Polymerization of acrylates by neutral palladium complexes. Isolation of complexes at the initial steps

Article abstract of DOI:10.1021/om030507t

The polymerization of methyl acrylate by pentafluorophenyl complexes [Pd₂(μ-X)₂(C₆F₅)₂ L₂] (L = tetrahydrothiophene (tht), X = Cl, 2; L = tht, X = Br, 3; L = AsPh₃, X = Br, 4) gives atactic polymers in good yields. Mechanistic studies reveal that the polymerization of methyl acrylate starts by insertion of methyl acrylate in the Pd-aryl bond of the precatalyst to give the alkyl complexes trans-[Pd₂(μ-Cl)₂{CH(CO₂Me) -CH₂C₆F₅}₂ (tht)₂] (5) and trans-[Pd₂-(μ-Cl)₂{CH(C₆F₅) CH₂CO₂Me}₂ (tht)₂] (6). These complexes can be isolated, and the X-ray crystal structure of 5 has been determined. Complexes 5 and 6 decompose mainly by β-H elimination but also by homolytic cleavage of the Pd-C bond in the light. In the presence of methyl acrylate, insertion of MA in hydrido-Pd species produces the alkyl complex trans-[Pd₂(μ-Cl)₂{CH(CO₂Me) CH₃}₂(tht)₂] (9). Then a radical polymerization is initiated by small amounts of the radicals generated from these complexes (5, 6, or 9). Formation of 9 is the regeneration pathway of radicals after a termination reaction has occurred by recombination of the growing radical with palladium and β-H elimination. The success of the polymerization requires a slow but steady supply of radicals by slow decomposition of alkyl complexes (5 and 6) or by slow generation of Pd-H species that provide new alkyl complexes (9), as well as an efficient recycling of the Pd-H generated in the termination step to 9.

Full text of DOI:10.1021/om030507t

4206
Organometallics 2003, 22, 4206-4212
P olym er iza tion of Acr yla tes by Neu tr a l P a lla d iu m
Com p lexes. Isola tion of Com p lexes a t th e In itia l Step s
Ana C. Albe´niz, Pablo Espinet,* and Raquel Lo´pez-Ferna´ndez
Departamento de Qu´ımica Inorga´nica, Facultad de Ciencias, Universidad de Valladolid,
47005 Valladolid, Spain
Received J uly 1, 2003
The polymerization of methyl acrylate by pentafluorophenyl complexes [Pd2(µ-X)2(C6F5)2L2]
(L ) tetrahydrothiophene (tht), X ) Cl, 2; L ) tht, X ) Br, 3; L ) AsPh3, X ) Br, 4) gives
atactic polymers in good yields. Mechanistic studies reveal that the polymerization of methyl
acrylate starts by insertion of methyl acrylate in the Pd-aryl bond of the precatalyst to
give the alkyl complexes trans-[Pd2(µ-Cl)2{CH(CO2Me)-CH2C6F5}2(tht)2] (5) and trans-[Pd2-
(µ-Cl)2{CH(C6F5)CH2CO2Me}2(tht)2] (6). These complexes can be isolated, and the X-ray
crystal structure of 5 has been determined. Complexes 5 and 6 decompose mainly by â-H
elimination but also by homolytic cleavage of the Pd-C bond in the light. In the presence of
methyl acrylate, insertion of MA in hydrido-Pd species produces the alkyl complex trans-
[Pd2(µ-Cl)2{CH(CO2Me)CH3}2(tht)2] (9). Then a radical polymerization is initiated by small
amounts of the radicals generated from these complexes (5, 6, or 9). Formation of 9 is the
regeneration pathway of radicals after a termination reaction has occurred by recombination
of the growing radical with palladium and â-H elimination. The success of the polymerization
requires a slow but steady supply of radicals by slow decomposition of alkyl complexes (5
and 6) or by slow generation of Pd-H species that provide new alkyl complexes (9), as well
as an efficient recycling of the Pd-H generated in the termination step to 9.
In tr od u ction
heteroatoms in the alkenes, because heteroatom coor-
dination competes with double-bond coordination and
prevents polymerization by an insertion route. In the
search for less acidic centers, attention has turned
toward late-transition-metal complexes. Following the
discovery of the high activity of diimine cationic pal-
ladium complexes in the polymerization of ethylene by
Brookhart et al.,6 acrylates were also tried. These
systems do not homopolymerize acrylates but incorpo-
rate them up to 12% in the ramifications of a polyeth-
ylene chain, at the expense of a significant lowering in
the polymer yields.7 Coordination of the carbonyl group
to the metal after insertion of a methyl acrylate unit
There is a great deal of interest in the development
of metal-based catalysts for the polymerization and
copolymerization of polar monomers, which complement
the commonly used radical or anionic processes, par-
ticularly for acrylate derivatives.1,2 Efforts are aimed
at finding systems tolerant to polar functionalities that
perform the polymerization and especially the copolym-
erization of these monomers with nonpolar olefins at
will. Metal complexes have been used to effect living
and better-controlled radical homopolymerizations of
acrylates.3 Some lanthanocenes and group 4 metal-
locenes are active catalysts for the polymerization of
acrylates. They follow mechanisms (enolate and group
transfer) different from the insertion route operating in
the same systems for nonpolar monomers.4 Thus, they
bring about block but not alternating copolymerization
of acrylates and ethylene.5 The hard Lewis acidic
character of most early-transition-metal catalysts makes
them incompatible with the presence of hard donor
(4) (a) Webster, O. W.; Sogah, D. Y. In Comprehensive Polymer
Science; Allen, G., Bevington, J . C., Eds.; Pergamon: New York, 1989;
Vol. 4, p 163. (b) Yasuda, H.; Yamamoto, H.; Yokota, K.; Miyake, S.;
Nakamura, A. J . Am. Chem. Soc. 1992, 114, 4908-4910. (c) Collins,
S.; Ward, D. G. J . Am. Chem. Soc. 1992, 114, 5460-5462. (d) Yasuda,
H.; Yamamoto, H.; Yamashita, M.; Yokota, K.; Nakamura, A.; Miyake,
S.; Kai, Y.; Kanehisa, N. Macromolecules 1993, 26, 7134-7143. (e)
Boffa, L. S.; Novak, B. M. Macromolecules 1994, 27, 6993-6995. (f)
Giardello, M. A.; Yamamoto, Y.; Brard, L.; Marks, T. J . J . Am. Chem.
