Keto-Enol Thermodynamics of Breslow Intermediates

Article abstract of DOI:10.1021/jacs.5b13236

Breslow intermediates, first postulated in 1958, are pivotal intermediates in carbene-catalyzed umpolung. Attempts to isolate and characterize these fleeting amino enol species first met with success in 2012 when we found that saturated bis-Dipp/Mes imida

Full text of DOI:10.1021/jacs.5b13236

Subscriber access provided by Weizmann Institute of Science
Article
Keto-Enol Thermodynamics of Breslow Intermediates
Mathias Paul, Martin Breugst, Jörg-Martin Neudörfl, Raghavan B. Sunoj, and Albrecht Berkessel
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b13236 • Publication Date (Web): 14 Feb 2016
Downloaded from http://pubs.acs.org on February 15, 2016
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 free 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 accessible to all
readers and 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.
Journal of the American Chemical Society 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 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Keto-Enol Thermodynamics of Breslow Intermediates
Mathias Paul,† Martin Breugst,† Jörg-Martin Neudörfl,† Raghavan B. Sunoj,‡ and Albrecht
Berkessel*,†
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 (Organic Chemistry), University of Cologne, Greinstrasse 4, 50939 Cologne, Germany;
‡Department of Chemistry, Indian Institute of Technology Bombay, Powai, Mumbai 400076, India
KEYWORDS: Carbenes; Umpolung; keto-enol tautomerization; DFT calculations; NMR spectroscopy
ABSTRACT: Breslow intermediates, first postulated in 1958, are pivotal intermediates in carbene-catalyzed Umpolung.
Attempts to isolate and characterize these fleeting amino enol species have first met with success in 2012 when we found
that saturated bis-Dipp/Mes imidazolidinylidenes readily form isolable, though reactive diamino enols with aldehydes
and enals. In contrast, triazolylidenes, upon stoichiometric reaction with aldehydes, gave exclusively the keto tautomer,
and no isolable enol. Herein, we present the synthesis of the "missing" keto tautomers of imidazolidinylidene-derived
diamino enols, and computational thermodynamic data for 15 enol-ketone pairs derived from various carbenes/aldehydes.
With one exception, the experimentally observed enol-ketone preferences correspond to the calculated thermodynamics
of the enol-ketone pairs. Electron-withdrawing substituents on the aldehyde favor enol formation, the same holds for
N,N'-Dipp [2,6-di(2-propyl)phenyl] and N,N'-Mes [2,4,6-trimethylphenyl] substitution on the carbene component. The
latter effect rests on stabilization of the diamino enol tautomer by Dipp-substitution, and could be attributed to disper-
sive interaction of the 2-propyl groups with the enol moiety. For three enol-ketone pairs, equilibration of the thermody-
namically disfavored tautomer was attempted with acids and bases, but could not be effected, indicating kinetic inhibi-
tion of proton transfer.
various aldehydes [e.g. benzaldehyde (A-1), Scheme 2], we
I. INTRODUCTION
observed rapid and exclusive formation of the ketone K-
11, and no corresponding enol (E or Z) E-11 (Scheme 2).5,6
Both in vitamin-B1-dependent enzymes and in organocat-
alytic Umpolung, catalysis by N-heterocyclic carbenes
(NHCs) hinges on the formation of the so-called Breslow-
intermediates1,2 [chemically: (di)amino enols] E (Scheme
1) in which the genuine polarity of e.g. an aldehyde sub-
strate is inverted from electrophilic to nucleophilic. At-
tack of the nucleophilic aminoenol E on various electro-
philes gives rise to benzoin condensation, Stetter reaction
and other well known NHC-catalyzed Umpolung reac-
tions. The Breslow intermediate was first postulated in
1958 for thiamine-catalyzed transformations (X = S,
Scheme 1).1,2 The first successful generation of diamino
enols E (X = NR) from aldehydes and imidazolidinyli-
denes, and their characterization by in situ NMR, was
reported by us in 2012, followed by isolation and X-ray
characterization in 2013.3,4
The ketone K-11 was shown to be catalytically incompe-
tent, indicating that its formation from benzaldehyde (A-
1) and carbene C-1 is irreversible.5 It is worthy of note that
in 2012, the first keto form of a thiamine-derived Breslow
intermediate has been identified in pyruvate oxidase.7 In
this enzyme, acetyl thiamine serves as an intermediate en
route to acetyl phosphate. As pyruvate decarboxylation
first affords the enol, the interconversion with its keto
form must be feasible in the enzyme.
Scheme 2. Comparison of the Reaction of Benzalde-
hyde (A-1) with the Triazolylidene C-1, and with the
Imidazolidinylidene C-2.
Scheme 1. The Breslow Intermediate as Enol (E) and
Keto (K) Tautomer.
The keto tautomer K (Scheme 1) of the Breslow-inter-
mediate has received considerably less attention than the
amino enol E. In our study on the stoichiometric interac-
tion of the 1,2,4-triphenyltriazolylidene carbene C-1 with
As briefly summarized in Scheme 2, the reactions of im-
idazolidinylidenes (such as C-2, SIPr) with aldehydes have
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 9
thus far led exclusively to diamino enols E, and no ke-
"Wanzlick dimer" with benzaldehyde at elevated temper-
ature.10
Figure 1 summarizes the ketones thus prepared, to-
tones K were observed.3,4,6 In contrast, the ketones K were
the only observable products in the reaction of triazolyli-
denes (such as C-1) with aldehydes, and no enols could be
isolated or traced spectroscopically.5 We decided to em-
bark on a joint experimental/computational approach to
elucidate the reasons (thermodynamics/kinetics) underly-
ing this divergent behavior. In here, we disclose a com-
bined experimental and computational study which (i) by
synthesis proves the existence and stability of the "miss-
ing ketones" of the saturated carbene/aldehyde combina-
tion, and (ii) analyzes [DFT calculations using M06-2X-
D3/def2-TZVPP/IEFPCM(THF)//M06-L-D3/631+G(d,p)]
the thermodynamics of the enol/keto system for both
saturated and unsaturated NHCs in combination with
various aldehydes.
