Selective catalytic C-H alkylation of alkenes with alcohols

Article abstract of DOI:10.1126/science.1208839

Alkenes and alcohols are among the most abundant and commonly used organic feedstock in industrial processes. We report a selective catalytic alkylation reaction of alkenes with alcohols that forms a carbon-carbon bond between vinyl carbon-hydrogen (C-H) and carbon-hydroxy centers with the concomitant loss of water. The cationic ruthenium complex [(C₆H₆)(PCy ₃)(CO)RuH]⁺BF₄^- (Cy, cyclohexyl) catalyzes the alkylation in solution within 2 to 8 hours at temperatures ranging from 75° to 110°C and tolerates a broad range of substrate functionality, including amines and carbonyls. Preliminary mechanistic studies are inconsistent with Friedel-Crafts-type electrophilic activation of the alcohols, suggesting instead a vinyl C-H activation pathway with opposite electronic polarization.

Full text of DOI:10.1126/science.1208839

Selective Catalytic C−H Alkylation of Alkenes with Alcohols
Dong-Hwan Lee et al.
Science 333, 1613 (2011);
DOI: 10.1126/science.1208839
This copy is for your personal, non-commercial use only.
If you wish to distribute this article to others, you can order high-quality copies for your
colleagues, clients, or customers by clicking here.
Permission to republish or repurpose articles or portions of articles can be obtained by
following the guidelines here.
The following resources related to this article are available online at
www.sciencemag.org (this information is current as of September 3, 2012 ):
Updated information and services, including high-resolution figures, can be found in the online
version of this article at:
http://www.sciencemag.org/content/333/6049/1613.full.html
Supporting Online Material can be found at:
http://www.sciencemag.org/content/suppl/2011/09/14/333.6049.1613.DC1.html
A list of selected additional articles on the Science Web sites related to this article can be
found at:
http://www.sciencemag.org/content/333/6049/1613.full.html#related
This article cites 30 articles, 2 of which can be accessed free:
http://www.sciencemag.org/content/333/6049/1613.full.html#ref-list-1
This article appears in the following subject collections:
Chemistry
http://www.sciencemag.org/cgi/collection/chemistry
Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the
American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright
2011 by the American Association for the Advancement of Science; all rights reserved. The title Science is a
registered trademark of AAAS.

REPORTS
undergo energetically more favorable alkoxylation
Selective Catalytic C–H Alkylation
and dehydrogenation reactions over the requisite
C–O bond-cleavage step (C–O bond dissociation
energy = 85 to 91 kcal mol−1) (10). Here, we report
a highly selective, operationally simple catalytic
alkylation method that successfully applies al-
cohols as the alkylation agent, tolerates a number
of common oxygen and nitrogen functional groups,
of Alkenes with Alcohols
Dong-Hwan Lee, Ki-Hyeok Kwon, Chae S. Yi*
Alkenes and alcohols are among the most abundant and commonly used organic feedstock in
industrial processes. We report a selective catalytic alkylation reaction of alkenes with alcohols that and liberates water as the only by-product.
forms a carbon-carbon bond between vinyl carbon-hydrogen (C–H) and carbon-hydroxy centers with
We recently discovered that the cationic ru-
thenium hydride complex [(C6H6)(PCy3)(CO)
–
the concomitant loss of water. The cationic ruthenium complex [(C6H6)(PCy3)(CO)RuH]+BF4 (Cy,
cyclohexyl) catalyzes the alkylation in solution within 2 to 8 hours at temperatures ranging from 75° RuH]+BF4– (1; Cy, cyclohexyl) is a highly effec-
to 110°C and tolerates a broad range of substrate functionality, including amines and carbonyls.
tive catalyst for a number of the coupling reac-
Preliminary mechanistic studies are inconsistent with Friedel-Crafts–type electrophilic activation of the tions that involve vinyl and arene C–H activation
alcohols, suggesting instead a vinyl C–H activation pathway with opposite electronic polarization.
