A frustrated-Lewis-pair approach to catalytic reduction of alkynes to cis-alkenes

Article abstract of DOI:10.1038/nchem.1693

Frustrated Lewis pairs are compounds containing both Lewis acidic and Lewis basic moieties, where the formation of an adduct is prevented by steric hindrance. They are therefore highly reactive, and have been shown to be capable of heterolysis of molecular hydrogen, a property that has led to their use in hydrogenation reactions of polarized multiple bonds. Here, we describe a general approach to the hydrogenation of alkynes to cis-alkenes under mild conditions using the unique ansa-aminohydroborane as a catalyst. Our approach combines several reactions as the elementary steps of the catalytic cycle: hydroboration (substrate binding), heterolytic hydrogen splitting (typical frustrated-Lewis-pair reactivity) and facile intramolecular protodeborylation (product release). The mechanism is verified by experimental and computational studies.

Full text of DOI:10.1038/nchem.1693

ARTICLES
PUBLISHED ONLINE: 7 JULY 2013 | DOI: 10.1038/NCHEM.1693
A frustrated-Lewis-pair approach to catalytic
reduction of alkynes to cis-alkenes
1
2
2
1
1
´
´
´
´
¨
Konstantin Chernichenko , Adam Madarasz , Imre Papai , Martin Nieger , Markku Leskela
and Timo Repo1
*
*
Frustrated Lewis pairs are compounds containing both Lewis acidic and Lewis basic moieties, where the formation of an
adduct is prevented by steric hindrance. They are therefore highly reactive, and have been shown to be capable of heterolysis
of molecular hydrogen, a property that has led to their use in hydrogenation reactions of polarized multiple bonds. Here, we
describe a general approach to the hydrogenation of alkynes to cis-alkenes under mild conditions using the unique ansa-
aminohydroborane as a catalyst. Our approach combines several reactions as the elementary steps of the catalytic cycle:
hydroboration (substrate binding), heterolytic hydrogen splitting (typical frustrated-Lewis-pair reactivity) and facile
intramolecular protodeborylation (product release). The mechanism is verified by experimental and computational studies.
ecently, a new approach to dihydrogen activation known as the bis(pentafluorophenyl)alkylboranes as well as (C6F5)2BH itself 32
frustrated Lewis pairs (FLPs) concept has been introduced1–3
together with the properly chosen Lewis bases, can split hydrogen
A combination of highly Lewis-acidic boranes and sterically heterolytically to give the respective onium borohydrides33.
,
.
R
hindered bases can split hydrogen heterolytically to generate However, upon heating of these compounds, only hydrogen
onium (for example, phosphonium, ammonium) borohydrides. release was observed, demonstrating the reversibility of H2 uptake
These compounds demonstrate reduction activities resembling by these FLPs. Our numerous attempts to realize the approach
those of inorganic borohydrides such as NaBH4; that is, they are depicted in Fig. 1b using (C6F5)2BH together with different bases
suitable predominantly for the reduction of polarized multiple and additives were unsuccessful (Supplementary Section S47).
bonds. Imines, enamines, silyl ethers4–6, a,b-enones7, ynones8 and
Recently, we have reported 2-[bis(pentafluorophenyl)boryl]-
N-alkylanilines9 have been hydrogenated using stoichiometric or N,N-dialkylanilines, exemplifying a new class of bridged frustrated
catalytic amounts of FLPs. Owing to the heterolytic nature of B/N Lewis pairs34. Interestingly, compound 6 exists as an intra-
FLP–H2 adducts, the hydrogenation of unactivated multiple C–C molecular Lewis adduct, containing a strained four-membered
bonds using FLPs has some natural limitations, because during C–N ꢀ B–C cycle. Owing to strain, the B–N bond in 6 is relatively
the relevant step of the catalytic cycle a proton transfer from catalyst weak, because at room temperature 6 reversibly reacts with hydro-
to substrate should take place (Fig. 1a). Although Greb et al. have gen to give ammonium borohydride 7 (Fig. 2).
implemented this approach in the hydrogenation of alkenes under
ambient conditions, this method is predictably restricted to Results and discussion
alkenes with high proton affinity10.
New ansa-aminohydroborane as a catalyst. In this work we
Combining the FLP approach and previous knowledge about report that, upon heating of aminoborane 6 at 80 8C under 2 bar
1
borane-catalysed hydrogenation of alkenes11–14 and polyarenes15–18
,
H2, new signals (different from those of 6 or 7) appear in the H,
we propose herein a new general catalytic pathway to the hydrogen- 19F and 10B NMR spectra, together with the formation of C6F5H
ation of unsaturated hydrocarbons (Fig. 1b) and demonstrate its (9). The new species was isolated as a greenish oil and identified
1
validity by the highly selective hydrogenation of alkynes into as hydroborane 8 (Fig. 2), producing in the H NMR spectrum
cis-alkenes. Stereoselective hydrogenation of alkynes is an important
a
characteristic partially relaxed quadruplet (d ¼ 4.35 ppm,
protocol in the synthesis of natural and industrially relevant com- J ¼ 105 Hz) attributed to BH signal. This reactivity is
pounds19–23. Heterogeneous as well as homogeneous metal catalysts unprecedented, because neither inter- nor intramolecular FLPs
for this stereoselective hydrogenation are known24–28; however, have been reported to undergo B–C6F5 hydrogenolysis as a result
metal-free catalytic hydrogenation of unactivated alkynes into of hydrogen activation35–37
.
alkenes has not been reported previously.
