Tying Up Loose Ends Some Mechanistic Aspects of Catalytic Chain

Document Sample
Tying Up Loose Ends Some Mechanistic Aspects of Catalytic Chain Powered By Docstoc
					Tying Up Loose Ends: Some Mechanistic Aspects of Catalytic Chain Transfer

R. Vollmerhaus, S. Pierik, A. M. v. Herk
Laboratory of Polymer Chemistry, Eindhoven Polymer Laboratories, Eindhoven University of Technology, P. O. Box 513, 5600 MB Eindhoven, The Netherlands

Mechanism of Chain Transfer
CH2 Pn CH2 CO 2 CH3 H3 CO 2 C growing polymer chain CH3 Co(II) C H2 CO 2 CH3

CH3 C• Pn C H2

transfer product


cobalt H hydride


[M] growing polymer chain

H2 C re-initiation CH3 H3 C Co(III) CO 2 CH3

Me O


Kinetics of Chain Transfer

• •

Both CLD and Mayo methods are used – essentially equivalent but Mayo method much simpler Mayo equation – chain transfer coefficient Ctr = ktr/kp slope of Mayo plot: 1/DPn vs [Co]/[M]

•]2 k t [R k tr [Co(II)][R•] 1 = (1 + λ ) + CM + k p [M][R•] k p [M][R•] DPn
• Last term is dominant in CCT – radical concentration cancels out, so there should be no dependence – confirmed when dodecanethiol is transfer agent

Standard Polymerisation
• MMA purified by distillation, passed through Alumina column • degassed with argon and stored over molecular sieves • Co and AIBN solutions prepared and stored in glovebox • oxygen < 1ppm, •COBF solutions are stable for months inside glovebox

0.004 1/DP M w

Ctr = 3613

0 0.00E+00


8.00E-07 [CTA]/[M



Typical Mayo plot for Co(meso-Ph4-porphyrin), CoP, with MMA at 80 ºC

Effect of impure AIBN

Increasing [AIBN] 0.015-0.087M



2.0E-07 [CTA]/[M]



plots are offset for clarity

•Retardation increases as more AIBN is added •Activity returns at sufficient catalyst loadings •Estimation of impurity is 0.002% •Manufacturing process is origin of impurity, easily removed by recrystallisation

Oxygen Impurities
0.012 0.01 0.008

0.8 ppm O2 3 ppm O2 9 ppm O2 19 ppm O2

Ctr = 38000

0.006 0.004 0.002 0 0.00E+00

Ctr = 21000



1.50E-07 [Co]/[M]




•Ctr is not affected as significantly as expected •1 ppm O2 = 3.1 x 10-5 M >>> [Co] •Even at high oxygen concentration transfer activity is still very high •However, rate of polymerisation is retarded, MW increases slightly •Exposure of Co to O2 was kept to a minimum before initiation

Previous Reports on Temperature Dependence
MAYO plot at 90°C 0.3 y = 18056x + 0.0166



0 0.00E+00


8.00E-06 [S]/[M]



• From examination of chain length dependence • Very high [Co] used to obtain low molecular weights ~300 • Not reasonable to obtain a slope from these data for Ctr O' Driscoll 1989

Ctr for specific # of points
Ctr vs # data points 45000

35000 60C Ctr 70C 80C 90C 25000

15000 1 3 # points 5 7


signifies values used by Heuts in reanalysis 1999, many slopes do not have a good fit for four points • Use of only the first 3 data points (better fit) gives a temperature trend exactly like what we see • From the graph, general trend is that Ctr is inversely related to temperature

No Temperature Dependence?
• In Heuts' work, data is only presented for the temperature set shown below • More data points are required

Mayo Plot of 60°C MMA data from Heuts 1999 0.016 y = 39675 x - 7E-05 Mn Mw y = 33575 x + 0.0006




0 0.00E+00


2.00E-07 [Co]/[M]



Heuts, Davis, Forster 1999

Arrhenius Plot for COBF and MMA
from Heuts and Davis
Arrhenius plot COBF and MMA Heuts 1999


/ Ea • 6 kJ m ol 17 A • 2 x 108 M-1s-1
ln ktr


/ Ea • 24 kJ m ol A • 2 x 1011 M-1s-1
16.6 0.0029 0.003 1/T 0.0031 0.0032

• Straight line fit of all data points gives values almost identical with ours, •similar to values previously obtained by Davis, 1998 • Well below (faster) than values for propagation

COBF and Temperature
[AIBN] = 3x10-2M
46000 Ctr = -351.49T + 60359

17.6 17.4

Ctr Mw 38000

ln k tr 17.2 y = -1702.1x + 22.3 17

30000 35 45 55 temp 65 75 85

16.8 0.0028


0.003 1/T



•transfer process: Ea = 14.2 kJ/mol A = 5.3 x 1010 M-1s-1 •propagation: Ea = 22.34 kJ/mol A = 2.65 x 106 M-1s-1 •Similar to O' Driscoll, Davis, definitely faster than propagation •CCT with COBF reported to be diffusion controlled, propagation is not •Ctr is inversely dependent on temperature since transfer has lower Ea than propagation does

