Activation of Molecular Hydrogen by Arylcarbenes

Article abstract of DOI:10.1002/chem.201804657

The hydrogenation reactions of diphenylcarbene 1, fluorenylidene 2, and dibenzocycloheptadienylidene 3 were investigated in solid H₂ and D₂ matrices and in H₂- and D₂-doped argon matrices at cryogenic temperatures. The reactivity of the carbenes towards H₂ increases in the order 1<3<2. Whereas 1 is stable in solid H₂, 2 and 3 react fast under the same conditions via quantum chemical tunneling. In D₂ both 1 and 3 are stable, whereas 2 slowly reacts. The different reactivity of the three carbenes is rationalized in terms of differing carbene stabilization energies.

Full text of DOI:10.1002/chem.201804657

A Journal of
Accepted Article
Title: Activation of Molecular Hydrogen by Arylcarbenes
Authors: Wolfram Sander, Enrique Mendez-Vega, Mika Maehara,
Akshay Hemant Raut, Joel Mieres-Perez, Masashi Tsuge,
and Yuan-Pern Lee
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201804657
Link to VoR: http://dx.doi.org/10.1002/chem.201804657
Supported by

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
Activation of Molecular Hydrogen by Arylcarbenes
Enrique Mendez-Vega,[a] Mika Maehara,[a] Akshay Hemant Raut,[a] Joel Mieres-Perez,[a] Masashi
Tsuge,[b] Yuan-Pern Lee,*[b,c] and Wolfram Sander*[a]
Abstract: The hydrogenation reactions of diphenylcarbene 1, fluo-
renylidene 2, and dibenzocycloheptadienylidene 3 were investigated
in solid H2 and D2 matrices and in H2- and D2-doped argon matrices
at cryogenic temperatures. The reactivity of the carbenes towards H2
increases in the order 1 < 3 < 2. While 1 is stable in solid H2, 2 and 3
react fast under the same conditions via quantum chemical tunneling.
In D2 both 1 and 3 are stable, whereas 2 slowly reacts. The different
reactivity of the three carbenes is rationalized in terms of differing car-
bene stabilization energies.
Scheme 1. Reactions of singlet and triplet carbenes with molecular hydrogen.
Introduction
It was demonstrated by electron paramagnetic resonance
Carbenes are highly versatile reactive intermediates that exhibit a
(EPR) and by trapping studies that arylcarbenes such as diphe-
nylcarbene 1, fluorenylidene 2, and dibenzocycloheptadienyli-
dene 3 exhibit triplet ground states, although with smaller S–T
gaps than the parent CH2.[4-5] Carbenes 1–3 are archetypal aryl-
broad range of reactivity patterns, depending on the substituents
at the carbene center. Closed-shell singlet carbenes can be de-
scribed as 1,1-zwitterions, bearing both nucleophilic and electro-
philic regions at the carbene center. In contrast, triplet carbenes
carbenes and therefore were thoroughly investigated in solution
exhibit radical-like reactivity, such as atom abstractions, to yield
by time resolved spectroscopy[6], by low-temperature spectros-
radical pairs that rapidly undergo secondary reactions (Scheme
copy in organic glasses,[7] and in inert gases matrices,[8] and by
1). A paradigm in carbene chemistry is that reactions of carbenes
computational methods. Recently, we reported reactions of these
are spin specific and occur from equilibrated spin states, depend-
three carbenes with H2O and CH3OH as hydrogen bond donors,
ing on the temperature.[1-3] It is therefore of importance not only to
CF3I as halogen bond donor, and BF3 as strong Lewis acid under
consider the ground state multiplicity of a carbene, but also the
energy gap EST between the ground state and lowest lying ex-
cited state of opposed multiplicity.
the conditions of matrix isolation.[9] Most notably, if triplet diphe-
nylcarbene T-1 interacts with H2O or CH3OH, hydrogen-bonded
complexes of the singlet carbene S-1 are formed. Thus, the
ground state of 1 is switched from triplet to singlet. This is not only
observed in cryogenic matrices at low temperature,[10-11] but also
in solution at room temperature.[12] Carbene 2 shows a similar be-
havior,[13] whereas in 3 the singlet–triplet splitting is too large to
be overcome by hydrogen bonding. Halogen bonding stabilizes
the singlet states of carbenes even more than hydrogen bond-
ing,[14] and BF3 forms strongly stabilized donor-acceptor com-
plexes,[15-16] similar to NH3-BF3. These results stress the high ba-
sicity of the closed-shell singlet states of carbenes.
