low overall reactivity of
Cum-IB1-Cl and the similar -St-IB-Cl end struc-
ture.
Received February 5, 2002
Revised Manuscript Received April 24, 2002
Exp er im en ta l Section
In tr od u ction
Ma ter ia ls. Syn th esis of (3-Ch lor o-1,1,3-tr im eth ylbu t-
yl)ben zen e (Cu m -IB1-Cl). Cum-IB1-Cl was prepared by
a slightly different procedure than that already reported by
Mayr and co-workers.5 Under a nitrogen atmosphere in a
glovebox hexanes (Hex), methyl chloride (MeCl), cumyl chlo-
ride (0.02 M), DTBP (0.003 M), and IB (1.03 M) were added
sequentially to a prechilled 400 mL round-bottom flask at -80
°C. BCl3 (0.06 M) dissolved in Hex/MeCl 60/40 (v/v) was added
last. After 40 min, the reaction was terminated with prechilled
methanol. The resulting solution was poured into NH4OH/
methanol to neutralize the reaction mixture. After evaporation
of the solvent, the residue was dissolved in CH2Cl2, and the
resulted solution was washed with water several times and
with 1 M hydrochloric acid solution. The organic phase was
dried over anhydrous magnesium sulfate and concentrated by
evaporation followed by drying in a vacuum to afford quantita-
tive yield of a colorless liquid.
The effect of a γ-substituent on the reactivity of the
active center is well established in radical polymeriza-
tion. That the reactivity of the propagating species may
be affected by the next to the last or penultimate
monomer unit is commonly referred to as penultimate
behavior and necessitates the use of a modified copoly-
mer composition equation with four reactivity ratios.1
The effect of the penultimate monomer structure on the
overall reactivity of the chain end in cationic polymer-
ization has recently been demonstrated.2 Compared to
conventional radical copolymerization, living cationic
copolymerization is complicated by the dynamic equi-
librium between dormant and active centers. We have
reported that the overall reactivity of ∼CH2CH(Ph)-
CH2C(CH3)2-Cl (-St-IB-Cl) end structure is much
lower than that of ∼CH2C(CH3)2-CH2C(CH3)2-Cl (-IB-
IB-Cl).2 Because of the low reactivity of -St-IB-Cl
ends, which arise in the living cationic synthesis of poly-
(styrene-b-isobutylene) when IB is added at less than
complete St conversion, coupling of the living diblock
ends is not the preferred technique for the synthesis of
poly(styrene-b-isobutylene-b-styrene) triblock copoly-
mers.2 The preferred simple and successful method
involves difunctional initiation and sequential monomer
addition.3 This method, however, is successful only when
St is added after the complete polymerization of IB.
When St is added at less than 100% IB conversion, the
polymerization of St is much slower,3b which again is
due to the formation and low reactivity of -St-IB-Cl
chain ends. For instance, when St is added after
complete polymerization of IB, St polymerization is
complete in 1 h. In contrast, when St is added at 94%
IB conversion, St conversion reaches only ∼50% in 1 h
at otherwise identical conditions.4
1H NMR: δ 7.30-7.07 (m, 5H, -C6H5), 2.30 (s, 2H, -CH2-),
1.35 (s, 6H, -2(CH3)), 1.21 (s, 6H, -2(CH3)). 13C NMR: δ 148.7,
128.0, 125.9, 125.6, 71.7, 58.1, 38.4, 34.2, 30.9.
Sources and purity of all other materials have been previ-
ously described.6
Gen er a l Rea ction P r oced u r e. All reactions were carried
out at -80 °C under a dry nitrogen atmosphere using [Cum-
IB1-Cl] ) 2.0 × 10-3 mol L-1, [TiCl4] ) 1.6 × 10-2 mol L-1
,
and [2,6-di-tert-butylpyridine, DTBP] ) 3.0 × 10-3 mol L-1
(unless otherwise noted). The solution of Cum-IB1-Cl, TiCl4,
and DTBP in Hex/MeCl 60/40 (v/v) was kept for ∼1 h to allow
complex inorganic salts to precipitate. Then, the solution was
filtered, and the reference spectrum was taken. Finally, the
π-nucleophile was added, and the accumulation of the UV-
vis spectra started.