1
2
3
4
5
6
7
8
gether with characteristic NMR data. As mentioned al-
ready, ketone K-11 results from the stoichiometric interac-
tion of the triazolylidene carbene C-1 with benzaldehyde
(A-1), and its published NMR data5 are included in Figure
1 for completeness. In the case of ketone K-31, we were
able to obtain crystals suitable for X-ray crystallography,
whereas all other ketones prepared in this study were
viscous liquids. The molecular structure of ketone K-31 is
shown in Figure 2.
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
II. RESULTS
2
1
II.1 Synthesis and Characterization of the "Missing
Ketones" Derived from Imidazolidinylidenes
3
5
Our synthetic approach to the ketones K-21 – K-26, K-31,
and K-41 is shown in Scheme 3. In the first step, aryl eth-
anones were converted to glyoxyl hydrates by SeO2 oxida-
tion, as described by Sutherland et al..8 For the para-
bromophenyl glyoxal hydrate, we were able to obtain the
first X-ray crystal structure of its monomeric gem-diol
form (see SI). Subsequent condensation with N,N'-
disubstituted ethylenediamines in the presence of molec-
ular sieves afforded the desired ketones (see SI for exper-
imental details), a method originally described by Schön-
berg et al..9 Note that ketone K-31 had already been ob-
tained in 1961 by Wanzlick and Schikora, by reacting the
4
Figure 2: X-Ray crystal structure of the ketone K-31.
As a most important structural feature, the crystal
structure of ketone K-31 revealed a syn-periplanar ar-
rangement of the aminal C2-H bond and the carbonyl C-
O bond [H-C-C-O = 2.9o]. Additionally, the phenyl rings
at N1 and N3 are more or less coplanar with the imidazol-
idine ring. Of the latter, the ring puckering places C4
somewhat below the plane defined by the other four im-
idazolidine ring atoms. NMR spectra of ketone K-31 were
already reported by Schönberg et al., and interpreted in
terms of the 5-ring dynamics in K-31.9 Our own NMR
investigation of K-31 and of the other ketones in this
study revealed that in addition to the syn-periplanar ar-
rangement of the aminal C-H and the carbonyl C-O bond
(as seen in the crystal), in solution also the anti-peri-
planar conformer is present. Both conformers readily
equilibrate at room temperature (see NOESY data in the
Supporting Information, exemplified for K-21). Addition-
ally, rapid rotation of the Dipp substituents is evident
through the exchange signals of the 2-propyl groups in
the NOESY spectra (see Supporting Information).11 In
summary, we were able to prepare and characterize, in
addition to the "Wanzlick ketone" K-31, the "missing
keto-tautomers" of all Breslow diaminoenols previously
reported by us.3,4
Scheme 3. Preparation of Aryl Glyoxyl Hydrates, and
Condensation with Ethylenediamines.
II.2 Computational Studies: Influence of the Carb-
ene Component on the Keto-Enol Thermodynamics
Having established that both the keto- and the enol-
tautomers exist for quite a number of C-2 (SIPr) or C-4
SIMes) combinations with aldehydes, this part of our
study aimed at calculating the thermodynamic relation of
the corresponding ketone-enol pairs. We furthermore
aimed at elucidating how the structural and electronic
Figure 1. Ketones prepared in this study, together with char-
acteristic 1H/13C shifts [ppm (298 K, C6D6)].
ACS Paragon Plus Environment

Page 3 of 9
Journal of the American Chemical Society
are depicted in the Supporting Information, selected op-
Table 1. Reaction Free Energies for the Formation of
1
2
3
4
5
6
7
8
the Breslow Intermediates [ΔGEnol] and the Corre-
sponding Ketones [ΔGKetone] from the Free Carbenes
timized structures are shown in Figure 3. In the lowest-
energy conformers of all Breslow intermediates obtained
from C-1–C-7 and benzaldehyde (A-1), the O–H proton is
not hydrogen bonded to an azole nitrogen atom, but is
pointing away from the heterocyclic ring (cf. Figure 2,
Supporting Information, and X-ray crystal structures in
refs. 3,4). In the corresponding ketones, both the syn- and
the anti-orientation of the H–C–C–O dihedral could be
located with the syn being preferred in most cases
(ꢀꢀGsyn/anti K-11: +2.8, K-21: +1.0, K-31: +2.5, K-41: +0.1, K-
51: –0.4, K-61: +3.8, K-71: –0.7 kcal mol–1, respectively).
and Benzaldehyde, as well as the Keto-Enol-Differen-
a
ce [ΔΔGKE
]
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
Entry
1
Carbene
ΔGEnol
ΔGKetone
ΔΔGKE
+3.6 (E-E-11)
+3.5 (Z-E-11)
–3.5 (K-11)
–7.0
2
3
4
5
–9.6 (E-21)
–4.9 (E-31)
–10.4 (E-41)
+0.1 (E-51)
+2.4 (E-61)
–7.2 (K-21)
–13.5 (K-31)
–5.4 (K-41)
+3.6 (K-51)
+2.5
–8.7
+5.0
+3.4
Z-E-11
K-11
E-21
K-21
6
7
–3.1 (K-61)
–5.6
–5.2
–7.5 (E-E-71)
–9.1 (Z-E-71)
–14.2 (K-71)
aall in kcal mol–1, M06-2XD3/def2-TZVPP/IEFPCM(THF)
//M06-L-D3/6-31+G(d,p).
