A Revised Mechanism for the Kinugasa Reaction

Article abstract of DOI:10.1021/jacs.8b04635

Detailed kinetic analysis for the Cu(I)-catalyzed Kinugasa reaction forming β-lactams has revealed an anomalous overall zero-order reaction profile, due to opposing positive and negative orders in nitrone and alkyne, respectively. Furthermore, the reaction displays a second-order dependence on the catalyst, confirming the critical involvement of a postulated bis-Cu complex. Finally, reaction progress analysis of multiple byproducts has allowed a new mechanism, involving a common ketene intermediate to be delineated. Our results demonstrate that β-lactam synthesis through the Kinugasa reaction proceeds via a cascade involving (3 + 2) cycloaddition, (3 + 2) cycloreversion, and finally (2 + 2) cycloaddition. Our new mechanistic understanding has resulted in optimized reaction conditions to dramatically improve the yield of the target β-lactams and provides the first consistent mechanistic model to account for the formation of all common byproducts of the Kinugasa reaction.

Full text of DOI:10.1021/jacs.8b04635

Subscriber access provided by Kaohsiung Medical University
Article
A Revised Mechanism for the Kinugasa Reaction
Thomas C Malig, Diana Yu, and Jason E Hein
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b04635 • Publication Date (Web): 02 Jul 2018
Downloaded from http://pubs.acs.org on July 2, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
A Revised Mechanism for the Kinugasa Reaction
7
8
9
Thomas C. Malig, Diana Yu, and Jason E. Hein*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Department of Chemistry, The University of British Columbia, Vancouver, BC V6T 1Z1, Canada
ABSTRACT: Detailed kinetic analysis for the Cu(I)ꢀcatalyzed Kinugasa reaction forming βꢀlactams has revealed an anomalous
overall zeroꢀorder reaction profile, due to opposing positive and negative orders in nitrone and alkyne respectively. Furthermore,
the reaction displays a secondꢀorder dependence on catalyst, confirming the critical involvement of a postulated bisꢀCu complex.
Finally, reaction progress analysis of multiple byproducts has allowed a new mechanism, involving a common ketene intermediate
to be delineated. Our results demonstrate that βꢀlactam synthesis through the Kinugasa reaction proceeds via a cascade involving
(3+2) cycloaddition, (3+2) cycloreversion and finally (2+2) cycloaddition. Our new mechanistic understanding has resulted in opꢀ
timized reaction conditions to dramatically improve the yield of the target βꢀlactams and provides the first consistent mechanistic
model to account for the formation of all common byproducts of the Kinugasa reaction.
remains a significant obstacle, often resulting in modest yields
of the desired βꢀlactams.
INTRODUCTION
βꢀlactams represent a privileged scaffold in organic synthesis
in addition to their important role as the core motif in βꢀlactam
antibiotics such as penicillins, cephalosporins, monobactams,
and carbapanems.1,2 More recently molecules containing a βꢀ
lactam subunit have demonstrated additional therapeutic efꢀ
fects by functioning as antifungal,3 anticancer,4,5 and antiviral
agents.5,6 Furthermore, βꢀlactams make valuable synthons as
they are readily amenable to subsequent functionalization to
yield other biologically relevant building blocks such as βꢀ
amino acids and amino alcohols.7 High yielding convergent
methods for the syntheses of βꢀlactams capable of imparting
both enantioꢀ and diastereoselectivity are therefore of great
value.
The mechanism of the Kinugasa reaction has been a topic of
study for decades but remains an area of active research. Reꢀ
cently a working mechanism based on computational studies
for the catalytic Kinugasa reaction was proposed (Scheme 2).11
A key (3+2) cycloaddition between a copper acetylide (5) and
nitrone (2) generates a transient cuprated heterocycle 6.11 It
has been proposed that protonation of 7 at the nitrogen center
facilitates NꢀO bond cleavage, forming an acyclic ketene inꢀ
termediate 8. Intramolecular cyclization of 8 yields a copper
coordinated enolate 9. Ligand exchange of copper, as well as a
proton transfer affords a mixture of βꢀlactam diastereomers
(3).
One such method for the direct syntheses of βꢀlactams was
discovered by Kinugasa and Hashimoto in 1972 when they
reported that treatment of a copper acetylide with a nitrone
afforded βꢀlactams (Scheme 1).8
Scheme 1. The Kinugasa reaction
The Kinugasa reaction provides an attractive option for the
syntheses of βꢀlactams because of its optimal atom economy,
its employment of readily accessible starting materials, and
due to the convergency of the approach. The utility of the
Kinugasa reaction was increased when it was reported that
corresponding βꢀlactams could be synthesized asymmetrically
with ee’s up to 93%.9 Despite the demonstrated usefulness of
the reaction, there still exist several major limitations preventꢀ
ing it from being more universally adopted. A recent review
article by Chmielewski in 2014 deemed the Kinugasa reaction
“an “ugly duckling” of βꢀlactam chemistry”.10 The formation
of multiple byproducts throughout the course of the reaction
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
Scheme 2. Kinugasa reaction mechanism by Himo et al11
points to a common intermediate for generation of both lactam
and imine/amide products.
