Determining the Origin of Rate-Independent Chemoselectivity in CuAAC Reactions: An Alkyne-Specific Shift in Rate-Determining Step

Article abstract of DOI:10.1002/anie.201612288

We report a kinetic and spectroscopic analysis of alkyne-dependent chemoselectivity in the copper-catalyzed azide–alkyne click (CuAAC) reaction. Studies of six alkyne subtypes reveal that the rate-determining step (RDS) of an aromatic ynamine class is shi

Full text of DOI:10.1002/anie.201612288

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201612288
German Edition: DOI: 10.1002/ange.201612288
Chemoselectivity
Determining the Origin of Rate-Independent Chemoselectivity in
CuAAC Reactions: An Alkyne-Specific Shift in Rate-Determining
Step
Ciaran P. Seath, Glenn A. Burley,* and Allan J. B. Watson*
Abstract: We report a kinetic and spectroscopic analysis of
alkyne-dependent chemoselectivity in the copper-catalyzed
azide–alkyne click (CuAAC) reaction. Studies of six alkyne
subtypes reveal that the rate-determining step (RDS) of an
aromatic ynamine class is shifted from acetylide formation to
the azide ligation/migratory insertion event allowing chemo-
selectivity independent of overall rate.
T
he Cu-catalyzed azide–alkyne cycloaddition (CuAAC) or
click” reaction is an essential transformation used through-
out medicinal chemistry, chemical biology, and the material
“
[1,2]
sciences.
The power and utility of the CuAAC reaction is
greatly enhanced by the design of multifunctional molecular
scaffolds containing several reactive sites that can be used for
[3]
sequential chemoselective reactions. The success of this
strategy relies on the exploitation of established reactivity
[
3–6]
profiles of the alkyne and azide partners.
[
10]
Scheme 1. Fokin’s mechanism of the CuAAC reaction. a) Chemo-
selectivity via acetylide formation RDS. b) Chemoselectivity via azide
ligation/insertion RDS.
Current strategies that enable chemoselective control of
reactive groups of the CuAAC reaction exploit differences in
the reactivity of the alkyne and azide reagents. Specifically,
1
) Zhu et al. have shown that chelating groups enable
[4]
chemoselective control over the azide component.
Here we analyze the mechanistic origins of rate-inde-
pendent chemoselectivity in CuAAC reactions of an aryl
ynamine class of alkyne. Kinetic and spectroscopic investiga-
tions support the transition of the RDS from acetylide
formation to azide ligation/migratory insertion.
2
) Carell and Leigh et al. have shown that silyl protecting
group strategies provide a robust method for chemoselective
control in multi-alkyne systems.
developed chemoselective CuAAC reactions of 1-iodo-
alkynes that, due to their mechanistically distinct operation,
have allowed for highly chemoselective reactions versus
conventional terminal alkynes.
Based on work by Finn and Fokin, and Zhu, acetylide
formation is the rate-determining step (RDS) of the CuAAC
[3,5]
3) Fokin et al. have
Recently we have shown that aromatic ynamines are
highly reactive alkyne substrates, which allow chemoselective
CuAAC reactions in the presence of aliphatic alkynes
[
6]
[7]
[8]
[12]
irrespective of the nature of the azide used in the reaction.
To probe and compare the reactivity of different classes of
alkyne more broadly, we performed a series of competition
experiments where equistoichiometric quantities of two
terminal alkynes compete for one equivalent of a common
[9]
reaction (Scheme 1). Chemoselective discrimination of
alkynes is therefore possible by exploiting differences in
their respective rates of acetylide formation. Modification of
the alkyne to incorporate groups that facilitate a more rapid
insertion of Cu into the alkyne CꢀH bond has been a strategy
azide component (benzyl azide, BnN ) using established
3
reaction conditions (Scheme 2, see the Supporting Informa-
[
4,8,12]
employed by Finn, Hsung, and others in order to leverage
tion (SI) for all triazole products).
[10]
chemoselective control in systems containing two alkynes,
with mechanistic investigations focusing on this first key
mechanistic event of the catalytic cycle.