UV-vis Sp ectr oscop y. For the UV-vis spectroscopic
measurements a quartz immersion probe 661.300-QX (Hellma,
optical path 0.02 cm) connected to a fiber-optic visible (tung-
sten light source, Ocean Optics) and UV (AIS model UV-2,
Analytical Instrument Systems, Inc.) light source and a Zeiss
MMS 256 photodiode array detector was used. The latter was
connected to a personal computer via a TEC5 interface, and
the spectra were recorded using the “Aspect Plus” software
(Zeiss).
Deter m in a tion of th e Ra te a n d Equ ilibr iu m Con sta n ts
via UV-vis Sp ectr oscop y. For the calculation of the appar-
ent rate constant of the capping reaction, the concentration of
5-(Cum-IB1)-2-PhFu+ carbenium ions was derived from the
measured absorbance at the absorption maximum λmax ) 340
A similar γ-substituent effect on the overall cationic
reactivity has already been reported by Mayr et al.5 The
rate of IB addition to (CH3)2C(Ph)CH2C(CH3)2Cl (Cum-
IB1-Cl) was found to be 20-35 times lower than the
rate of IB addition to Cum-IB2-Cl, which changed little
with further increase of n in Cum-IBn-Cl. This may
be ascribed to intramolecular solvation of the cationic
center and/or to the negative inductive effect of the
phenyl ring. In other words, the lower overall reactivity
nm and the corresponding molar adsorption coefficient ꢀmax
)
30 000. A more detailed description of the procedure has been
given.6
+
may be due to a lower reactivity of Cum-IB1 cations
Ch a r a cter iza tion . 1H NMR spectra were recorded on
Bruker 250 MHz spectrometer using CDCl3 as a solvent
(Cambridge Isotope Laboratories, Inc.).
compared to Cum-IBn+ cations, or it may be attributed
to a lower equilibrium constant of ionization of Cum-
IB1-Cl.
Resu lts a n d Discu ssion
* Corresponding author.
† Dedicated to Prof. J .P. Kennedy on the occasion of his 75th
birthday.
Kin etics of th e Ca p p in g Rea ction w ith 2-P h F u .
We have previously reported on the addition reactions
10.1021/ma020189u CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/24/2002

Macromolecules, Vol. 35, No. 13, 2002
Sch em e 1. Ion iza tion Equ ilibr iu m of Cu m -IB1-Cl (Ki ) ki/k-i) a n d Ca p p in g w ith 2-P h F u
Notes 5321
of non-(homo)polymerizable π-nucleophiles (π-Nu) 1,1-
bis(4-methylphenyl)ethylene (DTE), 1,1-bis(4-tert-but-
ylphenyl)ethylene, and 2-PhFu to TMPCl and other
hydrochlorinated isobutylene n-mers (n ) 3, 36) in the
presence of TiCl4 in hexanes (Hex)/methyl chloride
(MeCl) 60/40 (v/v) at -80 °C using on-line UV-vis
spectroscopy.6 The apparent rate constants of capping,
kcKi (where kc is the rate constant of capping and Ki is
the equilibrium constant of ionization), and the rate
constant of ionization, ki, have been determined using
eq 1.
[PIBCl][TiCl4]2
)
d
dt
([PIB-π-Nu + Ti2Cl9-] + [PIB-πNu+])
F igu r e 1. Reciprocal initial rate of capping vs 1/[π-Nu] (9,
2-PhFu; O, DTE; 4, DBE) for the capping reaction of Cum-
IB1-Cl and TMPCl in Hex/MeCl 60/40 (v/v) at -80 °C.