Z-E-71
K-71
Figure 3: Lowest energy structures for the enol-ketone pairs
E-11/K-11, E-21/K-21, and E-71/K-71.
features of the carbene and the aldehyde influence the
keto-enol thermodynamics.
In line with previous investigations,12 our computation-
al study again shows that sterically demanding N-sub-
stituents - e.g., Dipp in the bis-Dipp carbene C-2 (SIPr), or
Mes in C-4 (SIMes) - induce larger interplanar angles, i.e.
an almost perpendicular orientation compared to smaller
substituents (Figure 3 and Supporting Information).
Within the different classes of N-heterocyclic carbenes,
we first computationally analyzed the combination of
different triazole- (C-1), imidazolidine- (C-2, C-3, C-4),
imidazole- (C-5, C-6), and thiazole-derived (C-7) carbenes
with benzaldehyde (A-1). Table 1 collects the reaction free
energies for the formation of the Breslow intermediates
(aminoenols) and the tautomeric ketones as well as the
keto-enol energy difference ꢀꢀGKE. While all structures
Within the phenyl-substituted series (i.e., carbenes C-1,
C-3, C-6, and C-7), the reactions of the thiazole derived
carbene C-7 are most exergonic (ΔGEnol = –9.1 kcal mol–1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 9
and ΔGKetone –14.2 kcal mol–1; Table 1, entry 7) followed by
Scheme 4: Influence of the 2-Propyl Groups on the
Keto-Enol Energy Difference (all Values in kcal
mol-1).
1
2
3
4
5
6
7
8
the saturated "Wanzlick carbene" C-3 (ΔGEnol = –4.9 kcal
mol–1 and ΔGKetone = -13.5 kcal mol–1; Table 1, entry 3). For
both unsaturated carbenes C-1 and C-6, the formation of
the Breslow intermediate is endergonic (ΔGEnol = +3.5 and
+2.4 kcal mol–1) while the tautomeric ketone was calculat-
ed to be thermodynamically more stable than the reac-
tants (ΔGKetone = –3.5 and –3.1 kcal mol–1). Enol and ketone
formation are thermoneutral/endergonic for the combi-
nation of the unsaturated Dipp-substituted carbene C-5
(IPr) and benzaldehyde (A-1; Table 1, entry 5). Almost
identical reaction free energies were calculated for the
formation of the E- and Z-enol E-11 derived from the tria-
zole carbene C-1 and benzaldehyde (A-1; Table 1, entry 1),
indicating that the additional phenyl group at C3 has no
significant effect on the stabilities of the two diastere-
omers. In contrast, a significant preference for the Z- over
the E-configuration was calculated for the Breslow inter-
mediate E-71 formed from ne carbene C-7 and benzalde-
hyde (A-1; Table 1, entry 7). The unfavorable steric inter-
action of the two phenyl groups in the Z-orientation is
obviously compensated by a stabilizing π-π-interaction
between the aromatic groups (Figure 3). Within the series
of phenyl-substituted carbenes, the ketones are thermo-
dynamically preferred over the corresponding enols (Ta-
ble 1, entries 1,3,6,7). However, the nature of the carbene
C-1,3,6,7 has only a small effect, as evidenced by the rela-
tively narrow range of 3.5 kcal mol–1 in which all keto-
enol-differences are found (–8.7 < ΔΔGKE < –5.2 kcal
mol–1).
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
the hydrogen atoms added were allowed to relax during
the optimizations.As expected, the relative keto-enol
energy differences in the presence and absence of 2-
propyl groups are not identical. Closer inspection reveals
that the reaction energy for ketone formation is virtually
unaffected by the truncation (-23.9 kcal mol-1 for K-21, vs.
-23.8 kcal mol-1 for truncated K-21). However, for enol
formation, the reaction energy in the presence of the 2-
propyl groups is by ca. 2 kcal mol-1 more negative (-25.1
kcal mol-1 for E-21 vs. -23.0 kcal mol-1 for truncated E-21).
The calculated change in the keto-enol energy therefore
results exclusively from the energy change of the enol. As
the reaction energy for E-21 is 2 kcal mol–1 more exother-
mic, we have to conclude that there is an attractive, dis-
persive stabilization caused by the 2-propyl groups.
The most striking feature revealed by our study is the
influence of the substituents at the carbene-nitrogen-
atoms on the keto-enol energy difference. When the phe-
nyl substituents in C-3 (the "Wanzlick carbene") were
exchanged for 2,6-di(2-propyl)phenyl (C-2, SIPr) or me-
sityl (C-4, SIMes), the formation of both Breslow inter-
mediates and their corresponding ketones remained exer-
gonic (Table 1, entries 2,3,4). Closer inspection of the
reactions involving the carbenes C-2 and C-3 reveals,
however, that the reaction free energy for the formation
of the enols (E-21 vs. E-31) is 4.7 kcal mol–1 more favorable
for E-21 (Table 1, entries 2,3). In contrast, K-31 is formed
in a more exergonic reaction (ꢀꢀG = 6.3 kcal mol–1) than
K-21. The same trend, i.e. stabilization of the enol and
destabilization of the ketone, was found for the (saturat-
ed) imidazolidinylidene pair C-3/C-4 (Table 1, entries 3,4),
and for the (unsaturated) imidazolylidene pair C-5/C-6
(Table 1, entries 5,6).