(11, 12). Because these coupling reactions are
driven by dehydration, we reasoned that the de-
oncerns about climate change and envi- alytic alkylation and alkenylation processes, a hydrative coupling reaction with alcohols might
ronmental pollution have prompted ex- relatively limited substrate scope and the require- be achievable. To test the feasibility of alcohols
tensive research toward the development ment of a chelate-directing group, as well as stoi- as an alkylating agent, we initially used compound
C
of energy-efficient “green” catalytic methods chiometric oxidants and additives, remain hindrances 1 to probe the coupling reaction of simple a-olefins
that form desired products from readily available to applying these methods more broadly.
and ethanol. In a 100-ml Fisher-Porter pressure
and environmentally benign feedstock. Direct al- From both environmental and economic points bottle, a mixture of propene (30 mmol), ethanol
kylation of hydrocarbon compounds is a funda- of view, alcohols constitute a highly attractive (40 mmol), and the catalyst 1 [0.01 mole percent
mentally important C–C bond forming method that class of alkylating reagent because they are in- (mol %)] in chlorobenzene (C6H5Cl, 3 mL) solu-
encompasses a broad spectrum of applications rang- expensive (and often easily derived from natural tion was heated in an oil bath at 110°C, after which
ing from petrochemical processing to the synthe- sources), and water is the only by-product resulting the product mixture was analyzed by both 1H nu-
sis of pharmaceutical agents. Traditionally, the from the associated coupling reaction (Fig. 1D). clear magnetic resonance (NMR) spectroscopy
Friedel-Crafts reaction has been widely used in However, alcohols are seldom employed as al- and gas chromatography (GC) (Eq. 1 and fig. S1).
industrial-scale alkylation processes, but this meth- kylating agents on account of their tendency to The product mixture showed a 9:1 ratio of 2-
od suffers from requiring stoichiometric amounts
R
1 (0.01 to 0.02 mol%)
R
R
of both Lewis acid promoters and base for neutral-
ization of acid by-products (Fig. 1A) (1). Further-
more, the method typically forms a mixture of
linear and branched alkylation products resulting
from both primary and secondary carbocations.
Mizoroki-Heck coupling has also been used exten-
sively for the direct alkenylation of arenes, but this
catalytic method still requires stoichiometric base
and is generally limited to activated olefins (Fig.
1B) (2). More recently, catalytic C–H bond activ-
ation reactions have emerged as a promising meth-
od for the direct couplings of arenes and alkenes
(3). Since Murai et al.’s pioneering report on the
chelate-assisted catalytic C–H alkylation of arenes
that obviated the formation of stoichiometric by-
products (4), many nitrogen and oxygen directed
alkylation and alkenylation methods of arene com-
pounds have been successfully developed over the
past two decades (Fig. 1C) (5–7). In a seminal re-
port, Stuart and Fagnou reported a highly selective
palladium-catalyzed cross-coupling of two arene
compounds by using a stoichiometric amount of
metal oxidant (8). Wang et al. have recently suc-
ceeded in developing a regioselective arene C–H
alkenylation method catalyzed by a palladium cat-
alyst with a tailored amino acid ligand (9). The
direct C–H alkenylation is of particular synthetic
importance because alkenes are a highly valuable
class of precursors for a variety of chemical trans-
formations. Despite such notable advances in cat-
+
+
+
(1)
CH3CH2OH
H2O
C6H5Cl, 110 o
C
(30 mmol) (40 mmol)
3
2
R = CH3:
2a:3a = 9:1, initial TOF = 0.9 s-1, TON = 4000
Equation 1.