The alkyne selection was based on representative mem-
bers of six specific classes—aromatic ynamine (1), tertiary
propiolamide (2), ynamide (3), propargyl alkyne (4), aryl
alkyne (5), and alkyl alkyne (6). Previous work by Finn has
shown that 2 is a highly reactive substrate under conventional
CuAAC conditions, outcompeting 4–6 in competition exper-
[
7–9,11]
[
*] C. P. Seath, Dr. G. A. Burley, Dr. A. J. B. Watson
Department of Pure and Applied Chemistry, University of Strathclyde
Glasgow, G1 1XL (UK)
[10a]
[10b,c]
iments.
In addition, Hsung has shown that 3 > 5 > 6.
[
12]
Lastly, we have shown that 1 > 6. To assist in the selection
E-mail: glenn.burley@strath.ac.uk
of alkyne partners for chemoselective CuAAC applications,
allan.watson.100@strath.ac.uk
[
13]
we sought to generate a reactivity scale that may inform the
selection of alkyne partners for chemoselective click applica-
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
http://dx.doi.org/10.1002/anie.201612288.
[3,10a]
tions in multifunctional systems.
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Scheme 2. Competition CuAAC reactions of systems containing two
alkynes. Ratios are products arising from vertical number:horizontal
1
number. Determined by H NMR spectroscopy. See SI.
The data in Scheme 2 revealed the reactivity trend 1 > 2 >
3
> 4 > 5 > 6. Ynamine 1 was the most reactive alkyne
substrate, outcompeting all five other classes. Exclusive
formation of the ynamine triazole product was observed
over 3–6 and favored over 2 by approx. 8:1. Based on these
initial observations, we sought to establish the origins of the
rate acceleration of this chemotype in CuAAC reactions.
1
Kinetic measurements by H NMR allowed the establish-
ment of rate data for each of the six classes of alkyne. Charts
for alkynes 1 and 2 are provided in Figure 1; charts for 3–6 are
provided in the SI.
ꢀ
4
ꢀ1
Based on the extracted kobs, alkyne 2 (2.4 ꢀ 10 m s ) was
ꢀ5
ꢀ1
fastest followed by ynamine 1 (5.5 ꢀ 10 m s ), which was
ꢀ
5
ꢀ1
marginally faster than alkyne 3 (5.1 ꢀ 10 m s ). The relative
order of reactivity was in agreement with literature observa-
tions, including the requirement for reductant additives, such
as sodium ascorbate (NaAsc), to promote the reactions of 4 as
Figure 1. Reaction profiles for alkynes a) 1 and b) 2. Reaction profile
for competition reaction of 1 vs. 2 and overall rate data for alkynes
1
1
–6. Determined by H NMR spectroscopy, see SI. [a] Using NaAsc
[
14]
well as the less reactive aryl (5) and alkyl (6) alkynes.
(
10 mol%).
The kinetic data for 3–6 displayed a significant initiation
time (3, 16 min; 4–6, > 1 h), which, based on previous studies,
II
I
we attribute to reduction of Cu to a catalytically reactive Cu
species likely through oxidative homocoupling (Glaser cou-
only 1.17 was observed for 1 (Figure 2a), indicating that Cu-
acetylide formation was not rate limiting.
Further experiments revealed that the reaction rate of
1 was largely independent of [alkyne] (Figure 2b) but was
heavily influenced by variation of [BnN ] (Figure 2c). A zero-
order dependency on [alkyne], and approx. second order
[
8a]
pling) of the alkyne. This initiation time could partly assist
in explaining the origin of selectivity between 1 and 3, where
the reaction rates are almost identical. However, if the
3
II
initiation time is a function of Cu reduction, 1 clearly
I
[15]
generates Cu more rapidly (initiation time ꢁ 5 min), and this
dependency on [RN ] were extracted (see SI). This was
3
would then accelerate both CuAAC reactions in the com-
petition experiment, which was not observed. As such, the
chemoselectivity in the competitive reaction containing 1 and
a remarkable contrast to the data presented for more
conventional alkynes, which display first order kinetics with
respect to [alkyne] and a slight negative dependency in
[7]
3
must have a different origin. Similarly, 1 has a slower rate
azide. Taken together, all of these data are consistent with
a shift in the RDS of the CuAAC reaction using 1 as the
alkyne substrate from acetylide formation to the azide
ligation/migratory insertion event. These data assist in the
explanation of the discrepancy between the observations that
1 outcompetes the other alkyne classes in the competition
experiments in Scheme 2, despite displaying a rate that is
slower than 2 and similar to 3 and 4.
