RETRACTED ARTICLE: Mechanical activation of catalysts for C-C bond forming and anionic polymerization reactions from a single macromolecular reagent

Article abstract of DOI:10.1021/ja107620y

Coupling of pyridine-capped poly(methyl acrylate)s, PyP_M (where M corresponds to the number average molecular weight in kDa), to the SCS-cyclometalated dipalladium complex [(1)(CH₃CN)₂] afforded organometallic polymers [(1)(PyP_M)₂] with a concomitant doubling in molecular weight. Ultrasonication of solutions containing [(1)(PyP_M)₂] effected the mechanical scission of a palladium-pyridine bond, where the liberated PyP_M was trapped with excess HBF₄ as the corresponding pyridinium salt, harnessed to effect the stoichiometric deprotonation of a colorimetric indicator, or used to catalyze the anionic polymerization of α-trifluoromethyl-2,2,2- trifluoroethyl acrylate. The mechanically induced chain scission also unmasked a catalytically active palladium species which was used to facilitate carbon-carbon bond formation between benzyl cyanide and N-tosyl imines. Spectroscopic and macromolecular analyses as well as a series of control experiments demonstrated that the aforementioned structural changes were derived from mechanical forces that originated from ultrasound-induced dissociation of the polymer chains connected to the aforementioned Pd complexes.

Full text of DOI:10.1021/ja107620y

Published on Web 11/02/2010
Mechanical Activation of Catalysts for C-C Bond Forming
and Anionic Polymerization Reactions from a Single
Macromolecular Reagent
Andrew G. Tennyson, Kelly M. Wiggins, and Christopher W. Bielawski*
Department of Chemistry and Biochemistry, The UniVersity of Texas at Austin, Austin,
Texas 78712, United States
Received August 23, 2010; E-mail: bielawski@cm.utexas.edu
Abstract: Coupling of pyridine-capped poly(methyl acrylate)s, PyPM (where M corresponds to the number
average molecular weight in kDa), to the SCS-cyclometalated dipalladium complex [(1)(CH3CN)2] afforded
organometallic polymers [(1)(PyPM)2] with a concomitant doubling in molecular weight. Ultrasonication of
solutions containing [(1)(PyPM)2] effected the mechanical scission of a palladium-pyridine bond, where
the liberated PyPM was trapped with excess HBF4 as the corresponding pyridinium salt, harnessed to effect
the stoichiometric deprotonation of a colorimetric indicator, or used to catalyze the anionic polymerization
of R-trifluoromethyl-2,2,2-trifluoroethyl acrylate. The mechanically induced chain scission also unmasked
a catalytically active palladium species which was used to facilitate carbon-carbon bond formation between
benzyl cyanide and N-tosyl imines. Spectroscopic and macromolecular analyses as well as a series of
control experiments demonstrated that the aforementioned structural changes were derived from mechanical
forces that originated from ultrasound-induced dissociation of the polymer chains connected to the
aforementioned Pd complexes.
Introduction
recent efforts by Sijbesma have highlighted the promise for using
sonochemically induced changes in polymer structure to drive
Ultrasonication has recently received significant attention for
its ability to alter the chemical structures, properties, and
functions of polymeric materials via mechanical force, a process
termed “mechanochemistry”.1,2 During the application of ul-
trasonic energy to a solution of polymers, bubbles are created
and rapidly collapse which induces velocity gradients and
attendant stresses to the solvated polymer chains.3 For macro-
molecules of sufficient chain length, this mechanical force may
effect chain scission4 or isomerization5 at centrally located sites.1
Because mechanochemistry ultimately enables access to ther-
mally inaccessible pathways,4,5 there has been substantial interest
in developing mechanically activated catalysts. For example,
catalytic olefin metathesis and transesterification reactions.6
More broadly, mechanocatalysts4b are anticipated to uncover
fundamentally new chemistries, enable transformations not
currently feasible,7 and endow stimulus-responsive materials
with novel functions and applications.8 Achieving many of these
goals, however, hinges on using mechanical force to access new
reaction pathways, particularly those that are thermally inac-
cessible or prohibited.
To guide the design of such mechanically activated catalysts,
we first considered the components necessary for a mechano-
responsive material:1 (i) a mechanophore,1,4,5 which undergoes
an electronic or structural change in response to mechanical
force and (ii) an actuator, which translates exogenously supplied
energy into useful mechanical force at the mechanophore.9
Polymers have been extensively employed as actuators, given
that they respond to mechanical stress at both the macro- and
microscopic level and can be covalently linked to small
(1) Caruso, M. M.; Davis, D. A.; Shen, Q.; Odom, S. A.; Sottos, N. R.;
White, S. R.; Moore, J. S. Chem. ReV. 2009, 109, 5755.
