Redox control of a ring-opening polymerization catalyst

Article abstract of DOI:10.1021/ja2036089

The activity of an yttrium alkoxide complex supported by a ferrocene-based ligand was controlled using redox reagents during the ring-opening polymerization of l-lactide. The oxidized complex was characterized by X-ray crystallography and ¹H NMR, XANES, and Moessbauer spectroscopy. Switching in situ between the oxidized and reduced yttrium complexes resulted in a change in the rate of polymerization of l-lactide. Synthesized polymers were analyzed by gel permeation chromatography. Polymerization of trimethylene carbonate was also performed with the reduced and oxidized forms of an indium alkoxide complex. The indium system showed the opposite behavior to that of yttrium, revealing a metal-based dependency on the rate of polymerization.

Full text of DOI:10.1021/ja2036089

COMMUNICATION
pubs.acs.org/JACS
Redox Control of a Ring-Opening Polymerization Catalyst
Erin M. Broderick,† Neng Guo,‡ Carola S. Vogel,§ Cuiling Xu,|| J€org Sutter,§ Jeffrey T. Miller,‡
Karsten Meyer,§ Parisa Mehrkhodavandi,|| and Paula L. Diaconescu*,†
†Department of Chemistry & Biochemistry, University of California, Los Angeles, California 90095, United States
‡Chemical Sciences and Engineering Division, Argonne National Laboratory, Argonne, Illinois 60439, United States
§Department of Chemistry and Pharmacy, Friedrich-Alexander-University Erlangen-Nuremberg, 91058 Erlangen, Germany
University of British Columbia, Vancouver V6T1Z1, Canada
S Supporting Information
b
be important in tuning the effect of ligand oxidation/reduction on
ABSTRACT: The activity of an yttrium alkoxide complex
supported by a ferrocene-based ligand was controlled using
redox reagents during the ring-opening polymerization of
L-lactide. The oxidized complex was characterized by X-ray
the metal center of interest. In addition, the first example of
increasing the reactivity of a polymerization catalyst through its
oxidation is shown using an indium phosfen aryloxide (1-In-
(OPh)) and the oxidized analogue ([1-In(OPh)][BArF]).10 The
differentratesofpolymerizationbetweentheoxidized and reduced
forms of the yttrium and indium complexes indicate that the
control is metal dependent.
1
crystallography and H NMR, XANES, and M€ossbauer
spectroscopy. Switching in situ between the oxidized and
reduced yttrium complexes resulted in a change in the rate
of polymerization of L-lactide. Synthesized polymers were
analyzed by gel permeation chromatography. Polymeriza-
tion of trimethylene carbonate was also performed with the
reduced and oxidized forms of an indium alkoxide complex.
The indium system showed the opposite behavior to that of
yttrium, revealing a metal-based dependency on the rate of
polymerization.
n an effort to develop catalytic processes using external
I
switches, organometallic complexes capable of reversible pro-
cesses have been increasingly researched.1ꢀ15 One method to
modulate reactivity and selectivity is through the redox control
of the supporting ligand in a metal complex. The goal of this
research is to design a compound that exhibits orthogonal reac-
tivity for different substrates by switching between the oxidized
and reduced forms of a catalyst.
Although substrate selectivity has not been demonstrated in
this context, a change in reactivity has been observed for systems
containing ferrocene-derivatized ligands. Gibson et al.7 reported
that the rate of ring-opening polymerization of rac-lactide could
be altered by changing the redox state of a ferrocenyl unit in a
titanium salen bis(isopropoxide) catalyst. In addition, Plenio
et al.5 oxidized a ferrocenyl substituent in a Grubbs type catalyst,
leading to its precipitation during a ring-closing metathesis re-
action. In both cases, redox control was achieved by using a
ferrocenyl functionalized ligand, with the ferrocenyl unit distant
from the metal center, and a decrease in reactivity was observed
after the oxidation of the ferrocenyl group.