Soc. 1995, 117, 3276-3277. (g) Deng, H.; Shiono, T.; Soga, K.
Macromolecules 1995, 28, 3067-3073. (h) Li, Y.; Ward, D. G.; Reddy,
S. S.; Collins, S. Macromolecules 1997, 30, 1875-1883. (i) Chen, E.
Y.-X.; Metz, M. V.; Li, L.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc.
1998, 120, 6287-6305. (j) Nguyen, H.; J arvis, A. P.; Lesley, M. J . G.;
Kelly, W. M.; Reddy, S. S.; Taylor, N. J .; Collins, S. Macromolecules
2000, 33, 1508-1510. (k) Cameron, P. A.; Gibson, V. C.; Graham, A.
J . Macromolecules 2000, 33, 4329-4335. (l) Frauenrath, H.; Keul, H.;
Ho¨cker, H. Macromolecules 2001, 34, 14-19.
(5) Yasuda, H.; Furo, M.; Yamamoto, H.; Nakamura, A.; Miyake,
S.; Kibino, N. Macromolecules 1992, 25, 5115-5116.
(6) J ohnson, L. K.; Killian, C. M.; Brookhart, M.J . Am. Chem. Soc.
1995, 117, 6414-6415.
(7) (a) J ohnson, L. K.; Mecking, S.; Brookhart, M. J . Am. Chem. Soc.
1996, 118, 267-268. (b) Mecking, S.; J ohnson, L. K.; Wang, L.;
Brookhart, M. J . Am. Chem. Soc. 1998, 120, 888-899. (c) Gottfried,
A. C.; Brookhart, M. Macromolecules 2003, 36, 3085-3100.
* To whom correspondence should be addressed. Fax: 34-983423013.
E-mail: espinet@qi.uva.es.
(1) (a) Odian, G. Principles of Polymerization; Wiley: New York,
1991; p 630. (b) Handbook of Polymer Synthesis; Kricheldorf, H. R.,
Ed.; Marcel Decker: New York, 1992; Chapter 4.
(2) Polymerization of polar monomers using transition metals:
Boffa, L. S.; Novak, B. M. Chem. Rev. 2000, 100, 1479-1493.
(3) (a) Wayland, B. B.; Poszmik, G.; Mukerjee, S. L. J . Am. Chem.
Soc. 1994, 116, 7943-7944. (b) Nishikawa, T.; Ando, T.; Kamigaito,
M.; Sawamoto, M. Macromolecules 1997, 30, 2244-2248. (c) Matyjas-
zewski, K.; Shipp, D. A.; Wang, J .-L.; Grimaud, T.; Patten, T. E.
Macromolecules 1998, 31, 6836-6840. (d) Matyjaszewski, K.; Xia, J .
Chem. Rev. 2001, 101, 2921-2990. (e) Kamigaito, M.; Ando, T.;
Sawamoto, M. Chem. Rev. 2001, 101, 3689-3746. (f) Harrisson, S.;
Rourke, J . P.; Haddleton, D. M. Chem. Commun. 2002, 1470-1471.
10.1021/om030507t CCC: $25.00 © 2003 American Chemical Society
Publication on Web 09/12/2003

Polymerization of Acrylates by Neutral Pd Complexes
Organometallics, Vol. 22, No. 21, 2003 4207
Ch a r t 1
Sch em e 1
Ta ble 1. P olym er iza tion of Acr yla tes w ith
Com p lexes 2-4a
initiator
(amt,
mmol)
monomer
(amt,
mmol)
poly(MA)
solvent (% yield)
Mw/
Mn
b
b
entry
Mw
1
2
3
4
5
6
7
8
2 (0.0014)c MA (5.6)
89
3.2 × 106 6.2
1.9 × 105 2.2
1.8 × 105 2.2
2.5 × 105 2.4
1.9 × 105 3.1
3.0 × 105 2.1
5.1 × 105 4.8
1.8 × 105 2.9
2 (0.0028) MA (11.1) CH2Cl2
2 (0.0028) MA (11.1) CHCl3
2 (0.0028) MA (11.1) PhCl
82
81
93
85
73
66
6e
plex [Pd2(µ-Br)2(C6F5)2(PPh3)2] does not polymerize MA.
Complex 2 also polymerizes other acrylates such as tert-
butyl acrylate (tBuA) but not methyl methacrylate
(MMA), as shown in Table 1.
3 (0.0028)d MA (5.6)
4 (0.0028)d MA (5.6)
CH2Cl2
CH2Cl2
2 (0.0018)d tBuA (3.4) CH2Cl2
2 (0.0024)d MMA (4.7) CH2Cl2
Mech a n istic Stu d ies. The polymerizations of MA
using 2 as initiator and the stoichiometric reactions of
a
Reaction conditions: solvent, 2 mL; 24 h; room temperature.
b
1
2 with MA were followed by 19F and H NMR in CDCl3.
Determined by SEC in THF relative to polystyrene standards.
c SEC in CHCl3. Solvent, 0.5 mL. e mm:mr:rr ) 0.026:0.299:0.675,
d
Initially the insertion of MA into the Pd-C6F5 bond
gives the two alkyl palladium compounds 5 and 6
(Scheme 1),12 which are clearly seen in the 19F NMR
spectrum, where typical signals for a C-bound pen-
tafluorophenyl appeared (at about -140 ppm) at the
expense of the starting material (ca. -120 ppm).13
Coordination of MA to [Pd2(µ-Cl)2(C6F5)2L2] (L ) tht,
2) must occur prior to this insertion, probably by
substitution of tht in preference to bridge splitting. This
is suggested by the following observations: (i) the
polymerization activity of the dimeric complexes [Pd2-
(µ-X)2(C6F5)2L2] decreases steeply (L ) tht ≈ AsPh3 .
L ) PPh3) with increasing coordination ability of L (in
fact, the PPh3 complex does not react with MA); (ii) a
change in the halogen bridge does not produce a
noticeable difference (cf. complexes 2 and 3, Table 1,
entries 2 and 5), as should be expected if bridge cleavage
were involved (Scheme 1).
A yellow solid containing a mixture of 5 and 6 (5:6 )
15:1) was isolated from the reaction of 2 with MA (Pd:
MA ) 1:5) in CH2Cl2 at room temperature for 55 min.