1
1
ki
+
(1)
kcKi[π-Nu]
Ta ble 1. Kin etic a n d Th er m od yn a m ic P a r a m eter s for th e
Ca p p in g Rea ction of P IBCl, TMP Cl, a n d Cu m -IB1-Cl
w ith 2-P h F u a
The denominator on the left side of eq 1 contains the
sum of the concentrations of capped ion pairs and free
ions. The identical kcKi values obtained with all three
π-nucleophiles suggested diffusion-limited addition, and
from the known value of diffusion-controlled second-
order rate constant kc ) 3 × 109 L mol-1 s-1, Ki and k-i
were calculated. We used the same methodology in the
present research to determine ki, Ki, and k-i for Cum-
IB1-Cl. Scheme 1 describes the capping of Cum-IB1-
Cl with 2-PhFu as a reactive π-nucleophile.
Capping is a two-step process that involves the
ionization of the chain end and subsequent addition of
2-PhFu. 2-PhFu was chosen as a reactive π-nucleophile
because the nucleophilicity parameter N, which can be
used to predict kc based on the free energy relationship7
(eq 2, slope parameter, s = 1), is known for 2-PhFu
(N ) 3.6)8 but remains to be determined for 1,1-
diarylethylenes.
kcKi (L3
ki (L2
Ki (10-7 L2 k-i (10-7
initiator
PIBCl
TMPCl
Cum-IB1-Cl
mol-3 s-1
)
mol-2 s-1
)
mol-2
)
s-1
)
1430
400
47
16.4
6.5
1.4
4.8
1.3
0.16
3.4
5.0
8.7
a
Ki and k-i were calculated using kc ) 3 × 109 L mol-1 s-1
.
The kcKi and ki values determined from Figure 1 and
the calculated Ki and k-i values are shown in Table 1.
In addition to values for Cum-IB1-Cl, values reported
for PIBCl and TMPCl6 are also included.
Capping of Cum-IB1-Cl is 9 times slower than
capping of TMPCl and 30 times slower than that of
PIBCl. Since capping is diffusion-limited for all three,
the difference is due to a difference in Ki. This in turn
can be attributed mainly to a large difference in ki
values; i.e., the rate constant of ionization of Cum-IB1-
Cl is about 5 and 12 times lower than that of TMPCl
and PIBCl, respectively. Interestingly, for Cum-IB1-
Cl the rate constant of deactivation k-i is only 75%
higher than that of TMPCl. This in turn suggests that
the low overall reactivity of Cum-IB1-Cl is due mainly
to the low rate constant of ionization brought about by
the electronic withdrawing effect of the phenyl ring.
While intramolecular stabilization of the formed Cum-
IB1+ by the phenyl ring cannot be ruled out, judged from
log k ) s(N + E)
(2)
Based on N ) 3.6 and the reported electrophilicity
parameter of PIB+ (E ) 7.5),9 kc ∼ 1011 L mol-1 s-1 is
calculated. The reactivity of Cum-IB1-Cl is reportedly
20-35 times lower; however, even if the overall reactiv-
ity decrease is due solely to a decrease in the reactivity
of Cum-IB1+, kc ∼ 4 × 109 L mol-1 s-1 is calculated;
i.e., diffusion-limited addition of 2-PhFu is predicted.
Thus, Ki and k-i can be calculated from kcKi values using
+
the k-i values, the reactivity of Cum-IB1 is actually
somewhat higher than the reactivity of the TMP+ cation.
This relatively small reactivity difference, however,
cannot compensate for the large decrease in the ki value,
and the overall effect is that Cum-IB1-Cl is much less
reactive than TMPCl or PIBCl.
kc ) 3 × 109 L mol-1 s-1
.
The reciprocal initial rate of capping vs 1/[2-PhFu]
for the capping reaction of Cum-IB1-Cl in Hex/MeCl
60/40 (v/v) at -80 °C is shown in Figure 1. For
comparison, the plot obtained with TMPCl and reported
in ref 6 is also included.
Kin et ics of t h e Ca p p in g R ea ct ion w it h 1,1-
+
Dip h en yleth ylen e (DP E). The reactivity of Cum-IB1

5322 Notes
Macromolecules, Vol. 35, No. 13, 2002
slightly higher than that of TMP+. The penultimate
effect found in the copolymerization of IB and St may
be explained similarly. In this case, however, in addition
to the negative inductive effect of the phenyl ring, a
lower back strain (release of steric strain upon ioniza-
tion) in -St-IB-Cl compared to -IB-IB-Cl may
further decrease the rate end equilibrium constant of
ionization.
