Structure-mechanochemical activity relationships for cyclobutane mechanophores

Article abstract of DOI:10.1021/ja2086728

Ultrasound activation of mechanophores embedded in polymer backbones has been extensively studied of late as a method for realizing chemical reactions using force. To date, however, there have been few attempts at systematically investigating the effects of mechanophore structure upon rates of activation by an acoustic field. Herein, we develop a method for comparing the relative reactivities of various cyclobutane mechanophores. Through the synthesis and ultrasonic irradiation of a molecular weight series of poly(methyl acrylate) polymers in which each macromolecule has a single chain-centered mechanophore, we find measurable and statistically significant shifts in molecular weight thresholds for mechanochemical activation that depend on the structure of the mechanophore. We also show that calculations based on the constrained geometries simulate external force method reliably predict the trends in mechanophore reactivity. These straightforward calculations and the experimental methods described herein may be useful in guiding the design and the development of new mechanophores for targeted applications.

Full text of DOI:10.1021/ja2086728

ARTICLE
pubs.acs.org/JACS
StructureꢀMechanochemical Activity Relationships for Cyclobutane
Mechanophores
Matthew J. Kryger, Alexander M. Munaretto, and Jeffrey S. Moore*
Department of Chemistry and the Beckman Institute of Advanced Science and Technology,
University of Illinois at UrbanaꢀChampaign, Urbana, Illinios 61801, United States
S Supporting Information
b
ABSTRACT: Ultrasound activation of mechanophores em-
bedded in polymer backbones has been extensively studied
of late as a method for realizing chemical reactions using force.
To date, however, there have been few attempts at system-
atically investigating the effects of mechanophore structure
upon rates of activation by an acoustic field. Herein, we develop
a method for comparing the relative reactivities of various
cyclobutane mechanophores. Through the synthesis and ultra-
sonic irradiation of a molecular weight series of poly(methyl
acrylate) polymers in which each macromolecule has a single
chain-centered mechanophore, we find measurable and statistically significant shifts in molecular weight thresholds for
mechanochemical activation that depend on the structure of the mechanophore. We also show that calculations based on the
constrained geometries simulate external force method reliably predict the trends in mechanophore reactivity. These straightfor-
ward calculations and the experimental methods described herein may be useful in guiding the design and the development of new
mechanophores for targeted applications.
’ INTRODUCTION
induced polymerizations and catalysis,27,28 among others.29ꢀ31
While computational modeling has made an impact on our
understanding of mechanical activation,8,19,26,32ꢀ39 predicting
the ability of mechanophores to undergo mechanically induced
reactions, in addition to predicting relative chemical reactivities,
has been limited. Here we aim to experimentally validate a simple
computational technique, constrained geometries simulate ex-
ternal force (CoGEF),38 for predicting mechanophore reactivity
trends. Experimental validation involved the synthesis and soni-
cation of a molecular weight series of mechanophore-linked
polymers to compare their relative rates of reactivity. This paper
thus illustrates an example of SMARs and suggests its potential as
a useful technique for the rational development of mechano-
phores with a range of applications and uses.
Developing pharmaceutically active molecules is a lengthy and
difficult process with the average time from conception to FDA
approval being upward of 12 years.1 One method utilized in the
pharmaceutical world to reduce this time and aid in the targeted
development of biologically active molecules is the study of
structureꢀactivity relationships (SARs).2 SARs are commonly
used to correlate the three-dimensional chemical structure of a
molecule and its particular biological function. Utilizing both
computational modeling and biological assays, SARs are phar-
macophore models for the design of more active and specific
biological molecules. However, a method comparable to SAR has
never, to our knowledge, been utilized in the growing field of
mechanochemistry. Therefore, we considered the possibility of
applying a similar approach, structureꢀmechanochemical activ-
ity relationships (SMARs), to probe the importance of chemical
structure on mechanophore reactivity and streamline the devel-
opment of force-sensitive target mechanophores. Lessons
learned from the systematic study of a series of properly chosen
molecular structures will further increase our understanding of
force-activated reactions and improve predictive capabilities for
mechanophore design.