The X-ray diffraction structure of a single crystal taken
from this mixture corresponded to the major complex 5
and is shown in Figure 1. Selected distances and angles
are given in Table 2. The compound is a dimeric σ-alkyl
derivative resulting from insertion of MA into the Pd-
C6F5 bond. The coordination plane is defined by a chiral
carbon, the sulfur atom of a tht molecule, and two
bridging Cl atoms, while the carboxylate group is not
coordinated. Two diastereoisomers are produced, arising
from the presence in the dimer of two chiral carbons
formed unselectively. These are clearly seen in the 19F
NMR spectrum at low temperature, but the crystal
studied contained only one of them. The alkyl and tht
ligands are mutually trans across the bent bridges. The
dihedral angle between both palladium coordination
planes is 20.6°. The higher trans influence of the alkyl
groups is reflected in the longer trans Pd-Cl bond
determined from the 1H NMR spectrum in CHCl3.
was observed, and the stability of the metallacycle
formed (Chart 1) was deemed responsible for the inef-
ficient polymerization.
Other less well-defined palladium systems show
similar amounts of acrylate incorporation in the poly-
ethylene chain.8 Since the high electrophilicity of the
metal in cationic complexes favors oxygen coordination,
neutral derivatives have been tested. Grubbs et al. have
developed very active neutral nickel catalysts, tolerant
to polar groups, but they are ineffective for acrylates.9
Novak et al.10 and we and Sen11 described neutral
palladium complexes active in the polymerization of
acrylates. We studied the activity of the system [PdBr-
(C6F5)(NCMe)2] (1) in the presence of 1 equiv of PPh3
and proved that polymer growth is a radical process,
but insertion and â-H elimination reactions on the metal
center are very influential on the polymerization process
and activity of the system toward different monomers.11
We describe here new neutral palladium catalysts for
the polymerization of acrylates. They basically follow
the same mechanism described for 1 + PPh3, but the
new catalytic system is easier to prepare and facilitates
the control and monitoring of the reaction, which allows
a better understanding of the mechanism.
Resu lts a n d Discu ssion
Act ivit y in t h e P olym er iza t ion of Acr yla t es.
Dimeric palladium complexes [Pd2(µ-X)2(C6F5)2L2] (L )
tht (tht ) tetrahydrothiophene), X ) Cl (2); L ) tht, X
) Br, (3); L ) AsPh3, X ) Br (4)) are active precursors
for the polymerization of methyl acrylate (MA). Atactic
high-molecular-weight polymers were obtained in good
yields at room temperature in the light (Table 1).
Interestingly, as reported before,11 the analogous com-
(8) Drent, E.; van Dijk, R.; van Ginkel, R.; van Oort, B.; Pugh, R. I.
Chem. Commun. 2002, 744-745.
(12) For simplicity the dimeric palladium complexes are represented
in the scheme as homodimers, although mixed dimers are also
expected. Heterodimers were not detected in this case, but they can
be seen, for example, in the 19F NMR spectrum of a mixture of 2 and
5 in CDCl3.
(13) Albe´niz, A. C.; Espinet, P.; Foces-Foces, C.; Cano, F. H.
Organometallics 1990, 9, 1079-1085.
(9) Younkin, T. R.; Connor, E. F.; Henderson, J . I.; Friedrich, S. K.;
Grubbs, R. H.; Bansleben, D. A. Science 2000, 287, 460-462.
(10) Tian, G.; Boone, H. W.; Novak, B. M. Macromolecules 2001,
34, 7656-7663.
(11) Elia, C.; Elyashiv-Barad, S.; Sen, A.; Lo´pez-Ferna´ndez, R.;
Albe´niz, A. C.; Espinet, P. Organometallics 2002, 21, 4249-4256.

4208 Organometallics, Vol. 22, No. 21, 2003
Albe´niz et al.
Sch em e 2
Sch em e 3
F igu r e 1. Molecular structure of the complex [Pd2(µ-Cl)2-
{CH(CO2Me)CH2C6F5}2(tht)2] (5). Hydrogen atoms are
omitted for clarity.
Ta ble 2. Selected Bon d Dista n ces (Å) a n d An gles
(d eg) for Com p lex 5
Pd(1)-C(1)
Pd(1)-S(1)
Pd(1)-Cl(1)
Pd(1)-Cl(2)
2.063(7)
2.299(2)
2.3649(17)
2.4368(18)
Pd(2)-C(21)
Pd(2)-S(2)
Pd(2)-Cl(1)
Pd(2)-Cl(2)
2.066(6)
2.277(2)
2.4737(17)
2.3399(19)
immediately, and at this point, no polymer is observed
in the 1H NMR spectrum. Complexes 5 and 6 decompose
slowly to the substituted product 7. An undetected
palladium hydrido complex should be formed also, which
reacts with MA to form the new palladium complex [Pd2-
(µ-Cl)2{CH(CO2Me)CH3}2(tht)2] (9) by insertion into the
Pd-H bond (Scheme 3).12 This is shown by the appear-
ance of a doublet at 0.80 ppm, coupled to a signal at
3.40 ppm partially overlapped by other resonances in
C(1)-Pd(1)-Cl(1)
S(1)-Pd(1)-Cl(2)
Cl(1)-Pd(1)-Cl(2)
C(1)-Pd(1)-S(1)
89.7(2)
C(21)-Pd(2)-Cl(2) 89.4(2)
95.45(8) S(2)-Pd(2)-Cl(1)
85.04(6) Cl(1)-Pd(2)-Cl(2)
89.8(2) C(21)-Pd(2)-S(2)
97.82(7)
84.75(6)
88.2(2)
Pd(1)-Cl(1)-Pd(2) 92.33(6) Pd(1)-Cl(2)-Pd(2) 93.90(7)
lengths (about 0.1 Å). The preference for coordination
of the tht and bridge formation, rather than coordination
of the carbonyl oxygen, can be attributed to the unfa-
vorable four-membered metallacycle that should be
formed and the reduced electrophilicity of the metal
center in this neutral palladium complex compared to
that of the cationic derivatives.7
1
the spectrum but detected in H COSY. When 40% of 5
+ 6 had decomposed (13 h), the formation of a small
amount of polymer (3%) was clear in the 1H NMR
spectrum (Figure 2).