(2) Mason, T. J.; Lorimer, J. P. Applied Sonochemistry; Wiley-VCH: New
York, 2002.
(3) Basedow, A. M.; Ebert, K. H. AdV. Polym. Sci. 1977, 22, 83.
(4) (a) Berkowski, K. L.; Potisek, S. L.; Hickenboth, C. R.; Moore, J. S.
Macromolecules 2005, 38, 8975. (b) Karthikeyan, S.; Potisek, S. L.;
Piermattei, A.; Sijbesma, R. P. J. Am. Chem. Soc. 2008, 130, 14968.
(c) Kryger, M. J.; Ong, M. T.; Odom, S. A.; Sottos, N. R.; White,
S. R.; Martinez, T. J.; Moore, J. S. J. Am. Chem. Soc. 2010, 132,
4558.
(6) Piermattei, A.; Karthikeyan, S.; Sijbesma, R. P. Nature Chem. 2009,
1, 133.
(7) (a) Anastas, P. T.; Kirchhoff, M. M.; Williamson, T. C. Appl. Catal.,
A 2008, 221, 3. (b) Madhavan, N.; Jones, C. W.; Weck, M. Acc. Chem.
Res. 2008, 41, 1153.
(5) (a) Hickenboth, C. R.; Moore, J. S.; White, S. R.; Sottos, N. R.; Baudry,
J.; Wilson, S. R. Nature 2007, 446, 423. (b) Potisek, S. L.; Davis,
D. A.; Sottos, N. R.; White, S. R.; Moore, J. S. J. Am. Chem. Soc.
2007, 129, 13808. (c) Davis, D. A.; Hamilton, A.; Yang, J.; Cremar,
L. D.; Van Gough, D.; Potisek, S. L.; Ong, M. T.; Braun, P. V.;
Martinez, T. J.; White, S. R.; Moore, J. S.; Sottos, N. R. Nature 2009,
459, 68. (d) Lenhardt, J. M.; Black, A. L.; Craig, S. L. J. Am. Chem.
Soc. 2009, 131, 10818. (e) Wiggins, K. M.; Hudnall, T. W.; Shen,
Q.; Kryger, M. J.; Moore, J. S.; Bielawski, C. W. J. Am. Chem. Soc.
2010, 132, 3256. (f) Lenhardt, J. M.; Ong, M. T.; Choe, R.; Evenhuis,
C. R.; Martinez, T. J.; Craig, S. L. Science 2010, 329, 1057.
(8) (a) White, S. R.; Sottos, N. R.; Geubelle, P. H.; Moore, J. S.; Kessler,
M. R.; Sriram, S. R.; Brown, E. M.; Viswanathan, S. Nature 2001,
409, 794. (b) Xiangxu, C.; Dam, M. A.; Ono, K.; Mal, A.; Shen, H.;
Nutt, S. R.; Sheran, S. R.; Wudl, F. Science 2002, 295, 1698. (c) White,
S. R.; Caruso, M. M.; Moore, J. S. MRS Bull. 2008, 33, 766. (f) Wool,
R. P. Soft Mater. 2008, 4, 400. (d) Williams, K. A.; Dreyer, D. R.;
Bielawski, C. W. MRS Bull. 2008, 33, 759. (e) Kolmakov, G. V.;
Matyjaszewski, K.; Balazs, A. C. ACS Nano 2009, 3, 885.
(9) Luche, J. L.; Einhorn, J.; Einhron, J. J.; Sinisterra-Gago, J. V.
Tetrahedron Lett. 1990, 31, 4125.