’ RESULTS AND DISCUSSION
In order to probe the effects of molecular structure on the
fragmentation rates of different mechanophores, six cyclobutanes
of varying stereochemistry, electronic environment, and levels of
substitution were targeted. It has previously been shown with a
small number of examples that substituted cyclobutanes can be
sonochemically or mechanically cleaved in a manner consistent
with a formal retro [2 + 2] reaction to yield reactive alkenes.21,40,41
Building on this knowledge, we focused our attention on the cis
and trans (with respect to the pendant ester groups) stereoisomers
While there has been a long-standing interest in mechanically
induced reactions in polymers,3ꢀ15 the past five years has seen an
increased focus on the development of novel mechanophores
with a variety of functional targets including, but not limited
to, damage detection through color change,16ꢀ20 self-healing
applications,21,22 and novel reactivity23ꢀ26 as well as mechanically
Received: September 14, 2011
Published: October 27, 2011
r
2011 American Chemical Society
18992
dx.doi.org/10.1021/ja2086728 J. Am. Chem. Soc. 2011, 133, 18992–18998
|

Journal of the American Chemical Society
ARTICLE
Figure 2. CoGEF results for the DCC mechanophore. Upon pulling
apart the methyl ester groups, the molecule first elongates then abruptly
fragments into a pair of alkenes (all CoGEF calculations were done using
a DFT-B3LYP 6-31G* level of theory).
Table 1. CoGEF, Fmax, and Cleavage Threshold Values for
Mechanophores Studied Herein
mechanophore
CoGEF (kJ/mol)
Fmaxa (nN)
threshold (kDa)
Figure 1. Structures of (A) cis and trans dicyano-substituted cyclobu-
tanes (DCC and DCT), (B) cis and trans monocyano-substituted cy-
clobutanes (MCC and MCT), and (C) cis and trans cyclobutanes
having no cyano substituents (NCC and NCT).
DCC
DCT
MCC
MCT
NCC
NCT
239
266
269
275
336
386
3.30
5.03
3.67
4.72
4.37
5.90
11.7
23.5
23.9
29.3
23.6
33.9
of 1,2-dicyano-substituted cyclobutane (DCC and DCT, respec-
tively, Figure 1A) and the cis and trans stereoisomers of mono-
cyano-substituted cyclobutane (MCC and MCT, Figure 1B) as
well as the cis and trans stereoisomers of cyclobutane having no
cyano substituents (NCC and NCT, Figure 1C) as our putative
mechanophores. These particular cyclobutanes were chosen for
their variation in not only the degree of substitution but also for
their variation in the relative stereochemistry (cis and trans) of
the pendant ester groups as well as the electronics of the cy-
clobutane ring. This set of structures provided an appropriate
level of diversity for an initial foray into the development of both
computational and mechanochemical testing protocol for SMAR.
Before the synthesis of this series of mechanophores was
attempted, a sequence of CoGEF calculations was carried out to
predict the trends of fragmentation reactivity. CoGEF is a com-
putational method developed by Beyer that probes the effect of
molecular stretching deformations upon fragmentation reactions.38
By artificially constraining the distance between two points (i.e.,
two atoms) in the structure of a mechanophore and sequentially
increasing the distance between these points by discrete incre-
ments, it is possible to simulate the effects of induced molecular-
scale mechanical deformation in a controlled way. In the case of
the mechanophores studied herein, by sequentially increasing the
distance between the methyl carbons of the methyl ester groups
and measuring the corresponding change in ground state energy,
it is possible to estimate the energy necessary to fragment a par-
ticular mechanophore (see Supporting Information for details).
The relationship of this CoGEF energy to activation energy is
unclear, but one might anticipate the two to be strongly corre-
lated. An example of a CoGEF elongation energy curve for the
DCC mechanophore can be found in Figure 2 above. Starting
from the unconstrained, energy minimized structure, elongation
results in stretching of the molecule and an increase in the
a Fmax values were calculated using the method outlined in ref 38.
ground-state energy. At the peak of the curve (energy maximum),
the cyclobutane fragments and then relaxes to a lower energy
conformation, where a retro [2 + 2] reaction has occurred to
form two alkenes. In the case of all six of the cyclobutanes,
molecular elongation resulted in a retro [2 + 2] reaction with the
formation of two alkenes.