Because 8 is a potentially hydroborating BH species and can be
In contrast to classical FLP-catalysed reactions, the substrate is produced in situ from 6, we attempted to use 8 as a catalyst in the
bound to catalyst 1 by hydroboration before hydrogen activation hydrogenation of unactivated alkenes and alkynes following the
(Fig. 1b). The resulting borane 3, together with a Lewis base co-catalyst, strategy depicted in Fig. 1b. Hex-1-ene (12b), hex-1-yne (12a)
can activate hydrogen, producing adduct 4. In this onium borohydride and hex-3-yne (11a) were heated separately together with
4, a proton transfer can occur, liberating the initial borane 1, the Lewis 10 mol% of precatalyst 6 in C6D6 under 2 bar H2 at 80 8C. After
base and hydrogenated substrate 5. To the best of our knowledge, this 15 h, no products of hex-1-ene and hex-1-yne hydrogenation were
approach has not been studied experimentally or theoretically.
detected by NMR. In the case of 11a, no evidence of starting
Initially, we attempted to use Piers’ borane, (C6F5)2BH (ref. 29), alkyne was found; instead, a complex mixture of alkenes was
as a catalyst. (C6F5)2BH smoothly hydroborates different alkenes observed, mostly comprising cis-hex-3-ene 11b. Minor amounts
and alkynes30,31. Moreover, it has been shown that the resulting of trans-hex-3-ene and other hexenes were attributed to
1Department of Chemistry, Laboratory of Inorganic Chemistry, University of Helsinki PO Box 55, FIN-00014, Finland, 2Institute of Organic Chemistry, Research
Centre for Natural Sciences, Hungarian Academy of Sciences, PO Box 17, H-1525 Budapest, Hungary. e-mail: papai.imre@ttk.mta.hu; timo.repo@helsinki.fi
*
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
1
© 2013 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1693
ARTICLES
R1
b
R6
R4
R3
R5
R3
a
R5
R1
R2
R6
R5
R3
R6
R4
B
BH
R3
R4
H
H
H2 (H-H)
H
R2
H
5
1
2
Hydrogen
activation
Hydroboration
(substrate binding)
Hydride addition
(FLP reactivity)
Base
R3
R5
R1
R2
Protonative
elimination
R4
Base
R6
R5
R3
B
R5
R6
R4
R1
H
R3
H
B–
H
+
B–
R2
BaseH
+
R1
H
R1 B– R3
H
+
BaseH
3
R6
R4
R2
R2
R3
H
4
Substrate
protonation
Hydrogen
activation
(FLP reactivity)
R5
H2 (H-H)
R6
R3
R4
Figure 1 | FLP-catalysed hydrogenation of multiple C–C bonds. a, Traditional approach including heterolytic hydrogen splitting by the catalyst followed by
protonation of the substrate as the key steps. b, In the novel approach, substrate activation via hydroboration precedes hydrogen activation. Reduced
substrate is then released via protonation of the formed H2 adduct.
Formation of the active catalyst species
2 bar H2, 80 °C,
C6D5Br, 6 h, >90%,
or
N
2 bar H2, 80 °C,
toluene, 7 h, 80%
B
C6F5
6
Catalytic cycle and
catalyst degradation pathway
C6F5
R1
R2
Ar, 10 mol%/day
RT C6D6
–H2
2 bar H2, 12 h, 100%
RT C6D6
Quenching of alkenylboranes with D2
N
N
D
H
+
D2
H
N
–
+
H
F
F
B
N
R2
R1
R2
B
80 °C,
D
C6F5
C6F5
21c (R1, R2 = Ph)
26d:26e = 79:21
B
R1
F
–C6F5H (9)
B
F5C6
H
8-d
27
H
C6F5
F
F
27d:27e = 79:21
7
8
27, 28:
a (R1, R2 = Me)
b (R1, R2 = Et)
c (R1, R2 = Ph)
26, 27, 28:
d (R1 = CD3, R2 = 4-MePh)
e (R1= 4-MePh, R2 = CD3)
H2 (H–H)
N
H
+
H
H
k1
H
R2
–
R2
B
F
F
R1
H
F
R1
F
k2
F
28
N
B
H
C6F5H
+
R2
R1
H
29
Figure 2 | Mechanism of catalytic hydrogenation of alkynes into cis-alkenes. Formation of the active catalyst species 8 proceeds through H2 addition to the
precatalyst 6, followed by intramolecular protonative cleavage of the C6F5 ring. Produced catalyst 8 hydroborates alkyne (substrate binding), producing
vinylborane 27. The latter activates hydrogen by the FLP mechanism. Intramolecular protonation of vinyl carbon in 28 causes cycle propagation, while C6F5
group cleavage leads to active catalyst degradation. Reaction of hydroboration intermediates 27 with D2 results in selective formation of monodeuterated
cis-alkenes 21c, 26d–e and catalyst 8-d. Deuteration occurs selectively to the B–C carbon, giving solid support to the proposed mechanism.
isomerization of the initially produced cis-hex-3-ene via a hydro- 2 bar H2, 80 8C, 3 h (Table 1)—demonstrating the generality of
boration/retrohydroboration sequence, catalysed either by 8 or the approach and providing exceptional cis-stereoselectivity. Silyl-
other hydroborane species. When hydrogenation of 11a was protected esters, enynes, silyl-protected ynols and diynes (Table 1,
repeated for 3 h with 5 mol% 6, cis-hex-3-ene 11b was produced entries 9, 10, 19, 12, 13) were also successfully hydrogenated.
almost exclusively (according to NMR).