Herein we report that an yttrium alkoxide phosfen complex
(1-Y(OtBu), eq 1, phosfen =1,10-di(2-tert-butyl-6-diphenylpho-
sphiniminophenoxy)ferrocene)16 can be used to control the poly-
merization of L-lactide by using redox processes. The ferrocene
unit is central to the backbone of the ligand and near the metal
center; the proximity of the metal center to the ferrocene unit may
In order to achieve redox-switchable catalysis, the catalyst must
be able to change reversibly between the oxidized and reduced
forms. Electrochemical studies on 1-Y(OtBu) (E1/2= ꢀ0.28 V vs
ferrocene) indicated that ferrocenium salts would be appropriate
oxidants for the ferrocene backbone.16 Consequently, the addition
of 1 equiv of ferrocenium tetrakis(3,5-bis(trifluoromethyl)phenyl)
borate (FcBArF) to 1-Y(OtBu) in tetrahydrofuran resulted in a
dark green product, [1-Y(OtBu)][BArF] (eq 1), within minutes.
Received: April 19, 2011
Published: May 23, 2011
r
2011 American Chemical Society
9278
dx.doi.org/10.1021/ja2036089 J. Am. Chem. Soc. 2011, 133, 9278–9281
|

Journal of the American Chemical Society
COMMUNICATION
Figure 3. Plot of conversion (%) vs time for the polymerization of 100
equiv of L-lactide (0.2 M in tetrahydrofuran) with 1-Y(OtBu) using
in situ oxidation and reduction with FcBArF and CoCp2, respectively.
the SI for details) were carried out to determine the electronic
structure of 1-Y(OtBu) and [1-Y(OtBu)][BArF]. A pre-edge
peak characteristic for ferrocenium salts was observed in the
XANES spectra for [1-Y(OtBu)][BArF], but not for 1-Y(OtBu).
The M€ossbauer spectrum of [1-Y(OtBu)][BArF] exhibits two
singlets at δ = 0.29(1) and 0.70(1) mm sꢀ1. The spectrum is
consistent with the presence of iron(III) (the two singlets may be
due to a difference in packing of the borate salt or solvent in the
crystals). The M€ossbauer spectrum of 1-Y(OtBu) is similar to
Figure 1. Thermal-ellipsoid (50% probability) representation of
[1-Y(OtBu)][BArF]; hydrogen and anion atoms were removed for
clarity.
that of ferrocene and exhibits a doublet at δ = 0.52(1) mm sꢀ1
.
The data from XANES and M€ossbauer spectroscopy indicate
that [1-Y(OtBu)][BArF] contains an iron(III), while 1-Y(OtBu)
contains an iron(II) as part of the ligand backbone.
Once the ferrocene and ferrocenium-based compounds were
characterized, the polymerization of L-lactide with 1-Y(OtBu)
and [1-Y(OtBu)][BArF] was attempted since yttrium complexes
are known catalysts for this reaction (Table 1).17ꢀ22 At room
temperature in tetrahydrofuran, 1-Y(OtBu) polymerizes 100
equiv of L-lactide in 3 h with 74% conversion, while no conver-
sion was observed in the presence of [1-Y(OtBu)][BArF] un-
der the same conditions. In addition to polymerizing L-lactide,
1-Y(OtBu) polymerized 100 equiv of rac-lactide in tetrahydro-
furan at 70 °C. After 8 h, the polymerization reached 95%
conversion, but no tacticity control was observed (see the SI
for details).
Figure 2. XANES spectra of the iron K-edge for ferrocene, FcBArF,
[1-Y(OtBu)][BArF], and 1-Y(OtBu).