In the cases studied herein, it should be noted that the molec-
ular forces induced during CoGEF calculations directly affect the
bond that is expected to cleave. There may be cases where the
expected mechanophore activation results from very different
reaction geometries or mechanisms (for example, mechanophores
that would activate through bond rearrangements without con-
current chain scission). In these cases, an alternative modeling
method may be more appropriate, however, the concepts intro-
duced herein should still hold true.
Upon completion of these CoGEF calculations for all six
mechanophores (Table 1), two trends in the relative fragmenta-
tion energy emerged. The first trend noted was that for a given
level of substitution, the cis mechanophore was more reactive
than the trans mechanophore (for example, DCC was more reactive
than DCT). This trend can also be seen in the Fmax values (Table 1)
calculated for these mechanophores, where Fmax is the maximum
force imposed on the model mechanophores for bond rupture
to occur. We speculate that this difference in reactivity is possibly
due to steric interactions between pendant ester groups within
the cis diester that are not present in the trans diester.
The second trend was that mechanophore reactivity increased
as the level of mechanophore substitution increased (for example,
18993
dx.doi.org/10.1021/ja2086728 |J. Am. Chem. Soc. 2011, 133, 18992–18998

Journal of the American Chemical Society
ARTICLE
Scheme 1. Synthesis of Cyclobutane Mechanophore Coresa
a (A) Synthesis of dimethyl ester-DCC and dimethyl ester-DCT. (B) Synthesis of dimethyl ester-MCC and dimethyl ester-MCT. (C) Synthesis of
3-oxabicyclo[3.2.0]heptane-2,4-dione (NCC core). (D) Commercially available trans-cyclobutane-1,2-dicarboxylic acid (NCT core).
DCT was more reactive toward cleavage than MCT, which in
turn was more reactive than NCT). We speculate this higher
reactivity is a result of increased steric interactions upon the 1,2-
substituted cyclobutane bond, which leads to increased ring
strain. This notion is supported by analyzing the average bond
lengths of the 1,2-substituted structures;42 whereby longer bond
lengths reflect greater internal ring strain and increased suscept-
ibility to cyclobutane cleavage. The average 1,2-cyclobutane
bond length for the dicyano-substituted cyclobutanes was 1.61 Å,
whereas the average 1,2-cyclobutane bond length for the mono-
cyano-substituted cyclobutanes was 1.58 Å. Finally, bond lengths
averaged 1.55 Å for cyclobutanes having no cyano substituents.
Concurrently, this additional reactivity could also be a result of
increased stabilization of the transition states leading to diradical
intermediates that are formed during asynchronous cyclobutane
fragmentation.21 Regardless of the reasoning, we show below
that the predicted trends are testable through the synthesis of a
polymer molecular weight series for each of the cyclobutane
derivatives followed by a statistical comparison of their rates of
fragmentation in an acoustic field.
to yield a mixture of bis-alkenes 4. Ozonolysis followed directly
by methanolysis yielded the symmetric dicyano diester 5. Ring
closure gave a mixture of methyl ester substituted cyclobutanes 6
(DCC) and 7 (DCT) in a 1:3 ratio. The products were further
separated to give the pure isomers via sublimation (Scheme 1A),
where relative stereochemistry was determined through X-ray
crystallography (see Supporting Information). Enantiomers of 7
were tested as a racemic mixture of the DCT mechanophore.
MCC and MCT were synthesized by α-bromination and methyl-
esterification of adipic acid 8 to yield diester 9.44 This step was
followed by the addition of sodium cyanide to afford a racemic
mixture of cyclobutanes 10 (MCC) and diastereomer 11 (MCT),
which were further separated through selective crystallization of
the cis derivative (Scheme 1B). The structure of MCC was con-
firmed by X-ray crystallography (see Supporting Information). The
NCC mechanophore was synthesized from the [2 + 2] cycload-
dition of ethylene gas and maleic anhydride to form 3-oxabicyclo-
[3.2.0]heptane-2,4-dione45 14 (Scheme 1C), and the NCT
mechanophore was synthesized from the commercially available
enantiomeric mixture of trans-cyclobutane-1,2-dicarboxylic acid
15 (Scheme 1D).