The products were isolated in excellent yields in experiments
Various dialkyl-, diaryl-, arylalkylacetylenes were successfully scaled up to 10 mmol of substrate. Some of the substrates required
hydrogenated under standard conditions—5 mol% 6 in C6D6, prolonged reaction times and/or higher temperatures and catalyst
2
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2013 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1693
ARTICLES
Table 1 | Catalytic hydrogenation of alkynes using 6 as a precatalyst.
6 (5 mol%)
2.2 bar H2
R2
H
C6D6, 80 °C, 3 h
H
R2
R1
R1
Entry Substrate
Product
Conversion:
isolated yield (%)
Entry Substrate
Product(s)
Conversion:
product (%)
Ph
TMS
10a
1*
100
16
22b
22c
88 (22b)
12 (22c)
10b
Ph
Ph
TMS
22a
2
100
n.r.‡
17}
18
22a
30.5 (22b)
4.3 (22c)
11a
11b
TMS
3†
–
,20 (23b)
12a
23b
23a
4
–
–
n.r.
19**
23a
71 (23b)
9.5 (23c)
11a
+
23c
12b
1:1
Bu
20††
44 (24b)
26 (other alkenes)
Bu
5
11a þ 12a
1:1
n.r.
2
Bu
24b
24c
24a
Bu
13a
6*
7§
100
100
13b
14b
Bu
Bu
Bu
24d
Bu
21‡‡
24a
42 (24b)
10 (24c)
10.5 (24d)
10.2 (24e)
14a
15a
Ph
Ph
Bu
Ph
Ph
Bu
24e
Bu
8
9
100 (80)
100 (98)
15b
OTBS
TBSO
O
22
–
n.r.
16a
O
7
7
TMSO
25a
16b
7
7
p-Tol
CD3
H
H
Bu
10
100
23
100
100
Bu
17a
26a
26b
17b
p-Tol
CD3
D
D
24§§
26a
Cl
Pr
11
100 (95)
100
TES
26c
18a
TES
18b
Cl
Pr
p-Tol
CD3
OTMS
OTMS
12
19a
19b
26b
+
26c +
13
100 (94)
25ꢁꢁ
26a
14.4 (26b)
15.9 (26c)
18.7 (26d)
11 (26e)
4
H
D
20a
21a
20b
4
26d
p-Tol
CD3
H
+
D
14
50
Ph
Ph
21b
Ph
Ph
26e
CD3
p-Tol
15ꢁ
21a
21b
100 (91)
*7 mol% of 6; †Reaction time, 15 h; ‡No reaction; §10 mol% of 6; ꢁReaction time, 9 h; }Reaction time, 1.5 h; **10 mol% of 6, 120 8C, 15 h; ††20 mol% of 6, 18 h; ‡‡15 mol% of 6, 120 8C, 10 h; §§D2 was used instead
of H2, 10 mol% of 6, 5 h; ꢁꢁHD gas was used instead of H2. Low conversion (60%) is due to insufficient pressure of available HD (1.2 bar). Isotopes scrambling in ꢂ1:1:1:1 ratio. Bu, butyl; Ph, phenyl; Pr, propyl; p-Tol,
4-methylphenyl; TBS, tert-butyldimethylsilyl; TES, triethylsilyl; TMS, trimethylsilyl.
loading, and some were not hydrogenated at all. There are essen- conventional methods and the obtained silylacetylenes were
tially two substrate classes that are unreactive using the current smoothly hydrogenated (Table 1, entries 11, 16). Catalytic activity
method: terminal alkynes and alkynes comprising a terminal up to 31.6 h21 was estimated using 6 or 8 as the catalyst under stan-
double bond. Nevertheless, terminal alkynes can be silylated using dard conditions and 11a or 15a as substrates. Remarkably, the
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2013 Macmillan Publishers Limited. All rights reserved.
3

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1693
ARTICLES
Density functional theory (DFT) calculations (Supplementary
Section S49) carried out for the reaction of catalyst 8 with
but-2-yne (10a) predict a relatively small activation barrier
(16.2 kcal mol21) for the hydroboration process, and point to the
high exergonicity of this step (Fig. 3). These results suggest that
alkyne hydroboration is irreversible, so compound 8 can be recovered
only upon completion of the catalytic cycle. This irreversibility has
also been demonstrated experimentally (Supplementary Section S32).