than 2 but is the favored alkyne in the competition experi-
ments, again, indicating a different origin of the observed
chemoselectivity profile.
Deuterium kinetic isotope effect (KIE) experiments of
the alkyne CꢀH were informative. Zhu et al. reported
a primary KIE of 2.3 when using an aryl alkyne (such as 5)
and a marked increase in the initiation time associated with
[
8]
oxidative homocoupling. In agreement with these data, we
observed primary KIEs (e.g., KIE for 2 = 2.66) and a signifi-
cant increase in initiation time (see SI). In contrast a KIE of
Upon further analysis of the spectral data, the rate of Cu
acetylide formation could be delineated from the primary rate
data (Figure 2d). To further confirm our hypothesis, the rate
2
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Scheme 3. Competition CuAAC reaction between 9 and 2. Determined
by NMR spectroscopy. See SI.
negligible in the reaction of 9 vs. 2, implying a pivotal role of
N3 of 1.
NMR analysis was used to probe the role of the
I
benzimidazole N3 (Scheme 4). Treatment of 1 with Cu or
II
Cu results in acetylide formation, which hindered inter-
II
1
Figure 2. a) Kinetic isotope effect, b) the effect of [alkyne] variation,
Scheme 4. a) Coordination of Cu to N3 of 1. b) pK and H NMR
a
c) [BnN ] variation, and d) acetylide formation vs. product formation
for 1.
resonance of 1 and 1!LA complexes. c) Rationale for slower azide
ligation. LA=Lewis acid.
3
of acetylide formation was twice as fast as the reaction as
a whole. Accordingly, these data suggest that chemoselectiv-
ity is attained through a very rapid Cu insertion step, which is
followed by a comparatively slow azide ligation/insertion
step.
rogation of any N-Cu interaction. However, addition of TIPS-
protected ynamine 11 to Cu(OAc) resulted in the immediate
2
formation of a new complex (12, Scheme 4a). No change was
II
observed on treatment of 11 with CuOAc, suggesting a Cu -
To better understand the origin of these observations,
a series of control experiments were performed to understand
the influence of the benzimidazole in the reaction. We
postulated that an interaction between copper and the
specific interaction. Similarly, no change was observed upon
treatment of the equivalent TIPS-protected 9 (13) with
Cu(OAc) or CuOAc (Scheme 4a). The experiments suggest
2
II
coordination of Cu to N3 of 1.
benzimidazole through N3 lowers the acetylinic pK , resulting
The pK of the alkyne CꢀH of 1 was determined to be 27.9
a
a
1
in rapid Cu-acetylide formation. Indeed, synthesis and kinetic
evaluation of the corresponding indole ynamine 9 revealed
that the imidazole N3 is essential for chemoselectivity
(MeCN) by NMR titration with a H NMR resonance of the
alkyne CꢀH at 3.67 ppm (Scheme 4b). Treatment of 1 with
Lewis acids that do not undergo CꢀH insertion (AlCl ,
3
(
Scheme 3).
Sc(OTf) ) confirmed a substantial increase in the alkyne CꢀH
3
[16]
In contrast to the ca. 8:1 selectivity displayed by 1 in the
acidity, as demonstrated by a significant downfield shift.
same competition experiment, chemoselectivity was almost
Cu-based Lewis acids could not be used for this purpose due
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
to rapid acetylide formation. While the CuAAC process is Acknowledgements
I
catalyzed by Cu , UV analysis of reaction mixtures using
I
1
demonstrated that the reduction of Cu(OAc) to Cu only
We thank the Carnegie Trust for a PhD Scholarship (C.P.S.),
G.A.B. thanks the Alexander von Humboldt Foundation
(Visiting Research Fellowship, LMU Munich). We thank
Marine Hatit for the preparation of compound 1, Liam
McLean for useful discussions, Craig Irving for assistance with
NMR analyses, Julien Vantourout for assistance with mass
spectroscopy, and the EPSRC UK National Mass Spectrom-
etry Facility at Swansea University for analyses. We thank
a reviewer for an insightful point relating to azide depend-
ency.