Table 1. GPC Data for the Polymerization of L-Lactide
(0.2 M in Tetrahydrofuran) with 1-Y(OtBu) at Room
Temperature
a,b
equiv
time
(h)
con-
Mn
complex
of LLA
version
(kg/mol)
PDI
Switching in situ between the oxidized and reduced forms of the
yttrium complex was performed during the polymerization of 100
equiv of L-lactide (L-LA) in tetrahydrofuran at room temperature
(Figure 3). After 1 h, the polymerization with 1-Y(OtBu) reached
24% conversion. Upon the oxidation of 1-Y(OtBu) with FcBArF,
the polymerization halted. Once CoCp2 was added to the reaction
mixture, the polymerization resumed with the same rate as that
before the switch was performed (see the SI for details). In situ
switchingwas performed three consecutive times;itwas found that
there was minimal change in the rate of the reaction before or after
changing the iron oxidation states (see the SI for details).
As detailed above, the catalyst containing the oxidized ligand
decreases the rate of the polymerization reaction compared to the
catalyst containing the reduced ligand. The same observation was
made by the Gibson group when using a remote ferrocenyl unit.7
The lack of activity of the oxidized catalyst was investigated by
carrying out a stoichiometric reaction of [1-Y(OtBu)][BArF] with
L-LA. The addition of 1 equiv of L-LA to a diethyl ether solution of
[1-Y(OtBu)][BArF] led to the formation of a new species, which
was unreactive toward further L-LA polymerization. This species
was reduced with CoCp2 to produce a diamagnetic product. We
1-Y(OtBu)
100
200
300
500
100
3
8
74%
76%
70%
64%
51%
51%
87%
7.5
17.6
28.2
39.5
5.4
1.07
1.03
1.04
1.08
1.05
1.04
1.06
10
47
2
1-Y(OtBu)
add FcBArF
add CoCp2
0.5
4
5.4
7.6
a Samples were run in tetrahydrofuran with polystyrene standards. b A
MarkꢀHouwink factor23 of 0.58 was applied to the experimental
molecular weights.
1
The H NMR spectrum of the product was consistent with the
presence of a paramagnetic species, as expected. [1-Y(OtBu)]-
[BArF] could be reduced to 1-Y(OtBu) within minutes in
tetrahydrofuran through the addition of 1 equiv of CoCp2 (see
the Supporting Information (SI) for details).
X-ray quality crystals of [1-Y(OtBu)][BArF] were grown from
a concentrated diethyl ether solution layered with n-pentane
at ꢀ36 °C (Figure 1). X-ray absorption near edge structure
(XANES) (Figure 2) and M€ossbauer spectroscopic studies (see
9279
dx.doi.org/10.1021/ja2036089 |J. Am. Chem. Soc. 2011, 133, 9278–9281

Journal of the American Chemical Society
COMMUNICATION
1
Scheme 1. Synthesis of 1-In(OPh)
after minutes of stirring. [1-In(OPh)][BArF] had a H NMR
spectrum consistent with that of a paramagnetic species and could
be reduced with 1 equiv of CoCp2 to produce 1-In(OPh) (see the
SI for details). The M€ossbauer spectra for the indium complexes
display a doublet for 1-In(OPh), which is similar to the spectrum
for ferrocene, and a singlet for [1-In(OPh)][BArF], a character-
istic of an oxidized iron center. The M€ossbauer spectra support
the assignment that [1-In(OPh)][BArF] contains iron(III), while
1-In(OPh) contains iron(II) as part of the ligand backbone.
With 1-In(OPh) and [1-In(OPh)][BArF] in hand, ring-open-
ing polymerizations of cyclic esters were performed. The polymer-
izations of ε-caprolactone and L-lactide were slow with both 1-
In(OPh) and [1-In(OPh)][BArF]. As a result, the polymerization
of 100 equiv of trimethylene carbonate was carried out at room
temperature, in benzene-d6. After 1 day, 1-In(OPh) reached 2%
conversion, while [1-In(OPh)][BArF] reached 49% conversion
under the same conditions. These results suggest that withdrawing
electrons from indium through the oxidation of the ligand backbone
increased the rate of polymerization of trimethylene carbonate.