R = C(CH3)3: 2b:3b = 10:1, initial TOF = 0.7 s-1, TON = 2800
A
Friedel-Crafts Alkylation
R
H
X
R
Acid Promoter
BH+X-
+
+
+
+
B
R
B = organic or inorganic base; R = alkyl, aryl; X = Cl, Br, I, OTf
B
Mizoroki-Heck Coupling Reaction
R
X
R
Pd(0) Catalyst
BH+X-
+
+
B
+
H
B = organic or inorganic base; R = aryl, carbonyl group; X = Cl, Br, I, OTf
C
Chelate-Assisted Arene C-H Alkenylation
Z
Z
R
Pd(II) Catalyst
+
[O]
base or additive
H
R
H
Z = oxygen or nitogen chelate directing group; R = alkyl, aryl, carbonyl group
[O] = O2, H2O2 or metal oxidant
D
C-H Alkylation of Alkenes with Alcohols
R
OH
R
Ru Catalyst
+
+
H2O
R'
R'
H
R = R' = alkyl, aryl
Department of Chemistry, Marquette University, Milwaukee,
WI 53201–1881, USA.
Fig. 1. (A) Lewis acid–promoted Friedel-Crafts alkylation of arenes and alkyl halides. OTf, trifluoro-
methanesulfonate. (B) Mizoroki-Heck coupling of aryl halides and olefins. (C) Chelate-assisted oxidative
C–H alkenylation of arenes. (D) Dehydrative C–H alkylation of alkenes with alcohols.
*To whom correspondence should be addressed. E-mail:
chae.yi@marquette.edu
www.sciencemag.org SCIENCE VOL 333 16 SEPTEMBER 2011
1613

REPORTS
pentenes (2a) (E/Z = 20:1) and 2-methyl-1-butene
(3a). The initial turnover frequency (TOF) of the mixture of an alkene (1.0 mmol), an alcohol (1.1 mmol), and 1 (0.5 mol %) in CH2Cl2 (2 mL for <80°C)
Table 1. Scope of the C–H alkylation reaction of alkenes with alcohols under the following conditions. A
or C6H5Cl (2 mL for >80°C) in a 25 mL Schlenk tube was heated in an oil bath at the specified
temperature for the specified time period. The coupling product was isolated and purified by a
column chromatography on silica gel (280 to 400 mesh). Ar, C6H4-p-OMe; Me, CH3; Et, CH2CH3; n-Hexyl,
CH2(CH2)4CH3; Ph, phenyl.
combined alkylation products 2a and 3a after
20 min was measured to be 0.9 s−1, and the
turnover number (TON) of 4000 was easily
reached within 6 hours of reaction time (40%
isolated yield of 2a and 3a). Replacing propene
with 3,3-dimethyl-1-butene led to a considerably
lower TOF of 0.7 s−1 and a TON of 2800 in
yielding 2b and 3b (10:1) under similar condi-
tions (56% isolated yield of 2b and 3b with 0.02
mol % of 1). We found that the catalytic activity
of complex 1 is distinctly high for the alkylation
reaction, as none of the other catalysts we screened
showed any considerable activity under similar
reaction conditions (table S1). We also found that
employing a slight excess amount of alcohol was
beneficial in limiting alkene dimerization and
other side reactions.
Encouraged by these initial results, we next
surveyed the scope of the alkylation reaction cat-
alyzed by 1 (Table 1). In general, we found that
both aliphatic and aryl-substituted primary alco-
hols react smoothly with cyclic olefins to give the
products 4a through 4g (entries 1 to 7). Only
linear alkylation products were formed with pri-
mary alcohols; no branched products were de-
tected. We found secondary aliphatic alcohols,
such as 2-propanol and cyclohexanol, to be quite
sluggish, resulting in low yields under the stan-
dard applied conditions. Regioselective 2-alkylation
products 5a through 5g were formed from the
reaction of indene with both primary and second-
ary alcohols (entries 8 to 14). The alkylation of
heteroarene compounds such as N-methylindole
and benzopyrene with both aliphatic and ben-
zylic alcohols afforded the products 6 and 7,
respectively (entries 15 to 20), whereas the re-
action of indoline with 4-methoxybenzyl alcohol
yielded the dehydrogenative alkylation product
8 (entry 21). In support of the latter case, sim-
ilar types of catalytic dehydrogenative cou-
pling reactions of amines have been reported
recently (13).