Typical reactions of triplet carbenes with less polar reagents
are atom abstractions to form radical pairs, which subsequently
undergo a multitude of secondary reactions. Thus, photolysis of
diazodiphenylmethane 4 in cyclohexane at room temperature
predominantly leads to the dimer of the benzhydryl radical to-
gether with small amounts of the formal C–H insertion product
and diphenylmethane 5. In contrast, photolysis of 9-diazofluorene
6 in cyclohexane produces mainly (65%) the C–H insertion prod-
uct 9-cyclohexylfluorene and only smaller amounts of radical-re-
combination products.[17] This was taken as evidence that 1 re-
acts with hydrocarbons from its triplet ground state via hydrogen
abstraction, whereas 2 with a smaller S–T splitting undergoes di-
rect C–H insertions via its thermally populated singlet state.[18-19]
[a]
Dr. E. Mendez-Vega, M. Maehara, A. H. Raut, Dr. J. Mieres-Perez,
Prof. Dr. W. Sander
Lehrstuhl für Organische Chemie II, Ruhr-Universität Bochum
44780 Bochum, Germany
E-mail: wolfram.sander@rub.de
[b]
[c]
Dr. M. Tsuge, Prof. Dr. Y.-P. Lee
Department of Applied Chemistry and Institute of Molecular Science
National Chiao Tung University, Hsinchu 30010, Taiwan
E-mail: yplee@nctu.edu.tw
Prof. Dr. Y.-P. Lee
Center for Emergent Functional Matter Science
National Chiao Tung University, Hsinchu 30010, Taiwan
Institute of Atomic and Molecular Sciences, Academia Sinica
Taipei 10617, Taiwan
Supporting information for this article is available on the WWW un-
der http://www.chemeurj.org/.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
Both singlet and triplet carbenes are in principle able to insert
into the H–H bond, however, the mechanisms are entirely differ-
ent. Bertrand et al. reported that stable (alkyl)(amino)carbenes in-
sert into H2 with a barrier of 22–23 kcal mol–1.[20-21] These car-
benes have very high-lying triplet states, and therefore any triplet
reactivity is excluded. The activation of H2 by these carbenes
shows some similarity to the activation by transition metals: in the
transition state the electrophilic center of the carbene withdraws
electron density from the bonding  orbital of H2 whereas the nu-
cleophilic center donates electron density into a * orbital, thus
weakening the H–H bond (Scheme 1). Henkel and Sander
demonstrated that the more reactive singlet 1-azulenylcarbene in-
serts into H2 at cryogenic temperatures via a tunneling mecha-
nism.[22] So far, this is the only known singlet carbene that under-
goes H2 insertion at temperatures below 10 K.
Here, we describe the reactions of the three triplet ground
state carbenes 1–3 in H2 and D2 matrices and in H2/D2-doped ar-
gon. The reactivity of these three carbenes towards hydrogen is
very different, which allows us to gain detailed insights into struc-
tural and electronic prerequisites that govern the carbene-hydro-
gen reaction.
Results and Discussion
The reactions of carbenes 1–3 with molecular hydrogen in cryo-
genic matrices were investigated under two different conditions:
(i) in solid hydrogen matrices and (ii) in inert gas matrices doped
with hydrogen. In both cases, H2 and D2 isotopes were used to
investigate isotope effects on the reaction. Due to the high volatil-
ity of H2, temperatures below 4 K are necessary to prevent subli-
mation of H2, whereas the heavier isotope D2 is stable up to 6 K
under matrix isolation conditions. In solid Ne and Ar, H2 and D2
can be trapped at higher temperatures, and bimolecular reactions
can be studied up to 8 and 35 K, respectively.
In contrast, triplet carbenes are more reactive. Zuev and Sher-
idan studied a number of reactive singlet and triplet ground state
carbenes in H2-doped argon matrices at 30 K and concluded that
singlet ground state carbenes such as chloro(phenyl)carbene 7
do not react with H2 under these conditions whereas triplet car-
benes such as 8–10 readily produce the corresponding H2 inser-
tion products.[23] The triplet carbenes were assumed to react via
radical pairs (Scheme 1) which could not be directly observed in
these experiments due to the rapid recombination to the final
products. The formation of the radical pairs was calculated to be
mildly exothermic, however activation barriers of several kcal mol–
Diphenylcarbene. Triplet carbene 1 was generated in Ar ma-
trices by visible light photolysis (λ = 520–560 nm) of diphenyldi-
azomethane 4. The infrared (IR) spectrum shows two intense sig-
nals at 744 and 674 cm1, which are assigned to the out-of-plane
(oop) ring deformation modes of 1 and are in agreement with lit-
erature (Figure 1a).[26]
1
were predicted for the hydrogen abstraction. Since these barri-
ers are too high to be overcome at 30 K, a quantum chemical tun-
neling mechanism was proposed. If D2 was used instead of H2,
no reaction was observed, indicating a very large kinetic isotope
effect, as expected for a tunneling reaction. In contrast, highly
electrophilic triplet carbenes were found to undergo fast reactions
with D2 under similar conditions.[24]
This mechanistic assumption was later confirmed for 4-oxacy-
clohexadienylidene 9.[25] The radical pair could be detected by
EPR spectroscopy if the carbene was allowed to react in solid ar-
gon doped with H2, HD, or D2. Interestingly, with HD only D atoms
were observed in the radical pair, indicating a very large prefer-
ence for H abstraction over D abstraction. The experiments were
also performed in solid H2 and D2. While in H2 the insertion prod-
uct was already formed after photolysis of the precursor, in D2
carbene 9 could be observed in addition to the insertion product.