Another organometallic complex is slowly formed
during this process, with 19F NMR chemical shifts very
close to those of 5. It could be a cis alkyl derivative
across the bridge, an isomer of the trans complex 5 (the
new putative cis complex is about 30% of the total
amount of 5 after 14 h). Complexes 5 (cis and trans)
and 6 continue decomposing to 7, and no other fluorine-
containing species are formed. Complete decomposition
of 5 takes around 8 days (2 days for complex 6) under
these conditions (lower concentration of MA than in the
polymerization reactions in Table 1).
The amount of polymer formed in this experiment is
very low, suggesting that effective bulk polymerization
of MA with 2 requires the presence of light, as typically
happens in many radical processes. This was assessed
experimentally. Two NMR tubes were charged with 2
and MA (Pd:MA ) 1:220) in CDCl3. Only one of the
samples was protected from light. After the first 30 min
both samples were a mixture of 5 (87%), 6 (5%), and 7
(8%), and no polymer had been formed. When the
samples were checked again after 7 h more, the sample
protected from light contained 5 (62%), 6 (3%), and 7
(35%) but no polymer, whereas the sample in the light
showed 71% conversion of MA to poly(MA), in addition
to 5 (50%) and 7 (30%), and new broad 19F NMR signals
The minor Pd-migration isomer 6 (Scheme 1) shows
a low-field signal in the H NMR spectrum (4.15 ppm)
1
corresponding to the methine proton attached to the
Pd-bound carbon. The rest of the signals are overlapped
with the signals of the major complex 5 and the tht
resonances but could be identified in a 1H COSY
experiment.
Both 5 and 6 decompose in solution to give the
substituted olefin 7 and the alkane 8 (7:8 ) 3.9:1). The
latter is formed by hydride transfer between palladium
atoms in a dimer (Scheme 2).11,14 J ust a very small
amount of 8 is formed when excess MA is present (this
is the case for the polymerization reactions), since MA
can act as a ligand and disfavors the formation of the
dimeric complexes needed for H transfer.
Polymerization reactions inside the NMR probe (con-
sequently in the dark) using 2 as precatalyst and a ratio
of Pd to MA up to 1:220 in CDCl3 were monitored by
19F and 1H NMR. The formation of 5 and 6 occurs
(14) The formation of the alkene 7 leaves a Pd-H moiety, and the
H atom can transfer to another Pd by formation of a dimeric halo-
hydrido species. Reductive elimination of alkyl and H leads to 8. This
has been studied in detail: Albe´niz, A. C.; Espinet, P.; Lin, Y.-S.
Organometallics 1997, 16, 4030-4032.

Polymerization of Acrylates by Neutral Pd Complexes
Organometallics, Vol. 22, No. 21, 2003 4209
Sch em e 4
(13) in a 1.2:1.2:1.5:1 molar ratio. Derivatives 10-12
were separated by TLC and identified spectroscopically
and by MS, whereas 13 could only be identified in the
reaction mixture. Compounds 7 and 8 are the expected
products of decomposition by â-H elimination and are
formed also in the dark, but 10-13 reveal the formation
of radicals generated by cleavage of the Pd-C bond in
5 or 6, which are trapped by their very fast reaction
with O2.15 The peroxide radicals formed can either
couple or react with Pd-H species to give hydroperox-
ides. In both cases their eventual decomposition gives
compounds 10-13.15 Thus, complexes 5 and 6 behave
simultaneously as a slow source of Pd-H (via â-H
elimination) and as a reservoir of minute concentrations
of radicals. In the presence of MA, the radicals initiate
the polymerization. However, it is worth noting that
galvinoxyl does not give detectable amounts of radical
coupling Pf-containing species when it reacts with 5 or
6. The only observed products are 7 and a small amount
of 8.16 This means that, for the Pf radicals in that small
a concentration, the coupling with galvinoxyl is not
sufficiently fast to trap them. Probably when galvinoxyl
stops this polymerization (see below), its action as a
hydride trap is crucial.
These data support the polymerization mechanism
shown in Scheme 4. It starts with the detected insertion
of MA into the Pd-C6F5 bond. The resulting complexes
5 and 6, plus 9 formed by â-H elimination and MA
insertion, can provide the radicals that add to MA and
bring about polymer growth. Since 5 and 6 are fairly
stable, they act as a kind of reservoir of radical species
and of Pd-H (hence of 9), providing a slow but steady
supply of catalytically active species. Eventually the
growing radical will recombine with [Pd] and the
resulting alkyl will undergo â-H elimination and liber-
ate an olefin that accounts for the terminal olefin group
1
F igu r e 2. 19F (a) and H (b) NMR spectra of a mixture of
2 and MA (MA:Pd ) 25:1) after 4 h: (O) 5; (0) 6; ([) 7;
(w) 9; (b) poly(MA).
(Fortho: -140.5, -142, -143.2 ppm) accounting for the
remaining 20% of fluorine-containing species. The poly-
mer obtained in this reaction was isolated, recrystal-
lized, and characterized by 19F and H NMR. The same
1
broad fluorine signals detected in the reaction mixture
were present in the polymer, supporting that at least
part of the reaction involves the incorporation of a Pf-
containing fragment (Pf ) pentafluorophenyl; it was
estimated by 19F NMR that about 20% average polymer
chains contained Pf; see the Experimental Section). The
1H NMR spectrum showed signals at δ 6.8 (dd, J ) 17.6,
8.8 Hz) and 5.9 (d, J ) 17.6 Hz) typical of a (CO2Me)-
CHdCH- terminal group in the polymer. Additional
experiments showed that, as observed before for the
catalytic system 1 + PPh3,11 oxygen and radical traps
such as galvinoxyl and DPPH inhibit the reaction,
whereas 2,6-di-tert-butylphenol (TBP) does not. All these
features, along with the polymer characteristics, support
a radical mechanism for polymer growth (Scheme 4).
The initiating radical species have to be formed from
complexes 5 and 6, since these are formed from 2 before
any polymer appears. The radical species may be formed
by homolytic splitting of the Pd-C bond in 5 and 6 (and
eventually 9, once this complex has been formed). In
other words, the alkyl complexes should be in a dynamic
equilibrium with minute amounts of the radicals pro-
duced in this homolytic cleavage. In fact, mixtures of 5
and 6 in CDCl3 decompose under UV irradiation (365
nm) to 7 (and small amounts of its cis isomer), 8, and
other Pf-containing products (about 8%), which in-
creased their ratio (to 15%) if the decomposition was
carried out in the air. These were identified as the
hydroxy esters PfCH2CH(OH)CO2Me (10) and PfCH-
(OH)CH2CO2Me (11) and their corresponding keto
esters PfCH2C(O)CO2Me (12) and PfC(O)CH2(CO2Me)
(15) Boukouvalas, J .; Haynes, R. K. In Radicals in Organic Syn-
thesis; Renaud, P., Sibi, M. P., Eds.; Wiley-VCH: Weinheim, Germany,
2001; Vol. 2, Chapter 5.4.