1
2
3
4
5
6
7
8
9
0.1
0.08
0.06
0.04
0.02
0
2
3a
3b
10
11
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
15
30
45
Time (min)
60
75
90
Figure 1. Reaction profile when DIPA was used as the basic
additive. Reaction conditions: [1]0 = [2]0 = [DIPA]0 = 0.1 M;
[Cu(MeCN)4BF4] = [TCPTA] = 0.01 M.
The conserved 1:1 ratio between byproducts 10 and 11 likely
stems from decomposition of the Cuꢀdihydroisoxazolide 7
intermediate (Scheme 3). This decomposition pathway would
involve cleavage of 7 via a formal (3+2) retrocycloaddition,
giving imine 10 and ketene 12, the latter of which could be
rapidly captured by a nucleophile (such as DIPA) to generate
generic product 13. Such a pathway is similar to the proposed
rearrangement of Nꢀsulfonyl Cuꢀtriazolides 14, which decomꢀ
pose to give the corresponding ketenimines (15) and nitrogen
gas.21 It should be noted that the formation of imine 10 and
ketene 12 (by way of a Cuꢀynoate intermediate) was originally
proposed by Miura, who invoked a direct Oꢀinsertion between
nitrone and Cuꢀacetylide.19 However, our reaction progress
and KIE measurements (vide infra) do not support this proꢀ
posal.
Despite the current understanding and synthetic usefulness of
the Kinugasa reaction, to the best of our knowledge there exꢀ
ists no detailed kinetic analysis for this reaction. Herein, we
report the use of reaction progress kinetic analysis (RPKA)12,13
in conjunction with VTNA14,15 to help delineate the underlying
mechanism with the goal to increase the yield of lactams and
provide insight into the pathways that form the undesired byꢀ
products often associated with the reaction.
Reactions were monitored using a prototype robotic sampling
apparatus coupled to HPLCꢀMS.16 This device provides realꢀ
time quantitative reaction progress data, without the need for
collection and quenching of aliquots, which would then be
analyzed offꢀline following reaction completion.
Results and Discussion
Given this new mechanistic insight, unproductive formation of
imine 10 and amide 11 may be arrested by using a stronger,
nonꢀnucleophilic amine base, such as DBU (1,8ꢀ
diazabicyclo(5.4.0)undecꢀ7ꢀene). This switch immediately led
to an increased yield βꢀlactam (from 17% to 78%), with a conꢀ
comitant reduction in imine 10 formation (Figure 2). It was
also noted that DIPA and DBU conferred complementary diaꢀ
stereoselectivities of the βꢀlactam products. Using DBU the
trans βꢀlactam (3b) was favoured in a ratio >4:1, while DIPA
give the cis βꢀlactam (3a) preferentially. This change in diaꢀ
stereoselectivity has been reported by other researchers,22 and
has been attributed to epimerization from the cis lactam to the
thermodynamically favoured trans geometry, made possible
by the increased basicity of DBU. Using DBU eliminates the
undesired amide 11, however, the generation of imine 10 was
not fully suppressed, and alkynyl imine 20 appears as a new
byproduct (Scheme 4).
Critical Importance of the Base Additive
Our investigation began by selecting a model reaction, utilizꢀ
ing tetrakis(acetonitrile)copper(I) tetrafluoroborate as the
Cu(I)
source,
tris((1ꢀcyclopentylꢀ1Hꢀ1,2,3ꢀtriazolꢀ4ꢀ
yl)methyl)amine (TCPTA) as the ligand, and diisopropylaꢀ
mine (DIPA) as the base. This choice of base was based on
previous reports, where sterically encumbered secondary
amines were found to give superior yields of the desired βꢀ
lactam.17,18
The reaction profile shows the formation of both cis and trans
βꢀlactams (3a and 3b, respectively), with the cis diastereomer
being favoured (Figure 1). Unfortunately, the desired βꢀ
lactams were not the major product under these conditions,
with ca. 60% of the initial starting materials being converted
into the corresponding imine (10) and amide (11).
Imines are often reported as a major byproduct, with their
formation has been attributed to Cuꢀcatalyzed deoxygenation
of the nitrone reagent.19,20 Interestingly, 10 and 11 were conꢀ
sistently generated in a 1:1 ratio throughout the reaction. The
observation of the uniform byproduct ratio is not consistent
with the offꢀcycle deoxygenation hypothesis, and instead
Scheme 3. Proposed retrocycloaddition of 7 and analogous
reaction seen with Nꢀsulfonylꢀ1,2,3ꢀtriazoles (14)
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
of stronger nucleophiles, highly electrophilic ketene 12 is capꢀ
1
2
3
4
5
6
tured via the oxygen center of nitrone 2. This acylꢀnitrone (17)
then fragments to give nitrilium 19 and 2ꢀphenyl acetate (18),
consistent with a proposal by Heine23. Finally, the highly elecꢀ
trophilic nitrilium is captured by copper acetylide 4 to afford
alkynylimine 20.
Kinetic Analysis to Identify Catalyst Resting State
7
8
9
To identify the catalyst resting state and test for any potential
catalyst deactivation a series of same and different excess exꢀ
periments were carried out.12,13 Using the same excess protoꢀ
col, there is a clear overlap in the kinetic profiles for the decay
of [2] over time (Figure 3a) by applying a positive time shift to
an experiment. This correlation confirms that neither product
inhibition or catalyst deactivation is occurring under these
conditions. Furthermore, this result contradicts proposals inꢀ
voking direct Cuꢀcatalyzed NꢀO deoxygenation of the nitrone
to give either imine 10 or alkynylimine 20. In these mechaꢀ
nisms, deoxygenation is coupled to the generation of Cu2O.