The carbene slowly reacts with D2 to form the insertion product
under these conditions. The kinetics were monitored by IR and
EPR spectroscopy, and the rates at 3 and 6 K were found to be
identical, strongly supporting a tunneling mechanism for the inser-
tion reaction.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
light (365 and 450 nm) leads only to dimerization of the carbene
under formation of tetraphenylethylene (Figure S1–3, Table S1).
This is explained by the local heating of the matrix during UV–vis
irradiation, resulting in diffusion of 1.
Triplet carbenes are well known to react barrierless with mo-
lecular oxygen,[28] and therefore we studied the photolysis of 4 in
p-H2 matrices doped with 1% O2 (Figure S4, Scheme 2). Under
these conditions, carbene 1 is formed together with benzophe-
none 11 and diphenyldioxirane 12. Annealing of the p-H2 matrix
at 4.5 K results in the reaction of the remaining 1 to benzophe-
none O-oxide 13, similar to the reaction observed in O2-doped ar-
gon matrices.[26] These experiments clearly indicate that H2 acts
as an inert host matrix at temperatures of 4.5 K and below, and
no thermally or IR activated reactions occur.
Scheme 2. Reaction of diphenylcarbene 1 with O2 in p-H2 matrices.
Photolysis (560 nm) of a 5% H2-doped Ar matrix containing 4
at 3 K produces carbene 1 in high yields. In this matrix, 1 is indef-
initely stable at temperatures below 15 K, while warming to 20–
30 K results in a rapid decrease of all IR bands assigned to 1.
Simultaneously, diphenylmethane 5 with strong IR bands at 737,
700, 695, and 609 cm–1 is formed (Figure S5). Obviously, hydro-
genation of 1 requires temperatures considerably higher than
15 K.
The kinetics of the hydrogenation of 1 depends on the intrinsic
activation barrier for this process and on the diffusion of H2 in the
solid matrix. Therefore, the kinetics of the 1 + H2 reaction was
studied in H2-doped argon and neon matrices at temperatures of
30 K and 8 K, respectively. Since the diffusion of small molecules
such as H2 in solid matrices depends on the temperature and the
melting point of the matrix (rapid diffusion is generally observed
at approximately 1/3 of the melting point), a temperature of 30 K
in argon corresponds roughly to 8 K in neon. The reaction rates
were determined to be 1.2 ± 0.1 × 10–5 s–1 at 30 K in Ar, but only
1.7 ± 0.6 × 10–6 s–1 at 8 K in Ne (Figure S5). Assuming that diffu-
sion rates of H2 are similar in these two experiments, we conclude
that there is a considerable thermal activation barrier for the 1 +
H2 reaction. When D2-doped argon is used in the experiments, no
reaction is observed, indicating a very large kinetic isotope effect.
Moreover, if the Ar matrix is doped with both H2 and D2 (2.5%
each), only the H2-insertion product 5 is formed upon annealing,
Figure 1. IR spectra showing the isolation of diphenylcarbene 1 in solid Ar, H2,
and D2 as well as its selective reaction with H2 in Ar/H2/D2 matrices. a-c) IR
spectra obtained after irradiating diphenyldiazomethane 4 with λ = 560 nm in Ar
(a), D2 (b), and H2 (c) at 3 K. d) Difference IR spectrum obtained after annealing
a matrix containing 1 in Ar doped with both H2 and D2 (2.5% each) at 30 K for
24 h. e) IR spectra of diphenylmethane 5 (blue) and its isotopomer d2-5 (red) in
Ar at 3 K. Signals corresponding to D2- and H2-insertion products d2-5 and 5 are
not present in spectra (b) and (c), respectively.
Complete photolysis of 4 in H2 or D2 matrices at 3 K using
560 nm light results in the formation of carbene 1 in high yields,
while the H2- and D2-insertion products 5 and d2-5 are not de-
tected (Figure 1). This clearly indicates that carbene 1 does not
react with solid H2 and D2 at 3 K even during 560 nm irradiation.
Warming the H2 and D2 matrices to 4.5 and 6 K, respectively, lead
to no reaction. Solid H2 (both normal and para) shows a set of
sharp signals (Q, S(0), and S(1)) in the region of 5000–4000 cm-1,
due to quadrupolar interaction.[27] Irradiation into this region is ex-
pected to excite the H2 molecules, however, no reaction with 1
was observed during IR irradiation. Irradiation with UV or visible
This article is protected by copyright. All rights reserved.

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
stressing the very large difference in reactivity between H2 and D2
(Figure 1d, Scheme 3).
Scheme 3. Reaction of diphenylcarbene 1 with H2 and D2 in matrices.
Fluorenylidene. Irradiation of 9-diazofluorene 6 in solid H2 at
3 K with UV light (365 nm) produces fluorene 14 as the only prod-
uct (Figure 2d). Carbene 2 is not observed, indicating that this
carbene reacts very fast with H2 even at temperatures as low as
3 K (Figure S6, Scheme 4).
Figure 2. IR spectra showing the photochemistry of 9-diazofluorene 6 in H2 and
D2 matrices. a) IR spectrum obtained after irradiating 6 with λ = 365 nm for 2 h
in Ar at 3 K. b) IR spectrum obtained after irradiating 6 with λ = 365 nm for 1 h
in D2 at 3 K. c) Difference IR spectrum obtained after keeping the matrix at 3 K
in darkness for 24 h. d) IR spectrum obtained after irradiating 6 with λ = 365 nm
for 2 h in H2 at 3 K. e) IR spectra of fluorene 14 (blue) and its isotopomer d2-14
(red) in Ar at 3 K.