II.3 Computational Studies: Influence of the Alde-
hyde on the Keto-Enol Thermodynamics
We further studied how a variation of the aldehyde in-
fluences the keto-enol thermodynamics. For that purpose,
we chose the carbene system C-2 as our model system. As
this system had the keto-enol difference which was clos-
est to zero (Table 1, entry 2), this system lends itself best
as the starting point. The calculated reaction free energies
for the formation of the Breslow intermediates and the
corresponding tautomeric ketones from different alde-
hydes are summarized in Table 2, and selected structures
are depicted in Figure 4.
All reactions were found to be exergonic with the ex-
ception of the formation of the keto-adduct K-27 derived
from 2,4-bis(trifluoromethyl)benzaldehyde (A-7; entry 7,
Table 2). In the latter case, the additional ortho-CF3 sub-
stituent destabilizes the ketone due to unfavorable steric
interactions with the 2,6-di(2-propyl)phenyl groups (Ta-
ble 2, compare entries 2 and 7). When steric effects can be
neglected, e.g., within a series of para-mono-substituted
benzaldehydes (entries 1–6 in Table 2), the substituents
still have a substantial effect on the thermodynamic sta-
bilities of the Breslow intermediates (–12.5 < ΔG < –3.2
An explanation for the remarkable influence of the 2,6-
di(2-propyl)phenyl groups on the keto-enol energy differ-
ence may be seen in stabilizing dispersive interactions
between the 2-propyl groups and the enol substructure
within the Breslow intermediates.13 To assess dispersive
effects caused by the 2-propyl groups, we employed a
truncated model system. In the latter, we replaced the 2-
propyl groups of the carbene C-2 (SIPr), the enol E-21 and
the ketone K-21 by hydrogen atoms (Scheme 4). All atoms
present in the starting "full" system were frozen for the
optimization of the "truncated" model system, and only
ACS Paragon Plus Environment

Page 5 of 9
Journal of the American Chemical Society
Table 2. Reaction Free Energies for the Formation of
the Breslow Intermediates [ΔGEnol] and the Corre-
sponding Ketones [ΔGKetone] from SIPr (C-2) and Vari-
ous Aldehydes, together with the Keto-Enol-Differ-
ence [ΔΔGKE].a
1
2
3
4
5
6
7
8
E-27
K-27
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
Entry
1
Aldehyde
ΔGEnol
ΔGKetone
ΔΔGKE
–12.5 (E-22)
–7.8 (K-22)
+4.7
2
3
4
5
6
7
–10.7 (E-23)
–9.6 (E-24)
–9.6 (E-21)
–6.0 (E-25)
–3.2 (E-26)
–12.6 (E-27)
–7.4 (K-23)
–7.6 (K-24)
–7.2 (K-21)
–6.8 (K-25)
–6.4 (K-26)
+1.9 (K-27)
+3.3
+2.0
+2.5
–0.8
–3.2
+14.5
E-29
K-29
Figure 4: Lowest energy structures for the enol-ketone pairs
E-27/K-27 and E-29/K-29.
8
9
–10.0 (E-28) –8.6 (K-28)
–15.4 (E-29) –6.6 (K-29)
+1.4
Figure 5: Correlation of the ΔΔGKE and the corresponding
substituent constant σp [in kcal mol–1, M06-2X-D3/def2-
14
+8.8
TZVPP/IEFPCM(THF)//M06-L-D3/6-31+G(d,p)].
aall in kcal mol–1, M06-2XD3/def2-TZVPP/IEFPCM(THF)
//M06-L-D3/6-31+G(d,p)
Breslow intermediates. As a consequence of the relative
insensitivity of the ketone stabilities towards substituent
effects, the free energies for the keto-enol tautomerism
(ΔΔGKE) vary between –3.2 and +4.7 kcal mol–1 for para-
substituted benzaldehydes. Again, a good correlation
exists between the calculated differences in free energies
and the substituent constants σp (Figure 5, r2 = 0.94; see SI
for the additional correlations of ΔGKetone vs. σp, ΔGEnol vs.
σp, and dC=C vs. σp). The substituents’ influence on the
stability of the Breslow intermediates is also reflected in
the positive slope in Figure 5 which again indicates that
electron-withdrawing substituents stabilize the enol tau-
tomer. Entry 9 of Table 2 shows that the formation of the
diamino dienol E-29, derived from cinnamic aldehyde (A-
kcal mol–1). In contrast, the corresponding ketones are
significantly less affected by a change in substituents (–
7.8 < ΔG < –6.4 kcal mol–1). This can be rationalized by the
direct interaction of the aromatic substituent and the
diamino enol C-C double bond within the Breslow inter-
mediates, which is not present in the tautomeric ketones.