9
10.1021/ja107620y  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 16631–16636 16631

A R T I C L E S
Tennyson et al.
molecules.1 Conducive polymer-based actuators are those that
are viscoelastic and can be obtained by synthetic methods which
enable accurate prediction and control of molecular weight.1
Poly(methyl acrylate) (PMA), in particular, exhibits the desired
macromolecular and mechanical properties and can be prepared
in a wide range of predetermined molecular weights using the
SET-LRP method developed by Percec.10,11 Moreover, actuators
comprising PMA chains have been successfully utilized by
Moore,4c,5a-c and more recently by us,5e to exert force upon
mechanophores and to induce structural as well as functional
changes therein. Finally, because a complete catalytic cycle
necessitates both substrate binding and product release, a catalyst
generated in response to ultrasonically induced mechanical force
must retain this critical function.
with 0.5 equiv of [(1)(CH3CN)2] in DMF at room temperature
for 16 h resulted in a doubling of Mn, as determined by gel
permeation chromatography (GPC), that was consistent with the
facile displacement of CH3CN by pyridine-based ligands to
afford [(1)(PyPM)2] (step i, Figure 1; results summarized in
Table 1). Similarly, treatment of [(2)(CH3CN)] with PyP110
produced the end-capped derivative [(2)(PyP110)]. Additional
support for the binding of PyPM to Pd in [(1)(PyPM)2] and
[(2)(PyP110)] was obtained via 1H NMR spectroscopic analysis,
wherein the signals for the 2,6-protons in the pyridyl moieties
shifted upfield from 8.57 ppm (CDCl3) in free PyPM to 8.23
ppm in the Pd complexes, reflecting their proximity to arene
ring currents upon coordination (see Figures S1-S3, Supporting
Information).
Guided by these considerations, we concluded that a transition
metal complex, when coordinated to an actuating ligand, would
serve as an ideal mechanophore for achieving mechanically
activated catalytic functions. Pioneering studies by Weck12 and
Craig13 have demonstrated that certain SCS-, NCN-, and PCP-
cyclometalated palladium(II) complexes exhibit reversible
metal-pyridine interactions, even when both components are
appended with polymer chains. Moreover, these organometallic
systems have been shown to participate in the catalytic formation
of carbon-carbon bonds (e.g., Heck-type coupling), albeit under
thermal control. Inspired by these results, we reasoned that a
macromolecular complex, comprising a palladium(II) center
ligated by a pyridine-PMA conjugate, should exhibit sonochem-
ically induced structural dynamism that would be accompanied
by activation of metal- as well as base-mediated catalytic
reactions in response to applied mechanical force. Herein we
demonstrate the viability of such a mechanophore-actuator
construct to function as a mechanocatalyst, whereby ultrasoni-
cation liberates two catalytically active functionalities that effect
two orthogonal chemical reactivities: Pd-catalyzed C-C cou-
pling and base-catalyzed, anionic polymerization reactions. To
the best of our knowledge, our findings represent the first
demonstration of the activation of such catalytic reactions in
response to mechanical force and establish a foundation for
achieving the aforementioned goals.
Because the mechanophore in [(1)(PyPM)2] occupies the
center of the polymer chain, sonication of solutions thereof was
anticipated to afford [(1)(PyPM)(CH3CN)] and PyPM via Pd-
pyridine bond scission (step ii, Figure 1).15 Upon cessation of
the sonication, however, these species were found to recombine
in solution with no net effect on Mn (step iii, cf. Figure S7,
Supporting Information). Therefore, we selected HBF4 to trap
any liberated PyPM as its corresponding pyridinium tetrafluo-
roborate salt [HPyPM][BF4] (step iv). Sonication of CH3CN
solutions containing [(1)(PyPM)2] (where M ) 21, 25, and 66)
and excess HBF4 in a Suslick16 cell at 4 °C for 2 h afforded a
halving of the polymer’s Mn (see Figure 2A for M ) 66),
consistent with the anticipated cleavage of a single Pd-N bond.
Control experiments performed in the absence of sonication (at
both 4 and 25 °C) revealed that the polymers analyzed displayed
unaltered Mn values, indicating that the HBF4 did not cause chain
scission via protonation of the coordinated pyridine moieties.
Likewise, no change in the Mn was observed for the [(1)(PyPM)2]
containing shorter chains (M ) 7 and 11), consistent with the
minimum chain-length threshold requirement for a polymer to
experience sonication-induced tensile force. The Mn value
measured for [(2)(PyP110)] was also unaffected by sonication,
suggesting that the location of the scissile bond at the end of a
polymer chain precluded application of vectorially opposed
mechanomotive force on the Pd-pyridine bond.
