Alkyne mechanochemistry: Putative activation by transoidal bending

Article abstract of DOI:10.1039/c4cc03514c

We investigated the use of mechanical stress to bend carbon-carbon triple bonds. Formation of an isoquinoline after reaction with a benzyl azide trap points towards a nucleophilic addition mechanism, differentiating mechanochemical trans-bending of π bonds from the typical reactivity observed for cisoidal bending of triple bonds in strained cyclic alkynes.

Full text of DOI:10.1039/c4cc03514c

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Alkyne mechanochemistry: putative activation by
transoidal bending†
Cite this: Chem. Commun., 2014,
50, 13235
Charles E. Diesendruck, Lingyang Zhu and Jeffrey S. Moore*
Received 9th May 2014,
Accepted 30th July 2014
DOI: 10.1039/c4cc03514c
www.rsc.org/chemcomm
We investigated the use of mechanical stress to bend carbon–
carbon triple bonds. Formation of an isoquinoline after reaction
with a benzyl azide trap points towards a nucleophilic addition
mechanism, differentiating mechanochemical trans-bending of
p bonds from the typical reactivity observed for cisoidal bending
of triple bonds in strained cyclic alkynes.
Modern polymer mechanochemistry serves not only to study the
mechanical degradation of polymer chains under stress, but also
to explore new approaches to drive productive chemical change.1
Potential applications of non-degradative polymer mechano-
Fig. 1 Some examples of trans-bent alkynes previously studied by solid-
state crystallography.12
chemistry are in the development of self-healing2 and self- react as an electrophile or a di-radical.11 Trans-bent alkynes
reinforcing3 materials as well as other interesting uses, such as were observed in the solid state study of ortho-substituted phenyl
detection of the stress history in a material,4 changing the energy acetylenes as well as peri-substituted naphthalenes (Fig. 1).12
plane of chemical reactions to obtain products inaccessible by However, the bending angles were much smaller compared to
thermal mechanisms,5 pre-catalyst activation6 and more.
The design of new mechanochemically responsive functional of mechanical stress to bend alkynes to larger angles.
molecules (mechanophores) has mostly focused on the stretching Following polymer extension, a chain-centered triple-bond is
the angles in strained cyclooctynes.13 Here, we investigate the use
of weak bonds; however, Boydston and Craig have recently expected to be tilted with respect to the force axis, and therefore
demonstrated mechanochemical bond scission based on angle transoidal bending is expected (Fig. 2a). As a first step to test this
bending of bicyclic compounds.7 The energy involved in chemical notion, we used DFT calculations to simulate the stretching of a
bond stretching is significantly higher than the energy necessary model triple bond to examine the deformation state prior to bond
to bend bond or dihedral angles.8 An interesting, still unexplored scission. Model alkyne 1 was studied using the CoGEF (con-
mechanochemical reaction, is the bending of p bonds. Given the strained geometries simulate external force) technique.14 In the
recent success of the ‘‘copper-free click-chemistry’’9 where the minimized structure, deformation up to a 261 angle (1801 À a)
azide–alkyne Huisgen cycloaddition10 is significantly accelerated between the force vector and the triple bond is observed; there-
by using a cis-bent carbon–carbon triple bond, we decided to fore, stretching in the direction of the force vector should lead to
study if trans-bending of a triple bond would lead to enhanced bending of the angles in the triple bond away from linearity and
reactivity.
towards a trans-bending mode (Fig. 2). The terminal carbons in
Theoretical studies showed that during pyramidalization of the molecule were constrained at increasingly higher distances,
the triple bond, the energy of the anti-bonding p* is signifi- and the energy, as well as the two –H2C–CRC– angles were
cantly reduced while the energy of the p orbital remains mostly measured for each step. Before any bond scission event, the
unchanged, suggesting that the distorted triple bond could –H2C–CRC– angle bends to a minimum of 163.51 at relatively
low energies. At this point, the force vector and the triple bond
are almost aligned (Fig. 2b). Further stretching leads to a fast
energy increase towards bond scission, while the bond angle
Department of Chemistry, University of Illinois at Urbana-Champaign, Urbana,
USA. E-mail: jsmoore@illinois.edu
does not change. While 163.51 is larger compared to the angles
observed in cyclooctynes (152–1621),13 this result indicates
† Electronic supplementary information (ESI) available: Experimental procedures,
NMR spectra and CoGEF calculations. See DOI: 10.1039/c4cc03514c
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 13235--13238 | 13235

View Article Online
Communication
ChemComm
Solvodynamic shear from an acoustic field was used to
induce strain in the polymer chain, since during this type of
elongational flow unfolding of the chain is expected, leading to a
narrow distribution of angles between the force vector and the triple
bond (in the solid state all angles are possible).1c,17 A dilute solution
of polymer 5 (1 mg mLÀ1) and 4-nitrobenzylazide (12 000 equiv. to
triple bond) in acetonitrile was prepared and sonicated at 6–9 1C.