Irradiation of precursor 6 in D2 at 3 K results in the formation
of both carbene 2 and the insertion product d2-14 (Figure 2b and
S7). After keeping the matrix in the dark for several hours the
amount of 2 decreases while more d2-14 is formed, and after 24 h
approximately 70% of carbene 2 reacts to d2-14 (Figure S8).
This article is protected by copyright. All rights reserved.

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
The 2 + D2 reaction was also monitored by EPR spectroscopy
(Figure 4). Photolysis of matrix-isolated 6 at 5 K produces strong
EPR signals of triplet carbene 2.[29] After 24 h at 5 K in the dark
the EPR signals of 2 decreased by 71%, in good agreement with
the 73% loss of the IR signal intensity of 2 determined under sim-
ilar conditions. Simultaneously, the characteristic signals of D at-
oms at 3340, 3418, and 3496 G (hyperfine coupling 78 G) were
observed, as well as a small amount of H atoms (3147 and
3654 G, hyperfine coupling 507 G), presumably from small H2
contaminations.[30] The formation of D atoms concomitant with the
disappearance of 2 is consistent with an abstraction-recombina-
tion mechanism, as expected for triplet carbenes (Scheme 4).[23]
Scheme 4. Reactivity of fluorenylidene 2 in solid H2 and D2 matrices.
The kinetics of the reaction of carbene 2 in solid D2 was stud-
ied by following the decay of its characteristic IR band at 713 cm−1
and the rise of the 742 cm−1 absorption of product d2-14. The re-
action rate was determined to 2 – 5 × 10–5 s–1 in solid D2 in the
temperature range between 3 and 6 K and is thus almost inde-
pendent of temperature (Figure 3 and Table S2). This corre-
sponds to a near-zero Arrhenius activation barrier, strongly sup-
porting that the 2 + D2 reaction is governed by quantum chemical
tunneling.
Figure 4. EPR spectra showing the decay of triplet fluorenylidene 2 in D2 matri-
ces. a) EPR spectrum obtained after irradiating 9-diazofluorene 6 with λ =
365 nm for 30 min in D2 at 5 K. b) EPR spectrum obtained after keeping the
matrix at 5 K in darkness for 24 h. Inset shows the region of the spectra where
H and D radicals appear.
The 9-deuterofluorenyl radical d-15 could not be identified,
probably because of overlapping signals with that of D atoms. The
barrierless recombination of the radical pair to give d2-14 is ex-
pected to be the major reaction pathway, and therefore only low
concentrations of D atoms and d2-14 are expected. In agreement
with this, IR spectroscopy provides no evidence for radicals 15 or
d-15 in H2 and D2 matrices, respectively.
Additional experiments were performed in inert gas matrices
doped with H2 and D2. Annealing of a 5% H2-doped Ar matrix con-
taining carbene 2 at 30 K for only 5 min results in the complete
transformation of carbene 2 into 14 (Figure S9). Such results are
in line with the high reactivity observed in pure H2 at 3 K. The re-
action slows down in D2-doped argon to 4.4 ± 0.1 × 10–5 s–1 at
30 K, indicating a large kinetic isotope effect. This rate constant is
similar to the value determined in solid D2 at 6 K (4–4.5 × 10–5 s-1),
indicating that the diffusion of D2 in Ar at 30 K is quite efficient.
The similar rates observed at temperatures between 3 and 30 K
strongly support the assumption that the reaction proceeds via
tunneling.
Figure 3. Plot of the increasing intensity of the IR signal at 742 cm−1 (integrated
over 738.9–744.7 cm–1, with initial value subtracted) and the decreasing inten-
sity of the IR signal at 713 cm−1 (integrated over 711.6−714.2 cm1), recorded
in D2 matrices at 3 K. Inset shows the spectroscopic features used for the kinetic
analysis.
Dibenzocycloheptadienylidene. Visible light photolysis (λ =
530 nm) of matrix isolated diazodibenzocycloheptadienylidene 16
This article is protected by copyright. All rights reserved.

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
in argon at 3 K leads to the formation of triplet dibenzocyclohep-
tadienylidene 3 in good yields (Figure 5a). Carbene 3 was identi-
fied by comparing its IR spectrum with previously published spec-
tra.[31]
Scheme 5. Reactivity of dibenzocycloheptadienylidene 3 in H2 and D2 matrices
Additional experiments were performed in argon matrices
doped with H2 and D2. In 5% H2-doped argon matrices, precursor
16 was photolyzed (530 nm, 3 K) to produce carbene 3. Subse-
quent annealing of the matrix at 30 K leads to a decrease of car-
bene 3, and formation of the H2-insertion product 17 with a reac-
tion rate of 6.4 ± 0.2 × 10–5 s–1 (Figure S10). After 24 h, 67% of 3
reacted to 17, whereas no reaction was observed in 5% D2-doped
argon matrices at 30 K.