To study the kinetics of the chain scission process, the
decrease in the GPC refractive index (RI) signal attributed to
[(1)(PyPM)2] was monitored by withdrawing aliquots from the
reaction vessel at timed intervals during a typical sonication
experiment.15 Analysis of the aliquots from the sonication of
[(1)(PyP66)2] revealed a gradual decrease in Mn from 120 to 60
kDa that reached completion within 2 h (Figure 2B). Plotting
-ln(I/I0) vs time, where I0 and I correspond to the RI signal
for [(1)(PyP66)2] before sonication and at each time-point,
respectively, enabled determination of the first-order rate of 3.6
× 10-2 min-1 (black line, Figure 2C). Both [(1)(PyP21)2] and
Results and Discussion
To access the desired mechanoresponsive reagent, we first
synthesized the requisite pyridine-capped PMA-based actuators
(PyPM, Figure 1) of varying molecular weights (where M
corresponds to the number average molecular weight (Mn) in
kDa of the PMA chain PM; see Table 1) via SET-LRP of methyl
acrylate10 using 4-pyridinyl-2-bromoisobutyrate as the initiator.
In parallel, the SCS-cyclometalated dipalladium(II) salt
[(1)(CH3CN)2] was selected as a complementary mechanophore
and was prepared using established methods.14 Combining PyPM
[(1)(PyP25)2] exhibited slower rates (1.0 and 1.5 × 10-2 min-1
,
respectively), reflecting the decreased efficiency of cavitation-
induced scission for these shorter-chain polymers (blue and red
lines, respectively, Figure 2D). Consistent with the aforemen-
tioned bulk GPC analyses, neither [(1)(PyP7)2] nor [(1)(PyP11)2]
(10) Percec, V.; Guliashvili, T.; Ladislaw, J. S.; Wistrand, A.; Stjerndahl,
A.; Sienkowska, M. J.; Monteiro, M. J.; Sahoo, S. J. Am. Chem. Soc.
2006, 128, 14156.
(11) For examples of ultrasound-induced chain scission in poly(tetrahy-
drofuran)s containing coordinative Pd-P and Pt-P bonds, see: (a)
Paulusse, J. M. J.; Sibjesma, R. P. Angew. Chem., Int. Ed. 2004, 43,
4459. (b) Paulusse, J. M. J.; Huijbers, J. P. J.; Sibjesma, R. P.
Chem.sEur. J. 2006, 12, 4928. (c) Paulusse, J. M. J.; Sibjesma, R. P.
Chem. Commun. 2008, 37, 4416.
(15) General sonication conditions: pulsed ultrasound (1.0 s on, 1.0 s off)
was supplied at 23% power (10.1 W cm-2) for 2 h. All sonications
were performed on solutions with a final volume of 10 mL in a Suslick
cell under a positive pressure of argon. The external temperature was
maintained at 4 °C aided by the use of a cold room and ice-water
bath. The internal temperatures of the sonicated solutions were
monitored via a thermocouple and found to be e9 °C.
(12) (a) South, C. R.; Burd, C.; Weck, M. Acc. Chem. Res. 2007, 40, 63.
(b) Yang, S. K.; Ambade, A. V.; Weck, M. Chem.sEur. J. 2009, 15,
6605.
(13) (a) Loveless, D. M.; Jeon, S. L.; Craig, S. L. J. Mater. Chem. 2007,
17, 56. (b) Xu, D.; Craig, S. L. J. Phys. Chem. Lett. 2010, 1, 1683.
(14) Loeb, S. J.; Shimizu, G. K. H. Chem. Commun. 1999, 1395.
(16) Suslick, K. S.; Goodale, J. W.; Schubert, P. F.; Wang, H. H. J. Am.
Chem. Soc. 1983, 105, 5781.
9
16632 J. AM. CHEM. SOC. VOL. 132, NO. 46, 2010

Mechanical Activation of Catalysts
A R T I C L E S
Figure 1. Synthesis and chain scission of mechanoresponsive Pd-based polymers. Conditions: (i) DMF, room temperature, 16 h; (ii) sonication for 2 h at
4 °C of [(1)(PyPM)2] (10 mg) in CH3CN (10 mL), and (iii) recoordination of PyPM or (iv) trapping as [HPyPM][BF4] with 20 equiv of HBF4.
Table 1. Summary of Molecular Weight Data for Chain Coupling
exhibited any change in RI signal (overlapping purple and
orange lines, respectively; Figure 2D). Collectively, these
results indicate that (i) a minimum chain length of ap-
and Scission Experimentsa
[(1)(PyPM)2]
proximately 40 kDa was required for chain scission to occur
and (ii) increasing chain length beyond this threshold
correlated with an increased rate of chain scission.