Entry Alkene
Alcohol
Product(s)
Temp (°C) Time (h) Yield (%)
R
n
n
*
4a (n = 2, R = Et)
ethanol
6
6
4
5
4
4
4
88
79
90
91
92
84
87
1
2
3
4
5
6
7
90
90
75
75
75
75
75
4b (n = 2, R = n-Hexyl)
4c (n = 2, R = CH2Ar)
(+)-4d (n = 2, R = CH(CH3)Ar)
4e (n = 1, R = CH2-p-Tol)
4f (n = 3, R = CH2Ar)
4g (n = 4, R = CH2Ar)
1-hexanol
ArCH2OH
(+)-ArCH(CH3)OH
p-CH3-C6H4CH2OH
ArCH2OH
ArCH2OH
R
ethanol
5a (R = Et)
8
5
5
5
3
3
2
3
84
71
57
86
97
95
91
90
90
90
75
75
75
75
5b (R = n-Hexyl)
1-hexanol
9
5c (R = CH2CH(CH3)C4H9)
(+)-5d (R = CH(CH3)Ph)
(+)-5e (R = CH(CH3)Ar)
5f (R = CH2Ar)
2-ethyl-1-hexanol
(R)-PhCH(CH3)OH
(+)-ArCH(CH3)OH
ArCH2OH
10
11
12
13
14
5g (R = CH2C6H4-p-CO2Me)
p-CO2Me-C6H4CH2OH
Me
Me
N
N
R
1-hexanol
3
3
2
2
96
95
96
94
15
16
17
18
75
75
75
75
6a (R = CH2(CH2)4CH3)
6b (R = CH2CH(CH3)2)
6c (R = CH2Ar)
(CH3)2CHCH2OH
ArCH2OH
(+)-PhCH(CH3)OH
(+)-6d (R = CH(CH3)Ph)
O
R
O
19
20
ArCH2OH
(+)-p-Tol-CH(CH3)OH
7a (R = CH2Ar)
(+)-7b (R = CH(CH3)-p-Tol)
75
75
2
2
95
91
H
H
Ar
N
N
91
21
75
3
ArCH2OH
8
R
R
10
X
X
X
9
1-hexanol
ArCH2OH
1-hexanol
ArCH2OH
9a:10a = 8:1 (X = CH3, R = n-Hexyl)
9b:10b = 10:1 (X = CH3, R = CH2Ar)
9c:10c = 6:1 (X = Cl, R = n-Hexyl)
9d:10d = 9:1 (X = Cl, R = CH2Ar)
Ar
110
90
110
90
5
5
8
6
84
91
78
83
22
23
24
25
5
5
92
89
90
90
26
27
ArCH2OH
ArCH2OH
(5:1)
11
12
Ar
Ar
13
Styrene derivatives predominantly formed the
trans-alkylation products 9a to 9d (>98% trans),
along with a small amount of the hydrogenation
products 10a to 10d (10 to 15%) (entries 22 to 25).
The alkylation reaction of 1-alkenes typically
yielded a complex mixture of the coupling products
resulting from the isomerization of terminal to
internal olefins. In case of allylbenzene, a 5:1
OH
28
29
71
87
110
90
8
3
14
Ph
HO
Ph
(R)
(R)-15
*Due to the high volatility of 4a, this product yield was determined by GC.