Thus, if this pathway was operative the catalyst activity would
be modified as a function of reaction turnover as the speciation
of the Cuꢀcatalyst changes as it is converted to Cu2O.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.1
0.08
0.06
0.04
0.02
0
2
3a
3b
10
Next, a series of different excess experiments were conducted
to probe the driving force in each reagent. While the overall
reaction profile indicates the rate is nearly constant over time,
varying the initial concentration of either alkyne 1 or nitrone 2
does have a substantial impact on the observed rate (Figure
3b). Specifically, increasing the initial concentration of 2 acꢀ
celerates the reaction, indicting a positive order in [nitrone],
while an increase in 1 leads to a decrease in rate consistent
with a negative order in [alkyne]. These observed kinetic paꢀ
rameters suggest that interaction with 2 is involved the turnoꢀ
ver limiting step, while 1 participates in an unproductive equiꢀ
librium, producing an offꢀcycle reservoir which modulates the
concentration of available catalyst.
0
50
100
150
Time (min)
200
250
300
Figure 2. Reaction profile when DBU was used as the basic
additive. Reaction conditions: [1]0 = [2]0 = [DBU]0 = 0.1 M,
[Cu(MeCN)4BF4] = [TCPTA] = 0.01 M.
Scheme 4. Proposed reaction pathway for the formation of
imine 10 and alkynyl imine 20.
Finally, a series of experiments were carried out where the
concentration of catalyst was varied from 5 – 20 mM (Figure
3c). This series confirms the system displays a positive order
in catalyst, however initial inspection of the progress curves
indicates that the response is greater than first order. Analyzꢀ
ing the reaction progress data for both varied initial substrate
and catalysts (Figures 3b and c) using the variable time norꢀ
malization analysis (VTNA) method allows the order in each
component to be elucidated (see SI).14,15 Under our reaction
conditions the system obeys the rate law given in Equation 1.
(1) ꢀꢁꢂꢃ = ꢄꢅꢆꢇ[1]ꢈꢉ[2]ꢉ[ꢊꢁꢂ]ꢋ
While, structurally unique from amide 11, we believe the
mechanism for its formation is closely related. In the absence
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. RPKA/VTNA Experiment Data (a) Same excess data for the reaction described in Figure 2 used to probe for catalyst
deactivation and/or product inhibition. (b) Data from different excess experiments used to solve for order of 1 and 2. (c) Kinetic
data used of reactions outlined in Figure 2 but with various catalyst loadings, used to solve for catalyst order. (d) VTNA plot for
data
from
Figure
3c
indicating
that
the
system
is
second
order
in
[Cu].
Reductive elimination would liberate Cuꢀisoxazolide 7 and
ultimately yield the βꢀlactams.
These kinetic parameters allow key features of the mechanism
to be elucidated. In particular, the second order dependence on
the catalyst implies that two molecules of Cu are required in a
kinetically important step of the catalytic cycle. This is conꢀ
sistent with the proposal by Himo, where cycloaddition beꢀ
tween the nitrone and σꢀCuꢀacetylide was found to preferenꢀ
tially involve bisꢀCu intermediate 5.11 Invoking bisꢀCu comꢀ
plex 5 allows the cycloaddition to pass through 6ꢀmembered
transition state 6, which was predicted to be 4.6 kcal/mol lowꢀ
er in energy than the monoꢀCu equivalent. Himo’s calculation
further predicted that the cycloaddition between bisꢀCu comꢀ
plex 6 and nitrone 2 represented the highest energy barrier on
the reaction coordinate, and thus is likely the turnover limiting
step.
Scheme 5: Simplified mechanism representing the kinetically
relevant steps in the overall pathway.
4agg
Keq3
[Cu]
1
H+
Ph
[Cu]
Keq2
4
[Cu]
5
[Cu]
Ph
Keq1
1
2
By adapting Himo’s proposal it is possible to construct a simꢀ
plified kinetic model to recapitulate the observed kinetic beꢀ
haviors in the Kinugasa reaction (Scheme 5). First, the free
Cu(L)x complex would bind to alkyne 1, whereby deprotonaꢀ
tion with DBU forms σꢀCuꢀacetylide 4. Coordination of a seꢀ
cond Cu(L)x species leads to π activation of the alkynyl sysꢀ
tem and gives bisꢀCu complex 5. Alternatively, capture of the
σꢀCuꢀacetylide 4 with a second equivalent of alkyne would
generate an offꢀcycle and unreactive polyacetylide aggregate
(4agg). Productive catalysis would involve coordination of the
nitrone to bisꢀCu complex 5 followed by turnoverꢀlimiting
cycloaddition via Himo’s proposed 6ꢀmembered transition
state produces a transient bisꢀCuꢀintermediate, containing both
a Cu(I) and Cu(III) center (6).
k3
[Cu]
Ph
3
O N
6
Ph
N
[Culll]
[Cul]
Ph
Ph
Ph
O
Ph
[Cu]
7
[Cu]
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
Measurement of Competitive Secondary KIE
Deriving a steady state rate law for the proposed mechanism is
1
2
3
4
possible by applying the quasiꢀequilibrium approximation for
the three major processes involving the catalyst (Keq1, Keq2 and
Keq3). Using this approximation, expressions for each intermeꢀ
diate to be obtained as shown in Equations 2, 3 and 4.