GPC analysis showed incorporation of the trap to the polymer as
evidenced by the presence of a polymer peak at the UV-detector
following absorption at 280 nm (Fig. 3A). When the solution of
polymer and azide is left standing for 3 days at room temperature
(no sonication), no peak at 280 nm is observed (Fig. 3B), indicating
the reaction is not thermal. Comparison of the UV spectra of the
sonicated and non-sonicated polymers (obtained from the PDA
detector in the GPC), shows the addition of a new absorption band
between 250–300 nm, which is also present in the UV-spectra of the
trap azide, indicating the trap is chemically linked to the polymer
(Fig. 3C). To reinforce this result, we repeated the experiment with
1-(azidomethyl)pyrene as a trap. Again, the sonicated polymer shows
incorporation of the trap, clearly observed by the addition of the
four typical absorption bands of the pyrenyl group to the polymer
spectra (Fig. 3D and F). An incorporation of 14.4% was calculated
using the 344 nm absorption of pyrene (see the ESI†). When control
polymer 6 (having a triple bond at the end of the chain, which is not
mechanically stressed) is sonicated with the 1-(azidomethyl)pyrene
trap under similar conditions, no incorporation of the pyrenyl
trap is observed (Fig. 3E), supporting that the reaction is
mechanochemical (not thermal) and occurs at the triple-bond site.
Interestingly, polymers with Mw below 30 kDa are typically used
as controls, since they do not undergo chain scission during
sonication.17 However, when 5small (30 kDa, see the ESI†) is
sonicated with the 4-nitrobenzyl azide trap, incorporation is
still observed, albeit without chain scission.
Finally, we set out to study the product formed by the
mechanochemical reaction using 15N isotope labelling and
15N NMR spectroscopy to identify the mechanochemically
generated product. A 15N labelled trapping azide was prepared
by reaction of 4-nitrobenzyl bromide with sodium azide-1-15N,
producing two isotopic products, where the 15N labelled nitrogen
can be either bound to the carbon or terminal (Scheme 2).
Polymer 5 was sonicated with the labelled 4-nitrobenzyl azide
and analyzed by 15N NMR. However, given the low sensitivity of
15N NMR combined with the extremely low concentration of the
labelled atoms (a single atom per polymer chain), no peaks were
observed even after two days of data collection. Therefore, we
decided to use gel-phase NMR18 to reduce the scan time while
working at a high polymer concentration. The NMR sample was
in turn prepared by weighting 200 mg dry polymer in an NMR
tube, and adding 0.5 mL of CDCl3. The polymer was left to swell
for a day before analysis.
Fig. 2 (a) Mechanochemical trans-bending of triple bond (s represents
applied stress). (b) Stretched 1 from DFT calculations. (c) CoGEF calcula-
tions of model compound 1, where energy (blue) and –CRC–CH2– bond
angles (red) are shown.
that significant bending of the triple bond occurs before
bond scission.
If trans-bent alkynes react similarly to cis-bent alkynes, a
Huisgen cycloaddition between a polymer chain-centered alkyne
with a trapping azide would produce a 1,2,3-triazole attached to
the polymer at the 4,5 positions. An understanding of the mechano-
chemical stability of the product is needed prior to starting
experimental testing, given that Bielawski and co-workers demon-
strated the mechanochemical retrocycloaddition of 1,2,3-triazoles
attached to polymers at the 1,4 and 1,5 positions.15 We therefore
calculated the CoGEFs profiles for a model triazole, stretching in
the direction of the 1,4 vector and the 4,5 vector (see the ESI†). The
previously demonstrated retrocycloaddition occurs clearly in the
first case in a concerted manner, where two bonds are cleaved in a
single step, and the energy of the system decreases. However, when
the force is applied along the direction of the 4,5 vector, a single
bond is cleaved, and only at a much higher energy, causing a ring
opening reaction without scission of the polymer chain.