We sought to gain greater insight into the chemical process
underlying the observed chain scission by employing a base-
sensitive dye that would display a colorimetric response upon
exposure to free pyridine. To perform this task, 4-[(4-anili-
nophenyl)azo]benzenesulfonic acid (3) was selected given its
PyPM
Mn [kDa]
presonication
Mn [kDa] PDI
postsonication
Mn [kDa] PDI
PDI
P7
6.8
11
21
25
66
1.2
1.2
1.2
1.4
1.4
12
22
1.1
1.2
1.1
1.3
1.4
12
23
20
27
60
1.1
1.2
1.3
1.4
1.4
P11
P21
P25
P66
41
48
120
a Number average and weight average molecular weights (Mn and
Mw, respectively) were determined by GPC (eluent ) DMF, 0.1 M
LiBr) and are reported relative to polystyrene standards. Polydispersity
indices (PDIs) were calculated using the equation PDI ) Mw/Mn.
(17) Kolthoff, I. M.; Chantooni, M. K., Jr.; Bhowmik, S. Anal. Chem. 1967,
39, 315.
9
J. AM. CHEM. SOC. VOL. 132, NO. 46, 2010 16633

A R T I C L E S
Tennyson et al.
Figure 2. Representative examples of GPC traces and kinetic plots for the mechanochemical scission of [(1)(PyPM)2]. (A) GPC traces of [(1)(PyP66)2]
before and after sonication (red and blue, respectively) overlaid with PyP66 for comparison (black). (B) GPC data were acquired at timed intervals by
analyzing aliquots removed during the sonication of [(1)(PyP66)2] (t ) 0, topmost red); the gradual decrease in the peak intensity measured at retention time
)19.13 min (t ) 2-8 min, fading red) was accompanied by the appearance of a new peak measured at retention time ) 19.89 min that was complete after
2 h (t ) 16-120 min, darkening blue). All aliquots analyzed consisted of an initial polymer concentration of 4 mg mL-1. (C) First-order rate plot for the
sonication of [(1)(PyPM)2], where M ) 66 (black), 25 (red), 21 (blue), 11 (orange), and 7 (purple). I0 and I correspond to the RI signal prior to sonication
and at each time-point t thereafter, respectively. The RI intensity for each polymer was taken at the peak retention times (see Table S2, Supporting Information).
(D) Plot of rate coefficients determined in C as a function of Mn. All data points and error bars were calculated from the average and standard deviation,
respectively, of three separate experiments.
demonstrated viability17 for pH indication in CH3CN and the
detection of organic nitrogen bases.18 In CH3CN solution,
protonation of 3 by stoichiometric HBF4 afforded [3H][BF4]
which exhibited a deep purple color. Gratifyingly, subsequent
titration with 1 equiv of pyridine caused an immediate color
change to yellow, consistent with the reformation of 3 (Figure
3A), and indicated that the system would be viable for detecting
free PyPM generated under sonication. Thus, we envisioned that
a solution containing [(1)(PyPM)2] and [3H][BF4] would undergo
a similar color change upon sonication, reflecting the release
of PyPM upon cleavage of the Pd-pyridine bond.
In a typical experiment for the colorimetric detection of free
PyPM, a purple solution containing 10 µM [(1)(PyPM)2] and
10 µM [3H][BF4] in 10 mL of CH3CN was prepared and
transferred to a Suslick cell, whereupon chain scission was
induced using the aforementioned sonication conditions.15
Sonicating these polymers in the presence of [3H][BF4] afforded
a clear, yellow solution, consistent with its deprotonation by
free PyPM, and in good agreement with the reductions in
polymer molecular weight observed by GPC for [(1)(PyPM)2]
(M ) 21, 25, and 66) (Figure 3B). Subsequent addition of 1
equiv of HBF4 restored the solution to its original purple color,
evidence that sonication did not degrade the dye and generated
1 equiv of base. Because no changes in the optical or GPC
profiles were noted in the absence of sonication, we concluded
that thermally induced or proton-facilitated ligand displacement
did not occur. As negative control experiments, the aforemen-
tioned indicator experiments were performed via sonication of
(i) “sub-threshold” [(1)(PyP7)2] and [(1)(PyP11)2] (variants
shorter than the critical threshold needed to experience sonochem-
ical tensile force) to confirm that mechanical activation derived
from cavitation (for M ) 11, see Figure 3C); (ii) end-capped
[(2)(PyP110)], to demonstrate that an internally located mecha-
nophore was necessary for Pd-pyridine bond cleavage to occur
(Figure 3D); and (iii) mechanophore model complex
[(1)(PyPiv)2] (see Figure 1) in the presence of 90 kDa PMA,
to test whether the released base occurs only when the
mechanophore was covalently linked to PMA actuators (Figure
3E). No color change was observed during any of these control
experiments. Collectively, these results confirm that chain
scission in [(1)(PyPM)2] occurred at the Pd-pyridine bond to
liberate free PyPM and was only effected when the criteria for
delivery of ultrasonically induced mechanical force were
satisfied.