For comparison, at room temperature, 1-Y(OtBu) and [1-
Y(OtBu)][BArF] polymerized 100 equiv of trimethylene carbo-
nate within minutes in benzene-d6. When the polymerizations were
performed at ꢀ78 °C in THF, 1-Y(OtBu) transformed 29% of the
starting material, while [1-Y(OtBu)][BArF] reached only 3.5%
conversion. Therefore, for the yttrium complexes, the rate of
polymerization decreased when the complex containing ferroce-
nium was employed; these results are the opposite of those
observed for indium. The difference in reactivity between the
yttrium and indium complexes suggests that the change in reactivity
observed as a result of oxidizing the ferrocene backbone is metal
dependent. We hypothesize that, in the case of yttrium, although
substrate coordination may be more favorable with the more
electrophilic metal center ([1-Y(OtBu)][BArF]), the activation
barriers for alkoxide migration and ring opening are higher than for
the less electrophilic metal center (1-Y(OtBu)). In the case of
indium, it is possible that the effect is reversed; however, one study
with aluminum also indicates that electron-withdrawing substitu-
ents can decrease polymerization activity.37 These data suggest that
the explanation for the change in the rate of polymerization due to
the electron-donating or -withdrawing abilities of a ligand may not
be general. In-depth DFT calculations could shed light on this
proposal and will be reported in due course.
hypothesize that the 1H NMR spectrum of this product corresponds
to the insertion of one lactide molecule (see the SI for details);
further studies will be directed to the isolation and full characteriza-
tion of this intermediate. This observation suggests that the insertion
of the monomer is possible, but propagation with this new species is
unfavorable. Single crystals suitable for X-ray crystallography could
not be obtained.
The polymers obtained from the reactions with 1-Y(OtBu)
were characterized by gel permeation chromatography (GPC).
The molecular weights correlate well with the corresponding
theoretical molecular weights, and the PDIs are ca. 1.1; these data
indicate a controlled polymerization process. 1H NMR spectros-
copy of a polymer formed from the reaction of 1-Y(OtBu) and 50
equiv of L-lactide displayed tert-butoxide end groups, which
suggests that the monomer inserts into the yttrium-alkoxide bond
(see the SI for details). GPC analysis of the polymers produced by
switching in situ between 1-Y(OtBu) and [1-Y(OtBu)][BArF],
once the reaction had reached 51% conversion, shows that the
polymerization is also controlled when using redox agents, since
the molecular weights correlate well with corresponding theore-
tical molecular weights, and the PDIs are ca. 1.1 (Table 1).
As mentioned above, our results indicate that the oxidation of
ferrocene caused a decrease in the reaction rate compared to the
reduced form of the metal complex; a similar result was reported
by Gibson et al.7 We became interested in determining whether
the reverse effect (i.e., increased reactivity with a ferrocenium-
containing metal complex) is possible. Reports of using electron-
withdrawing groups to increase the rate of polymerization for
aluminum complexes24ꢀ28 indicated that a group 13 analogue of
1-Y(OtBu) should be targeted.
In conclusion, we described the first example of redox control
with a catalyst containing ferrocene in the proximity of a metal
center. This arrangement may allow tuning the reactivity of the
oxidized and reduced forms of a metal complex. The rate of ring-
opening polymerization of L-lactide with an yttrium alkoxide
phosfen catalyst was altered by switching in situ the redox state of
the ferrocene moiety. In addition, the first experimental evidence
of increasing the activity of a metal complex during the ring-
opening polymerization of a cyclic ester by oxidizing the ferro-
cene substituent is presented. The results discussed in this paper
are an important step toward the ultimate goal of achieving
substrate selectivity through redox-switchable catalysis.