In order to test the computational results, all six of the cyclo-
butanes were synthesized. The cis and the trans dicyano-sub-
stituted cyclobutanes were synthesized starting with the com-
mercially available methyl cyanoacetate 1 and 3-pentanone 2.21 A
Knoevenagel condensation of these two starting materials resulted
in α,β-unsaturated ester 3,43 which was reacted with dibromoethane
After successful synthesis of the mechanophores, the installa-
tion of initiator groups was necessary for the growth of pendant
polymer chains via single-electron-transfer living radical polym-
erization (SET-LRP).46 This installation of initiator groups was
accomplished through transesterification of the mechanophores
18994
dx.doi.org/10.1021/ja2086728 |J. Am. Chem. Soc. 2011, 133, 18992–18998

Journal of the American Chemical Society
Scheme 2. General Synthesis of Initiator Functionalized Mechanophores and Control for the Formation of Chain-Centered PMA
ARTICLE
with ethylene glycol to form the bis-hydroxy functionalized
derivatives followed by reaction with α-bromoisobutyryl bro-
mide to install the needed α-bromoester functionalized initiators
for polymerization (Scheme 2). In addition to these mechan-
ophore-containing initiators, two control initiators were used.
The first of these (CON1) consisted of a simple alkane bridge to
replace the active cyclobutane mechanophores. This initiator was
synthesized in a manner similar to the mechanophore initiators
above (Scheme 2). Additionally, the simple monofunctional
initiator methyl 2-bromopropionate was used in the synthesis
of homopolymer containing no center-functionalized mechan-
ophore or alkane bridge as a second control (CON2).
From these eight unique initiators (six active mechanophore
initiators as well as two control initiators) was produced a series
of center- or end-functionalized polymers of varying molecular
weight. Specifically, a series of low polydispersity (PDI) poly-
(methyl acrylate) (PMA) polymers ranging from 35 to 125 kDa
was synthesized using SET-LRP (see Supporting Information for
specific molecular weights and PDIs). The molecular weight
series of center-functionalized polymers containing the various
mechanophores makes it possible to probe relative reactivities by
measuring the relative rates of mechanophore scission in an
acoustic field.21 Molecular weight values, which were critical for
the analysis described below, were measured by gel permeation
chromatography (GPC) and are calculated relative to mono-
disperse polystyrene standards.
In order to determine the relative rates of polymer cleavage,
dilute solutions of polymer were exposed to ultrasonic irradiation
at a controlled temperature (6ꢀ9 °C) under an argon atmo-
sphere. During each sonication, aliquots were removed at regular
intervals and were subsequently analyzed by GPC. Every sonication
experiment was performed in duplicate to ensure reproducibility.
Rates of molecular cleavage can be calculated using the
method utilized by Malhotra and co-workers.47,48 Based on the
assumption that, during mechanical cleavage, random chain
scission occurs along the backbone of the polymer, Malhotra
used the theoretically derived49ꢀ51 eq 1 to model time-depen-
dent change in molecular weight:
the polymer, Mt is the number average molecular weight of the
sonicated sample at time t, and k is the rate constant of polymer
cleavage with initial molecular weight Mi. While the aforementioned
model assumes random scission along the backbone of the polymer,
it has been shown to also be applicable to systems where center-
selective scission is taking place (as is the case with polymers
sonicated in solution).21,47 Each of the mechanophore-containing
polymers was exposed to ultrasonic irradiation, and the
change in molecular weight distribution as a function of
ultrasound irradiation time was recorded by GPC (see exam-
ple, Figure 3A). A plot of the number average molecular
weight values obtained by GPC vs ultrasonic irradiation time
shows an asymptotic decay profile (Figure 3B). The cleavage
rate constant (k) for a mechanophore at a given initial mole-
cular weight is calculated by nonlinear least-squares fitting of
eq 1 to this data. As shown in Figure 3B, the fitted decay profile
(solid curve) is in good agreement with the experimental data
(points). A similar procedure was used to calculate the rate
constant for each polymer synthesized.