The reactivities of carbenes 1–3 towards H2 and D2 are re-
markably different (Table 1). Carbene 2 reacts with H2 and D2 un-
der all experimental conditions, and only with D2 the reaction is
slow enough to be directly monitored. In contrast, carbene 1 is
stable in solid H2 and only reacts at higher temperatures in H2-
doped argon matrices. Carbene 3 reacts rapidly with H2, but is
stable in D2 matrices, displaying a reactivity between that of car-
benes 1 and 2.
Table 1. Reactivity of arylcarbenes 1–3 with H2 and D2.
H2
(3 K)
D2
(6 K)
Ar/H2
(30 K)
Ar/D2
(30 K)
Figure 5. IR spectra showing the photochemistry of diazodibenzocycloheptadi-
enylidene 16 in H2 and D2 matrices. a-c) IR spectra obtained after irradiating 16
with λ = 530 nm in Ar (a), D2 (b), and H2 (c) at 3 K. d) IR spectra of dibenzosu-
berane 17 in Ar at 3 K.
Carbene
No
reaction
No
reaction
No
reaction
1
2 × 10–5
No
reaction
No
reaction
3
2
>10–2
>10–2
6 × 10–5
>10–2
Under similar conditions, photolysis of 16 in solid H2 at 3 K
results in the disappearance of 16, and new signals assigned to
dibenzosuberane 17 are formed concomitantly (Figure 5c). This
suggests that carbene 3 instantaneously reacts with surrounding
H2 molecules after it is generated, similar to carbene 2.
4 × 10–5
4 × 10–5
Pseudo-first-order rate constants (with β = 0.75) are given in (s–1).
When D2 is used as matrix, 530 nm irradiation of 16 at 3 K
produces carbene 3 as the only product (Figure 5b). To induce a
reaction between 3 and D2, broadband IR irradiation (730–
2500 nm) was carried out aiming at populating excited vibrational
levels of D2. However, neither IR irradiation nor annealing of the
matrix at 6 K for several hours resulted in a reaction of 3. This
indicates a very large H2/D2 kinetic isotope effect (Scheme 5).
Calculations. Carbene stability can be assessed by using the
concept of carbene stabilization energy (CSE), as described by
O´Ferral et al.[32-33] The CSE is defined as the enthalpy of the ho-
modesmotic reaction converting the divalent carbene center into
a tetravalent carbon atom, with methylene as reference (Scheme
6). By this definition, a positive CSE indicates that the substituent
provides greater stabilization to the carbene than the H atoms in
CH2 (Table 2).
Scheme 6. Carbene stabilization energy (CSE).
This article is protected by copyright. All rights reserved.

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
Table 2. Calculated carbene stabilization energies (CSE) and singlet-triplet split-
tings (ΔEST) of arylcarbenes 1–3
CSE
(Singlet)
CSE
(Triplet)
Carbene
Method
ΔEST
DFT
CCSD(T)
DFT
37.3
30.3
28.7
24.6
36.5
29.6
31.6
21.0
22.2
15.4
35.4
24.8
5.7
3.3
1
4.9
2
3
CCSD(T)
DFT
3.4
10.2
7.8
CCSD(T)
Energies calculated at the CCSD(T)/cc-PVDZ//B3LYP-D3/def2-TZVP level of
theory are given in kcal mol–1
.
DFT as well as CCSD(T) energy calculations reveal smaller
CSEs for both the singlet and triplet states of carbene 2 than for
the singlet and triplet states of 1 and 3 (Table 2). The relative
lower stability of S-2 compared to S-1 is attributed to the contribu-
tion of antiaromatic destabilization in S-2. T-2 is less stable than
T-1 and T-3 because of the smaller C–C–C angle in the five-mem-
bered ring which destabilizes the triplet state. According to DFT
calculations, T-2 has a central bond angle of 112.1°, which is
much smaller than that of T-1 (141.6°) and T-3 (142.6°). The sta-
bilization of T-3 with respect to T-1 results from the additional or-
tho alkyl substituents in 3 which are absent in 1.
The reaction profiles for the hydrogenation of carbenes 1–3
were studied by intrinsic reaction coordinate (IRC) calculations
(Figure 6). Similar IRC calculations were previously used to esti-
mate tunneling probabilities of carbene reactions.[34-36] The first
step of the reaction is the formation of the triplet radical pair by H-
abstraction. The radical pair subsequently undergoes ISC to the
singlet surface and recombines to the formal hydrogen insertion
product. This triplet pathway is found to be significantly lower in
energy than reactions at the singlet surface (Figure S11–13).
