65 - Makromol Chem Rap Comm 12 1991
65 - Makromol Chem Rap Comm 12 1991
65 - Makromol Chem Rap Comm 12 1991
Introduction
The advent of acyclic diene metathesis (ADMET) polymerization as a viable
synthetic route to high-molecular-weight linear polymers -), and copolymers 3, has
provoked a unique and currently important examination of metathesis chemistry.
Whereas ring-opening metathesis polymerization4) (ROMP) is a chain growth,
addition type polymerization driven by the alleviation of ring strain, acyclic diene
metathesis is a step growth, equilibrium condensation type polymerization and thus
provides the opportunity to shift the equilibrium between monomer and polymer,
depending upon reaction conditions.
We have found that a highly active Lewis acid free tungsten alkylidene catalyst
of the type ((CF,),CH,CO),(NAr)-W =CHC(CH,), ,in combination with ethylene,
depolymerizes unsaturated polymers, yielding low-molecular-weight oligomers and
a,malkadienes. In this paper we report the facile reversibility of acyclic diene
metathesis (ADMET) polymerization and its potential applicability to other polymer
systems.
By driving this equilibrium, condensation type polymerization toward monomer
with an excess of ethylene, polymers synthesized by ADMET and commercially
available unsaturated elastomers can be depolymerized successfully as depicted in
Scheme I.
Scheme 1:
Experimental part
Poly(1-octenylene) (polyoctenamer, weight-average molecular weight MW = 108000 by GPC)
was prepared by the ADMET polymerization of 1,9-decadiene as described previously ').
Polybutadiene (M,= 200000, ratio of weight-to number-average molecular weights, Mw/a,, =
2,1, 55% trans, 42% cis, 8% vinyl) and poly-trans-isoprene (M,= 2,5. lo5, GPCJersus poly-
styrene standards) were purchased from Polysciences Inc. The KratonTM*)sample (M, = 86000,
17% poIystyrene block, M, = 14600,40% 1,2-linkagesin polybutadiene block) was generously
donated by SHELL. Polynorhornene was prepared with the same catalyst as that used for depoly-
merization according to published methods '). Ethylene (Polymer Grade 99,9%) was obtained
from Matheson Gas Products. The catalyst ((CF3),CH3CO),(NAr)-W=CHC(CH3],) was
prepared by published methods6). In general, a mole ratio repeating unitkatalyst of 20: 1 was
used unless otherwise stated. The polymer samples were purified prior to use by several
precipitations with different solvents in methanol and dried under high vacuum for several days.
All solvents were distilled from potassium and stirred over potassium-mirrors.
Depolymerization:The polyoctenamer experiment was carried out by adding the polymer to a
custom ADMET reaction vessel equipped with magnetic stirrer, and a breakseal ampule
containing catalyst in the repeating unitkatalyst mole ratio of 43 : 1. Toluene was introduced to
dissolve the polymer, and the solution was subjected to four freezekhaw vacuum cycles. The
catalyst was added, followed by the addition of excess ethylene with a pressure of 3 3 atm. Depoly-
merization was carried out for 48 h at room temperature, after which the volatile reaction products
were vacuum-transferred and analyzed by GUMS.
The reactions at atmospheric pressure were performed in a 100 mL reaction flask equipped with
a gas bubbler and breakseal containing catalyst. The flask was flame-dried under high vacuum,
and the polymer was added under an argon flow after which a high vacuum mmHg) was
applied for several hours. Toluene was vacuum-transferred in the flask after which the catalyst was
washed down. Ethylene was bubbled through the solution at a slow consistent rate. In the case of
the depolymerization of polynorbornene at atmospheric pressure, the polynorbornene sample
was made in situ (under argon) after which the sample was exposed to an additional amount of
catalyst and ethylene.
The high pressure reactions were performed in a thickwall glass pressure bottle fitted with a
metalcap with valve. The reaction mixture was prepared and added under argon after which the
vessel was closed and pressurized to 4-6 atm. After several days, the reaction was stopped by
exposing the mixture to air and separating the volatiles from the non-volatile oligomers by vacuum
transfer.
Characterization: ‘H NMR 200 MHz and I3C NMR 50 MHz spectra were obtained with a
Varian XLSeries NMR superconducting spectrometer system. Size-exclusion chromatography
data were obtained using a Waters Associates liquid chromatograph apparatus equipped with a
RI detector and Phenogel columns (polybutadiene standards were used for the polybutadiene
samples, in the other cases polystyrene standards were used). GC/MS data were obtained with a
Finnegan 4500 gas chromatograph/mass spectrometer. Identification of products were confirmed
by comparison with an authentic sample. GC data were obtained with a Hewlett Packard 5880a
gas chromatograph equipped with a HP 5 % phenylmethylsilicone column.
solves these degradation problems, since the forward and reverse mechanisms of
ADMET mutually enjoy a clean and well-defined path and the chemistry is void of any
competing reactions. Under these conditions, the reverse reaction of acyclic diene
metathesis is depolymerization, not degradation.