The zwitterionic reaction intermediates 28b–e formed in the
hydrogen activation step were not detectable, which can be ration-
alized in light of the computed energetics. Although the formation
of 28a takes place via a low barrier (TSsplit lies only15.8 kcal mol21
above 27a þ H2 in free energy), this step is found to be slightly
endergonic. Notably, heterolytic H2 splitting with B/N FLPs contain-
ing only one electron-withdrawing C6F5 group on the Lewis acceptor
site is unprecedented due to the reduced acidity of the resulting
borane. However, as pointed out previously, the ortho-phenylene
linker between the B/N centres provides significant electrostatic
stabilization in the zwitterionic species formed upon H2 cleavage39.
The calculations predict facile intramolecular protonation of the
vinyl substituent in 28a, which proceeds as direct protodeboryla-
tion, leading to the elimination of cis-alkene 10b and regeneration
of the catalyst. The barrier to product elimination is rather low
(14.8 kcal mol21 relative to 28a) and is thermodynamically
favoured. The mutual position of the B/N sites of the catalyst
core plays a crucial role in this process because it determines the
feasibility of the proton shift. In this particular case, a low-lying
isomer of 28a could be identified computationally with the NH
TShydr (16.2)
20.0
10.0
0.0
TSsplit (–7.3)
TSprot (–5.6)
8 + 10a + H2 (0.0)
–10.0
–20.0
–30.0
28a (–20.4)
27a + H2 (–23.1)
8 + 10b (–31.0)
Figure 3 | Solution-phase Gibbs free energy diagram computed for the
hydrogenation of but-2-yne (10a). Optimized structures of transition states
identified for hydroboration (TShydr), heterolytic hydrogen splitting (TSsplit
)
and protonation (TSprot) steps are shown in the upper part of the figure. The
energetics of the elementary steps identified computationally is consistent
with the proposed reaction mechanism. A detailed description of the
structure and energetics of species involved in the catalytic cycle is given in
Supplementary Section S49.
catalytic hydrogenation proceeds at room temperature, although 20
times slower than at 80 8C. Meanwhile, high-pressure H2 (30 bar) bond oriented towards the alpha carbon atom of the vinyl group
causes almost tenfold acceleration of hydrogenation up to 296 h21
(Supplementary Section S49).
Protodeborylation of 28, resulting in B–C6F5 cleavage and degra-
(Supplementary Section S7).
dation of the active catalyst into inactive vinylborane 29 and C6F5H,
No over-reduction to alkanes was detected. Under standard
conditions cis-alkenes were produced exclusively, with traces of is an alternative reaction pathway in this phase of the catalytic cycle
1
(Fig. 2), which has also been explored computationally. The calcu-
lated barrier of aryl elimination is notably higher than that of
alkene formation (19.1 kcal mol21 relative to 28a), but this
process is still feasible at the applied reaction conditions. The proto-
other products such as trans-alkenes barely detected by H NMR.
The only exception was 1-trimethylsilyl-2-phenylacetylene 22a,
where a substantial amount of trans-alkene 22c was produced
(12 mol%), independently of the conversion level (Table 1, entries
16, 17). Product 22c is likely to be produced directly during hydro- deborylation of 28a is analogous to that taking place in the gener-
ation of 8 from precatalyst 6, for which computations predict an
genation. Accumulation of trans-alkenes as a result of isomerization
was observed when prolonged heating and/or high temperature
(120 8C) was applied to force hydrogenation of poorly reactive sub-
strates (Table 1, entries 19, 21).
N
B
Mechanistic insight into the catalytic cycle. The reaction
mechanism for the catalytic hydrogenation of alkynes was
investigated in a combined experimental/theoretical study. The
basic steps of the envisioned catalytic cycle are depicted in Fig. 2,
and are consistent with the new concept outlined in Fig. 1b.
The initial step of the catalytic cycle, that is, the hydroboration of
alkynes with 8, was verified by the isolation of respective intermedi-
ates 27b–e. Additionally, relative rates of hydroboration of different
alkynes and alkenes by 8 were measured in competitive experi-
ments. The relative rate is descending in the order hex-3-yne :
hex-1-yne : but-2-yne : hex-1-ene : cis-hex-3-ene : prop-1-yn-1-
ylbenzene (a þ b-positions) : prop-1-yn-1-ylbenzene (b-position) :
prop-1-yn-1-ylbenzene (a-position) : diphenylacetylene : cis-but-1-
en-1-yl benzene ¼ 136 : 109 : 57 : 44 : 5.5 : 5 : 4 : 1 : no reaction (25 8C) :
no reaction (80 8C) (Supplementary Section S31). These rates are in
R2
12b:
rapid, irreversible
10b, 11b:
slow, reversible
15b, 18b, 21b:
no reaction
C6D6
C6D5Br /
PhCH3
/
C6F5
R1
RT or 80 °C,
C6D6
2 bar H2,
80 °C,
3–5h
>80%
–C6F5H
30a, R1 = H, R2 = Bu
30b, R1, R2 = Me
30c, R1, R2 = Et
12b
Bu
12b, R1 = H, R2 = Bu
R1
10b, R1, R2 = Me
R2
or
6
+
11b, R1, R2 = Et
(1–5 equiv.)