The barrier for the H-abstraction is predicted to 2.6 kcal mol–1
for 2 and ~7 kcal mol–1 for both 1 and 3 at the B3LYP-D3/def2-
TZVP level of theory. These barriers are much higher than the
thermal energy available at 3 K, and therefore any thermal reac-
tion should be efficiently suppressed at such low temperatures,
suggesting that these reactions proceed via quantum chemical
tunneling. For 2 the H-abstraction is predicted to be exothermic
by −8.5 kcal mol–1, while it is slightly endothermic for 1 and 3. The
radical stabilization energy of radical pairs 15 and 18 can be esti-
mated by comparing the C–H bond dissociation energies (BDE)
of the corresponding hydrocarbons 5 and 14. The BDEs of 5 and
14 are quite similar with 82.0 and 84.5 kcal mol–1, respectively,[37]
and therefore the stability of the radical pairs 15 and 18 should
also be similar. Consequently, the strikingly different H-abstrac-
tion enthalpies calculated for carbenes 1 and 2 result from the
different carbene stabilities and not from different stabilities of the
product radical pairs.
Figure 6. Hydrogenation of diphenylcarbene 1 (blue), fluorenylidene 2 (red),
and dibenzocycloheptadienylidene 3 (black) on the triplet PES along the intrin-
sic reaction path. Energy profiles were calculated at the B3LYP-D3/def2-TZVP
level of theory. Values in parentheses include ZPE correction. Energies are
given in kcal mol–1 and selected bond lengths and angles are given in Å and
degrees, respectively. Relative energy of the complexes between the triplet car-
benes 1–3 and H2 is set to zero, as reference.
The high reactivity of 2 towards H2 and D2 at low temperatures
can be rationalized by (i) the large exothermicity resulting in an
This article is protected by copyright. All rights reserved.

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
early transition state, (ii) the low activation barrier, and (iii) the
short tunneling path through the barrier (Figure 6). Consequently,
the transition state T-2···H is formed at a larger intermolecular C–
H distance (1.53 Å) and a shorter H–H distance (0.84 Å) com-
pared to the other carbenes. Overall, the calculations predict a
considerably lower and narrower barrier for 2 compared to 3 and
1.
H) with solid argon and H2 as well as heterogeneous lattice struc-
tures might cause the reaction of 1 to become slightly exothermic
in H2-doped solid argon. Furthermore, as the matrix softens at
higher temperatures, it can better accommodate the structural
changes required for the tunneling of 1.
Experimental Section
Synthesis. For synthesis details and characterization of the compounds,
see the Supporting Information.
Conclusions
While the arylcarbenes 1–3 are structurally related, their reactivity
towards molecular hydrogen shows pronounced differences. Di-
phenylcarbene 1 is indefinitely stable in solid H2 at 3 K, whereas
fluorenylidene 2 and dibenzocycloheptadienylidene 3 react too
fast to be observable (Table 1). In solid D2, both 1 and 3 are stable,
while 2 reacts slowly with a rate of 4 × 10–5 s–1. These results sug-
gest a pronounced kinetic isotope effect and increasing reactivity
in the order 1 < 3 < 2. For all three carbenes, the first, rate-deter-
mining step is the H/D abstraction from the H2/D2 molecule via
quantum chemical tunneling to form a radical pair. The second
step is the rapid radical recombination to produce the formal H2
insertion product. This mechanism is in line with all spectroscopic
and computational evidence.
The closed-shell singlet states of carbenes 1–3 lie several
kcal mol–1 above their triplet ground states, and therefore reac-
tions from thermally populated excited singlet states can be ex-
cluded under the experimental conditions. Interestingly, coupled
cluster calculations predict almost identical ΔEST for 1 and 2 in the
gas phase (3.3 and 3.4 kcal mol–1, respectively, Table 2), in con-
tradiction to reactivity studies in solution at room temperature
which conclude that ΔEST for 2 is much smaller than that of 1.[17]
The calculated ΔEST for 3 is considerably larger than that of the
other two carbenes.
Matrix Isolation. Experiments were performed by standard techniques[38]
using Sumitomo Heavy Industries two-staged closed-cycle helium refrig-
erator systems (3 K). Matrices were generated by co-deposition of the sub-
stance of interest along with an excess of host gas onto a cesium iodide
window (IR) or an oxygen-free high-conductivity copper rod (EPR). Inert
gases used for matrix isolation were Ar, Ne, H2, O2 (Air Liquide, 99.999%),
and D2 (Merck Schuchardt, 99.5%). Broadband photolysis of the matrices
was performed using LEDs with  = 530, 450, 405 or 365 nm (max. 5 W).
For visible ( = 560 nm) and IR irradiation (>4000 cm–1), a lamp-pumped
OPO laser (InnoLas Laser model SpitLight 600 midband) was used. A de-
scription of the set-up and procedure used for preparing p-H2 is given in
the Supporting Information.
Low-Temperature Spectroscopy. IR spectra were recorded with a
Bruker Vertex V70 FTIR spectrometer with a resolution of 0.5 cm–1. The
IR beam of the spectrometer was passed through IR long pass interfer-
ence filters (passing  2400 or 9800 nm) to avoid the formation of vibra-
tionally excited H2 or D2 upon absorption of the unfiltered IR light.[39] Matrix-
isolation EPR spectra were recorded on a Bruker Elexsys 500 X-band
spectrometer. The kinetic data was fitted to a pseudo-first order by using
the equation of Wildman and Siebrand.[40] This model assumes a distribu-
tion of reaction rates by introducing a dispersion coefficient β into the first
order expression.