A poly(methyl acrylate) (PMA) containing a chain-centered
triple-bond (5) was prepared by SET-LRP16 (Scheme 1). For
control experiments, a similar PMA having the triple bond at
the end of the chain was prepared. The alkyne at the end of the
chain in 6 is not affected by ultrasound’s solvodynamic shear
forces, and therefore should be mechanochemically unreactive
toward the trap. As traps, 4-nitrobenzyl azide and 1-(azidomethyl)-
pyrene were prepared.
Searching the literature, we were surprised to find a single
example of gel-phase NMR using 15N.19 Furthermore, impor-
tant experimental details such as the delay time used were not
provided. Measuring T1 and T2 directly is not possible using our
polymer with a single labelled nitrogen per chain. To choose an
appropriate delay and acquisition time, we decided to use
Scheme 1 Synthesis of polymers and azide traps used in the study.
13236 | Chem. Commun., 2014, 50, 13235--13238
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
Fig. 3 (A) GPC traces of 5 after sonication with 4-nitrobenzyl azide (Mn = 67.8 kDa, PDI = 1.30). (B) GPC trace of 5 and 4-nitrobenzyl azide after 3 days at
room temperature. (C) UV spectrum of 4-nitrobenzyl azide (red) and 5 (black) after sonication with 4-nitrobenzyl azide (RT = 28.3 min). (D) GPC trace of 5
after sonication with 1-(azidomethyl)pyrene (Mn = 70.1 kDa, PDI = 1.24). (E) GPC trace of control polymer 6 after sonication with 1-(azidomethyl)pyrene
(Mn = 63.9 kDa, PDI = 1.31). (F) UV spectrum of 1-(azidomethyl)pyrene (red) and 5 (black) after sonication with 1-(azidomethyl)pyrene (RT = 28.3 min).
Fig. 4 15N NMR of (A) 5 after sonication with labelled 4-nitrobenzyl azide;
(B) model polymer with chain-centered labelled 1,2,3-triazole; and (C) model
polymer (Mn = 93 kDa, PDI = 1.17) with chain-centered labelled 1,2,3-triazole
Scheme 2 Synthesis of 15N labelled azide trap and model labelled polymer.
after sonication (1 mg mLÀ1, pulsed ultrasound 0.5 s on, 1.0 s off, 8.7 W cmÀ2
,
3 h). After sonication Mn = 50.8 kDa, PDI = 1.33.
T1 and T2 values determined experimentally from a 15N-labelled
protein in a gel.20 The actual values for our experiment are around À20 and À130 ppm, corresponding to the different
likely smaller; however, the values from Ju¨rgen Sassa et al. possible positions of the 15N labelled nitrogen (Fig. 4B and ESI†).
significantly decrease our measurement time. Also the extra delay On the other hand, polymer 5 sonicated with the 15N labelled
time would be plenty for complete decay. Delay and acquisition 4-nitrobenzyl azide, showed a single peak at À69 ppm (indicating
times were set to 175 ms and 115 ms, respectively. After two days, loss of N2 during sonication). We identified other possible
we were able to obtain a 15N signal with a reasonable signal to products by examining the extensive 15N NMR chemical shift
noise ratio (Fig. 4A).
study by Witanowski et al.21 Remarkably, few chemical function-
To identify the product, we prepared model polymers with alities exhibit signals at this chemical shift most are 6-membered
chain-centered labelled 1,2,3-triazole (Scheme 2). All labelled aromatic heterocycles where the nitrogen atom is linked to two
triazole model molecules presented two distinct 15N NMR signals, carbon atoms, such as quinolines and isoquinolines.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 13235--13238 | 13237

View Article Online
Communication
ChemComm
This material is based upon work supported by the National
Science Foundation (CHE-1300313). The authors acknowledge
Brendan Mar and Koushik Ghosh for helpful discussions.
Notes and references
1 (a) K. M. Wiggins, J. N. Brantley and C. W. Bielawski, Chem. Soc.