15b, R1 = Ph, R2 = Et
18b, R1 = Et3Si, R2 = (CH2)3Cl
21b, R1, R2 = Ph
R1
R2
+ 8
10a, R1, R2 = Me
11a, R1, R2 = Et
Hydroboration of alkenes by
the active catalyst
Alkylboranes from precatalyst 6
under hydrogenation condition
agreement with common rules of hydroboration38. Hydroboration of Figure 4 | Reaction of 6 and 8 with alkenes. Alkenes hydroboration
10a appears instantly at room temperature. Replacement of a methyl outcome depends on the level of steric crowding of the double bond:
group with a phenyl substituent leads to 14-fold retardation of rapid and irreversible for terminal alkenes (12b); slow and reversible for
hydroboration to the sterically less hindered site of prop-1-yn-1- cis-di(n-alkyl)ethenes (10b, 11b), no reaction for 15b, 18b, 21b. Alternatively,
ylbenzene and 57-fold to the more hindered site. Eventually, diphenyl- alkylboranes 30 are produced from alkenes or alkynes and 6 under
acetylene 21a remains intact with 8 at room temperature and hydrogenation conditions. The one-pot reaction includes formation of 8
requires heating at 80 8C, apparently making the hydroboration the in situ, followed by hydroboration of the alkene loaded as is or produced
rate-limiting step in the overall slow hydrogenation of this substrate. in situ from the respective alkyne.
4
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2013 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1693
ARTICLES
4 equiv.
Bu
H
different exergonicities for these substrates (211.9 and
220.5 kcal mol21 for cis-but-2-ene and ethylene, respectively),
which is in qualitative agreement with the observed reactivity. The
subsequent heterolytic hydrogen cleavage is a kinetically and ther-
modynamically allowed elementary step; however, the intramolecu-
lar protonation of the alkyl substituent in alkylborohydride
intermediate is hindered by a large activation barrier. Actually,
this step represents the only limiting factor towards the hydrogen-
ation of alkenes using the present approach, and it is associated
with the lack of a p-system in the zwitterionic intermediate
formed in the H2 activation step.
1H, 11B NMR:
Three new
species
C6D6, H2,
3 h, 80 °C
N
Bu
H
NH
+
C6D6 or CD2Cl2
1 h at RT
or 5 min at 80 °C
100%
–
B
B
C6F5
C6F5
19F NMR:
C6F5H only
C6F5
F5C6
31
6
Bu
Figure 5 | Reaction of aminoborane 6 with hex-1-yne and H2. Initially,
adduct 31 is formed upon addition of hex-1-yne to precatalyst 6. Upon
heating 6 or 31 with excess hex-1-yne under H2 pressure, three new boron-
containing species are formed together with full cleavage of C6F5 groups.
Without C6F5 groups the catalyst core cannot activate H2.
The limitation of the current approach to non-terminal alkynes
also required additional studies. It is known from previous publi-
cations that FLPs react with terminal alkynes via deprotonative bory-
even higher barrier (23.0 kcal mol21; Supplementary Section S48). lation pathway, producing the respective onium alkynylborates40–45
.
Assuming the C6F5H elimination to be the only catalyst degradation Indeed, 6 reacts with hex-1-yne giving the respective adduct 31
pathway, the ratio of the reaction rates of these two intramolecular (Fig. 5). Upon heating 6 with excess hex-1-yne at 80 8C under 2 bar
protonation pathways corresponds to the maximum turnover H2, 31 remains the major product after 1 h. However, after 3 h
number, which was found to be 91 for hydrogenation of hex-3-yne three new aminoborane species were formed in the ratio 3:3:2, each
(Supplementary Section S9). Eventually, the exceptional cis-selectivity containing the hex-1-ynyl group, as evident from 1H and 11B NMR.
observed in the hydrogenation reactions results from exclusive syn- The 19F NMR spectrum revealed complete cleavage of the C6F5-
hydroboration and the configuration is retained during subsequent group into C6F5H. Evidently, the inability to hydrogenate terminal
elementary steps, particularly protodeborylation38.
alkynes is a result of catalyst degradation into species inert to hydrogen
Additional experimental support for the proposed mechanism due to complete elimination of the perfluorophenyl groups.
was collected in isotope-labelling experiments. 1-Methyl-4-prop-
1-ynylbenzene-d3 26a, a model substrate, was hydrogenated with Conclusion
HD, resulting in all four possible isotopomeric cis-styrenes 26b–e In conclusion, we have developed a new strategy for the catalytic
in nearly equal ratio (Table 1, entry 25). Isotope scrambling is in metal-free hydrogenation of unactivated multiple C–C bonds
full accordance with the proposed mechanism, because the BH(D) using frustrated Lewis pairs. Using catalyst 8, formed in situ from
hydrogen in 8 originates from a preceding catalytic cycle (Fig. 2). aminoborane 6, the approach was implemented as highly chemo-
In addition, when 26a was treated with an equimolar amount of and stereoselective hydrogenation of internal alkynes into the
8, products of hydroboration were isolated as a 79:21 mixture of respective cis-alkenes under mild conditions (2 bar H2, 80 8C).