Calculations. Geometry optimizations, vibrational frequencies, and IRC
profiles were calculated using the B3LYP[41-42] hybrid functional with D3
empirical dispersion correction.[43] Basis sets such as def2-TZVP[44] and
cc-pVDZ[45] were employed. CCSD(T) single point calculations at DFT op-
timized geometries were performed to recalculate S–T energy gaps, en-
thalpies, and activation barriers of arylcarbenes. DFT and CCSD(T) calcu-
lations were performed with the programs Gaussian 09[46] and Molpro
2012,[47] respectively. Experimental EPR spectra were analyzed using the
Easyspin program package.[48]
The stability of 1 in solid H2 could either result from very low
tunneling rates (depending on the height and width of the activa-
tion barrier) or from unfavorable thermodynamics. The fact that
with IR irradiation, 1 still did not react in solid H2 really indicates
that this reaction is endothermic. The IRC calculations (Figure 6)
indicate that both, thermodynamics and kinetics, favor the reac-
tion of 2. The formation of the radical pair 15 is clearly exothermic,
and the barrier is lower and narrower than for the other two car-
benes. The differences between carbenes 1 and 3 are subtler.
The heights of the barriers (7.2–7.3 kcal mol-1) and also the
widths are very similar, however, for 1 the transition state is later,
and the reaction energy is slightly positive, whereas that of 3 is
nearly thermal neutral. The calculations also reveal that the higher
reactivity of T-2 compared to the other two triplet carbenes results
from its lower carbene stabilization, which in turn is a conse-
quence of the small carbene C-C-C angle in the five-membered
ring. This destabilization is only observed in carbene 2 but not in
radical 15, which explains the larger exothermicity of the hydro-
gen abstraction reaction of 2 compared to that of 1 and 3.
To understand the reaction of 1 with H2 in H2-doped argon at
30 K, we assume that the reaction becomes favorable by subtle
changes in the matrix environments. Varied interactions of H2 (or
Acknowledgements
This work was supported by the Cluster of Excellence RESOLV
(EXC 1069) funded by the Deutsche Forschungsgemeinschaft
(DFG), by the Ministry of Science and Technology, Taiwan (grant
No. MOST106-2745-M009-001-ASP and MOST107-3017-F009-
003), and by the Center for Emergent Functional Matter Science
of National Chiao Tung University from The Featured Areas Re-
search Center Program within the framework of the Higher Edu-
cation SPROUT Project by the Ministry of Education (MOE) in
Taiwan. YPL thanks the Alexander von Humboldt Foundation for
a Humboldt Research Award.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
[26]
[27]
W. W. Sander, J. Org. Chem. 1989, 54, 333-339.
H. P. Gush, E. J. Allin, H. L. Welsh, W. F. J. Hare,
Can. J. Phys. 1960, 38, 176-193.
Keywords: carbenes • hydrogen activation • IR spectroscopy •
matrix isolation • tunneling
[28]
[29]
[30]
W. Sander, Angew. Chem. Int. Ed. 1990, 29, 344-
354.
C. A. Hutchison, G. A. Pearson, J. Chem. Phys.
1965, 43, 2545-2546.
L. W. Anderson, F. M. Pipkin, J. C. Baird, Phys.
Rev. Lett. 1960, 4, 69-71.
chem.201803695
S. Gronert, J. R. Keeffe, R. A. M. O'Ferrall, J. Am.
Chem. Soc. 2011, 133, 3381-3389.
S. Gronert, J. R. Keeffe, R. A. M. O’Ferrall, in
Contemporary Carbene Chemistry, Vol. 7 (Eds.: R.
A. Moss, M. P. Doyle), John Wiley & Sons, Inc,
Hoboken, NJ, 2013, pp. 3-39.
P. R. Schreiner, H. P. Reisenauer, F. C. Pickard, A.
C. Simmonett, W. D. Allen, E. Matyus, A. G.
Csaszar, Nature 2008, 453, 906-U942.
S. Amiri, H. P. Reisenauer, P. R. Schreiner, J. Am.
Chem. Soc. 2010, 132, 15902-15904.
D. Gerbig, P. R. Schreiner, in Contemporary
Carbene Chemistry, Vol. 7 (Eds.: R. A. Moss, M. P.
Doyle), John Wiley & Sons, Inc, Hoboken, NJ,
2013, pp. 193-215.
Y.-R. Luo, Comprehensive Handbook of Chemical
Bond Energies, CRC Press, Boca Raton, FL, 2007.
I. R. Dunkin, Matrix-Isolation Techniques:ꢀ A
Practical Approach., Oxford University Press, New
York, 1998.
P. L. Raston, D. T. Anderson, Phys. Chem. Chem.
Phys. 2006, 8, 3124-3129.
W. Siebrand, T. A. Wildman, Acc. Chem. Res.
1986, 19, 238-243.
[1]
[2]
W. Kirmse, Carbene Chemistry, Academic Press,
New York, London, 1971.
K. B. Eisenthal, R. A. Moss, N. J. Turro, Science
1984, 225, 1439-1445.
R. A. Moss, Acc. Chem. Res. 1989, 22, 15-21.
K. Hirai, T. Itoh, H. Tomioka, Chem. Rev. 2009,
109, 3275-3332.