mixture of the products 11 and 12 was formed, 28). The treatment of indene with (R)-2-phenyl-1-
We observed a markedly high degree of
with the b-alkylation product 11 being pre- propanol directly formed an optically pure product functional-group tolerance in the alkylation
dominant (entry 26). The alkylation reaction of (R)-15 without any racemization or branched reaction of a number of biologically active alkene
vinylcyclohexane exclusively yielded the tetra- alkylation products (entry 29). With the use of substrates (Fig. 2). The alkylation of both choles-
substituted olefin product 13, apparently resulting rigorously purified starting materials, we have terol and progesterone with 4-methoxybenzyl al-
from olefin isomerization to 2-ethylcyclohexene been able to achieve a maximum TON of 102,000 cohol under the standard conditions afforded
before alkylation (entry 27). Ru catalysts are well at an initial TOF of 2.7 s−1 from the alkylation the products (–)-16 and (+)-17 respectively, with-
known for their ability to mediate facile olefin of indene with 4-methoxybenzyl alcohol under out affecting either alcohol or carbonyl functional
isomerization reactions (14, 15), and we have the standard reaction conditions at 150°C after groups. The alkylation of N-methoxycarbonyl-L-
also observed a rapid rate of olefin isomerization 48 hours (table S2). In all cases, the alkylation tryptophan methyl ester and (–)-strychnine resulted
catalyzed by 1 (16). In an intramolecular exam- reaction predictably and selectively afforded the in the chemoselective formation of the products
ple, the treatment of geraniol exclusively led to coupling products without any additives or for- (–)-18 and (–)-19, while tolerating both amino
the formation of p-cymene 14 in 71% yield (entry mation of wasteful by-products.
and carbonyl functional groups. The alkylation of
1614
16 SEPTEMBER 2011 VOL 333 SCIENCE www.sciencemag.org

REPORTS
O
of the C-2 indenyl carbon was deuterated, as
O
determined by both 1H and 2H NMR (fig. S2).
We also detected the formation of indan with
15% deuteration of the a carbon specifically.
Selective hydrogen-deuterium exchange at the
indenyl carbon suggests a facile vinyl C–H ac-
tivation step in the catalysis (18).
N
O
HN
OCH3
OCH3
H
H
H
H
Ar
N
H
H
H
H
O
H
O
HO
O
N
Ar
H
Ar
Ar
(+)-17, 81% (11% rsm)
(-)-18, 84% (7% rsm)
(-)-19, 73% (12% rsm)
(-)-16, 57% (28% rsm)
To explore the possibility of an electrophilic
substitution mechanism, we compared the rates
of the alkylation reaction for primary, secondary,
and tertiary alcohols. Thus, the treatment of in-
dene (1.5 mmol) with an equimolar mixture of
PhCH2OH, PhCH(CH3)OH, and PhC(CH3)2OH
(0.5 mmol each) (Ph, phenyl) in the presence of
1 (0.5 mol %) in CH2Cl2 (3 mL) led to a 12:4:1
ratio of the respective alkylation products under
the standard conditions at 75°C after 30 min. The
results are clearly opposite of those typically ob-
served with Friedel-Crafts–type electrophilic al-
kylation reactions, for which the rate has been
shown to be fastest for tertiary alcohols (19, 20).
Furthermore, in Friedel-Craft type of alkylations
with alcohols, a stoichiometric amount of acid is
required to generate secondary and tertiary car-
bocations, which precludes employing substrates
with an acid-sensitive functional group, as ex-
emplified by a recent report (21). Also, as listed
in Table 1 (entry 11), the formation of racemic
product (T)-5d from the treatment of indene with
an optically pure chiral alcohol (R)-PhCH(OH)CH3
is inconsistent with an SN2-type displacement
mechanism in the C–C bond formation step.
To further probe electronic influences asso-
ciated with both the alcohol and alkene substrates,
we constructed two separate Hammett plots by
measuring the rates of a series of para-substituted
benzyl alcohols p-X-C6H4CH2OH (where X =
OMe, CH3, H, F, Cl) with indene and from para-
substituted styrene derivatives p-Y-C6H4CH=CH2
(where Y = OMe, CH3, H, F, Cl) with benzyl
alcohol (Fig. 3B). The rate correlation with para-
substituents of benzyl alcohols led to a highly
negative Hammett r value of –3.3 T 0.4, whereas
we obtained a positive r value of +1.3 T 0.3
from the correlation with styrene derivatives (fig.