Rev., 2013, 42, 7130; (b) Z. S. Kean and S. L. Craig, Polymer, 2012,
53, 1035; (c) M. M. Caruso, D. A. Davis, Q. Shen, S. A. Odom,
N. R. Sottos, S. R. White and J. S. Moore, Chem. Rev., 2009, 109, 5755.
2 (a) J. Hentschel, A. M. Kushner, J. Ziller and Z. Guan, Angew. Chem.,
Int. Ed., 2012, 51, 10561; (b) X. Chen, M. A. Dam, K. Ono, A. Mal,
H. Shen, S. R. Nutt, K. Sheran and F. Wudl, Science, 2002, 295, 1698;
Scheme 3 Possible mechanism for formation of iso-quinoline during
mechanochemical reaction of alkyne with azide. ‘‘P’’ represents attach-
ment to the polymer.
´
(c) P. Cordier, F. Tournilhac, C. Soulie-Ziakovic and L. Leibler,
Nature, 2008, 451, 977; (d) C. E. Diesendruck and J. S. Moore,
in Self-Healing Polymers: From Principles to Applications, ed. W. H.
Binder, Wiley-VCH, Weinheim, 2013, ch. 8, pp. 191–211.
3 A. L. Black-Ramirez, Z. S. Kean, J. A. Orlicki, M. Champhekar,
S. M. Elsakr, W. E. Krause and S. L. Craig, Nat. Chem., 2013, 5, 757.
4 (a) D. A. Davis, A. Hamilton, J. Yang, L. D. Cremar, D. Van Gough,
Interestingly, there are numerous examples in the literature
on the radical and thermal decomposition of 1,2,3-triazoles
yielding quinolines and isoquinolines.22 We considered the
possibility of a secondary mechanochemical reaction giving
such a product. To test this possibility, we sonicated the model
polymer containing the chain-centered labelled 1,2,3-triazole.
After purification of the polymer, a new small peak at À69 ppm
could be observed by gel-phase 15N NMR (Fig. 4C).
Given that the À69 ppm peak is the only observed peak in
the sonication of 5 with the labelled azide, and not a secondary
peak as observed in the sonication of the triazole, it was clear
to us that the mechanochemical reaction does not cause a
Huisgen cycloaddition.
´
S. L. Potisek, M. T. Ong, P. V. Braun, T. J. Martınez, S. R. White,
J. S. Moore and N. R. Sottos, Nature, 2009, 459, 68; (b) Y. Chen,
A. J. H. Spiering, S. Karthikeyan, G. W. M. Peters, E. W. Meijer and
R. P. Sijbesma, Nat. Chem., 2012, 4, 559.
5 C. R. Hickenboth, J. S. Moore, S. R. White, N. R. Sottos, J. Baudry and
S. R. Wilson, Nature, 2007, 446, 423.
6 (a) C. E. Diesendruck, B. D. Steinberg, N. Sugai, M. N. Silberstein,
N. R. Sottos, S. R. White, P. V. Braun and J. S. Moore, J. Am. Chem.
Soc., 2012, 134, 12446; (b) A. Piermattei, S. Karthikeyan and R. P.
Sijbesma, Nat. Chem., 2009, 1, 133; (c) A. G. Tennyson, K. M. Wiggins
and C. W. Bielawski, J. Am. Chem. Soc., 2010, 132, 16631.
7 (a) M. B. Larsen and A. J. Boydston, J. Am. Chem. Soc., 2013,
135, 8189; (b) G. R. Gossweiler, G. B. Hewage, G. Soriano,
Q. Wang, G. W. Welshofer, X. Zhao and S. L. Craig, ACS Macro Lett.,
2014, 3, 216.
8 R. P. Wool and R. H. Boyd, J. Appl. Phys., 1980, 5116.
9 J. M. Baskin, J. A. Prescher, S. T. Laughlin, N. J. Agard, P. V. Chang,
I. A. Miller, A. Lo, J. A. Codelli and C. R. Bertozzi, Proc. Natl. Acad.
Sci. U. S. A., 2007, 16793.