A. Nemirowski, P. R. Schreiner, J. Org. Chem.
2007, 72, 9533-9540.
M. S. Platz, V. M. Maloney, in Kinetics and
Spectroscopy of Carbenes and Biradicals (Ed.: M.
S. Platz), Springer US, Boston, MA, 1990, pp. 239-
352.
M. S. Platz, in Kinetics and Spectroscopy of
Carbenes and Biradicals (Ed.: M. S. Platz), Springer
US, Boston, MA, 1990, pp. 143-211.
W. Sander, G. Bucher, S. Wierlacher, Chem. Rev.
1993, 93, 1583-1621.
[3]
[4]
[31]
[32]
[5]
[6]
[33]
[34]
[7]
[8]
[35]
[36]
[9]
[10]
chem.201803695
P. Costa, M. Fernandez-Oliva, E. Sanchez-Garcia,
W. Sander, J. Am. Chem. Soc. 2014, 136, 15625-
15630.
P. Costa, W. Sander, Angew. Chem. Int. Ed. 2014,
53, 5122-5125.
[37]
[38]
[11]
[12]
J. Knorr, P. Sokkar, S. Schott, P. Costa, W. Thiel,
W. Sander, E. Sanchez-Garcia, P. Nuernberger,
Nat. Commun. 2016, 7, 12968pp.
P. Costa, I. Trosien, M. Fernandez-Oliva, E.
Sanchez-Garcia, W. Sander, Angew. Chem. Int. Ed.
2015, 54, 2656-2660.
S. Henkel, P. Costa, L. Klute, P. Sokkar, M.
Fernandez-Oliva, W. Thiel, E. Sanchez-Garcia, W.
Sander, J. Am. Chem. Soc. 2016, 138, 1689-1697.
P. Costa, J. Mieres-Perez, N. Özkan, W. Sander,
Angew. Chem. Int. Ed. 2017, 56, 1760-1764.
P. Costa, I. Trosien, J. Mieres-Perez, W. Sander, J.
Am. Chem. Soc. 2017, 139, 13024-13030.
T. G. Savino, V. P. Senthilnathan, M. S. Platz,
Tetrahedron 1986, 42, 2167-2180.
H. Tomioka, in Reactive Intermediate Chemistry
(Eds.: R. A. Moss, M. S. Platz, M. S. Jones), John
Wiley & Sons, Inc., Hoboken, NJ, 2004, pp. 375-
462.
G. B. Schuster, in Adv. Phys. Org. Chem., Vol. 22
(Eds.: V. Gold, D. Bethell), Academic Press, 1986,
pp. 311-361.
G. D. Frey, V. Lavallo, B. Donnadieu, W. W.
Schoeller, G. Bertrand, Science 2007, 316, 439-
441.
D. Martin, M. Soleilhavoup, G. Bertrand, Chem. Sci.
2011, 2, 389-399.
S. Henkel, W. Sander, Angew. Chem. Int. Ed. 2015,
54, 4603-4607.
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[13]
[14]
C. T. Lee, W. T. Yang, R. G. Parr, Phys. Rev. B
1988, 37, 785-789.
K. Raghavachari, Theor. Chem. Acc. 2000, 103,
[15]
[16]
[17]
[18]
361-363.
S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem.
Phys. 2010, 132.
F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys.
2005, 7, 3297-3305.
T. H. Dunning, J. Chem. Phys. 1989, 90, 1007-
1023.
M. J. T. Frisch, 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.; Caricato, M.; Li, X.; Hratchian, H. P.;
Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg,
J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.;
Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, Jr., J.
A.; Peralta, J. E.; Ogliaro, 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, N. J.; 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.;
[19]
[20]
[21]
[22]
[23]
[24]
[25]
P. S. Zuev, R. S. Sheridan, J. Am. Chem. Soc.
2001, 123, 12434-12435.
W. Sander, R. Hubert, E. Kraka, J. Grafenstein, D.
Cremer, Chem. Eur. J. 2000, 6, 4567-4579.
S. Henkel, M. Ertelt, W. Sander, Chem. Eur. J.
2014, 20, 7585-7588.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.;
[47]
[48]
H. J. Werner, P. J. Knowles, G. Knizia, F. R.
Manby, M. Schutz, WIREs Comput. Mol. Sci. 2012,
2, 242-253.
Farkas, Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski,
J.; Fox, D. J. , Revision A.1 ed., Gaussian, Inc.,
Wallingford CT, 2009.
S. Stoll, A. Schweiger, J. Magn. Reson. 2006, 178,
42-55.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804657
Chemistry - A European Journal
FULL PAPER
FULL PAPER
The hydrogenation reactions of three
arylcarbenes were studied in low-
temperature matrices. Despite having
similar scaffolds, their reactivity to-
ward H2 and D2 is very contrasting.
Enrique Mendez-Vega, Mika Maehara,
Akshay Hemant Raut, Joel Mieres-Pe-
rez, Masashi Tsuge, Yuan-Pern Lee,*
and Wolfram Sander*
Page No. – Page No.
Activation of Molecular Hydrogen by
Arylcarbenes
This article is protected by copyright. All rights reserved.