Ta ble 2. Ra te Con sta n ts of Ca p p in g for th e Ca p p in g
Rea ction of TMP Cl a n d Cu m -IB1-Cl w ith DP E
initiator
TMPCl
Cum-IB1-Cl
kcKi (L3 mol-3 s-1
)
kc (L mol-1 s-1
)
24
4.7
1.9 × 108
2.9 × 108
and TMP+ cations can also be compared in capping
reactions using a non-(homo)polymerizable π-nucleo-
phile, which adds slower than the diffusion limit. We
have recently reported that capping of IB n-mers is
approximately 15 times slower with DPE than with
DTE or 2-PhFu, which was ascribed to a lower nucleo-
philicity of DPE compared to that of DTE and 2-PhFu.10
Therefore, we have determined the kcKi value in the
capping reaction of Cum-IB1-Cl with DPE and com-
pared it to values reported for TMPCl in ref 10. Using
the Ki values listed in Table 1, we have calculated the
kc values, which are listed in Table 2. The rate constants
of capping are not substantially different, indicating
that the reactivities of Cum-IB1+ and TMP+ cations are
similar.
Ack n ow led gm en t. Partial support from the NSF
(CHE-0131106) is gratefully acknowledged. We also
thank Prof. Herbert Mayr (Ludwig-Maximilians-Uni-
versity, Munchen, Germany) for valuable comments.
Refer en ces a n d Notes
(1) Odian, G. Principles of Polymerization; J ohn Wiley and
Sons: New York, 1991; p 503.
(2) Cao, X.; Faust, R. Macromolecules 1999, 32, 5487.
(3) (a) Kaszas, G.; Puskas, J .; Kennedy, J . P.; Hager, W. G. J .
Polym. Sci., Polym. Chem. Ed. 1991, 29 (1), 427. (b) Storey,
R. F.; Chisholm, B. J . Macromolecules 1993, 26, 6727. (c)
Gyor, M.; Fodor, Zs.; Wang, H.-C.; Faust, R. J . Macromol.
Sci. 1994, A31 (12), 2053.
(4) Kim, M. S.; Faust, R., unpublished results.
(5) Mayr, H.; Roth, M.; Deters, M. Macromolecules 1997, 30,
3965.
Con clu sion
(6) Schlaad, H.; Kwon, Y.; Sipos, L.; Faust, R.; Charleux, B.
Macromolecules 2000, 33, 8225.
The nature of a γ-substituent may have a large effect
on the rate of initiation in cationic polymerization. As
initiation is a two-step process in cationic polymeriza-
tion, ionization and cationation may be influenced to a
different degree. Replacement of a methyl group in
TMPCl with a phenyl group in Cum-IB1-Cl results in
an overall reactivity of Cum-IB1-Cl that is close to an
order of magnitude lower than that of TMPCl in cationic
initiation. This can be explained by the much lower
ionization rate and equilibrium constant of Cum-IB1-
Cl, attributed to the negative inductive effect of the
(7) Equation 2 is valid only for k < 107-108 L mol-1 s-1; at k )
108-5 × 109 L mol-1 s-1 the real rate constants are
somewhat smaller than the calculated ones. Mayr, H.; Bug,
T.; Gotta, M. F.; Hering, N.; Irrgang, B.; J anker, B.; Kempf,
B.; Loos, R.; Ofial, A. R.; Remennikov, G.; Schimmel, H. J .
Am. Chem. Soc. 2001, 123, 9500.
(8) Herrlich, M.; Mayr, H., unpublished results.
(9) Roth, M.; Mayr, H. Angew. Chem., Int. Ed. Engl. 1995, 34,
2250.
(10) Schlaad, H.; Kwon, Y.; Faust, R.; Mayr, H. Macromolecules
2000, 33, 743.
+
phenyl ring, while the reactivity of Cum-IB1 is only
MA020189U