2
II
proceeds to ca. 50%: Cu persists in the reaction mixture.
HRMS of reaction mixtures supported these observations—
a Cu (1) complex was detected while no interaction was
II
2
II
observed with Cu and 9 (see SI).
We hypothesize that the origin of the observed slower
azide ligation event is due to the inherent electronics of 1.
Specifically, the alkyne is comparatively electron-rich due to
the resonance contribution of the azole. This will, in turn,
render Cu less electrophilic and slow down ligation/insertion
of the electron-rich azide (Scheme 4c). The comparatively
electron-rich nature of the alkyne of 1 is readily observed by
1
3
C NMR spectroscopy.
Conflict of interest
Collectively, we propose that the origin of chemoselec-
II
tivity of 1 vs. other alkyne classes arises from Cu coordina-
tion to the benzimidazole N3 resulting in an increase in
alkyne CꢀH acidity facilitating a rapid Cu acetylide formation
The authors declare no conflict of interest.
Keywords: chemoselectivity · click reactions · CuAAC ·
reaction kinetics · ynamine
(
i.e., k > k ; Scheme 5). The subsequent azide ligation/
3
1
insertion event is then comparatively slower based on the
electronics of the alkyne (i.e., k < k ) and, as such, the RDS
4
3
transitions to this event. Based on these data, the chemo-
selectivity-determining event is acetylide formation with RDS
dependent on alkyne class and subtype.
[
[
1] For seminal reports, see: a) H. C. Kolb, M. G. Finn, K. B.
Sharpless, Angew. Chem. Int. Ed. 2001, 40, 2004 – 2021; Angew.
Chem. 2001, 113, 2056 – 2075; b) C. W. Tornøe, C. Christensen,
M. Meldal, J. Org. Chem. 2002, 67, 3057 – 3064.
2] a) J. E. Hein, V. V. Fokin, Chem. Soc. Rev. 2010, 39, 1302 – 1315;
b) M. Meldal, C. W. Tornøe, Chem. Rev. 2008, 108, 2952 – 3015;
c) J. E. Moses, A. D. Moorhouse, Chem. Soc. Rev. 2007, 36,
1249 – 1262.
[
3] P. M. E. Gramlich, S. Warncke, J. Gierlich, T. Carell, Angew.
Chem. Int. Ed. 2008, 47, 3442 – 3444; Angew. Chem. 2008, 120,
3
491 – 3493.
4] Z. Yuan, G. C. Kuang, R. J. Clark, L. Zhu, Org. Lett. 2012, 14,
590 – 2593.
[
2
[
[
5] V. Aucagne, D. A. Leigh, Org. Lett. 2006, 8, 4505 – 4507.
6] For seminal reports see: a) J. E. Hein, J. C. Tripp, L. B. Krasnova,
K. B. Sharpless, V. V. Fokin, Angew. Chem. Int. Ed. 2009, 48,
Scheme 5. Comparison of the chemoselectivity-determining and rate-
determining events for standard alkynes and aryl ynamines.
8
018 – 8021; Angew. Chem. 2009, 121, 8162 – 8165; b) D. N.
Barsoum, N. Okashah, X. Zhang, L. Zhu, J. Org. Chem. 2015,
0, 9542 – 9551; For synthetic examples see: c) K. Yamamoto, T.
Bruun, J. Y. Kim, L. Zhang, M. Lautens, Org. Lett. 2016, 18,
644 – 2647; d) W. Wang, F. Wei, Y. Ma, C.-H. Tung, Z. Xu, Org.
Lett. 2016, 18, 4158 – 4161.
