Effect of aromaticity on the rate of azaquinone methide-mediated release of benzylic phenols

Article abstract of DOI:10.1002/poc.3129

Herein we use small molecule models of stimuli-induced degradable/ depolymerizable polymers to demonstrate that less aromatic releasing units provide faster rates of azaquinone methide-mediated release of benzylic phenols (a surrogate for a group released in a polymer) than highly aromatic releasing units. Copyright 2013 John Wiley & Sons, Ltd. In small molecule models of stimuli-induced degradable polymers, less aromatic releasing units provide faster rates of azaquinone methide-mediated release of benzylic phenols than highly aromatic releasing units. Copyright

Full text of DOI:10.1002/poc.3129

Short Communication
Received: 16 February 2013,
Accepted: 31 March 2013,
Published online in Wiley Online Library: 20 May 2013
(wileyonlinelibrary.com) DOI: 10.1002/poc.3129
Effect of aromaticity on the rate of azaquinone
methide-mediated release of benzylic phenols
Kyle M. Schmida and Scott T. Phillipsa*
Herein we use small molecule models of stimuli-induced degradable/depolymerizable polymers to demonstrate
that less aromatic releasing units provide faster rates of azaquinone methide-mediated release of benzylic phenols
(a surrogate for a group released in a polymer) than highly aromatic releasing units. Copyright © 2013 John Wiley &
Sons, Ltd.
Supporting information may be found in the online version of this paper.
Keywords: aromaticity; azaquinone methide; controlled release; solvent effects
X-ray crystal structures of these model compounds reveal a
direct relationship between the lengths of the carbon-carbon
INTRODUCTION
bonds within the aromatic ring and the rate of release – that
is, the longer the carbon–carbon bonds, the faster the rate of
azaquinone methide-mediated release.
Polymers that degrade or depolymerize in response to specific
applied stimuli have the potential to transform the magnitude,
selectivity, and rate of response of stimuli-responsive materials
composed of these polymers. Such polymers ultimately may
impact diverse fields such as diagnostics, controlled drug release,
and self-healing materials.[1–6] To realize many of these applica-
tions, however, depolymerizable and degradable polymers must
be capable of responding to an applied stimulus quickly in
environments that are less polar than water (e.g., in the shells
of responsive microscale and nanoscale capsules,[3,5] films,[7]
plastics,[8] or other solid state materials). A common strategy
for enabling controlled degradation and/or depolymerization in
responsive polymers involves the use of moieties that release a
benzylic phenol or carbamic acid via formation of azaquinone
methide when a specific stimulus is applied to the polymer
(Fig. 1).[1,2] In small molecule model studies, we demonstrated
that the rate of this type of azaquinone methide elimination
reaction is directly proportional to the polarity of the chemical
environment: the more polar the environment, the faster the
rate of elimination,[9,10] whereas in low polarity environments,
azaquinone methide-mediated release may not proceed at all.
Ideally, the structure of the releasing group that forms
azaquinone methide could be modified in a logical way that
enables predictable and tunable rates of azaquinone methide-
mediated release in nearly any environment. However, the
design principles to enable these types of structural modifica-
tions have yet to be fully established. In this work we test
one new strategy for increasing azaquinone methide-mediated
release (i.e., we alter the aromaticity of the releasing unit) and
demonstrate its potential for increasing the rate of release of
benzylic phenols. The results of this study move us one step closer
to a set of predictable design rules that can be applied in the context
of azaquinone methide-mediated degradable and depolymerizable
polymers.
RESULTS AND DISCUSSION
Our design strategy is based on the hypothesis that the rate of
release in systems such as those depicted in Fig. 1 is slow, because
the releasing moiety must proceed through a less aromatic transi-
tion state (i.e., azaquinone methide) than that of the ground state.
This less aromatic transition state presumably results in a large
energy penalty for depolymerization/degradation. We reasoned that
by decreasing the aromatic character of the releasing moiety we
could raise the ground state energy of the unit and decrease the
activation energy barrier associated with the azaquinone methide-
mediated release reaction, thus increasing the rate of release.
To test this hypothesis, we developed three small molecule
model compounds (Fig. 2) to evaluate the effect of aromaticity
on the rate of azaquinone methide-based elimination reactions
under controlled circumstances without the added complexities
associated with studying responses in polymers. The compounds
each contain an allyl carbamate (Alloc) reaction-based detection
unit (bold), which is cleaved by the stimulus Pd(0), and a
pendant phenol (gray), which serves as a surrogate for a phenol
that may be released in a degradable[5] or depolymerizable
polymer.[1,2] Phenol also provides a convenient spectroscopic
probe for following the kinetics of the release reaction by using
a liquid chromatograph coupled to a mass spectrometer (LCMS).
Model compounds 2 and 3 were prepared through relatively
short syntheses (Schemes 1 and 2) whereas compound 1 was
* Correspondence to: Scott T. Phillips, Department of Chemistry, The Pennsylvania
State University, University Park, PA 16802, USA.
E-mail: sphillips@psu.edu
Specifically, we use small molecule model systems to
demonstrate that the level of aromaticity of the aromatic unit
(which could be a repeating unit in a polymer) has a substantial
effect on the rate of stimuli-initiated release of a pendant phenol.
a K. M. Schmid, S. T. Phillips
Department of Chemistry, The Pennsylvania State University, University Park,
PA, 16802, USA
J. Phys. Org. Chem. 2013, 26 608–610
Copyright © 2013 John Wiley & Sons, Ltd.