It is well-known that there is a molecular weight dependence
in the rate of polymer scission induced by sonication.4 Longer
polymer chains result in faster rates of cleavage. As long as the
initial molecular weight is not above ca. 150 kDa, a linear relation-
ship has empirically been found between molecular weight and
cleavage rate. Additionally, there is a molecular weight threshold
below which polymer chains do not fragment under sonication
conditions.3 Our group has previously shown that inclusion of a
single chain-centered mechanophore in the polymer backbone
lowers the molecular weight threshold due to selective cleavage
of the mechanophore.29 By comparing the sonochemically deter-
mined thresholds for the different mechanophores synthesized,
we can investigate the relationship between mechanophore struc-
ture and reactivity. This comparison should allow us to test the
effectiveness of CoGEF calculations in predicting mechano-
chemical trends.
Comparison of rate values was accomplished by sonicating all
eight of the unique polymer molecular weight series. Using the
method of Malhotra and co-workers outlined above, the rates of
cleavage were determined for each polymer of initial molecular
weight Mi for each of the different mechanophores and the
control polymers. The differences in mechanophore reactivity
are best seen in plots of k vs Mi. As described below, the different
mechanophores result in changes in threshold molecular weight
1
1
¼
þ k0t
ð1Þ
Mt
Mi
where k0 = k/M0, M0 is the molecular weight of the monomer
unit, Mi is the initial number average molecular weight (Mn) of
18995
dx.doi.org/10.1021/ja2086728 |J. Am. Chem. Soc. 2011, 133, 18992–18998

Journal of the American Chemical Society
ARTICLE
Figure 3. GPC traces and modeled data depicting the effect of ultra-
sonic irradiation on a mechanophore containing polymer. A 0.75 mg/mL
solution of polymer in acetonitrile was subjected to ultrasound at 20 kHz,
8.7 W/cm2, and 6ꢀ9 °C under argon (0.5 s on, 1.0 s off; irradiation time
refers to total time both off and on). (A) Change in GPC trace for a
sonicated DCT polymer having an Mi of 90 kDa. The inset colors code
the samples according to ultrasound irradiation time. (B) Experimen-
tally obtained number average molecular weight (Mt) with respect to
time of sonication (data points) and best fit (solid curve) to the Malhotra
model. The curve was derived by adjusting the k0 value in eq 1 to the
experimental data using a nonlinear least-squares fitting routine.
(i.e., x-axis intercept, also called elevation) rather than slope
(Table 1). Specifically, it was found that the rate of change due to
molecular weight differences was not statistically different for any
of the polymer series studied (see Supporting Information for
details).52
Most importantly, we wanted to know whether the changes in
molecular weight threshold were significantly different for each
of the mechanophore-containing polymers. Using the statistical
method outlined by Zar, it was found that many of the perceived
differences in threshold were in fact statistically significant at the
95% confidence value (see Supporting Information for detailed
statistical analysis and t-test values).52 Thus, an important general
conclusion from this study is that the inclusion of a single, center-
linked mechanophore results in measurable and statistically
significant shifts in molecular weight thresholds for polymer
cleavage. As detailed below, these threshold molecular weight
differences depend on the specific structure of the mechano-
phore, which correlate favorably to reactivity trends predicted by
CoGEF calculations.
Figure 4. Experimentally determined rate constants of polymer clea-
vage as a function of initial polymer molecular weight, R2 values for the fit
lines are included in parentheses. (A) Cleavage of DCC (R2 = 0.974) and
DCT (R2 = 0.962) polymers. (B) Cleavage of MCC (R2 = 0.958) and
MCT (R2 = 0.961) polymers. (C) Cleavage of NCC (R2 = 0.965) and
NCT (R2 = 0.972) polymers. (D) Cleavage of DCT, MCT, and NCT
polymers. (E) Cleavage of DCC, MCC, and NCC polymers.