Considering that the sonication of [(1)(PyPM)2] generated
[(1)(PyPM)(CH3CN)] and PyPM in situ, we envisioned exploiting
these species for use in both metal- and base-catalyzed reactions.
Elegant studies by Weck and Jones have shown that polymer-
and silica-supported derivatives of [(2)(CH3CN)] are capable
of facilitating Heck-type coupling chemistry, albeit at 120 °C
in DMF, and only upon reductive generation and release of
zerovalent Pd.19 Given the success of these SCS-cyclopalladated
systems at catalyzing carbon-carbon bond formation reactions,
we envisioned that [(1)(PyPM)2] could catalyze a comparably
illustrative coupling reaction under sonochemical conditions (i.e.,
nonthermal, in dilute solutions below 10 °C), optimally without
disrupting the Pd oxidation state or coordination sphere. An ideal
coupling reaction was recently described by Szabo´, wherein
[(2)(CH3CN)] was found to catalyze carbon-carbon bond
(18) (a) Pudipeddi, M.; Zannou, E. A.; Vasanthavada, M.; Dontabhaktuni,
A.; Royce, A. E.; Joshim, Y. M.; Serajuddin, A. T. M. J. Pharm. Sci.
2008, 97, 1831. (b) Karam, H.; El Kousy, N.; Towakkol, M. Anal.
Lett. 1999, 32, 79.
(19) Yu, K.; Sommer, W.; Richardson, J. M.; Weck, M.; Jones, C. W.
AdV. Synth. Catal. 2005, 347, 161.
9
16634 J. AM. CHEM. SOC. VOL. 132, NO. 46, 2010

Mechanical Activation of Catalysts
A R T I C L E S
Figure 3. Strategy for and representative examples of colorimetric indicator experiments demonstrating generation of free PyPM with corresponding negative
control experiments. (A) Chemical reaction underlying the colorimetric response: the purple solution containing protonated dye [3H][BF4] was converted to
a yellow solution containing 3 via deprotonation by PyPM generated upon sonication. (B-E) Results of selected colorimetric indicator experiments and
associated control experiments for [(1)(PyPM)2]. The solution containing [(1)(PyP66)2] changed from purple before sonication (pre) to yellow afterward
(post) (B), indicating release of free PyP66. No color change was observed during any of the negative control experiments (C-E). Conditions: sonication was
performed for 2 h at 4 °C on a 10 mL CH3CN solution containing 1 equiv of [3H][BF4] with respect to 10 µM analyte: (B) [(1)(PyP66)2], (C) [(1)(PyP11)2],
(D) [(2)(PyP110)], (E) [(1)(PyPiv)2].
Scheme 1. Palladium-Catalyzed Carbon-Carbon Bond Formation
formation between allyl20 or benzyl21 cyanides and N-tosyl
imines at ambient or lower temperatures. Moreover, van Koten
has shown that these SCS-cyclopalladated complexes can be
affixed to mesoporous silica and still exhibit their C-C bond
forming catalytic function, suggesting that tethering a polymer
chain could also preserve this activity.22 Because
[(1)(PyPM)(CH3CN)] features a Pd center structurally homolo-
gous to that in [(2)(CH3CN)], we reasoned that its generation
conversion (70% isolated yield) to the known21 coupled product
via sonication of [(1)(PyPM)2] would enable comparable cata-
6a (Table 2). Performing this reaction with a more electron-
deficient imine (5b) resulted in 98% conversion to 6b, which
was subsequently isolated in 71% yield. Alternatively, incor-
porating an electron-donating group into the imine (5c) com-
pletely quenched its reactivity, and no conversion to 6c was
noted. Given that this reaction formally proceeds via nucleo-
philic attack of a benzylic carbanion on an imine, it was
unsurprising that increasing the electron density at the latter
lowers its reactivity.21 No formation of 6a or 6b was catalyzed
by [(1)(PyP66)2] in the absence of sonication, consistent with
negligible thermal contribution to the observed conversion.23
Analogous to the indicator experiments described above, nega-
tive control experiments were performed on the sonochemically
activated coupling reaction for 6b with (i) sub-threshold
lytic function.