regioisomers 27d and 27e. Upon treatment with D2, a mixture of The catalytic pathway includes three steps: hydroboration of
26d and 26e in the ratio 79:21 was produced, together with an equi- alkyne (substrate binding), heterolytic H2 cleavage with formed
molar amount of deuterated 8-d (Fig. 2). Thus, deuteration of vinylborane, followed by intramolecular protodeborylation of
alkynes appears exclusively at the carbon adjacent to the boron vinyl substituent, recovering 8 and releasing cis-alkene. High cis-
atom in the alkenyl borane intermediate 27. A similar experiment selectivity is a result of exclusive syn-hydroboration, and is retained
performed with diphenylacetylene led to exclusive formation of during subsequent steps, particularly the intramolecular protona-
monodeuterated cis-stilbene-d 21c (Fig. 2).
tion. Mutual ansa-B/N geometry plays a key role in all elementary
steps, especially during protodeborylation, which proceeds in a
single step, rather than including carbocation intermediates. The
mechanism was supported by isolation of some intermediates,
including the active catalyst species 8, isotope-labelling studies
and quantum-chemical calculations. Substrate restrictions associ-
ated with the presence of the terminal double and triple bonds
were studied and rationalized. The computational analysis provides
solid support for the proposed mechanism as all elementary steps
could be identified and the obtained energetics is in full accordance
with experimental findings. In principle, alkenes could be hydrogen-
ated using the current approach; however, at the final stage of the
catalytic cycle the C6F5 group is cleaved more easily than an alkyl
group, causing catalyst degradation rather than alkane release.
Some early examples of the catalytic hydrogenation of unactivated
alkynes were reported by us as part of a review46 (and are included
again as entries 2, 6–8 and 11–15 in Table 1). Compared to other
ansa-aminoboranes discussed there, 8 is unique enabling the cataly-
tic hydrogenation of unactivated alkynes.
Alkenes under hydrogenation conditions. In an attempt to apply
the new synthetic approach (Fig. 1b) to alkene hydrogenation, the
reactivities of various terminal and cis-disubstituted alkenes with
6, 8 and H2 were examined; however, no alkane products were
detected (Supplementary Section S8). In stoichiometric reactions,
terminal alkene 12b is hydroborated rapidly and irreversibly by
catalyst 8, cis-di(n-alkyl)ethenes (10b, 11b) react at a slower rate
and reversibly, whereas 8 remains intact with more sterically
hindered cis-alkenes (Fig. 4). Alkylboranes 30a–c can be produced
directly via hydrogenolysis of 6 in the presence of 12b or alkynes
10a, 11a (Fig. 4) and found to be particularly stable to further
hydrogenolysis: 30c is the major boron-containing component of
the reaction mixture after 15 h at 2 bar H2. As a result, alkynes
that contain a terminal double bond cause partial or complete
deactivation of the catalyst via formation of alk-1-ylboranes,
which are unable to propagate the catalytic cycle (in Table 1
compare entry 2 and 4, 10 and 18, 12 and 22; Supplementary
Sections S24,25). Although at the end of 11a hydrogenation the
catalyst is present as alkylborane 30c, the latter can easily
dissociate to give active catalyst species 8, pointing again to the
reversibility of hydroboration in the reactions with cis-di(n-
alkyl)ethenes (Supplementary Section S33).
Methods
Standard protocol. A quantity of 0.2–0.5 mmol of an alkyne were placed into a
25 ml Schlenk tube, followed by 5 mol% of 6 and 0.7 ml of C6D6. The tube was filled
with 2–2.2 bar of hydrogen by two freeze–pump–thaw cycles and vigorously stirred
at 80 8C for 3 h, or longer if needed. The reaction mixture was transferred into an
NMR tube and analysed.
These results can be interpreted readily in terms of the free
energy profiles computed for the hydrogenation of cis-but-2-ene
and ethylene as model substrates for internal and terminal alkenes
Computational details. DFT, with the dispersion-corrected, range-separated hybrid
vB97X-D exchange-correlation functional47, was used to examine possible reaction
(Supplementary Sections S50,51). Calculations predict rather pathways relevant to the title reaction. For geometry optimizations, vibrational
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
5
© 2013 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.1693
ARTICLES
analysis and the estimation of solvent effects, the 6-311G(d,p) polarized triple-z
basis set was used, and additional single-point energy calculations were carried out
for each located stationary point with the larger 6-311þþG(3df,3pd) basis set. The
energy values reported in the paper correspond to solution-phase Gibbs free
energies. Additional computational details are provided in the Supplementary
Information (Computational Protocol).
27. Van Laren, M. W. & Elsevier, C. J. Selective homogeneous palladium(0)-
catalyzed hydrogenation of alkynes to (Z)-alkenes. Angew. Chem. Int. Ed. 38,
3715–3717 (1999).
28. Radkowski, K., Sundararaju, B. & Fu¨rstner, A. A functional-group-tolerant
catalytic trans hydrogenation of alkynes. Angew. Chem. Int. Ed. 52,
355–360 (2013).
29. Parks, D. J., von H. Spence, R. E. & Piers, W. E. Bis(pentafluorophenyl)borane:
synthesis, properties, and hydroboration chemistry of a highly electrophilic
borane reagent. Angew. Chem. Int. Ed. 34, 809–811 (1995).
30. Parks, D. J., Piers, W. E. & Yap, G. P. A. Synthesis, properties, and hydroboration
activity of the highly electrophilic borane bis(pentafluorophenyl)borane,
HB(C6F5)2. Organometallics 17, 5492–5503 (1998).