As kinetic studies (vide supra) suggest that numerous equilibꢀ
rium binding events precede the turnover limiting coupling
with nitrone 2, a secondary interrogation of the key transition
state was sought. By creating both H and D isotopologues of
the nitrone, a series of kinetic isotope experiments (KIE) could
be performed, monitoring both reaction rate and product selecꢀ
tivity using 1H NMR. As the KIE experiments were to be perꢀ
formed utilizing NMR timeꢀcourse analysis, the reaction beꢀ
havior must match the HPLC analysis if the results were to be
compared.24 Performing the reaction in a static NMR tube
produced a profile which is nearly identical to the HPLC trend
(Figure 4), confirming that data acquisition using the onꢀline
HPLC monitoring prototype accurately captures the reaction
profile. Furthermore, the strong correlation between these two
analytical methods validates the application of the NMR time
course for KIE investigations.
5
6
7
8
[ꢐ]
ꢏ
ꢖ
ꢘ
[ ]
ꢏ
[
]
(2) ꢌꢍꢎ
(3) ꢌꢍꢎ
(4) ꢌꢍꢎ
=
=
=
ꢐ = ꢌꢍꢎ ꢔꢕ [ꢓ]
[
[
]
ꢑꢒ [ꢓ]
[ꢗ]
[ ]
]ꢋ[ ]
ꢗ = ꢌꢍꢎ ꢌꢍꢎ ꢔꢕ ꢓ
ꢏ
ꢖ
[
]
ꢑꢒ [ꢐ]
[ꢐ
]
ꢙꢚꢚ
9
ꢋ
ꢏ
ꢘ
[
]
[ꢓ]ꢛꢅꢜꢝ = ꢌꢍꢎ ꢌꢍꢎ ꢔꢕ [ꢓ]
[
]
ꢐ [ꢓ]
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Expressing the steady state rate law requires the free catalyst
term [Cu] to be related to the total catalyst added via the mass
balance equation (Equation 5). However, evaluating this equaꢀ
tion is complicated as the complete form takes a quadratic
form. However, the equilibrium between alkyne 1 and σꢀCuꢀ
acetylide 4 is anticipated to be strongly product favored due to
the inclusion of 1 equivalent DBU, while the coordination of
the second Cu will be comparatively much weaker. Thus, the
catalyst mass balance equation can be simplified to the form
shown in Equation 6.
0.1
2 (NMR Data)
3b (NMR Data)
2 (LC Data)
0.08
0.06
0.04
0.02
0
3b (LC Data)
ꢏ
ꢖ
[
]ꢋ[ ]
[
]
ꢏ
[ ]
(5) [ꢔꢕ]ꢞꢅꢞꢟꢜ = ꢌꢍꢎ ꢌꢍꢎ ꢔꢕ ꢓ + ꢔꢕ (1 + ꢌꢍꢎ ꢓ +
ꢋ
ꢏ
ꢘ
ꢌꢍꢎ ꢌꢍꢎ [ꢓ] )
ꢋ
[
]
ꢏ
[
][ ]
ꢏ
ꢘ
[
]
(6) [ꢔꢕ]ꢞꢅꢞꢟꢜ ≅ ꢔꢕ + ꢌꢍꢎ ꢔꢕ ꢓ + ꢌꢍꢎ ꢌꢍꢎ ꢔꢕ [ꢓ]
0
50
100
150
200
250
300
Time (min)
The first order relationship in [nitrone] indicates the cycloadꢀ
dition between 5 and 2 represents the first irreversible step,
and all subsequent rearrangements are not kinetically meanꢀ
ingful. Thus, the rate of reaction can be written with respect to
this turnover limiting event (Equation 7) Solving for the unꢀ
bound catalyst term [Cu] and substituting into the rate law
gives the simplified rate law, Equation 8.
Figure 4. Comparison of reaction data gathered using HPLC
prototype and by performing NMR timeꢀcourse experiment in
a static NMR tube. Reaction conditions are described in Figure
2.
A competition reaction was performed between proteoꢀ and
deuteroꢀ nitrones, 2 and 21 respectively, with alkyne 1. The 1H
signals from the TCPTA ligand were clearly resolved and
could be used as an internal standard, allowing the rate of
change for each isotopologue of nitrone, βꢀlactam (both cis
and trans diastereomers) and imine to be delineated. The ratio
of nitrone isotopologues was plotted as a function of time,
indicating that the rate of consumption of Dꢀnitrone 21 is
greater than the corresponding rate for Hꢀnitrone 2 (Figure 5a).
Thus, the catalytic reaction exhibits an inverse secondary KIE
under these reaction conditions.