EFFECT OF AROMATICITY
Alloc group), then diluting a 10 mL aliquot of the solution in a
1:1 mixture of MeCN and pH 7.1 buffered water (to enable
azaquinone methide-mediated release of phenol) (Fig. 3). The
rate of disappearance of the aniline intermediate was monitored
by repetitive injections at 20-min intervals into an LCMS.[12] The
aniline intermediate is stable (i.e., will not proceed through
azaquinone methide on the time-scale of the experiments) when
dissolved in THF, but once the aniline is dissolved in the MeCN :
water mixture (a more polar solvent mixture than THF), the
azaquinone methide elimination reaction becomes favorable.
Thus, our assay measures the release reaction independent
of the kinetics of cleavage of the reaction-based detection
unit (Alloc).
Figure 1. General mechanism of a stimulus-induced depolymerization
and/or degradation reaction of
a polymer functionalized with a
reaction-based detection unit (often referred to as an endcap or trigger).
Once the reaction-based detection unit is cleaved by a specific stimulus,
the next portion of the polymer is released via an azaquinone methide
elimination reaction
Based on the relative aromaticity values depicted in Fig. 2, we
expected the phenanthrene derivative 3 to release phenol faster
than the naphthalene or benzene model systems (2 and 1,
Figure 2. Three model compounds used to measure whether ground
state aromaticity of the releasing group affects the rate of azaquinone
methide-mediated release of a pendant phenol. The numbers written on
each compound correspond to the calculated relative aromaticity values
for each ring (the values correspond to rings without substituents)[11]
Figure 3. Reaction conditions used for the cleavage of the Alloc group
and monitoring the kinetics for release of phenol from the three model
compounds in Fig. 2
Scheme 1. Synthesis of model compound 2. Reagents and conditions:
(a) phenol, K2CO3, dimethylformamide, 50ꢀC; (b) n-BuLi, CO2, tetrahydro-
furan, À78ꢀC; and (c) diphenylphosphoryl azide, triethylamine, allyl alco-
hol, 85ꢀC (40% over three steps)
Scheme 2. Synthesis of model compound 3. Reagents and conditions:
(a) phenol, K2CO3, dimethylformamide, 50ꢀC; (b) n-BuLi, CO2, tetrahydro-
furan, À78ꢀC; and (c) diphenylphosphoryl azide, triethylamine, allyl alco-
hol, 90ꢀC (19% over three steps)
Figure 4. Comparison of the relative aromaticity values for compounds
1, 2, and 3 with: (i) the distribution of carbon–carbon bond lengths within
the aromatic ring where azaquinone methide is generated during the
release reaction; and (ii) the half-life for release of phenol after the Alloc
group is removed using Pd(0). The blue bars represent the lengths for
the five carbon–carbon bonds within the central aromatic ring for each
structure indicated at the bottom of the graph.[13] The green data points
represent the relationship between relative aromaticity value[11] and the
half-life for release of phenol. These data points correspond to the
structures depicted below the graph and are the average of three
measurements. The error bars are smaller than the data points
prepared previously.[9] In all cases, a Curtius rearrangement was
used as the last step to introduce the reaction-based detection
unit to the test compounds, which proved to be a convenient
strategy for introducing the orthogonal functionality required
in these compounds.
The rates of release of phenol from compounds 1, 2, and 3
were determined by treating each compound with Pd(PPh3)4,
Bu3SnH, and acetic acid in tetrahydrofuran (THF) (to cleave the
J. Phys. Org. Chem. 2013, 26 608–610
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