Electron-withdrawing substituents such as –CN or –CF3
stabilize the diamino enol significantly, while electron-
donating groups such as –OMe or –NMe2 cause a small
destabilization. This effect can also be deduced from the
correlation between the substituent constants σp14 and the
calculated reaction free energies for the formation of the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 9
9) and SIPr (C-2) is the most exergonic of all the calculat- system composed of C-2 (SIPr) and p-methoxybenzalde-
ed systems (–15.4 kcal mol–1). This can be explained by the
elongation of the π-system, while at the same time unfa-
vorable steric interactions become less when compared
with the system derived from benzaldehyde (E-21, Entry
4, Table 2).
hyde (A-5): As shown in Table 2, entry 5, the formation of
both K-25 and E-25 from carbene and aldehyde is exer-
gonic by 6-7 kcal mol-1. The ketone is favored by 0.8 kcal
mol-1, corresponding to a ca. 80/20 ketone-enol equilibri-
um mixture at room temperature. The experiment, how-
ever, affords the pure diamino enol E-25, and no ketone
K-25.3 This observation can be explained by initial, i.e.
kinetically favored formation of the enol E-25, and kinet-
ically inhibited subsequent tautomerization to K-25. An
even more clearcut preference for ketone formation
should be expected for the system C-2 (SIPr) plus p-
dimethylamino benzaldehyde (A-6), with ꢀꢀGKE amount-
ing to -3.2 kcal mol-1 (Table 2, entry 6). Unfortunatley, for
this extremely electron-rich benzaldehyde derivative, the
reaction with the carbene C-2 was very slow, and led to
the formation of a multitude of products. An unambigu-
ous assignment of the newly emerging resonances to the
enol E-26 or the ketone K-26 was not possible.
1
2
3
4
5
6
7
8
III. DISCUSSION
III.1 Keto-Enol Systems Derived from the Saturated
Imidazolidinylidenes C-2, C-3, and C-4
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
As mentioned in the introduction, Breslow intermediates
are formed smoothly and quantitatively when the bis-
Dipp- and bis-Mes-substituted imidazolidinylidenes C-2
(SIPr) and C-4 (SIMes) are reacted with aldehydes.3 As
shown in entries 2 and 4 of Table 1, this experimental
result is in perfect agreement with the reaction thermo-
dynamics: for both carbenes, enol and ketone formation
are significantly exergonic, but again for both carbenes,
the enol is the energetically more favorable product tau-
tomer. For the ketone-enol pair E-41 - K-41, the ꢀꢀGKE
amounts to 5 kcal mol-1, meaning basically exclusive enol
tautomer in equilibium at room temperature. For the
ketone-enol pair E-21 - K-21, the ꢀꢀGKE amounts to 2.5
kcal mol-1, translating to there being ca 98 % of the enol
in equilibrium at room temperature. According to our
analysis, the favorable energetics of enol formation are
mostly due to stabilizing intramolecular dispersive inter-
actions in the enol, between the 2,6-di(2-propyl) groups
(or analogously the 2,6-methyl groups) and the enol sub-
structure. For non-heteroatom substituted enols, the
stabilizing effect of e.g. Dipp- and Mes-substitution has
been recognized and experimentally quantified earlier.4,15
Both conjugative and steric effects have been discussed as
the sources of stabilization. In view of our computational
assessment of intramolecular dispersive interaction (vide
supra), it appears reasonable to assume that this hitherto
neglected aspect also contributes to the enol stabilization
observed earlier for e.g. Dipp-substituents.4,15
As reported by Wanzlick and Schikora already in 1961,
heating of the dimer of C-3 (the "Wanzlick dimer") with
benzaldehyde (A-1) cleanly affords the ketone K-31.10 This
result is in perfect agreement with the large negative
ꢀGKetone found for this system (-13.5 kcal mol-1; Table 1,
entry 3), as compared to its ꢀGEnol (-4.9 kcal mol-1). It is
tempting to assume that in this system, generation of the
Breslow intermediate may in fact precede the formation
of the ketone - the unsubstituted N-phenyl substituents
should not impede enol-to-ketone tautomerization. Un-
fortunately, the experimental (NMR) approach to answer-
ing this question is barred by the non-existence of mon-
omeric Wanzlick carbene (C-3). In contrast, repetition of
the Wanzlick/Schikora experiment with SIPr (C-2) as the
carbene component led to the formation of benzoin in
high yield, but the ketone K-31 could not be detected (see
Supporting Information). The benzoin condensation is
indicative for the formation of the Breslow intermediate
E-21, but once again, the tautomerization of E-21 to the
thermodynamically favored ketone K-21 did not occur.
The fact that the (themodynamically unfavorable) ke-
tones K-41 and K-21 can be prepared in pure form
(Scheme 3, Figure 1) raised our suspicion that their keto-
enol tautomerization may be extremely slow, i.e. kinet-
ically inhibited. In fact, preliminary experiments (see
Supporting Information) revealed that neither the addi-
tion of acids nor bases effects any conversion of K-41/K-21
to E-41/E-21. Even for the ketone-enol pair K-27/E-27 with
the highest ꢀꢀGKE (-14.5 kcal mol-1; Table 2, entry 7), no
ketone-to-enol tautomerization could be effected. For
keto-enol tautomerization to occur, a proton needs to be
added to/removed from the carbon atom flanked by the
N-atoms, i.e. the former carbene center. We attribute the
kinetic inhibition to the extreme shielding of this carbon
atom by the N-substituents' 2,4,6-trimethylphenyl or 2,6-
di(2-propyl)phenyl groups (see Figure 3 for the E-21 - K-21
pair). In fact, treatment of ketone K-21 with equimolar
acetic acid-d1 did not result in H/D-exchange at the ke-
tone's α-carbon atom over three days (see Supporting
Information). The assumption of kinetically inhibited
keto-enol-tautomerization is further supported by the
III.2 Keto-Enol Systems Derived from the Unsaturat-
ed Triazolylidene C-1, Imidazolylidenes C-5, C-6, and
the Thiazolylidene C-7
Triazolylidene C-1: For this carbene, we experimentally
observed ketone formation when exposed to benzalde-
hyde (A-1) in a stoichiometric fashion.5 This earlier find-
ing is in good agreement with the calculated ꢀꢀGKE of -7.0
kcal mol-1 (Table 1, entry 1). It is interesting to note that
for this highly active carbene catalyst, the formation of
the Breslow intermediate is in fact endergonic. The ꢀGEnol
of +3.5 kcal mol-1 explains why all efforts to detect Breslow
intermediates derived from C-1 have thus far met with
frustration. The thermodynamic data shown in Table 1,
entry 1 furthermore explain why for slow catalytic pro-
cesses (e.g. benzoin condensations), ketone formation
may occur as a competing catalyst deactivation pathway.