The first trend analyzed was the relationship between stere-
ochemistry and reactivity. CoGEF calculations predicted that the
cis derivatives would be more reactive than the trans derivatives
for a given level of substitution. In comparing the DCC mechanophore
18996
dx.doi.org/10.1021/ja2086728 |J. Am. Chem. Soc. 2011, 133, 18992–18998

Journal of the American Chemical Society
ARTICLE
to the DCT mechanophore (Figure 4A), there was a statistically
significant threshold molecular weight difference, with the DCC
mechanophore being more reactive (i.e., smaller x-axis intercept
value). This result is seen by the higher elevation of the fitted line
for the cis derivative in the plot of k vs Mi. This means that, for a
given molecular weight, a polymer containing the DCC mechan-
ophore cleaves faster than a polymer containing the DCT
mechanophore, and consequently, the DCC mechanophore is
more mechanochemically reactive. Comparing MCC and MCT,
the cis mechanophore again appears, by visual inspection of the
plot, to be more reactive; however, the difference in the inter-
cepts was not statistically significant at the 95% confidence
interval (Figure 4B). On the other hand, the NCC mechan-
ophore, using the statistical criterion outlined above, was more
reactive than the NCT mechanophore, consistent with the
CoGEF calculations (Figure 4C).
The relationship between level of cyclobutane substitution
and mechanophore reactivity was the second trend analyzed.
CoGEF calculations predicted that the more highly substituted
mechanophores will cleave more readily than the less-substituted
derivatives; this behavior is presumably due to increased ring
strain present within the cyclobutane rings in the higher-sub-
stituted derivatives, which is consistent with the ring’s carbonꢀ
carbon bond lengths. Upon analysis of the trans series of
mechanophores (Figure 4D), we found that this trend was
indeed the case. By comparison of threshold values, the DCT
mechanophore was the most reactive, followed by the MCT
mechanophore, and finally, the NCT mechanophore was the
least reactive. Analyzing the cis series, the trend in reactivity is less
clear. The DCC mechanophore is significantly more reactive
than the MCC and NCC mechanophores; however, threshold
values for the latter two are statistically indistinguishable from
each other (Figure 4E). Nonetheless, the general reactivity
trends concerning degree of substitution predicted through
CoGEF are, again, generally validated by the rate data presented
in Figure 4.
useful mechanophores have been developed, few systematic
studies have been conducted that combine computational and
experimental approaches toward predicting trends in reactivity.
Herein, we illustrate the SMAR method for investigating these
trends. By probing the effects of mechanophore structure through
the systematic study of a class of putative mechanophores, our
aim is to open the way for the rational design and development of
new mechanically activated molecules.
The systematic synthesis and study of a molecular weight
series of cyclobutane-containing polymers experimentally vali-
dated CoGEF as a computational tool for the investigation of
SMARs. Both stereochemical and substitutional reactivity trends
in a series of mechanophores were predicted using CoGEF, and
the trends were compared to the behavior of a series of center-
functionalized polymers that fragment when subjected to an
acoustic field. Plots of cleavage rate constant vs initial molecular
weight gave threshold molecular weight values that differed in a
statistically significant way for many of polymers, even though
each macromolecule contains just one cyclobutane mechano-
phore. The trends in threshold molecular weight generally agreed
with reactivity predicted by CoGEF calculations.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental details, synthetic
b
procedures, GPC data, statistical analysis, crystallographic data,
and details on computational modeling. This material is available
free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
jsmoore@illinois.edu
’ ACKNOWLEDGMENT
The mechanophore-containing polymers were also compared
to the control polymers that were synthesized in order to verify
that they were indeed more reactive than polymers that did not
contain a cyclobutane ring in the center (CON1) as well as
simple PMA homopolymers (CON2). Based on statistical
comparison of threshold values, it was found that all of the
cyclobutane mechanophores studied herein are more reactive
than the control polymers, with the exception of the NCT
mechanophore. As the least reactive of the mechanophores,
the NCT derivative had threshold molecular weights that were
statistically indistinguishable from the thresholds of both CON1
and CON2. Consequently, it is plausible that the polymers
containing the NCT mechanophore are not selectively cleaved
at the cyclobutane moiety (possibly due to the decreased internal
ring strain present in the NCT mechanophore when compared
to the other cyclobutanes). General and unselective chain
scission of the polymer backbone is likely occurring in addition
to cleavage of the mechanophore in the case of NCT (See
Supporting Information for details).
This work was supported by the Army Research Office MURI
(grant W911NF-0701-0409). The authors thank Koushik Ghosh,
Preston May, and Doug Davis for helpful discussions relating to
this project.