K. M. SCHMID AND S. T. PHILLIPS
respectively). Indeed, when the half-lives for the release reac-
tions were compared with the calculated relative aromaticity
values, we observed a direct relationship between relative
aromaticity value[11] and half-life for release (Fig. 4). For example,
the phenanthrene derivative 3 (relative aromaticity value =
0.813)[11] releases phenol with a half-life of 27 min, which is
~51Â faster than the benzene derivative 1 (half-life = 23 h;
relative aromaticity value = 1.0).[11]
X-ray crystallographic analysis of the lengths of the carbon–
carbon double bonds within the central rings in 1À3 reveal an
inverse relationship between the distributions of bond lengths
and the calculated relative aromaticity value for the ring.[13] In
other words, aromatic releasing moieties with lower relative
aromaticity values have longer carbon–carbon bonds than
derivatives with higher relative aromaticity values and corre-
spondingly faster rates of release of phenol (Fig. 4).
and phosphate buffer (0.1 M, pH 7.1, 0.5 mL). The solution was
shaken for 10 s and then was filtered through a syringe filter
(polytetrafluoroethylene, 0.22 mm). The rate of release of phenol
was inferred[12] by monitoring the disappearance of the aniline
intermediate by LCMS by using an ultraviolet detector set at
254 (for compounds 1 and 2) or 330 nm (compound 3).
SUPPORTING INFORMATION
Detailed synthetic procedures, compound characterization
data, crystal structure data, tables of primary data, and figures
of NMR spectra.
Acknowledgements
This work was supported in part by the American Chemical
Society (PRF no. 51618-DNI7), the Arnold and Mabel Beckman
Foundation, the Camille and Henry Dreyfus Foundation, and Louis
Martarano. S.T.P. acknowledges support from the Alfred P. Sloan
Research Fellows Program.
CONCLUSION
In conclusion, we have established a new strategy for increasing
the rate of azaquinone methide-mediated release of benzylic
phenols. This strategy involves increasing the ground state en-
ergy of the aromatic releasing unit by decreasing the aromaticity
of this group. Until now, the available strategies for increasing
the rate of analyte-triggered azaquinone methide-mediated
release of a benzylic leaving group have involved adding
electron density to the aromatic ring of the releasing unit,[9,10,14]
introducing substituents at the benzylic position,[15] and increas-
ing the polarity of the medium in which the release reaction
occurs.[9,10] In the context of stimuli-responsive solid-state polymeric
materials, only the former two strategies are generalizable. Thus,
these results add to the short but growing list of design principles
that are available to guide the creation of polymers that efficiently
degrade or depolymerize via azaquinone methide pathways in
nonpolar environments when exposed to a specific stimulus.
REFERENCES
[1] G. I. Peterson, M. B. Larsen, A. J. Boydston, Macromolecules 2012, 45,
7317–7328.
[2] A. P. Esser-Kahn, S. A. Odom, N. R. Sottos, S. R. White, J. S. Moore,
Macromolecules 2011, 44, 5539–5553.
[3] A. P. Esser-Kahn, N. R. Sottos, S. R. White, J. S. Moore, J. Am. Chem.
Soc. 2010, 132, 10266–10268.
[4] R. Weinstain, A. Sagi, N. Karton, D. Shabat, Chem. Eur. J. 2008, 14,
6857–6861.
[5] N. Fomina, C. McFearin, M. Sermsakdi, O. Edigin, A. Almutairi, J. Am.
Chem. Soc. 2010, 132, 9540–9542.
[6] E. K. Y. Chen, R. A. McBride, E. R. Gillies, Macromolecules 2012, 45,
7364–7374.
[7] H. Zhang, K. Yeung, J. S. Robbins, R. A. Pavlick, M. Wu, R. Liu, A. Sen,
S. T. Phillips, Angew. Chem. Int. Ed. 2012, 51, 2400–2404.
[8] W. Seo, S. T. Phillips, J. Am. Chem. Soc. 2010, 132, 9234–9235.
[9] K. M. Schmid, L. Jensen, S. T. Phillips, J. Org. Chem. 2012, 77,
4363–4374.
EXPERIMENTAL
[10] J. S. Robbins, K. M. Schmid, S. T. Phillips, J. Org. Chem. 2013, 78,
3159–3169.
[11] M. Randić, Chem. Rev. 2003, 103, 3449–3605.
[12] Alternatively, we could have monitored the release of phenol,
but the aniline intermediate absorbed more strongly in the HPLC
spectra than phenol, and therefore the peak for the aniline interme-
diate was easier to integrate and quantify than the peak for phenol.
[13] The crystal structure data for 1 is available in reference.[9]
[14] E. E. Weinert, R. Dondi, S. Colloredo-Melz, K. N. Frankenfield,
C. H. Mitchell, M. Freccero, S. E. Rokita, J. Am. Chem. Soc. 2006,
128, 11940–11947.
General Conditions for Measuring the Release Kinetics of 1, 2
and 3. An Alloc-protected compound (1, 2, or 3) (0.01 mmol)
and tetrakis(triphenylphosphine)palladium(0) (1.1 mg) were
dissolved in 0.1 mL of tetrahydrofuran in a 2-mL vial. Acetic acid
(1.5 mL) and tributyltin hydride (5 mL) were added to the reaction
mixture, and the solution was shaken for 3 min. An aliquot
(10 mL) of the solution was added to an high-performance liquid
chromatography vial and was diluted with acetonitrile (0.5 mL)
[15] M. P. Hay, B. M. Sykes, W. A. Denny, C. J. O’Connor, J. Chem. Soc.-
Perkin Trans. 1 1999, 19, 2759–2770.
wileyonlinelibrary.com/journal/poc
Copyright © 2013 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2013, 26 608–610