ꢋ
[ ][ ]
ꢏ
ꢖ
(7) ꢀꢁꢂꢃ = ꢄꢠ ꢡ ꢗ ≅ ꢄꢠꢌꢍꢎ ꢌꢍꢎ [ꢓ][ꢡ][ꢔꢕ]
ꢖ
ꢢ
ꢣ
ꢣ
[ꢓ][ꢡ][ꢑꢒ]
ꢖ
ꢏ
ꢘ
ꢦꢧꢦꢙꢨ
ꢤꢥ
ꢤꢥ
(8) ꢀꢁꢂꢃ ≅ (ꢉꢩꢣ
ꢖ ꢖ
[ꢓ] )
ꢘ
ꢤꢥ
[
]
ꢓ ꢩꢣ
ꢣ
ꢏ
ꢏ
ꢤꢥ
ꢤꢥ
The negative order relationship with respect to [alkyne] may
reflect that, under strongly basic conditions of this reaction,
the equilibrium between 1 and 4 is strongly product favored,
while generation of the offꢀcycle polyacetylide complex inꢀ
To quantify the magnitude of the inverse KIE, another compeꢀ
tition reaction was carried out, varying only the initial concenꢀ
trations of Hꢀand Dꢀnitrone isotopologues. Assuming the reacꢀ
tion of both 2 and 21 obey the simplified rate law given in
Equation 9, rates of reaction for each isotopologues will be
modified both by [nitrone] but also the independent rate conꢀ
stants for the turnover limiting step associated with each
isotopologue (k3 in Scheme 7). By manipulating Equation 9,
the KIE for any experiment with varied initial concentrations
of 2 and 21 can be expressed as Equation 10.
1
volves a comparatively weaker interaction (Keq >> 1 while
Keq3 << 1 respectively). By applying these assumption, the rate
equation can be simplified to Equation 9, which reconciles the
experimentally observed kinetic dependencies. It should be
noted that the observation of an overall apparent zeroꢀorder
reaction profile and second order in [Cu] is specific to the
particular reagent concentrations from our study. Significant
deviation from our simplified power law is expected when
large excesses in either [alkyne] or [nitrone] are employed.
ꢖ
[
]
ꢣ
ꢡ [ꢑꢒ]
(9) ꢀꢁꢂꢃ ≅ ꢢ
ꢖ
ꢘ
ꢦꢧꢦꢙꢨ
ꢤꢥ
ꢢ
ꢢ
ꢮꢟꢞꢍ [ꢯꢰꢞ
]
]
ꢬ
ꢬ
ꢭ
(10)
ꢌꢪꢫ =
=
∘ = 0.95
ꢣ
[ꢓ]
ꢏ
ꢮꢟꢞꢍ [ꢯꢰꢞ
ꢤꢥ
ꢭ
ꢭ
ꢬ
∘
The rate of consumption for each isotopologue follows a zeroꢀ
order profile, and therefore is a constant given by the slope of
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
reaction profile (See SI for derivation). The reaction was turnoverꢀlimiting step, which is consistent with both the posiꢀ
1
2
3
4
5
6
7
8
found to display a secondary inverse KIE of 0.95, which sugꢀ
gests that the turnover limiting step involves a change in hyꢀ
bridization from sp2 to sp3 at the carbon bearing the isotopic
label.25 Applying the same analysis to our preliminary experiꢀ
ment with equimolar concentrations of both isotopologues
(from Figure 5A) we observe an identical KIE of 0.95. These
observations are consistent with the previous RPKA kinetic
data, confirming that the initial (3+2) cycloaddition representꢀ
ing the turnover limiting step in the catalytic cycle.
tive order in [nitrone] and inverse secondary KIE observed
with isotopologue 21.
From intermediate 6, rapid reductive elimination followed by
protonation and (3+2) cycloreversion gives ketene and imine
(12 and 10 respectively). The fate of these key intermediates
governs the observed chemoselectivity of the reaction. If sufꢀ
ficiently nucleophilic reagents are present (such as DIPA or
nitrone) capture of the highly electrophilic ketene dominates,
diverting the reaction towards byproducts such as amide 11,
alkynylimine 20 and imine 10. Alternatively, recombination of
10 and 12 via a Lewis acid catalyzed (2+2) cycloaddition
yields the desired βꢀlactams. Under this revised mechanism,
Cu(I)ꢀcatalyzed βꢀlactam generation between terminal alkynes
and nitrones is better classified as a cascade reaction, terminatꢀ
ing with a formal (2+2) cycloaddition, more classically known
as the Staudinger synthesis of βꢀlactams.26 This revised pathꢀ
way is analogous to the formation of Nꢀsulfonylazetidinꢀ2ꢀ
imines via capture of ketenimine 14 with imines.21
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1.04
a)
1.03
1.02
1.01
1
Scheme 6. Proposed reaction mechanism based on kinetic data
highlighting all key equilibria and binding events
Negative order
4agg
Keq3
1
in
Ph
Ph
O
H+
N
1
Ph
[Cu]
[Cu]
1
[Cu]
+
0
20
40
60
80
100
120
Ph
3
Keq1
Byproducts
11, 18, 20
b)
0.1
0.08
0.06
0.04
0.02
0
y = (-6.53 x 10-4)x +0.105
R2 = 0.9998
Keq2
Ph
Second Order
in Catalyst
2
Determines
stereoselectivity
NuH
[Cu]
21
[Cu]
[Cu]
5
O
•
Ph
2
+
N
Turnover
limitingstep
k3
H
Ph Ph
y = (-3.87 x 10-4)x + 0.0589
R2 = 0.9994
12'
10
Ph
0
50
100
Time (min)
150
200
Ph
O N
6
[CuIlI]
[CuI]
Ph
N
Ph
O
Figure 5. a) Variation in the ratio of nitrone isotopologues
over the course of the KIE competition experiment. Reaction
conditions: [1]0 = [2]0 = [21]0 = [DBU]0 = 0.1 M.