Imidazolylidenes C-5, C-6: For the bis-Dipp imidazol-
ylidene C-5, ketone formation with benzaldehyde (A-1) is
endergonic by +3.6 kcal mol-1, and the driving force for
ACS Paragon Plus Environment

Page 7 of 9
Journal of the American Chemical Society
enol formation is almost zero, resulting in an overall pref-
Breslow intermediates is crucial, has been reported earlier
by Bode et al..16
erence for enol formation (ꢀꢀGKE = +3.4 kcal mol-1; Table
1, entry 5). As may be expected from the thermodynamic
data of the C-5 - A-1 system, NMR-monitoring of the car-
bene's interaction with benzaldehyde (A-1) did not point
to the accumulation of either E-51 or K-51. Instead, the
conversion of the benzaldehyde to benzoin was observed.
The enol-stabilizing and ketone-destabilizing effect of the
Dipp-substituents in C-5 (IPr) becomes apparent again
when the thermodynamic data of the C-5 plus benzalde-
hyde (A-1) system are compared with those of the N,N'-
diphenyl imidazolylidene C-6. For the latter, the for-
mation of ketone K-61 is exergonic by -3.1 kcal mol-1, while
the formation of the enol E-61 becomes significantly en-
dergonic (+2.4 kcal mol-1), resulting in an overall prefer-
ence for ketone formation (ꢀꢀGKE = -5.6 kcal mol-1; Table
1, entry 6). Unfortunately, the experimental assessment of
this prediction is thus far frustrated by the non-
availability of the monomeric carbene C-6.
1
2
3
4
5
6
7
8
We furthermore discovered a remarkable case of kinetic
inhibition of proton translocation: For the carbene/alde-
hyde combinations for which the Breslow intermediate is
the thermodynamically favored tautomer, and for which
the Breslow intermediates have been characterized exper-
imentally, the keto tautomers were prepared by an alter-
native synthetic route. With these materials in hand, the
ketone-to-enol tautomerization was attempted, both in
the presence and absence of acid and base additives. Thus
far, no ketone-enol interconversion could be effected.
This kinetic inhibition of proton transfer is attributed to
the steric shielding of the ketone's α-/enol's β-carbon
atom by the 2,6-di(2-propyl)phenyl- or 2,4,6-trimethyl-
phenyl substituents on the carbene's N-atoms.17
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
V. EXPERIMENTAL
Details of the preparation and characterization of the
ketones K-11, K-21-26, K-31 and K-41 are summarized in
the Supporting Information, see also ref. 5 for the isola-
tion and characterization of ketone K-11.
Thiazolylidene C-7: Within the series of phenyl-substi-
tuted carbenes studied, both enol and ketone formation
are most exergonic when the N-phenyl thiazolylidene C-7
acts as carbene component. For benzaldehyde (A-1) as the
aldehyde component, ketone formation is preferred by ca.
5 kcal mol-1, well in line with the other N-phenyl substi-
tuted carbenes C-1 (triazolylidene), C-3 (imidazolidinyli-
dene), and C-6 (imidazolylidene).
VI. COMPUTATIONAL METHODS
First, the conformational space of all structures was ex-
plored using the OPLS-2005 force field18 and a modified
Monte Carlo search routine implemented in MacroModel
10.2.19 An energy cutoff of 20 kcal mol–1 was used during
the conformational analysis and structures with a heavy
atom root-mean-square deviation (RMSD) less than 1–2 Å
after the initial force field optimization were considered
to be the same conformer. All remaining structures were
then optimized in the gas phase employing the local me-
ta-GGA functional M06-L20 with Grimme’s dispersion-
correction D321 and the double-ζ split-valence basis set 6-
31+G(d,p) as well as density fitting. Furthermore, we in-
cluded geometries obtained from previously determined
crystal structures as additional starting points for the
geometry optimization. Subsequent vibrational analysis
verified that each structure was a minimum. Thermal
corrections were calculated from unscaled harmonic vi-
brational frequencies at the same level of theory for a
standard state of 1 mol L–1 and 298.15 K. The entropic
contributions to the reported free energies were derived
from partition functions evaluated using Truhlar’s quasi-
harmonic correction.22 This method uses the same ap-
proximations as the usual harmonic oscillator except that
all vibrational frequencies lower than 100 cm–1 are set
equal to 100 cm–1 to correct for the breakdown of the
harmonic oscillator approximation for low frequencies.
Electronic energies were subsequently obtained from
single-point calculations of the M06-L geometries em-
ploying the meta-hybrid M06-2X functional,23 Grimme’s
dispersion-correction D3 (zero damping),21 the large tri-
ple-ζ def2-TZVPP basis set,24 a level which is expected to
result in accurate energies.25 Solvation by tetrahydrofu-
ran, a solvent frequently used in carbene-catalyzed re-
actions, was taken into account by using the integral
equation formalism polarizable continuum model
IV. CONCLUSIONS
Our analysis of reaction free energies for the formation of
various Breslow intermediates and their keto tautomers
provides rationalization for previously unexplained exper-
imental observations:
(i) Ketone formation from triphenyltriazolylidene (C-1) and
aldehydes: The failure to observe Breslow intermediate E-
11 when reacting triphenyltriazolylidene (C-1) with ben-
zaldehyde (A-1) can be explained by the 7.0 kcal mol–1
thermodynamic preference for ketone formation, and the
absence of kinetic inhibition of enol/ketone tautomeriza-
tion by the carbene's N-phenyl substituents. Once at the
ketone stage, the concentration of enol E-11 will be much
too low to be detected.