’ REFERENCES
(1) GlaxoSmithKline Common Questions Website; http://us.gsk.com
(accessed August 13, 2011).
(2) Peltason, L.; Bajorath, J. J. Med. Chem. 2007, 50, 5571.
(3) 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.
(4) Basedow, A. M.; Ebert, K. H. Adv. Polym. Sci. 1977, 22, 83.
(5) Nguyen, T. Q.; Kausch, H.-H. Adv. Polym. Sci. 1992, 100, 73.
(6) Rubner, M. F. Macromolecules 1986, 19, 2129.
(7) Nallicheri, R. A.; Rubner, M. F. Macromolecules 1991, 24, 517.
(8) Beyer, M. K.; Clausen-Schaumann, H. Chem. Rev. 2005, 105, 2921.
(9) Schmidt-Naake, G.; Drache, M.; Weber, M. Macromol. Chem.
Phys. 2002, 203, 2232.
(10) Devries, K. L.; Roylance, D. K.; Williams, M. L. J. Polym. Sci,
Part B: Polym. Phys. 1972, 10, 599.
(11) Devries, K. L.; Roylance, D. K.; Williams, M. L. J. Polym. Sci,
Part A: Polym. Chem. 1970, 8, 237.
(12) Sohma, J. Prog. Polym. Sci. 1989, 14, 451.
(13) Encina, M. V.; Lissi, E.; Sarasꢀua, M.; Gargallo, L.; Radic, D.
J. Polym. Sci. Polym. Lett. Ed. 1980, 18, 757.
(14) Sheiko, S. S.; Sun, F. C.; Randall, A.; Shirvanyants, D.;
Rubinstein, M.; Lee, H.-I.; Matyjaszewski, K. Nature 2006, 440, 191.
’ CONCLUSIONS
The development of mechanically activated small molecules
(i.e., mechanophores) with targeted molecular level responses is
a relatively new field of study that has only come to prominence
in the last five years.3 Although many interesting and potentially
18997
dx.doi.org/10.1021/ja2086728 |J. Am. Chem. Soc. 2011, 133, 18992–18998

Journal of the American Chemical Society
ARTICLE
(49) Sato, T.; Nalepa, D. E. J. Appl. Polym. Sci. 1978, 22, 865.
(15) Zysman, V.; Nguyen, T. Q.; Kausch, H.-H. J. Polym. Sci, Part B:
Polym. Phys. 1994, 32, 1257.
(50) Casassa, E. J. Polym. Sci. 1949, 4, 405.
(16) Beiermann, B. A.; Davis, D. A.; Kramer, S. L. B.; Moore, J. S.;
Sottos, N. R.; White, S. R. J. Mater. Chem. 2011, 21, 8443.
(17) Kingsbury, C. M.; May, P. A.; Davis, D. A.; White, S. R.; Moore,
J. S.; Sottos, N. R. J. Mater. Chem. 2011, 21, 8381.
(18) Lee, C. K.; Davis, D. A.; White, S. R.; Moore, J. S.; Sottos, N. R.;
Braun, P. V. J. Am. Chem. Soc. 2010, 132, 16107.
(51) Jellinek, H. H. G. J. Polym. Sci. 1949, 264.
(52) Zar, J. H. Comparing Simple Linear Regression Equations.
Biostatistical Analysis, 5th ed.; Pearson Prentice Hall: Upper Saddle
River, NJ, 2010; p 363.
(19) Davis, D. A.; Hamilton, A.; Yang, J.; Cremar, L. D.; Van Gough,
D.; Potisek, S. L.; Ong, M. T.; Braun, P. V.; Martínez, T. J.; White, S. R.;
Moore, J. S.; Sottos, N. R. Nature 2009, 459, 68.
(20) Potisek, S. L.; Davis, D. A.; Sottos, N. R.; White, S. R.; Moore,
J. S. J. Am. Chem. Soc. 2007, 129, 13808.
(21) 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.
(22) Black, A. L.; Orlicki, J. A.; Craig, S. L. J. Mater. Chem. 2011,
21, 8460.