31. Parks, D. & Piers, W. Hydroboration of vinyl silanes with bis-(pentafluoro-
phenyl)borane: ground state b-silicon effects. Tetrahedron 54,
15469–15488 (1998).
32. Jiang, C., Blacque, O. & Berke, H. Metal-free hydrogen activation by the
frustrated Lewis pairs of ClB(C6F5)2 and HB(C6F5)2 and bulky Lewis bases.
Organometallics 28, 5233–5239 (2009).
33. Jiang, C., Blacque, O., Fox, T. & Berke, H. Reversible, metal-free hydrogen
activation by frustrated Lewis pairs. Dalton Trans. 40, 1091–1097 (2011).
34. Chernichenko, K., Nieger, M., Leskela¨, M. & Repo, T. Hydrogen activation by
2-boryl-N,N-dialkylanilines: a revision of Piers’ ansa-aminoborane. Dalton
Trans. 41, 9029–9032 (2012).
Received 22 February 2013; accepted 13 May 2013;
published online 7 July 2013
References
1. Welch, G. C., San Juan, R. R., Masuda, J. D. & Stephan, D. W. Reversible, metal-
free hydrogen activation. Science 314, 1124–1126 (2006).
2. Stephan, D. W. ‘Frustrated Lewis pairs’: a concept for new reactivity and
catalysis. Org. Biomol. Chem. 6, 1535–1539 (2008).
3. Stephan, D. W. & Erker, G. Frustrated Lewis pairs: metal-free hydrogen
activation and more. Angew. Chem. Int. Ed. 49, 46–76 (2010).
4. Stephan, D. W. et al. Metal-free catalytic hydrogenation of polar substrates by
frustrated Lewis pairs. Inorg. Chem. 50, 12338–12348 (2011).
5. Stephan, D. W. ‘Frustrated Lewis pair’ hydrogenations. Org. Biomol. Chem. 10,
5740–5746 (2012).
6. Sumerin, V. et al. Highly active metal-free catalysts for hydrogenation of
unsaturated nitrogen-containing compounds. Adv. Synth. Catal. 353,
2093–2110 (2011).
7. Eros, G. et al. Expanding the scope of metal-free catalytic hydrogenation through
frustrated Lewis pair design. Angew. Chem. Int. Ed. 49, 6559–6563 (2010).
8. Xu, B-H. et al. Reaction of frustrated Lewis pairs with conjugated ynones-
selective hydrogenation of the carbon–carbon triple bond. Angew. Chem. Int. Ed.
50, 7183–7186 (2011).
9. Mahdi, T., Heiden, Z. M., Grimme, S. & Stephan, D. W. Metal-free aromatic
hydrogenation: aniline to cyclohexyl-amine derivatives. J. Am. Chem. Soc.
134, 4088–4091 (2012).
35. Chase, P. A. & Stephan, D. W. Hydrogen and amine activation by a frustrated
Lewis pair of a bulky N-heterocyclic carbene and B(C6F5)3. Angew. Chem. Int.
Ed. 47, 7433–7437 (2008).
36. Robertson, A. P. M. et al. Experimental and theoretical studies of the potential
.
interconversion of the amine-borane iPr2NH BH(C6F5)2 and the aminoborane
iPr2N¼B(C6F5)2 involving hydrogen loss and uptake. Eur. J. Inorg. Chem. 2011,
5279–5287 (2011).
37. Erdmann, M. et al. Functional group chemistry at intramolecular frustrated
Lewis pairs: substituent exchange at the Lewis acid site with 9-BBN. Dalton
Trans. 42, 709–718, (2013).
10. Greb, L. et al. Metal-free catalytic olefin hydrogenation: low-temperature H2
activation by frustrated Lewis pairs. Angew. Chem. In. Ed. 51,
10164–10168 (2012).
11. Ko¨ster, R. Neue pra¨parative Mo¨glichkeiten in der Bor- und Silicium-Chemie.
Angew. Chem. 68, 383 (1956).
38. Brown, H. C. Hydroboration (W. A. Benjamin, 1962).
´
39. Rokob, T. A., Hamza, A. & Papai, I. Rationalizing the reactivity of frustrated
Lewis pairs: thermodynamics of H2 activation and the role of acid–base
properties. J. Am. Chem. Soc. 131, 10701–10710 (2009).
40. Dureen, M. A., Brown, C. C. & Stephan, D. W. Deprotonation and addition
reactions of frustrated Lewis pairs with alkynes. Organometallics 29,
6594–6607 (2010).
41. Dureen, M. A. & Stephan, D. W. Terminal alkyne activation by frustrated and
classical Lewis acid/phosphine pairs. J. Am. Chem. Soc. 131, 8396–8397 (2009).
42. Jiang, C., Blacque, O. & Berke, H. Activation of terminal alkynes by frustrated
Lewis pairs. Organometallics 29, 125–133 (2010).
43. Moemming, C. M. et al. Formation of cyclic allenes and cumulenes by
cooperative addition of frustrated Lewis pairs to conjugated enynes and diynes.
Angew. Chem. Int. Ed. 49, 2414–2417 (2010).