[Cu(MeCN)4BF4] = [TCPTA] = 0.01 M. b) Second KIE comꢀ
petition experiment used to solve for KIE from Equation 10.
Reaction conditions: [1]0 = [2]0 = [DBU]0 = 0.1 M, [21]0 =
0.06 M, [Cu(MeCN)4BF4] = [TCPTA] = 0.01 M.
Ph
[Cu]
Ph
7
[Cu]
Our revised mechanism is not only consistent with the availaꢀ
ble kinetic data but also resolves two important inconsistencies
with previous models. First, the intermediacy of the ketene 12
provides the first consistent model rationalizing both the forꢀ
mation of target βꢀlactams and common byproducts, including
imine 10, amide 11, alkynylimine 20, and benzyl carboxylate
18. While similar byproducts have been documented, underlyꢀ
ing mechanistic rationale for their appearance has been variaꢀ
ble.19 For example, Miura found that alkynylimines could be
Expanded Mechanistic Model
By combining our reaction progress and kinetic data, it is now
possible to propose a complete mechanistic model that exꢀ
plains both the productive and unproductive reaction pathways
(Scheme 6). First, the cycloaddition begins with the generation
of the key σꢀCu(I) acetylide 4 from the terminal alkyne 1. At
high concentration of alkyne an offꢀcycle reservoir is created,
restricting the Cu(I) available for productive cycloaddition and
manifests as the negative order in [alkyne]. Further activation
of the σꢀCu acetylide 4 by way of a second Cu(I) is required to
effect cycloaddition and gives rise to the second order behavꢀ
ior in [Cu]. Coordination of the nitrone dipole and cycloaddiꢀ
tion to produce metallocyclic intermediate 6 constitutes the
chemoselectively
formed
by
using
dppe
(1,2ꢀ
bis(diphenylphosphino)ethane). However, their proposed
pathway required direct addition of the σꢀCu(I) acetylide to the
nitrone sp2 carbon, followed by deoxygenation to give Cu2O.
Our RPKA data is not consistent with Miura’s proposal, as
there is no evidence of catalyst deactivation or change in cataꢀ
lyst activity (Figure 3a) which would be apparent if CuBF4
was being transformed to Cu2O during the reaction.
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
To summarize, we have reported the use of RPKA to delineate
A second important consequence of our modified mechanism
1
2
3
4
5
6
7
8
relates to the stereocontrolling step. The geometry of the
asymmetric center in the βꢀlactam is set by the (2+2) cycloadꢀ
dition between ketene 12 and imine 10, not the initial (3+2)
cycloaddition involving the nitrone and σꢀCu(I) acetylide as is
conventionally proposed. Thus, ligands, such as PyBOX, genꢀ
erate a Cu(I) complex that acts as a chiral Lewis acid to give
stereodiscrimination akin to an asymmetric Staudinger syntheꢀ
sis. This modification resolves an issue noted by Himo, who
found that incorporating chiral ligands onto the key bisꢀCuꢀ
intermediate 5 creates numerous steric penalties, raising the
energy of the chiral transition state above that of the uncataꢀ
lyzed thermal (3+2) Huisgen dipolar cycloaddition.
the underlying mechanism of the Kinugasa reaction. These
experiments reveal that the system displays an overall zeroꢀ
order profile, brought about by a near integer order of +1 and ꢀ
1 for [nitrone] and [alkyne] respectively. Furthermore, the
kinetic data reveal that an initial (3+2) cycloaddition between
doubly activated bisꢀCu(I) acetylide and nitrone constitutes the
turnover limiting step. This conclusion is supported by the
second order behavior with respect to [Cu] as well as a secꢀ
ondary inverse kinetic isotope effect of 0.95 for deuterated
nitrone isotopologues. The observed kinetic and chemoselecꢀ
tivity data allow us to propose a modified mechanism, involvꢀ
ing a novel (3+2) cycloreversion followed by Lewis acid cataꢀ
lyzed (2+2) cycloaddition. Most importantly, our new catalytic
pathway, which proceeds via a common ketene intermediate,
provides the first consistent mechanistic model for the generaꢀ
tion of all commonly observed byproducts of the Kinugasa
reaction. By understanding the pivotal role of the cycloaddiꢀ
tion reaction cascade and the intermediacy of the reactive keꢀ
tene we have been able to derive reaction conditions using
nonꢀnucleophilic bases (such as DBU), which improves the
yield of the desired βꢀlactam products to ~80%.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
One final consideration relates to the work by Shintani and Fu,
who designed conditions to trap the proposed Cuꢀenolate,
which would be formed after intramolecular cyclization of
intermediate 8.27 This result is still consistent in the context of
our modified mechanism, as both Fu’s conditions and ours
operate under stoichiometric strong base additives. Thus, it is
possible that the allylation occurs after the formation of the βꢀ
lactam via base promoted enolate formation.