(ii) Formation of Breslow intermediates from the imidazol-
idinylidenes C-2 (SIPr) and C-4 (SIMes) and aldehydes:
The introduction of N-Mes and N-Dipp substituents at
the carbene significantly stabilizes the enol-form, and
destabilizes the ketone. The synergism of these effects
makes the enol the thermodynamically favored tautomer
for both the saturated [imidazolidinylidenes C-2 (SIPr)
and C-4 (SIMes)] and the unsaturated [imidazolylidene C-
5 (IPr)] carbenes studied. The enol-stabilizing effect of the
Dipp-substituents could be attributed to intramolecular
dispersive interactions between the 2-propyl groups and
the enol moiety. In other words, for all carbene applica-
tions where fostering of enol formation is desired, resort-
ing to N-Dipp and/or N-Mes substitution is advisable
Note that the beneficial effect of mesityl substitution, for
triazolylidene catalytic reactions where the formation of
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 9
(IEFPCM)26 during the single point calculations. In gener-
al, solvation effects were found to be rather small as
demonstrated for the relative stabilities of E-21 and K-21
in the Supporting Information (Table S1, p. S8). Through-
out this investigation, an ultrafine grid corresponding to
99 radial shells and 590 angular points was used for the
numerical integration of the density.27 All density func-
tional theory calculations were performed with Gaussian
09.28
(4) For excellent and comprehensive discussion of the genera-
1
2
3
4
5
6
7
8
tion, reactivity, and spectroscopic/structural properties of non-
heteroatom substituted enols, see: a) Rappoport, Z. Ed. The
Chemistry of Enols, John Wiley & Sons: New York, 1990. b) Rap-
poport, Z.; Biali, S. E., Acc. Chem. Res. 1988, 21, 442-449.
(5) Berkessel, A.; Elfert, S.; Yatham, V. R.; Etzenbach-Effers,
K.; Teles, J. H. Angew. Chem. Int. Ed. 2010, 49, 7120 – 7124.
(6) C-1 to C-n denotes the carbenes discussed in this study,
A-1 to A-m the aldehydes. Of the ketones and enols formed, K-
nm indicates the ketone formed from carbene C-n and aldehyde
A-m, E-nm the corresponding enol composed of C-n and A-m.
(7) Meyer, D.; Neumann, P.; Koers, E.; Sjuts, H.; Lüdtke, S.;
Sheldrick, G. M.; Ficner, R.; Tittmann, K. Proc. Natl. Acad. Sci.
USA 2012, 109, 10867 – 10872.
9
VII. ASSOCIATED CONTENT
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
Supporting Information: Experimental details of the prep-
aration and characterization of the ketones K-11, K-21-26, K-
31 and K-41; X-ray data of p-bromophenyl glyoxal hydrate
and of the ketone K-31,29 Cartesian coordinates and energies
of all calculated structures, graphical visualization of lowest-
energy conformers, and details of computational methods.
(8) Reid, C. M.; Ebikeme, C.; Barrett, M. P.; Patzewitz, E.-M.;
Müller, S.; Robins, D. J.; Sutherland, A. Bioorg. Med. Chem. Lett.
2008, 18, 2455 – 2458.
(9) Schönberg, A.; Singer, E.; Eckert, P. Chem. Ber. 1980, 113,
2823 – 2826.
(10) Wanzlick, H.-W.; Schikora, E. Chem. Ber. 1961, 94, 2389 –
2393. The carbene C-3 is not persistent as monomer, and once
formed spontaneously dimerizes to the "Wanzlick dimer".
(11) Similar 2-propyl rotation occurs in the corresponding di-
aminoenols: see refs. 3,4. In these cases, different rates of rota-
tion are found depending on the cis/trans position of the N-
Dipp-substituent at the enol's C=C double bond.
(12) Maji, B.; Breugst, M.; Mayr, H. Angew. Chem. Int. Ed. 2011,
50, 6915 – 6919.
(13) a) Grimme, S.; Huenerbein, R.; Ehrlich, S. ChemPhysChem
2011, 12, 1258–1261. b) Wagner, J. P.; Schreiner, P. R. Angew.
Chem. Int. Ed. 2015, 54, 12274–12296.
(14) a) Hansch, C.; Leo, A.; Hoekman, D. Exploring QSAR: Hy-
drophobic, Electronic, and Steric Constants, American Chemical
Society, Washington, D.C., 1995. b) McDaniel, D. H.; Brown, H.
C. J. Org. Chem. 1958, 23, 420–427.
VIII. AUTHOR INFORMATION
Corresponding Author
*E-mail: berkessel@uni-koeln.de. Fax: (+49) 221-4705102
IX. ACKNOWLEDGMENT
This work was supported by the Deutsche Forschungsge-
meinschaft (DFG), Priority Program "Control of London
Dispersion Interactions in Molecular Chemistry" (SPP 1087,
grant no. BE 998/14-1) and by the Fonds der Chemischen
Industrie (Liebig Fellowship to M.B.). Computations were
performed on the DFG-funded Cologne High Efficiency
Operating Platform for Sciences (CHEOPS).