Polymers we have depolymerized by this method include polyoctenamer, poly-
norbornene and three commercially obtained elastomers: polybutadiene, polyisoprene
and a KratonTMsample (polystyrene/polybutadiene diblock copolymer). KratonTM
was studied with the intent of producing functionalized prepolymers from
commercially available thermoplastic elastomers, since only the unsaturated blocks
would be affected. The depolymerization results are summarized in Tab. 1.
Scheme 2:
A large decrease in molecular weight was observed (Tab. 1) and 1,9-decadiene was
detected by G U M S . The oligomers exhibit an increase in average trans-olefin content
from 88% to 96% as measured by quantitative I3C NMR. This increase probably
originates from the known higher reactivity of cis-olefins or a polymer-polymer
interchange4) (transmetathesis). The I3C NMR data of the starting material and
unpurified oligomers obtained exemplifies the very clean depolymerization pathway.
The 13CNMR spectra are shown in Fig. 1.
I Ifcis)
(bl
1,
__r_
1LO
6 in ppm
120 100 80 60 LO
2'
-
20 0
Fig. 1. '3C NMR of (a) polyoctenamer starting material and (b) octenamer oligomers after
depolymerization
Scheme 3:
Catalyst
+ oligomers
Fig. 2 shows the gel-permeation trace of several oligomeric products. The end groups
of these oligomers are also evident in the 'H and I3C NMR spectra of the reaction
mixture. The complexity of the product mixture arises primarily from the presence of
1,2-addition linkages in the starting material.
A C
Fig. 2. Gel-permeation
chromatography (GPC) trace of
starting material and residue
from the depolymerization of
polybutadiene. Peak A: Starting
material (M,= 200000,
a,/&& = 2,i); peak B: M, =
2000, peak C: catalyst; peak D
a, = 60 (molecular weights
relative to polybutadiene
standards)
Elution volume in mL
Scheme 4:
wx + excess H,C=CH,
Catalyst
+ oligomers
This reaction is only successful when carried out under an increased ethylene
pressure of 5 - 6 atm and at 50 "C. These more stringent conditions are required due
to the reduced reactivity of substituted olefins toward metathesis". Apparently, the
equilibrium is sufficiently shifted to depolymerization by elevated temperature and
pressure, whereas concurrent results in our labs indicate that ADMET polymerization
(the forward reaction) of 1,l-disubstituted olefins does not occur ll).
In order to prove once more that the depolymerization reaction proceeds along a
well-defined and clean pathway, a sample of squalene was depolymerized under the
same conditions as the polyisoprene depolymerization. This reaction is shown in
Scheme 5.
After the reaction, 2-methylpropene, 1,5-hexadiene, and 2-methy1-1,Shexadiene
were identified by GC and/or G U M S (Tab. 1). The 13CNMR spectrum of the non-
volatile residue shows only signals from the starting material and unsubstituted and
methyl-substituted vinyl end groups. No loss of unsaturation was observed.
424 K. B. Wagener, R. D. Puts, D. W. Smith, Jr.
Scheme 5:
Scheme 6:
v C6H5
+ excess H,C=CH, Catalyst
+
C6H5
+ M + oligomers
A B
1;
, /
I / Fig. 3. Gel-permeation
chromatography (GPC) trace of starting
I material and residue from the depoly-
I merization of KratonTM.Peak A:
starting materg (a, = 86000 17%
PS); peak B: M, = 15400, a,/@
1,Ol; peak C: a,= 1 OOO; peak D: M,
= 80 (molecular weights relative to
polystyrene standards)
1 1 1 ’ 1 . ’ 1 1 ’
5 10 15 20
Elution volume in mL
Scheme 7:
Catalyst
Catalyst CJ-4
b low-molecular-weight polymer
Acyclic diene metathesis depolymerization of elastomers 425
A significant decrease in the polymer’s molecular weight is observed (lkb. 1). The
structure of the polymeric product is the same as the starting material according to
NMR spectroscopy. We have not yet been able to show with GC/MS the presence of
the expected monomer, 1,3-divinylcyclopentane.
Without question, acyclic diene metathesis polymerization (ADMET) is a true equi-
librium process. Products at any extent of conversion can be isolated and subjected to
further polymerization, or depolymerization, by shifting the concentration of the
condensed olefin. This process is further applicable to a variety of unsaturated
polymers, and this result constitutes a meaningful opportunity to generate useful
chemicals from waste polymers. The synthesis of vinyl functionalized prepolymeric
products and pure dienes represent but two examples of numerous possibilities
resulting from ADMET depolymerization, which may offer a unique incentive to
recycling. Further research will concentrate on quantitation of the depolymerization
reaction and the tolerance of this catalyst toward impurities and functionalities, as well
as investigations involving new catalyst systems.
We thank the National Science Foundation (DMR-8912026) for their support of this work. We
also thank Dr. .
IG. Nel for his pioneering efforts in this area, the SHELL Development Company
for providing the KratonTM sample, Dr. J M. Boncella and Dr. A. S. Gamble for catalyst
synthesis, the University Research Initiative Program (Grant +N00014-864-0224) as
conducted by DARPA and the Naval Research Office.