(23) Wiggins, K. M.; Hudnall, T. W.; Shen, Q.; Kryger, M. J.; Moore,
J. S.; Bielawski, C. W. J. Am. Chem. Soc. 2010, 132, 3256.
(24) Wiggins, K. M.; Syrett, J. A.; Haddleton, D. M.; Bielawski, C. W.
J. Am. Chem. Soc. 2011, 133, 7180.
(25) Hickenboth, C. R.; Moore, J. S.; White, S. R.; Sottos, N. R.;
Baudry, J.; Wilson, S. R. Nature 2007, 446, 423.
(26) Lenhardt, J. M.; Ong, M. T.; Choe, R.; Evenhuis, C. R.;
Martinez, T. J.; Craig, S. L. Science 2010, 329, 1057.
(27) Tennyson, A. G.; Wiggins, K. M.; Bielawski, C. W. J. Am. Chem.
Soc. 2010, 132, 16631.
(28) Karthikeyan, S.; Potisek, S. L.; Piermattei, A.; Sijbesma, R. P.
J. Am. Chem. Soc. 2008, 130, 14968.
(29) Berkowski, K. L.; Potisek, S. L.; Hickenboth, C. R.; Moore, J. S.
Macromolecules 2005, 38, 8975.
(30) Huang, Z.; Yang, Q.-Z.; Khvostichenko, D.; Kucharski, T. J.;
Chen, J.; Boulatov, R. J. Am. Chem. Soc. 2009, 131, 1407.
(31) Yang, Q.-Z.; Huang, Z.; Kucharski, T. J.; Khvostichenko, D.;
Chen, J.; Boulatov, R. Nat. Nanotechnol. 2009, 4, 302.
(32) Ong, M. T.; Leiding, J.; Tao, H.; Virshup, A. M.; Martínez, T. J.
J. Am. Chem. Soc. 2009, 131, 6377.
(33) Lenhardt, J. M.; Ogle, J. W.; Ong, M. T.; Choe, R.; Martinez,
T. J.; Craig, S. L. J. Am. Chem. Soc. 2011, 133, 3222.
(34) Dopieralski, P.; Ribas-Arino, J.; Marx, D. Angew. Chem., Int. Ed.
2011, 50, 7105.
(35) Dopieralski, P.; Anjukandi, P.; R€uckert, M.; Shiga, M.; Ribasꢀ
Arino, J.; Marx, D. J. Mater. Chem. 2011, 21, 8309.
(36) Ribas-Arino, J.; Shiga, M.; Marx, D. J. Am. Chem. Soc. 2010,
132, 10609.
(37) Ribas-Arino, J.; Shiga, M.; Marx, D. Angew. Chem., Int. Ed. 2009,
48, 4190.
(38) Beyer, M. K. J. Chem. Phys. 2000, 112, 7307.
(39) Schmidt, S. W.; Beyer, M. K.; Clausen-Schaumann, H. J. Am.
Chem. Soc. 2008, 130, 3664.
(40) Cho, S.-Y.; Kim, J.-G.; Chung, C.-M. Sens. Actuators, B 2008,
134, 822.
(41) Chung, C.-M.; Roh, Y. S.; Cho, S. Y.; Kim, J. G. Chem. Mater.
2004, 16, 3982.
(42) Bond Lengths were determined from crystal structures in the
case of DCC, DCT, and MCC. Due to MCT, NCC, and NCT being
liquids, bond lengths were calculated using a computational minimiza-
tion of their structures (see Supporting Information for details).
(43) Cope, A. C.; Hofmann, C. M.; Wyckoff, C.; Hardenbergh, E.
J. Am. Chem. Soc. 1941, 63, 3452.
(44) Guha, P. C.; Sankaran, D. K. Org. Syn. 1946, 26, 57.
(45) Faure, S.; Jensen, A. A.; Maurat, V.; Gu, X.; Sagot, E.; Aitken,
D. J.; Bolte, J.; Gefflaut, T.; Bunch, L. J. Med. Chem. 2006, 49, 6532.
(46) 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.
(47) Malhotra, S. J. Macromol. Sci., Part A 1986, 37.
(48) Malhotra, S. J. Macromol. Sci., Part A 1982, 18, 1055.
18998
dx.doi.org/10.1021/ja2086728 |J. Am. Chem. Soc. 2011, 133, 18992–18998