X. ABBREVIATIONS
(15) Hart, H.; Rappoport, Z.; Biali, S. E. in Rappoport, Z. Ed.
The Chemistry of Enols, John Wiley & Sons: New York, 1990, 481-
589.
Dipp: 2,6-bis(2-propyl)phenyl; Mes: 2,4,6-trimethylphenyl.
XI. REFERENCES
(16) Mahatthanachai, J.; Bode, J.W. Chem. Sci.2012, 3, 192–197.
(17) Note that besides steric hindrance, other intrinsic barriers
to proton transfer may be operative in keto-enol systems: Rich-
ard, J.P.; Amyes, T. L. Acc. Chem. Res. 2001, 34, 981-988.
(18) Banks, J. L.; Beard, H. S.; Cao, Y.; Cho, A. E.; Damm, W.;
Farid, R.; Felts, A. K.; Halgren, T. A.; Mainz, D. T.; Maple, J. R.;
Murphy, R.; Philipp, D. M.; Repasky, M. P.; Zhang, L. Y.; Berne,
B. J.; Friesner, R. A.; Gallicchio E.; Levy, R. M. J. Comput. Chem.
2005, 26, 1752 – 1780.
(19) MacroModel and V. 10.2, Schrödinger, LLC, New York,
NY, 2013.
(20) Zhao, Y.; Truhlar, D. G. J. Chem. Phys. 2006, 125, 194101.
(21) Grimme, S.; Antony, J.; Ehrlich S.; Krieg, H. J. Chem. Phys.
2010, 132, 154104.
(22) Ribeiro, R. F.; Marenich, A. V.; Cramer C. J.; Truhlar, D. G.
J. Phys. Chem. B 2011, 115, 14556 – 14562.
(23) Zhao Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215 –
241.
(24) Weigend F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005,
7, 3297 – 3305.
(25) Goerigk L.; Grimme, S. Phys. Chem. Chem. Phys. 2011, 13,
6670 – 6688.
(1) Reviews and recent mechanistic studies on N-heterocyclic
carbene catalysis: (a) Chiang, P.-C.; Bode, J. W. TCI MAIL, 2011,
149, 2–17. (b) Grossmann, A.; Enders, D. Angew. Chem. Int. Ed.
2012, 51, 314–325. (c) Izquierdo, J.; Hutson, G. E.; Cohen, D. T.;
Scheidt, K. A. Angew. Chem. Int. Ed. 2012, 51, 11686−11698. (d)
Bugaut, X.; Glorius, F. Chem. Soc. Rev. 2012, 41, 3511–3522. (e)
Douglas, J.; Churchill, G.; Smith, A. D. Synthesis 2012, 2295–2309.
(f) Ryan, S. J.; Candish, L.; Lupton, D. W. Chem. Soc. Rev. 2013,
42, 4906−4917. (g) Sarkar, S. D.; Biswas, A.; Samanta, R. C.; Stu-
der, A. Chem. Eur. J. 2013, 19, 4664−4678. (g) Collett, C. J.;
Massey, R. S.; Maguire, O. R.; Batsanov, A. S.; O'Donoghue, A. C.;
Smith, A. D. Chem. Sci. 2013, 4, 1514-1522. (h) Hopkinson, M. N.;
Richter, C.; Schedler, M.; Glorius, F. Nature 2014, 510, 485–496.
(i) Mahatthananchai, J.; Bode, J. W. Acc. Chem. Res. 2014, 47,
696–707. (j) Martin, D.; Canac, Y.; Lavallo, V.; G. Bertrand, G. J.
Am. Chem. Soc. 2014, 136, 5023–5030 (k) D. M. Flanigan, F. Ro-
manov-Michailidis, N. A. White, T. Rovis, Chem. Rev. 2015, 115,
9307-9387. (l) Collett, C. J.; Massey, R. S.; Taylor, J. E.; Maguire,
O. R.; O'Donoghue, A. C.; Smith, A. D. Angew. Chem. Int. Ed.
2015, 54, 6887-6892.
(2) a) Breslow, R. J. Am. Chem. Soc. 1957, 79, 1762 – 1763. b)
Breslow, R. J. Am. Chem. Soc. 1958, 80, 3719 – 3726.
(26) Cancès, E.; Mennucci B.; Tomasi, J. J. Chem. Phys. 1997,
107, 3032 – 3041.
(27) Wheeler S. E.; Houk, K.N. J. Chem. Theory Comput. 2010,
6, 395 – 404.
(28) Gaussian 09: Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.;
Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.;
Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Carica-
(3) a) Berkessel, A.; Elfert, S.; Yatham, V. R.; Neudörfl, J.-M.;
Schlörer, N. E.; Teles, J. H. Angew. Chem. Int. Ed. 2012, 51, 12370 –
12374. b) Berkessel, A.; Yatham, V. R.; Elfert, S.; Neudörfl, J.-M.;
Angew. Chem. Int. Ed. 2013, 52 11158-11162.
ACS Paragon Plus Environment

Page 9 of 9
Journal of the American Chemical Society
to, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng,
G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda,
1
R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.;
Nakai, H.; Vreven, T.; Montgomery, Jr., J. A.; Peralta, J. E.; Oglia-
ro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.;
Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.;
Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.
E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J.
W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Far-
kas, Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.;
Gaussian, Inc., Wallingford CT, 2009.
(29) CCDC 1437685 (p-bromophenylglyoxal hydrate) and
CCDC 1437684 (ketone K-31) contain supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac. uk/data_request/cif.
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
Table of Contents artwork:
ACS Paragon Plus Environment