(16) The presence of galvinoxyl dramatically reduces the amount
of 8, since it reacts with the palladium hydride species formed and
halts the H transfer process.18

4210 Organometallics, Vol. 22, No. 21, 2003
Albe´niz et al.
Sch em e 5
Sch em e 6
faster for 14. Eventually the equilibria afford the most
stable olefin, 15, which appears slowly because it needs
to be formed by decoordination from a particularly
stable 6.5-membered palladacycle.20 Brookhart and
Hauptman have found a very similar effect in the
dimerization of MA catalyzed by rhodium complexes.21
The processes described are favored when less free MA
is present. Otherwise MA can associatively substitute
either the coordinated MA in A, producing a fast
exchange before rotation and readdition, or the coordi-
nated oxygen in the chelating metallacycles, thus halt-
ing the whole process.
Methyl methacrylate (MMA) is not polymerized by 2.
The reaction of MMA and 2 using the ratio Pd:MMA )
1
1:220 was monitored by 19F and H NMR. No interme-
diate palladium alkyl species was observed, and only
the substituted olefin 16 was formed (Scheme 6). After
40 min complex 2 had disappeared, and 16 was the only
fluorine-containing product present in solution. This
indicates that â-H elimination from the methyl group
in MMA is much faster than the â-H elimination from
the methylene group (the only â-H source existing for
MA). The fast formation of palladium hydride from E
produces a high concentration of hydride species and
leads to its immediate decomposition. Thus, E cannot
play the role of a slow but sustained source of radicals
and hydrides that 5, 6, and 9 play in the reaction with
MA, and the initiator is consumed irreversibly in the
early stages of the reaction.
observed in the polymer by 1H NMR. In this termination
process Pd-H is formed, which reenters the cycle by
reaction with MA to give 9. Radical disproportionation
could also be a termination pathway forming the
1
terminal olefin group observed in the H NMR, but the
stability of acrylate radicals makes this pathway less
favorable than for other radical species.17
The effect of the radical traps used in the polymeri-
zation (galvinoxyl, DPPH, TBP) parallels the observed
reactivity of these compounds with palladium hy-
drides.18 TBP does not react with Pd-H species and
does not affect the polymerization process. Galvinoxyl
and DPPH halt the polymerization of MA by a 2-fold
mechanism. First, they can destroy the palladium
hydrido complexes formed in the reaction medium either
from 5 or in the termination step at the bottom of
Scheme 4, avoiding in the latter case the regeneration
of 9 and reducing seriously the TON. This detrimental
effect is important whether the radical traps are added
at the beginning or at any stage of the reaction. In
addition, but probably with less incidence in the halting
of the reaction, these additives can trap the growing
polymeric radical species, disfavoring the propagation
step (Scheme 4).
It is worth noting that the addition of MMA to a
running MA polymerization reaction will also halt the
process. A three-experiment batch was set up under the
same conditions. After 30 min batch 1 was quenched,
yielding 0.1 g of polymer; MMA was added to batch 2
to reach a ratio MA:MMA ) 1:1; a volume of inert
solvent, identical to the volume of MMA used for batch
2, was added to batch 3. After 11 days batches 2 and 3
were quenched, recovering 0.12 g of polymer in batch 2
and 0.5 g of polymer in batch 3. The recovered polymer
in batch 2 had incorporated some MMA, as determined
After the formation of poly(MA) has been going on for
several hours, the formation of the MA dimer (MeO2C)-
CH2CHdCHCH2(CO2Me) (14), as a mixture of E and
1
by H NMR. Since radical polymerization of MMA is a
favorable process, this result indicates that MMA
incorporation in the growing radical results in termina-
tion of the polymerization as soon as the equilibrium of
recombination of the new radical with palladium allows
fast â-H elimination (Scheme 4). In addition, MMA will
avoid the formation of new radicals (Scheme 6).
1
Z isomers (E:Z ) 4:1), is observed by H NMR. Only
later, partial isomerization of 14 to the 2-alkene
(MeO2C)CHdCHCH2CH2(CO2Me) (15) is observed. The
selective dimerization of MA to 14 (and not to a mixture
of regioisomers) indicates that it is produced from a
palladium linear alkyl complex (B, Scheme 5). Some-
what surprisingly, the first product observed is 14,
although the formation of 14 implies a pathway through
the precursor of 15: namely, C.19 This means that the
reversible Pd migration connecting C and D is very fast,
and which olefin is formed first depends on the deco-
ordination rate, which, from the results observed, is
In conclusion, the polymerization of methyl acrylate
by these systems is initiated by insertion of methyl
(19) For economy the alkene complexes are represented uniformly
as monomeric chelated with terminal Cl, although probably equilibria
exist with their Cl-bridged dimeric isomers with monodentate alkenes
(bonded either by the ketone or by the double bond). The dominant
species will vary with the size of the cycle.
(20) For closely related palladium migration processes see: (a)
Albe´niz, A. C.; Espinet, P.; Lin, Y.-S. Organometallics 1996, 15, 5010-
5017. (b) Albe´niz, A. C.; Espinet, P.; Lin, Y.-S. Organometallics 1997,
16, 4138-4144. (c) Albe´niz, A. C.; Espinet, P.; Lin, Y.-S. Organome-
tallics 1997, 16, 5964-5973.
(21) Brookhart, M.; Hauptman, E. J . Am. Chem. Soc. 1992, 114,
4437-4439.
(17) Kochi, J . K. Free Radicals; Wiley: New York, 1973; Vol. I, p
50.
(18) Albe´niz, A. C.; Espinet, P.; Lo´pez-Ferna´ndez, R.; Sen, A. J . Am.
Chem. Soc. 2002, 124, 11278-11279.