CoP and Temperature

3600 [AIBN]= 3 x 10-2M


y = -2622.3x + 22.746 14.6


3300 35 55 75

14.2 0.0028





Ctr vs Temperature

lnktr vs 1/T

•Ctr essentially invariant with temperature •However, Arrhenius plot for chain transfer is almost linear, •Ea = 21.8 kJ/mol; A = 7.56 x 109 M-1s-1 • Since Ea is similar for propagation, there is little variance of Ctr with temperature

Interpretation of Temperature Trends
Mn vs [Co] for CoP 60000 50000 60°C 80°C

40000 Mn 30000


10000 2.0E-06 6.0E-06 [Co] 1.0E-05 1.4E-05

•Important information is lost if only Ctr is examined •For CoP: MW is significantly lower at 80°C, although Ctr does not change • For COBF: Ctr decreases with T, MW is approximately constant •Ctr trends alone, do not comment on MW due to influence of termination term. MW is important measure for industry

Effect of Initiator Concentration k t [R•] k tr [Co(II)] 1 = (1 + λ ) + CM + DPn k p [M] k p [M]
• Equation predicts that chain transfer constant, ktr/kp, will be independent of initiator concentration • Studied by Heuts, 1999 for dodecanethiol and MMA over large range of [I] • No [I] dependence was seen • transfer agent is not regenerated,instead, incorporated into chain

COBF at Different Initiator Concentrations

ln k tr 17.2

[AIBN] = 2 x 10-3M [AIBN] = 3 x 10-2M
Ea = 8.2 kJ/mol A = 8.5 x 108 Ea = 14.2 kJ/mol A = 5.3 x 1010

16.8 0.0028


0.003 1/T



• ktr values are significantly higher for polymerisations done at lower initiator concentrations • Ctr: values for ktr increase since Ctr increased over lower [AIBN]

CoP at Different Initiator Concentrations


[AIBN] = 2 x 10-3M
ln ktr 14.7

Ea = 17.7 kJ/mol A = 2.0 x 109 Ea = 22.7 kJ/mol A = 1.1 x 1010

[AIBN] = 3 x 10-2M

14.3 0.0029

0.003 1/T



•Same as for COBF, the values of Ctr and (by consequence) ktr increase with decreasing [initiator]. •Unexpected based on kinetics

Kinetic Reason for [Radical] Dependence k t [R•] k tr [Co(II)][R•] 1 = (1 + λ ) + CM + DPn k p [M][R•] k p [M][R•]
CH2 CH3 C• Pn C H2 CO 2 CH3 H3 CO 2 C growing polymer chain CH3 Co(II) transfer product Pn CH2 C H2 CO 2 CH3


cobalt H hydride


[M] growing polymer chain

H2 C re-initiation CH3 H3 C Co(III) CO 2 CH3

Me O


Catalytic Cycle

• • • • • • • • •

Co(II) + R• ---> Co(III)-H + R (olefin ) Co(III)-H + M ---> Co(II) + M•

(1) (2)

Therefore there are at least two steps in this cycle, possibly more The first step is the only step in conventional chain transfer, therefore default RDS In CCT, it is assumed to be RDS, Reaction is accepted to be diffusion controlled, or in DC region Possible that further reactions in the cycle will add to the rate, leading to the dependence on initiator concentration Further reactions can not be first order in [initiator] ==> net dependence of rate of transfer (MW) on initiator concentration

Consequences of [Initiator] Dependence
• • CCT has a further variable to control MW non-conventional polymerisation – conventional polymerisations: molecular weight of polymer is inversely dependent on [initiator] due to bimolecular termination – CCT MW is directly dependent on [initiator], significant dependence Control (important for industry) can be varied by changing temperature and [initiator] – small amounts of initiator produce lower molecular weights, but slower reaction times, high temperature for industrial process can offset the slower times environmentally friendlier – less chain transfer agent since catalytic – less initiator to produce a more effective catalyst





Several factors have been examined to determine effect on CCT
– AIBN trace impurities poison the catalyst – Oxygen, slow reaction with Co? not as important as initially feared


– – – – previous reports are unclear appears that Ctr is inversely dependent on temperature ktr and MW are more effective as gauges needs to be studied at constant [R•]


Radical Concentration
– definite effect by changing radical concentration – lower radical concentration leads to faster transfer – direct dependence of MW on [radical]


Wieb Kingma

Alex van Herk Anton German

Shared By:
Description: Tying Up Loose Ends Some Mechanistic Aspects of Catalytic Chain