[7] V. O. Rodionov, V. V. Fokin, M. G. Finn, Angew. Chem. Int. Ed.
005, 44, 2210 – 2215; Angew. Chem. 2005, 117, 2250 – 2255.
8
In conclusion, we have demonstrated that the RDS of the
CuAAC reaction is non-constant and can be shifted away
from the generally observed acetylide formation RDS. A
benzimidazole class of aromatic ynamine undergoes chemo-
selective CuAAC reaction over a range of alkynes even in the
presence of alkyne substrates that react, in isolation, at
a faster overall rate (e.g., 2). Kinetic investigation shows that
the RDS of the reaction has shifted from Cu-acetylide
formation to azide ligation due to a rapid Lewis acid-assisted
acetylide formation step. To the best of our knowledge this is
the only report of chemoselectivity despite a lower overall
rate. Up to this point, interrogation of this key mechanistic
event has been limited, due to its rapidity with respect to
acetylide formation. We anticipate that the unusual reactivity
presented here will enable more detailed studies into the
azide ligation/insertion step of this cornerstone reaction.
2
2
[
8] a) G. C. Kuang, P. M. Guha, W. S. Brotherton, J. T. Simmons,
L. A. Stankee, B. T. Nguyen, R. J. Clark, L. Zhu, J. Am. Chem.
Soc. 2011, 133, 13984 – 14001; b) X. Zhang, P. Liu, L. Zhu,
Molecules 2016, 21, 1697.
9] B. T. Worrell, J. A. Malik, V. V. Fokin, Science 2013, 340, 457 –
460.
[10] a) A. A. Kislukhin, V. P. Hong, K. E. Breitenkamp, M. G. Finn,
Bioconjugate Chem. 2013, 24, 684 – 689; b) X. Zhang, H. Li, L.
You, Y. Tang, R. P. Hsung, Adv. Synth. Catal. 2006, 348, 2437 –
[
2
2
442; c) X. Zhang, R. P. Hsung, L. You, Org. Biomol. Chem.
006, 4, 2679 – 2682.
[
11] a) L. Jin, D. R. Tolentino, M. Melaimi, G. Bertrand, Sci. Adv.
015, 1, e1500304; b) J. Jin, E. A. Romero, M. Melaimi, G.
Bertrand, J. Am. Chem. Soc. 2015, 137, 15696 – 15698.
2
4
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
[
12] M. Z. C. Hatit, J. C. Sadler, L. A. McLean, B. C. Whitehurst,
[15] A reviewer suggested that this second order dependency may be
II
C. P. Seath, L. D. Humphreys, R. J. Young, A. J. B. Watson, G. A.
Burley, Org. Lett. 2016, 18, 1694 – 1697.
due to an azide-promoted decomplexation of the N3-bound Cu .
See SI for a schematic description.
[
[
13] For examples of reactivity scales, see: H. Mayr, A. R. Ofial, Acc.
Chem. Res. 2016, 49, 952 – 965.
14] F. Himo, T. Lovell, R. Hilgraf, V. V. Rostovtsev, L. Noodleman,
K. B. Sharpless, V. V. Fokin, J. Am. Chem. Soc. 2005, 127, 210 –
[16] D. G. Lee, R. Cameron, Can. J. Chem. 1972, 50, 445 – 448.
Manuscript received: December 19, 2016
216.
Final Article published: && &&, &&&&
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Chemoselectivity
C. P. Seath, G. A. Burley,*
A. J. B. Watson*
&&& — &&&
Determining the Origin of Rate-
Independent Chemoselectivity in CuAAC
Reactions: An Alkyne-Specific Shift in
Rate-Determining Step
The old switcheroo: A kinetic and spec-
troscopic analysis of alkyne-dependent
chemoselectivity in the copper-catalyzed
azide–alkyne click (CuAAC) reaction is
reported. Studies of six alkyne subtypes
reveal that the rate-determining step
(RDS) of an aromatic ynamine class is
shifted from acetylide formation to the
azide ligation/migratory insertion event
allowing chemoselectivity independent of
overall rate.
6
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!