Polymerization of Acrylates by Neutral Pd Complexes
Organometallics, Vol. 22, No. 21, 2003 4211
Ta ble 3. Cr ysta l Da ta for Com p lex 5
acrylate in the Pd-aryl bond of the precatalyst. A
radical polymerization can be initiated by the radicals
generated from these alkyl complexes (5, 6) but also
from the alkyl complex 9 generated by insertion of MA
in hydrido-Pd species. The latter is the regeneration
pathway of radicals after a termination reaction has
occurred. For this reason, the main pathway by which
the radical traps halt this polymerization seems to be
the annihilation of H-Pd species (hydride trapping),
more than radical trapping itself.
empirical formula
C28H28Cl2F10O4Pd2S2
966.32
fw
temo
293(2) K
wavelength (Mo KR)
0.710 69 Å
monoclinic
P21/c
cryst syst
space group
unit cell dimens
a
b
c
15.114(5) Å
12.028(5) Å
19.611(5) Å
90.000(5)°
102.628(5)°
90.000(5)°
R
â
Exp er im en ta l Section
γ
V, Z
3479(2) Å3, 4
1.845 g/cm3
Gen er a l Con sid er a tion s. 19F and 1H NMR spectra were
recorded on Bruker AC-300 and ARX-300 instruments. Chemi-
cal shifts are reported in δ (ppm) and referenced to Me4Si (1H)
or CFCl3 (19F). The spectra were recorded at 293 K unless
otherwise noted. Size exclusion chromatography (SEC) was
carried out on a Waters SEC system using a two-column bed
(Styragel 7.8 × 300 mm columns: 50-100 000 D and 2000-
4 000 000 D) and Waters 410 differential refractometer. SEC
samples were run in THF at room temperature and calibrated
to polystyrene standards. Solvents were dried over CaH2,
distilled, and deoxygenated before use. Methyl acrylate, tert-
butyl acrylate, and methyl methacrylate (Aldrich) were dis-
tilled and deoxygenated prior to use. Galvinoxyl, 2,2-bis(4-tert-
octylphenyl)-1-picrylhydrazyl (DPPH), 2,6-di-tert-buylphenol
(TBP), and AsPh3 were purchased from Aldrich. 1,3,5-C6F3Cl3
was purchased from Fluorochem Ltd. [PdBr(C6F5)(NCMe)2]
(1),13 [Pd2(µ-Cl)2(C6F5)2(tht)2] (2),22 and [Pd2(µ-Br)2(C6F5)2(tht)2]
(3)22 were synthesized by literature procedures. All the reac-
tions described were carried out under an inert atmosphere.
Polymer tacticity was determined by integration of the meth-
ine and the methylene signals in the 1H NMR spectrum.
tr a n s-[P d 2(µ-Br )2(C6F 5)2(AsP h 3)2] (4).23 A solution of As-
Ph3 (0.054 g, 0.175 mmol) in CH2Cl2 (20 mL) was added
dropwise at room temperature to a stirred solution of [PdBr-
(C6F5)(NCMe)2] (1; 0.076 g, 0.175 mmol) in CH2Cl2 (10 mL).
The mixture was stirred for 10 min, and the solvent was
evaporated to dryness. The residue was triturated in EtOH
(10 mL) to yield a yellow solid which was filtered, washed with
EtOH, and air-dried. Yield: 0.103 g (90%). Anal. Calcd for
density (calcd)
abs coeff
1.394 mm-1
F(000)
1904
abs cor method;
transmissn factor max, min
cryst size
SADABS; 1.000 000,
0.804 450
0.05 × 0.13 × 0.15 mm3
1.38 e θ e 23.27
16 355, 5004
99.9%
scan range, deg
no. of rflns collected, indep
completeness to θ ) θmax
no. of data/restraints/params
goodness of fit on F2
final R indices (I > 2σ(I))
R indices (all data)
largest diff peak and hole
5004/0/435
1.042
R1 ) 0.0424, wR2 ) 0.0974
R1 ) 0.0617, wR2 ) 0.1083
1.335 (close to Pd)
and -0.603 e Å3
X-r a y Cr ysta l Str u ctu r e Deter m in a tion of 5. Crystals
of 5 suitable for X-ray analysis were obtained by slow diffusion
of n-hexane into a solution of 5 in CH2Cl2. The crystal selected
was mounted on the tip of a glass fiber. X-ray measurements
were made using a Bruker SMART CCD area-detector dif-
fractometer with graphite-monochromated Mo KR radiation.
Reflections were collected, intensities were integrated, and the
structure was solved by the direct methods procedure.24 Non-
hydrogen atoms were refined anisotropically, and hydrogens
were constrained to ideal geometries and refined with fixed
isotropic displacement parameters. Relevant crystallographic
data are gathered in Table 3.
Gen er a l P r oced u r e for th e P olym er iza tion of Acr y-
la tes. Methyl acrylate (0.956 g, 11.1 mmol) was added to a
solution of [Pd2(µ-Cl)2(C6F5)2(tht)2] (2; 0.0022 g, 0.0028 mmol)
in CH2Cl2 (2 mL). The reaction proceeded in daylight at room
temperature for 24 h, although the high viscosity that was
generated halted the stirring after 5 h. The polymer was
precipitated on pouring the mixture onto MeOH (30 mL). The
MeOH was decanted off, and the polymer was dried under
C
48H30As2Br2F10Pd2: C, 43.70; H, 2.29. Found: C, 43.78; H,
2.56. 19F NMR (282 MHz, δ, CDCl3): -162.9 (m, 2Fmeta), -160.4
1
(t, 1Fpara), -118.2 (m, 2Fortho). H NMR (300 MHz, δ, CDCl3):
7.52 (m, 2H; Ph), 7.35 (m, 3H; Ph).
tr a n s-[P d 2(µ-Cl)2{CH(CO2Me)CH2C6F 5}2(th t)2] (5) a n d
tr a n s-[P d 2(µ-Cl)2{CH(C6F 5)CH2CO2Me}2(th t)2] (6). To a
solution of [Pd2(µ-Cl)2(C6F5)2(tht)2] (2; 0.200 g, 0.252 mmol) in
CH2Cl2 (8 mL) was added methyl acrylate (0.217 g, 2.521
mmol). The mixture was stirred for 55 min at room temper-
ature and evaporated to dryness. The residue was washed with
n-hexane (3 × 1.5 mL) and then triturated in Et2O (5 mL).
The yellow solid was filtered, washed with Et2O, and air-dried.