Confirmation of the Intermediacy of Ketene
To assess the participation of a Staudingerꢀlike (2+2) coupling
in the formation of the βꢀlactam products, a crossꢀover experꢀ
iment was conducted. The reaction of 1 and 2 was performed
in the presence of imine 23, giving both the expected lactam
3b as well as the newly observed crossꢀover product lactam 24
(Scheme 7). Control experiments have been conducted conꢀ
firming that oxygen exchange between imine 23 and nitrone 2
does not appear to be occurring on the timescale of the cyꢀ
cloaddition (see supporting information for details). This result
suggests that an exogenously dosed imine can participate in a
formal (2+2) cycloaddition and provides additional evidence
that ketene 12 is an intermediate in the productive lactamꢀ
forming pathway. The relatively low concentration of the
crossꢀover product 24 further suggests that if a formal (2+2)
reaction is ultimately responsible for the generation of the βꢀ
lactam, that it proceeds via an inner sphere mechanism involvꢀ
ing rapid coupling between 12’ and 10, which does not allow
facile exchange of the exogenous imine coupling partner. Simꢀ
ilar lack of positive crossꢀover results have been reported in
TsujiꢀTrost like processes where molecular fragments remain
tightly bound prior to recapture.28 This proposal is further supꢀ
ported by our kinetic evidence, which suggests that the cyꢀ
cloreversion and subsequent recapture are rapid relative to the
initial (3+2) coupling between nitrone and Cuꢀacetylide.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website.
Experimental procedures and characterization data
AUTHOR INFORMATION
Corresponding Author
* jhein@chem.ubc.ca
ORCID
Jason E. Hein: 0000ꢀ0002ꢀ4345ꢀ3005
ACKNOWLEDGMENT
The authors gratefully acknowledge Prof. Paul Cheong for helpful
discussion and insight into experimental design. In addition, we
acknowledge MettlerꢀToledo Autochem for generous donation of
process analytical equipment (EasyMax). Financial support for
this work was provided by the University of British Columbia and
the Natural Sciences and Engineering Resource Council of Canaꢀ
da for both research (2016ꢀRGPINꢀ04613) and student support
(NSERCꢀCGS to T. C. M.).
Scheme 7. Crossover experiment designed to probe for interꢀ
mediacy of ketene 12. Ar = pꢀtolyl. [1]0 = [2]0 = [DBU]0 = 0.1
M, [23]0 = 0.025 M, [CuI] = [TCPTA] = 0.01 M.
REFERENCES
(1)
(2)
(3)
Galletti, P.; Giacomini, D. Monocyclic βꢀLactams: New
Structures for New Biological Activities. Curr. Med.
Chem. 2011, 18, 4265–4283.
Singh, G. S. βꢀLactams in the New Millennium. PartꢀI:
Monobactams and Carbapenems. MiniꢀReviews Med.
Chem. 2004, 4, 69–92.
Keri, R. S.; Hosamani, K. M.; Reddy, H. S.; Shingalapur,
R. V. Synthesis, inꢀVitro Antimicrobial and Cytotoxic
Studies of Novel Azetidinone Derivatives. Arch. Pharm.
2010, 343, 237–247.
CONCLUSIONS
(4)
Indrani Banik; Frederick F. Becker, A.; Banik, B. K.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
Stereoselective
Synthesis
of
βꢀLactams
with
Hultin, P. G.; Hein, J. E. RealꢀTime HPLCꢀMS Reaction
Polyaromatic Imines: Entry to New and Novel
Anticancer Agents. J. Med. Chem 2002, 46, 12–15.
Progress Monitoring Using an Automated Analytical
Platform. React. Chem. Eng. 2017, 2, 309–314.
1
2
3
4
5
6
7
8
9
(5)
(6)
(7)
Bonneau, P. R.; Hasani, F.; Plouffe, C.; Malenfant, E.;
LaPlante, S. R.; Guse, I.; Ogilvie, W. W.; Plante, R.;
Davidson, W. C.; Hopkins, J. L.; Morelock, M. M.;
Cordingley, M. G; Deziel, R.. Inhibition of Human
Cytomegalovirus Protease by Monocyclic βꢀLactam
(17)
(18)
(19)
Ye, M. C.; Zhou, J.; Tang, Y. Trisoxazoline/Cu(II)ꢀ
Promoted Kinugasa Reaction. Enantioselective Synthesis
of βꢀLactams. J. Org. Chem. 2006, 71, 3576–3582.
Ye, M.ꢀC.; Zhou, J.; Huang, Z.ꢀZ.; Tang, Y. Chiral
tris(oxazoline)/Cu(II) Catalyzed Coupling of Terminal
Alkynes and Nitrones. Chem. Commun. 2003, 0, 2554.
Derivatives: Kinetic Characterization Using
a
Fluorescent Probe. J. Am. Chem. Soc. 1999, 121, 2965–
2973.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Miura, M.; Enna, M.; Okuro, K.; Nomura, M. Copperꢀ
Catalyzed Reaction of Terminal Alkynes with Nitrones.
Selective Synthesis of 1ꢀAzaꢀ1ꢀButenꢀ3ꢀYne and 2ꢀ
Azetidinone Derivatives. J. Org. Chem. 1995, 60, 4999–
5004.