To test this hypothesis, a CH3CN solution containing 2-fluo-
robenzyl cyanide (4), N-tosylbenzylimine (5a), and [(1)(PyP66)2]
was subjected to the aforementioned sonication conditions
(Scheme 1). After 2 h, an aliquot was removed and subsequently
analyzed by gas chromatography (GC) which revealed 93%
(20) Aydin, J.; Szabo´, K. J. Org. Lett. 2008, 10, 2881.
(21) Aydin, J.; Conrad, C. S.; Szabo´, K. J. Org. Lett. 2008, 10, 5175.
(22) (a) Mehendale, N. C.; Sietsma, J. R. A.; de Jong, K. P.; van Walree,
C. A.; Gebbink, R. J. M. K.; van Koten, G. AdV. Synth. Catal. 2007,
349, 2619. (b) Gagliardo, M.; Selander, N.; Mehendale, N. C.; van
Koten, G.; Gebbink, R. J. M. K.; Szabo, K. J. Chem.sEur. J. 2008,
14, 4800. (c) Kruithof, C. A.; Dijkstra, H. P.; Lutz, M.; Spek, A. L.;
Gebbink, R. J. M. K.; van Koten, G. Organometallics 2008, 27, 4928.
9
J. AM. CHEM. SOC. VOL. 132, NO. 46, 2010 16635

A R T I C L E S
Tennyson et al.
Table 2. Selected Conversions and Yields for the Reactions
CH3CN was sonicated for 2 h, to ensure quantitative release of
PyP66, and then allowed to stir at 4 °C for 22 h to allow
sufficient chain growth, the corresponding polymer p(7) (Mn )
23 kDa, PDI ) 1.4; Table 2) was obtained in 42% yield with
respect to monomer consumption.26 No polymer was obtained
in the absence of sonication under otherwise identical conditions,
representing negligible thermal background formation of p(7).
Similarly, no polymerization occurred when a CH3CN solution
of 7 was sonicated with (i) sub-threshold [(1)(PyP7)2], (ii) end-
capped [(2)(PyP110)], or (iii) model complex [(1)(PyPiv)2] in
the presence of 90 kDa PMA. As observed for the Pd-catalyzed
reactions, these results indicate that the anionic polymerization
of 7 to p(7) was activated by mechanical force and proceeded
via release of PyP66 from [(1)(PyP66)2] during sonication.
Initiation of comparable polymerization reactions in analogous
mechanoresponsive materials could be harnessed to facilitate
self-repairing functions in response to the formation or micro-
cracks or other damage.6,8
Shown in Schemes 1 and 2a
5a f 6ab d
5b f 6bb d
,
,
c e
,
7 f p(7)
(X ) H)
(X ) F)
[(1)(PyP66)2]
[(1)(PyP66)2] (no sonication)
[(1)(PyP7)2]
[(2)(PyP110)]
[(1)(PyPiv)2]+ 90 kDa PMA
93% (70%) 98% (71%) 42%
3%
s
3%
3%
9%
3%
s
s
s
s
s
s
a Unless specified otherwise, the sonication experiments were
performed for 2 h at 4 °C. For 5 f 6 (a, X ) H; b, X ) F; c, X )
NMe2). Pd complex (i.e., [(1)(PyP66)2], [(1)(PyP7)2], [(2)(PyP110)], or
[(1)(PyPiv)2] + 90 kDa PMA; 1.0 µmol) was combined with the
following (either b or c, as indicated in table): b 4 (0.3 mmol, 1.5
equiv), 5 (0.2 mmol), Cs2CO3 (0.04 mmol, 20 mol %), and powdered 3
Å mol sieves (15 mg) in 10 mL of CH3CN. c 3.0 mL of 7 (3.8 mmol)
and 7 mL of CH3CN. d Percent conversions were determined by GC
using an internal standard (mesitylene); the numbers in parentheses
correspond to isolated yields. For 5c f 6c (X ) NMe2), less than 3%
conversion was observed under all conditions explored. e Percent yield
determined relative to monomer consumption. See Tables S4 and S5,
Supporting Information, for additional details.