Yield: 0.087 g (36%). Anal. Calcd for C28H28Cl2F10O4S2Pd2: C,
34.80; H, 2.92. Found: C, 34.84; H, 3.02.
vacuum to yield 0.786 g of poly(methyl acrylate) (82%): Mw
)
1
1.9 × 105, Mw /Mn ) 2.2. H NMR (300 MHz, δ, CDCl3): 3.68
(s, 6H; OCH3), 2.34 (br, 2H; CH(CO2CH3)s, CH(CO2CH3)i), 1.95
(br, 1H; CHHi), 1.68 (br, 2H; CH2s), 1.51 (br, 1H; CHHi). In
this and all following data lists, i ) isotactic and s )
syndiotactic.
The same procedure was used for other monomers and/or
solvents (see Table 1) and also for the tests of the influence of
additives (molar ratio Pd:additive ) 1:1).
Data for 5 are as follows. 19F NMR (282 MHz, δ, CDCl3):
-163.5 (m, 2Fmeta), -158.5 (t, 1Fpara), -142.6 (m, 2Fortho). 1H
NMR (300 MHz, δ, CDCl3): 3.65 (s, 3H; OCH3), 3.48 (m, 1H;
CHCO2CH3), 3.10 (br, 4H; SCH2), 2.90 (m, 1H; CHHC6F5), 2.58
(m, 1H; CHHC6F5), 2.08 (br, 4H; SCH2CH2).
P oly(ter t-bu tyl a cr yla te): 66% yield; atactic polymer; Mw
) 5.1 × 105, Mw/Mn ) 4.8. 1H NMR (300 MHz, δ, CDCl3): 2.23
(br, 2H; CH(CO2tBu)s, CH(CO2tBu)i), 1.81 (br, 1H; CHHi), 1.60
(br, 1H; CHHi), 1.51 (br, 2H; CH2s), 1.43 (s, 18H; tBu).
P oly(m eth yl m eth a cr yla te): 6% yield; mm:mr:rr ) 0.026:
0.299:0.675. Mw ) 1.8 × 105, Mw/Mn ) 2.9. 1H NMR (300 MHz,
δ, CDCl3): 3.6 (s, 6H; OCH3), 2-1.7 (br, 4H; CH2), 1.2 (s, 3H;
CH3mm), 1.0 (s, 3H; CH3mr), 0.8 (s, 3H; CH3rr).
Data for 6 are as follows. 19F NMR (282 MHz, δ, CDCl3):
-164.1 (m, 2Fmeta), -159.8 (t, 1Fpara), -138.7 (m, 2Fortho). 1H
NMR (300 MHz, δ, CDCl3, 273 K): 4.15 (t, J ) 7.80 Hz, 1H;
CHC6F5), 3.65 (s, 3H; OCH3), 3.10 (br, 4H; SCH2), 3.04 (m,
1H; CHHCO2CH3), 2.62 (m, 1H; CHHCO2CH3), 2.08 (br, 4H;
SCH2CH2).
P olym er iza tion of Meth yl Acr yla te in th e P r esen ce of
Meth yl Meth a cr yla te. To a solution of [Pd2(µ-Cl)2(C6F5)2-
(22) Uso´n, R.; Fornie´s, J .; Navarro, R.; Garc´ıa, M. P. Inorg. Chim.
Acta 1979, 33, 69-75.
(23) Uso´n, R.; Fornie´s, J .; Nalda, J . A.; Lozano, M. J .; Espinet, P.;
Albe´niz, A. C. Inorg. Chim. Acta 1989, 156, 251-256.
(24) The data analysis and drawings were performed with the
following programs: SMART V5.051, 1998. SAINT V6.02, 1999.
Sheldrick, G. M. SHELXTL V5.1; Bruker AXS, Inc. Madison, WI, 1998.

4212 Organometallics, Vol. 22, No. 21, 2003
Albe´niz et al.
(tht)2] (2; 0.005 g, 0.007 mmol) in CH2Cl2 (3 mL) was added
methyl acrylate (2.390 g, 27.8 mmol). The resulting solution
was divided into three equal samples. After 30 min the samples
were treated as follows: sample 1 was quenched by adding it
to a large excess of MeOH, resulting in the precipitation of
0.098 g (12%) of polymer (Mw ) 2.6 × 105, Mw/Mn ) 4.3).
Methyl methacrylate (1.0 mL, 9.3 mmol) was added to sample
2, and the reaction was allowed to continue for 11 days, after
which time quenching with MeOH afforded 0.116 g of copoly-
mer (Mw ) 2.9 × 105, Mw/Mn ) 5.0; MA:MMA ratio in polymer,
1:1.9). CH2Cl2 (1 mL) was added to sample 3, to keep the same
concentration of sample 2, and quenching after 11 days yielded
0.503 g (66%) of polymer (Mw ) 2.6 × 105, Mw/Mn ) 2.7).
The extent of monomer incorporation in the copolymer was
determined by integration of the methoxy signals (methyl
acrylate/methyl methacrylate copolymer) in the 1H NMR
spectra.
Mon itor in g of th e P olym er iza tion of Meth yl Acr yla te
by NMR. [Pd2(µ-Cl)2(C6F5)2(tht)2] (2; 0.004 g, 0.005 mmol) was
dissolved in CDCl3 (0.6 mL) in an NMR tube. Methyl acrylate
(0.191 g, 2.221 mmol) was added to the solution. The polym-
erization was monitored by 19F and 1H NMR, and the forma-
tion of 5, 6, 7,11 8,11 9, 14, and 15 was observed as described
in the text. The identities of 14 and 15 were checked by
comparison with independently synthesized samples.
Data for 7 are as follows.25 19F NMR (282 MHz, δ, CDCl3):
-162.0 (m, 2Fmeta), -151.5 (t, 1Fpara), -139.9 (m, 2Fortho).
Data for 8 are as follows.25 19F NMR (282 MHz, δ, CDCl3):
-162.9 (m, 2Fmeta), -157.2 (t, 1Fpara), -144.1 (m, 2Fortho).
Data for 9 are as follows. 1H NMR (300 MHz, δ, CDCl3):
3.60 (s, 3H; OCH3), 3.40 (1H; CHCO2CH3), 0.80 (d, J ) 8.90
Hz, 3H; CH3).