In order to ensure the formation of a mononuclear species, a
larger metal center than aluminum, indium (ionic radius 0.80 Å),29
was chosen; such complexes would be analogous to the yttrium
(ionic radius 0.90 Å) complexes 1-Y(OtBu) and [1-Y(OtBu)]-
[BArF]. Also, indium complexes are known catalysts for the
ring-opening polymerization of lactide.30ꢀ36 1-In(OPh) was
synthesized similarly to 1-Y(OtBu): the salt metathesis reaction
between InCl3(THF)2 and [Na(THF)]2[phosfen] (1-Na2-
(THF)2) led to the formation of InCl(phosfen) (1-InCl), which
underwent a second salt metathesis reaction with sodium phen-
oxide (NaOPh) to give 1-In(OPh) (Scheme 1). The oxidized
complex [1-In(OPh)][BArF] was obtained upon the addition of1
equiv of FcBArF to 1-In(OPh) at room temperature in toluene,
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures, NMR
b
spectroscopy, M€ossbauer spectroscopy, and X-ray crystallogra-
phy data. This material is available free of charge via the Internet
at http://pubs.acs.org.
9280
dx.doi.org/10.1021/ja2036089 |J. Am. Chem. Soc. 2011, 133, 9278–9281

Journal of the American Chemical Society
COMMUNICATION
’ AUTHOR INFORMATION
(27) Pang, X.; Du, H.; Chen, X.; Wang, X.; Jing, X. Chem.—Eur.
J. 2008, 14, 3126.
(28) Gong, S.; Ma, H. Dalton Trans. 2008, 3345.
(29) Shannon, R. Acta Crystallogr. A. 1976, 32, 751.
(30) Douglas, A. F.; Patrick, B. O.; Mehrkhodavandi, P. Angew.
Chem., Int. Ed. 2008, 47, 2290.
Corresponding Author
pld@chem.ucla.edu
(31) Pietrangelo, A.; Hillmyer, M. A.; Tolman, W. B. Chem. Commun.
2009, 19, 2736.
’ ACKNOWLEDGMENT
This work was supported by the UCLA, DOE (Grant ER15984),
Sloan Foundation, the University of Erlangen-Nuremburg, DFG,
and the Bavarian California Technology Center (BaCaTec). Use of
the Advanced Photon Source was supported by the U.S. Depart-
ment of Energy, Office of Science, Office of Basic Energy Sciences,
under Contract No. DE-AC02-06CH11357. MRCAT (Sector 10)
operations are supported by the Department of Energy and the
MRCAT member institutions.
(32) Peckermann, I.; Kapelski, A.; Spaniol, T. P.; Okuda, J. Inorg.
Chem. 2009, 48, 5526.
(33) Buffet, J. C.; Okuda, J.; Arnold, P. L. Inorg. Chem. 2010, 49, 419.
(34) Pietrangelo, A.; Knight, S. C.; Gupta, A. K.; Yao, L. J.; Hillmyer,
M. A.; Tolman, W. B. J. Am. Chem. Soc. 2010, 132, 11649.
(35) Blake, M.; Schwarz, A. D.; Mountford, P. Organometallics 2011,
30, 1202.
(36) Acosta Ramírez, J. A.; Douglas, A. F.; Yu, I.; Patrick, B. O.;
Diaconescu, P. L.; Mehrkhodavandi, P. Inorg. Chem. 2010, 49, 5444.
(37) Alcazar-Roman, L. M.; O’Keefe, B. J.; Hillmyer, M. A.; Tolman,
W. B. Dalton Trans. 2003, 3082.
’ REFERENCES
(1) Lorkovic, I. M.; Duff, R. R., Jr.; Wrighton, M. S. J. Am. Chem. Soc.
1995, 117, 3617.
(2) Hembre, R. T.; McQueen, J. S.; Day, V. W. J. Am. Chem. Soc.
1996, 118, 798.