Conclusion
Scheme 2. Pyridine-Catalyzed Anionic Polymerization
In summary, we have constructed a mechanoresponsive
reagent, comprising a cyclometalated dipalladium-derived mech-
anophore and pyridine-capped PMA-based actuator, and dem-
onstratedthatchainscissionoccursselectivelyatthemetal-pyridine
bond. This cleavage process produced catalytically active
palladium and pyridine species with reactivities that could be
harnessed to effect Pd-catalyzed C-C coupling and base-
catalyzed anionic polymerization reactions, respectively. None
of the aforementioned structural changes were observed (i) in
the absence of sonication, or when (ii) the total mechanophore-
actuator chain length was below the threshold required for the
system to experience ultrasonication-induced mechanical force,
(iii) the mechanophore was located at the end of the actuator
chain, or (iv) the mechanophore was not covalently linked with
an actuator. Collectively, our findings confirm that polymer
chain scission occurred at the mechanophore-actuator bond and
may be used to activate stoichiometric and catalytic reactions.
The associated processes were unambiguously established to
be mechanical in origin and may be used to access catalytic
reaction pathways that are not attainable with the thermal energy
supplied by the ambient reaction temperature. In a broader
context, our results show that judicious selection of comparable
mechanoresponsive reagents and methodologies could enable
unknown chemical reactivities, as well as unprecedented
advances in diagnostic, self-healing, and other types of stimulus-
responsive materials.
[(1)(PyP7)2], (ii) end-capped [(2)(PyP110)], and (iii) mechano-
phore model [(1)(PyPiv)2] in the presence of 90 kDa PMA.
Because no formation of 6b was observed in any of these control
experiments, we conclude that the C-C coupling catalysis
performed by [(1)(PyP66)2] was activated exclusively by me-
chanical force. The mechanical activation of such types of cross-
coupling catalysts is envisioned to find utility in diagnostic and
damage-reporting applications, where donor-acceptor conju-
gates with intense optical transitions may be formed in response
to mechanical stress.5c
Encouraged by the success with the carbon-carbon bond-
forming reactions, we envisioned that the PyPM released from
[(1)(PyPM)2] during sonication may be similarly harnessed for
base-mediated catalysis. Although pyridine-containing systems
have been shown to catalyze a variety of polymerization
reactions, these heteroaromatic species typically comprise
electron-donating substituents that significantly enhance their
nucleophilicities.24 Drawn to a recent report by Willson who
demonstrated25 that pyridine is capable of initiating the anionic
polymerization of R-trifluoromethyl-2,2,2-trifluoroethyl acrylate
(7) at -78 °C in THF (Scheme 2), we envisioned that PyPM,
liberated upon sonication of [(1)(PyPM)2], may exhibit similar
reactivity. Indeed, when a solution of [(1)(PyP66)2] and 7 in
Acknowledgment. We are grateful to the Army Research Office
(grant No. W911NF-07-1-0409) and the Robert A. Welch Founda-
tion (grant No. F-1621) for their generous financial support.
Supporting Information Available: Tables and figures sum-
marizing control experiments and additional data. This material
is available free of charge via the Internet at http://pubs.acs.org.
(23) We surmise that the catalytically active species sonochemically
generated in solution from [(1)(PyPM)2] was [(1)(PyPM)(CH3CN)] in
the presence of stoichiometric PyPM, which would have different
stereoelectronic properties and better organic solubility than those of
synthetic precursors [(1)(CH3CN)2] or [(2)(CH3CN)]. Indeed,
[(1)(CH3CN)2] and [(2)(CH3CN)] did not catalyze the coupling of 4
and 5 under strictly thermal conditions (0 °C, 0.5 mol% catalyst, 20
mol% Cs2CO3).
JA107620Y
(26) Monomer consumption may have been curtailed by early termination
processes due to the unoptimized polymerization reaction conditions
employed. For example, p7 (Mn ) 30 kDa; PDI ) 1.3) was obtained
in 50% yield with respect to monomer consumption when PyP66 was
used as the pyridine source and similar conditions to the polymerization
reaction performed under ultrasound described above were employed
(see Table S5, Supporting Information).
(24) Kamber, N. E.; Joeng, W.; Waymouth, R. M.; Pratt, R. C.; Loheijer,
B. G. G.; Hedrick, J. L. Chem. ReV. 2007, 107, 5813.
(25) Strahan, J. R.; Adams, J. R.; Jen, W.; Vanleenhove, A.; Neikrik, C. C.;
Rochelle, T.; Willson, C. G. J. Micro/Nanolithog. MEMS MOEMS
2009, 8, 043011.
9
16636 J. AM. CHEM. SOC. VOL. 132, NO. 46, 2010