The polymerization experiments controlling the light source
were carried out similarly. New broad signals (that are present
in the final polymer) were observed in the 19F NMR spectra
when the polymerization was carried out in the presence of
light. After 5 days, this sample was added to a large excess of
MeOH. The polymer obtained was filtered, dissolved in
CH2Cl2, and precipitated in MeOH. 19F and 1H NMR spectra
of the polymer were taken to analyze the end groups. 1H NMR
(300 MHz, δ, CDCl3): 6.8 (dd, J ) 17.6, 8.8 Hz, 1H; (CO2Me)-
CHdCHP), 5.9 (d, J ) 17.6 Hz, 1H; (CO2Me)CHdCHP). 19F
NMR (282 MHz, δ, CDCl3): -162.6 (b, 2Fmeta), -156.3 (b,
1Fpara), -143.2 (b, 2Fortho). The rest of the signals cannot be
assigned to distinct Pf groups. Broad signals in the charac-
teristic 19F regions indicated in parentheses are observed as
follows: -161.5 (b, Fmeta), -162.35 (b, Fmeta), -155 (b, Fpara),
-140.5 (b, Fortho), -142 (b, Fortho).
achieved at the solubility limit of the polymer the Teflon
contained in the NMR probe distorts the baseline and makes
the adjustment of the baseline extremely difficult.
P h otoch em ica l Decom p osition of Mixtu r es of 5 a n d
6. A mixture of trans-[Pd2(µ-Cl)2{CH(CO2Me)CH2C6F5}2(tht)2]
(5) and trans-[Pd2(µ-Cl)2{CH(C6F5)CH2CO2Me}2(tht)2] (6; 0.021
g, 0.021 mmol) (5:6 ) 15:1) was dissolved in CDCl3 in the
presence of air. The NMR tube was irradiated for 9 h using a
365 nm (6 W) lamp. A mixture of 7 (and its corresponding cis
isomer) and 8 as major products, in addition to 10-13, was
observed in the 19F NMR spectra. The sample was evaporated
to dryness and the residue chromatographed by preparative
TLC (silica plate) using a mixture of ethyl acetate and
n-hexane (1:3) as eluent. Compounds 10-12 were separated
and characterized by NMR and GC-MS.
Data for 10 are as follows. 19F NMR (282 MHz, δ, CDCl3):
-162.9 (m, 2Fmeta), -156.2 (t, 1Fpara), -143.0 (m, 2Fortho). 1H
NMR (300 MHz, δ, CDCl3): 4.41 (m, 1H; CHCO2Me), 3.85 (s,
3H; CHCO2CH3), 3.20 (m, J ) 14.3, 5.2 Hz, 1H; CHHC6F5),
3.05 (m, J ) 14.3, 8.1 Hz, 1H; CHHC6F5), 2.85 (d, J ) 6.1 Hz,
1H; CH(OH)). MS (EI, m/z (relative intensity)): 271 (M - H+,
1), 252 (55) 250 (50), 211 (90), 181 (97), 163 (100), 89 (34), 59
(6).
Data for 11 are as follows. 19F NMR (282 MHz, δ, CDCl3):
-161.8 (m, 2Fmeta), -154.4 (t, 1Fpara), -143.0 (m, 2Fortho). 1H
NMR (300 MHz, δ, CDCl3): 5.53 (m, 1H; CHC6F5), 3.75 (s, 3H;
CHCO2CH3), 3.22 (d, J ) 5.7 Hz, 1H; CH(OH)), 3.15 (dd, J )
16.7, 9.7 Hz, 1H; CHHCO2CH3), 2.75 (dd, J ) 16.7, 4.1 Hz,
1H; CHHCO2CH3). MS (EI, m/z (relative intensity)): 270 (M+,
1), 197 (100), 177 (65), 167 (6), 74 (21), 59 (3).
Data for 12 are as follows. 19F NMR (282 MHz, δ, CDCl3):
-162.2 (m, 2Fmeta), -154.4 (t, 1Fpara), -142.2 (m, 2Fortho). 1H
NMR (300 MHz, δ, CDCl3): 4.28 (s, 2H; CH2C6F5), 3.93 (s, 3H;
CO2CH3). MS (EI, m/z (relative intensity)): 268 (M+, 9), 209
(7), 181 (100), 59 (10).
Data for 13 are as follows. 19F NMR (282 MHz, δ, CDCl3):
-160.1 (m, 2Fmeta), -147.7 (t, 1Fpara; chemical shift character-
istic of the C6F5CO group), -140.7 (m, 2Fortho).
The decomposition of mixtures of 5 and 6 under nitrogen,
in the presence of galvinoxyl and with the light source
controlled (365 nm and in the dark), were carried out in the
same way.
Mon itor in g of th e Rea ction w ith Meth yl Meth a cr yla te
by NMR. [Pd2(µ-Cl)2(C6F5)2(tht)2] (2; 0.003 g, 0.004 mmol) was
disolved in CDCl3 (0.6 mL) in an NMR tube, and methyl
methacrylate (0.187 g, 1.870 mmol) was added to the solution.
The reaction was monitored by 19F and 1H NMR. The formation
of 16 was observed.11
Estim a tion of th e C6F 5 Con ten ts of P oly(m eth yl a cr y-
la te). A sample of poly(methyl acrylate) was repeatedly
dissolved and precipitated to remove any traces of nonpoly-
meric materials. A 16.5 mg amount of of polymer was
completely dissolved in 0.5 mL of acetone, and 0.1 mL of a
solution of 1,3,5-C6F3Cl3 (8.5 × 10-4 M, used as internal
standard) was added. A 19F NMR spectrum of the solution was
recorded, using a d6-acetone capillary for the lock. The number
of equivalents of C6F5, estimated by integration, was 3.8 ×
10-5. For Mn ) 8.6 × 104 this corresponds to 20% of C6F5-
containing chains. However, this result has to be considered
just as an estimate, because at the low level of F-concentration
Ack n ow led gm en t. We thank Prof. Ayusman Sen
and his group for useful discussions and assistance with
the GPC characterization of polymers. Financial support
from the Direccio´n General de Investigacio´n (BQU2001-
2015) and the J unta de Castilla y Leo´n (Projects VA057/
03, VA058/03 and VA120/01) is gratefully acknowledged.
We also thank the Spanish MECD for a fellowship to
R.L.-F.
Su p p or tin g In for m a tion Ava ila ble: Tables giving crys-
tallographic data for 5. This material is available free of charge
via the Internet at http://pubs.acs.org.
(25) 19F NMR spectral data for 7 and 8 were mistaken in ref 11.
The corrected chemical shifts are given here.
OM030507T