44. Voss, T. et al. Frustrated Lewis pair behavior of intermolecular amine/B(C6F5)3
pairs. Organometallics 31, 2367–2378 (2012).
45. Winkelhaus, D., Neumann, B., Stammler, H-G. & Mitzel, N. W. Intramolecular
Lewis acid–base pairs based on 4-ethynyl-2,6-lutidine. Dalton Trans. 41,
9143–9150 (2012).
12. Ko¨ster, R., Bruno, G. & Binger, P. Borverbindungen, V Hydrierung von
Bortrialkylen und -triarylen. Justus Liebigs Ann. Chem. 644, 1–22 (1961).
13. DeWitt, E. J., Ramp, F. L. & Trapasso, L. E. Homogeneous hydrogenation
catalyzed by boranes. J. Am. Chem. Soc. 83, 4672 (1961).
14. Ramp, F. L., DeWitt, E. J. & Trapasso, L. E. Homogeneous hydrogenation
catalyzed by boranes. J. Org. Chem. 27, 4368–4372 (1962).
15. Yalpani, M. & Ko¨ster, R. Partial hydrogenation: from anthracene to coronene.
Chem. Ber. 123, 719–724 (1990).
16. Ko¨ster, R., Schu¨ßler, W. & Yalpani, M. Reduktion kondensierter Arene mitBH-
Boranen, I Reaktionen von Naphthalin, Anthracen und Phenanthren mit
Tetraalkyldiboranen (6). Chem. Ber. 122, 677–686 (1989).
17. Yalpani, M., Lunow, T. & Ko¨ster, R. Reduction of polycyclic arenes by-boranes,
II. Borane catalyzed hydrogenation of naphthalenes to tetralins. Chem. Ber. 122,
687–693 (1989).
´
18. Haenel, M. W., Narangerel, J., Richter, U-B. & Rufinska, A. The first liquefaction
46. Sumerin, V. et al. Amine-borane mediated metal-free hydrogen activation and
catalytic hydrogenation. Top. Curr. Chem. 332, 111–155 (2013).
47. Chai, J-D. & Head-Gordon, M. Long-range corrected hybrid density functionals
with damped atom–atom dispersion corrections. Phys. Chem. Chem. Phys. 10,
6615–6620 (2008).
of high-rank bituminous coals by preceding hydrogenation with homogeneous
borane or iodine catalysts. Angew. Chem. 45, 1061–1066 (2006).
19. Siau, W-Y., Zhang, Y. & Zhao, Y. Stereoselective synthesis of Z-alkenes.
Top. Curr. Chem. 327, 33–58 (2012).
20. Jain, S. C. et al. Polyene pheromone components from an arctiid moth (Utetheisa
ornatrix): characterization and synthesis. J. Org. Chem. 48, 2266–2270 (1983).
21. Fu¨rstner, A., Guth, O., Rumbo, A. & Seidel, G. Ring closing alkyne metathesis.
Comparative investigation of two different catalyst systems and application to
the stereoselective synthesis of olfactory lactones, azamacrolides, and the
macrocyclic perimeter of the marine alkaloid nakadomarin A. J. Am. Chem. Soc.
121, 11108–11113 (1999).
22. Ghosh, A. K., Wang, Y. & Kim, J. T. Total synthesis of microtubule-stabilizing
agent (2)-laulimalide. J. Org. Chem. 66, 8973–8982 (2001).
23. Fu¨rstner, A. & Davies, P. W. Alkyne metathesis. Chem. Commun.
2307–2320 (2005).
24. Caggiano, T. J., Siegel, S., King, A. O. & Shinkai, I. Encyclopedia of Reagents
for Organic Synthesis Vol. 6 (ed. Paquette, L. A.) 3694–3869; 3861–3865;
3966–3867 (Wiley, 1995).
Acknowledgements
The authors acknowledge financial support from the Academy of Finland (139550) and the
Hungarian Research Foundation (OTKA, grant K-81927) and COST action CM0905
(Organocatalysis). The authors also thank A. Reznichenko for discussions and corrections
during the preparation of the manuscript, M. Lindqvist for corrections and S. Heikkinen for
help with NMR measurements.
Author contributions
K.C. and T.R. conceived and K.C. carried out the experiments. A.M. and I.P. designed and
performed the DFT studies. All authors discussed and co-wrote the paper.
Additional information
Supplementary information and chemical compound information are available in the
online version of the paper. Reprints and permissions information is available online at
www.nature.com/reprints. Correspondence and requests for materials should be addressed
to T.R. and those related to computational studies to I.P.
25. Schrock, R. R. & Osborn, J. A. Catalytic hydrogenation using cationic rhodium
complexes. II. The selective hydrogenation of alkynes to cis olefins. J. Am. Chem.
Soc. 98, 2143–2147 (1976).
26. Sodeoka, M. & Shibasaki, M. New functions of (arene)tricarbonylchromium(0)
complexes as hydrogenation catalysts: stereospecific semihydrogenation of
alkynes and highly chemoselective hydrogenation of ab-unsaturated carbonyl
compounds. J. Org. Chem. 50, 1147–1149 (1985).
Competing financial interests
The authors declare no competing financial interests.
6
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2013 Macmillan Publishers Limited. All rights reserved.