GeronaꢀNavarro, G.; Pérez De Vega, M. J.; Garcíaꢀ
López, M. T.; Andrei, G.; Snoeck, R.; De Clercq, E.;
Balzarini, J.; GonzálezꢀMuñiz, R. From 1ꢀAcylꢀβꢀ
Lactam Human Cytomegalovirus Protease Inhibitors to
1ꢀBenzyloxycarbonylazetidines with Improved Antiviral
(20)
Mames, A.; Stecko, S.; Mikołajczyk, P.; Soluch, M.;
Furman, B.; Chmielewski, M. Direct, Catalytic Synthesis
of Carbapenams via Cycloaddition/Rearrangement
Cascade Reaction: Unexpected Acetylenes’ Structure
Effect. J. Org. Chem. 2010, 75, 7580–7587.
Activity.
A Straightforward Approach to Convert
Covalent to Noncovalent Inhibitors. J. Med. Chem. 2005,
48, 2612–2621.
Ojima, I.; Delaloge, F. Asymmetric Synthesis of
BuildingꢀBlocks for Peptides and Peptidomimetics by
Means of the βꢀLactam Synthon Method. Chem. Soc.
Rev. 1997, 26, 377–386.
(21)
(22)
Whiting, M.; Fokin, V. V. CopperꢀCatalyzed Reaction
Cascade: Direct Conversion of Alkynes intoNꢀ
Sulfonylazetidinꢀ2ꢀImines. Angew. Chem. Int. Ed. 2006,
45, 3157–3161.
(8)
(9)
Kinugasa, M.; Hashimoto, S. The Reactions of copper(I)
Phenylacetylide with Nitrones. J. Chem. Soc. Chem.
Commun. 1972, 0, 466.
Zhang, X.; Hsung, R. P.; Li, H.; Zhang, Y.; Johnson, W.
L.; Figueroa, R. A Highly Stereoselective Synthesis of
Chiral αꢀAminoꢀβꢀLactams via the Kinugasa Reaction
Employing Ynamides. Org. Lett. 2008, 10, 3477–3479.
Lo, M. M.ꢀC.; Fu, G. C. Cu(I)/Bis(azaferrocene)ꢀ
Catalyzed Enantioselective Synthesis of βꢀLactams via
Couplings of Alkynes with Nitrones. J. Am. Chem. Soc.
2002, 124, 4572–4573.
(23)
(24)
Heine, H. W.; Zibuck, R.; VandenHeuvel, W. J. A.
Mechanisms for the Reactions of Nitrones with Aroyl
Chlorides. J. Am. Chem. Soc. 1982, 104, 3691–3694.
(10)
(11)
Stecko, S.; Furman, B.; Chmielewski, M. Kinugasa
Reaction: An “ugly Duckling” of βꢀLactam Chemistry.
Tetrahedron 2014, 70, 7817–7844.
Foley, D. A.; Dunn, A. L.; Zell, M. T. Reaction
Monitoring Using Online vs Tube NMR Spectroscopy:
Seriously Different Results. Magn. Reson. Chem. 2016,
54, 451–456.
Santoro, S.; Liao, R.ꢀZ.; Marcelli, T.; Hammar, P.; Himo,
F. Theoretical Study of Mechanism and Stereoselectivity
of Catalytic Kinugasa Reaction. J. Org. Chem. 2015, 80,
2649–2660.
(25)
(26)
Anslyn, E. V.; Dougherty, D. A. Modern Physical
Organic Chemistry; University Science, 2006.
(12)
Blackmond, D. G. Reaction Progress Kinetic Analysis: A
Powerful Methodology for Mechanistic Studies of
Complex Catalytic Reactions. Angew. Chem. Int. Ed.
2005, 44, 4302–4320.
Staudinger,
H.
Zur
Kenntniss
Der
Ketene.
Diphenylketen. Justus Liebig’s Ann. der Chemie 1907,
356, 51–123.
(27)
(28)
Shintani, R.; Fu, G. C. Catalytic Enantioselective
Synthesis OfβꢀLactams: Intramolecular Kinugasa
Reactions and Interception of an Intermediate in the
Reaction Cascade. Angew. Chem. Int. Ed. 2003, 42,
4082–4085.
(13)
(14)
Blackmond, D. G. Kinetic Profiling of Catalytic Organic
Reactions as a Mechanistic Tool. J. Am. Chem. Soc.
2015, 137, 10852–10866.
Burés, J. Variable Time Normalization Analysis: General
Graphical Elucidation of Reaction Orders from
Concentration Profiles. Angew. Chem. Int. Ed. 2016, 55,
16084–16087.
AlꢀAmin, M.; Roth, K. E.; Blum, S. A. Mechanistic
Studies of Gold and Palladium Cooperative Dualꢀ
Catalytic CrossꢀCoupling Systems. ACS Catal. 2014, 4,
622–629.
(15)
Burés, J. A Simple Graphical Method to Determine the
Order in Catalyst. Angew. Chem. Int. Ed. 2016, 55,
2028–2031.
(16)
.
Malig, T. C.; Koenig, J. D. B.; Situ, H.; Chehal, N. K.;
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Ph
O
Ph
O
N
N
Cu(I)/Base
Ligand
Ph
Ph
H
Ph
Ph
(3+2)
(2+2)
Ph
N
O
•
Ph
retro (3+2)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ph
Ph
O
+
N
Ph
H
Ph
[Cu]
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
ACS Paragon Plus Environment