(3) Slone, C. S.; Mirkin, C. A.; Yap, G. P. A.; Guzei, I. A.; Rheingold,
A. L. J. Am. Chem. Soc. 1997, 119, 10743.
(4) Allgeier, A. M.; Mirkin, C. A. Angew. Chem., Int. Ed. 1998, 37, 894.
(5) Sussner, M.; Plenio, H. Angew. Chem., Int. Ed. 2005, 44, 6885.
(6) Fujiwara, M.; Terashima, S.; Endo, Y.; Shiokawa, K.; Ohue, H.
Chem. Commun 2006, 4635.
(7) Gregson, C. K. A.; Gibson, V. C.; Long, N. J.; Marshall, E. L.;
Oxford, P. J.; White, A. J. P. J. Am. Chem. Soc. 2006, 128, 7410.
(8) Ringenberg, M. R.; Kokatam, S. L.; Heiden, Z. M.; Rauchfuss,
T. B. J. Am. Chem. Soc. 2008, 130, 788.
(9) Liu, G.; He, H.; Wang, J. Adv. Synth. Catal. 2009, 351, 1610.
(10) Tennyson, A. G.; Lynch, V. M.; Bielawski, C. W. J. Am. Chem.
Soc. 2010, 132, 9420.
(11) Yoon, H. J.; Kuwabara, J.; Kim, J. H.; Mirkin, C. A. Science 2010,
330, 66.
(12) Peeck, L. H.; Leuthausser, H.; Plenio, H. Organometallics 2010,
29, 4339.
(13) Kwak, Y.; Matyjaszewski, K. Macromolecules 2010, 43, 5180.
(14) Tanabe, M.; Vandermeulen, G. W. M.; Chan, W. Y.; Cyr, P. W.;
Vanderark, L.; Rider, D. A.; Manners, I. Nat. Mater. 2006, 5, 467.
(15) Magenau, A. J. D.; Strandwitz, N. C.; Gennaro, A.; Matyjaszewski,
K. Science 2011, 332, 81.
(16) Broderick, E. M.; Thuy-Boun, P. S.; Guo, N.; Vogel, C. S.;
Sutter, J.; Miller, J. T.; Meyer, K.; Diaconescu, P. L. Inorg. Chem. 2011,
50, 2870.
(17) Broderick, E. M.; Diaconescu, P. L. Inorg. Chem. 2009, 48, 4701.
(18) Bouyahyi, M.; Ajellal, N.; Kirillov, E.; Thomas, C. M.; Carpentier,
J. F. Chem.—Eur. J. 2011, 17, 1872.
(19) Zhang, F.; Zhang, J.; Song, H.; Zi, G. Inorg. Chem. Commun.
2011, 14, 72.
(20) Luo, Y.; Li, W.; Lin, D.; Yao, Y.; Zhang, Y.; Shen, Q. Organo-
metallics 2010, 29, 3507.
(21) Platel, R. H.; White, A. J. P.; Williams, C. K. Chem. Commun
2009, 27, 4115.
(22) Arnold, P. L.; Buffet, J. C.; Blaudeck, R. P.; Sujecki, S.; Blake,
A. J.; Wilson, C. Angew. Chem., Int. Ed. 2008, 47, 6033.
(23) Save, M.; Schappacher, M.; Soum, A. Macromol. Chem. Phys.
2002, 203, 889.
(24) Cameron, P. A.; Jhurry, D.; Gibson, V. C.; White, A. J. P.;
Williams, D. J.; Williams, S. Macromol. Rapid Commun. 1999, 20, 616.
(25) Hormnirun, P.; Marshall, E. L.; Gibson, V. C.; Pugh, R., I.;
White, A. J. P. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15343.
(26) Qian, F.; Liu, K.; Ma, H. J. Chem. Soc., Dalton Trans. 2010,
39, 8071.
9281
dx.doi.org/10.1021/ja2036089 |J. Am. Chem. Soc. 2011, 133, 9278–9281