Previous studies on bent triple bonds argue that the reduction
in the energy of the p* orbital, without change in the p orbital,
allows the alkyne to react as an electrophile. Scheme 3 presents a
possible reaction mechanism to form isoquinoline. Interestingly,
1H-azirine was proposed as an intermediate in the pyrolysis of
1,2,3-triazoles to isoquinolines. According to the mechanism
10 R. Huisgen, Proc. Chem. Soc., 1961, 357.
1
proposed by Gilchrist et al., H-azirine undergoes ring opening 11 (a) A. D. Walsh, J. Chem. Soc., 1953, 2288; (b) W. F. Maier, G. C. Lau
and A. B. McEwen, J. Am. Chem. Soc., 1985, 107, 4724; (c) W. E.
to form the iminocarbene, which undergoes cyclization, leading
Kammer, Chem. Phys. Lett., 1970, 6, 529; (d) S. Y. Chu and
L. Goodman, J. Am. Chem. Soc., 1975, 97, 7.
to the formation of the 1,4-dihydroisoquinoline which rapidly
oxidizes by air to the isoquinoline product.22b
12 (a) P. C. Bell, W. Skranc, X. Formosa, J. O’Leary and J. D. Wallis,
J. Chem. Soc., Perkin Trans. 2, 2002, 878; (b) C. R. Riceand and J. D.
Wallis, J. Chem. Soc., Chem. Commun., 1993, 572; (c) M. Pilkington,
S. Tayyip and J. D. Wallis, J. Chem. Soc., Perkin Trans. 2, 1994,
2482.
13 C. G. Gordon, J. L. Mackey, J. C. Jewett, E. M. Sletten, K. N. Houk and
C. R. Bertozzi, J. Am. Chem. Soc., 2012, 134, 9199.
We have studied the mechanochemical trans-bending of the
carbon–carbon triple bond. The reactive intermediates were
trapped using azides during sonication, and adequate controls
were used to rule out the possibility of thermal reactivity. While
Huisgen cycloaddition reactions occur easily with cisoidal alkynes,
14 M. K. Beyer, J. Chem. Phys., 2000, 112, 7307.
15N labelled traps pointed to isoquinolines and not 1,2,3-triazoles 15 (a) J. N. Brantley, K. M. Wiggins and C. W. Bielawski, Science, 2011,
333, 1606; (b) J. N. Brantley, S. S. M. Konda, D. E. Makarov and
C. W. Bielawski, J. Am. Chem. Soc., 2012, 134, 9882.
16 B. M. Rosen and V. Percec, Chem. Rev., 2009, 109, 5069.
as products. Based on theoretical and experimental studies, we
propose the bent and stretched alkyne reacts as an electrophile.
An alternative mechanism was proposed to explain the formation 17 P. A. May and J. S. Moore, Chem. Soc. Rev., 2013, 42, 7497.
18 H. Sternlicht, G. L. Kenyon, E. L. Packer and J. Sinclair, J. Am. Chem.
Soc., 1971, 93, 199.
19 E. E. Swayze, Tetrahedron Lett., 1997, 38, 8643.
of isoquinolines, based on previously proposed mechanisms for
the photolysis and pyrolysis of 1-benzyl-1,2,3-triazoles, however,
a mechanism involving sonochemically produced nitrenes or 20 H.-J. Sass, G. Musco, S. J. Stahl, P. T. Wingfield and S. Grzesiek,
J. Biomol. NMR, 2000, 18, 303.
21 M. Witanowski, L. Stefaniak and G. A. Webb, Annu. Rep. NMR
nitrenyl radicals cannot be discarded. This study points towards
a new way to activate triple bonds, and further studies on electronic
Spectrosc., 1993, 25, 1.
effects, as well as different nucleophiles and other electron rich 22 (a) S. J. Barker, G. B. Jones, K. R. Randles and R. C. Storr, Tetra-
hedron Lett., 1988, 29, 953; (b) T. L. Gilchrist, G. E. Gymer and
C. W. Rees, J. Chem. Soc., Perkin Trans. 1, 1975, 1; (c) A. R. Katritzky
dipoles are underway. This study also demonstrated the power of
15N gel-phase NMR as a key analytical technique to investigate the
`
and B. Yang, J. Org. Chem., 1998, 63, 1467; (d) M. P. Serve and
mechanism of mechanochemical reactions.
H. M. Rosenberg, J. Heterocycl. Chem., 1973, 10, 245.
13238 | Chem. Commun., 2014, 50, 13235--13238
This journal is ©The Royal Society of Chemistry 2014