S3). The strong promotional effect of electron-
releasing groups on benzyl alcohol suggests a
cationic transition state involving an electrophilic
ruthenium catalyst. The relatively small Hammett
r value from the correlation of para-substituted
styrenes is consistent with a diminished promo-
tional effect due to the formation of a terminal
Ru-alkenyl species.
conditions: 90 oC, 6 h
conditions: 90 oC, 6 h
conditions: 90 oC, 4 h
conditions: 90 oC, 5 h
H3CO
H3CO
Ar
HO
H
HO
ArCH2OH
110 oC, 6 h
MeO
HO
N
HO
ArCH2OH
90 oC, 6 h
N
H
N
CH3
N
MeO
CH3
O
O
Ar
OCH3
OCH3
N
N
(-)-quinine
(-)-20, 58% (24% rsm)
(-)-sinomenine
(-)-21, 62% (21% rsm)
Fig. 2. Alkylation of bioactive alkene compounds with 4-methoxybenzyl alcohol in C6H5Cl. rsm, recovered
starting material; Ar, C6H4-p-OMe; Me, methyl.
D
(15%)
1
A
B
D3C
D3C
1 (2 mol %)
2
+
+
CDOD
D
toluene-d8, 25 oC
(21%)
3
(0.6 mmol, 99% D)
(0.2 mmol)
(1 : 1.1)
1 (0.5 mol %)
C6H5Cl, 80 oC
OH
+
+
X
5
X = OMe, CH3, H, F, Cl
ρ = -3.3
X
1 (0.5 mol %)
C6H5Cl, 80 oC
OH
9
Y
Y
ρ = +1.3
Y = OMe, CH3, H, F, Cl
C
0.999
CH3
OH
1.000
1.002
1.001
1.000 (reference)
1 (1.0 mol %)
C6H5Cl, 75 oC
1.049
1
6
+
2
3
MeO
5e
1.001
OMe
Fig. 3. (A) Hydrogen-deuterium exchange experiment from the treatment of indene with 2-propanol-d8
in toluene-d8 at 25°C. (B) Hammett studies of the alkylations of indene with p-X-C6H4CH2OH (X = OCH3,
CH3, H, F, Cl) and p-Y-C6H4CH=CH2 (Y = OCH3, CH3, H, F, Cl) with benzyl alcohol. (C) Carbon isotope effect
measurement for the alkylation of indene with 1-(4-methoxyphenyl)ethanol by using Singleton’s method.
The 13C isotope ratio of 5e at 95% conversion was compared with the product isolated at 11 to 18%
conversion (average of three runs).
(–)-quinine formed the product (–)-20 in a highly tion of the amino group, but without coupling at
We successfully measured the 12C/13C kinetic
stereoselective fashion, which apparently resulted the enaminyl carbon itself. In all cases, the alkyl- isotope effect from the coupling reaction of in-
from regioselective alkylation of the internal ation products are formed predictably without dene and 1-(4-methoxyphenyl)ethanol by em-
olefinic carbon followed by a diastereoselective evidencing any racemization.
ploying Singleton’s NMR technique (Fig. 3C)
hydrogenation of the alkene. Karplus coupling We conducted the following experiments to (22). We observed the most pronounced carbon
constant analysis between two vicinal hydrogens gain mechanistic insight into the alkylation reac- isotope effect on the indenyl carbon when we
(three-bond coupling constant 3J = 5.8 Hz) led to tion. First, we examined the hydrogen-deuterium compared the 13C ratio of the product 5e at 95%
our stereochemical assignment at the hydrogenated exchange pattern of the coupling reaction, using conversion to the product obtained from three
carbon atom (17). The alkylation of a morphine NMR spectroscopy to monitor the reaction of separate low conversions [(13C at 95% conversion)/
analog, (–)-sinomenine, selectively formed the indene (0.2 mmol) with 2-propanol-d8 (0.6 mmol) (average of 13C at 11, 15 and 18% conversion) at
coupling product (–)-21 in 62% yield. In this case, and 1 (2 mol %) in toluene-d8 (0.4 mL) (Fig. 3A). C-2 = 1.049] (table S3). The results are consistent
the alkylation is accompanied by dehydrogena- After 1 hour at 25°C, we found that nearly 21% with C–C bond formation as the rate-limiting step
www.sciencemag.org SCIENCE VOL 333 16 SEPTEMBER 2011
1615

REPORTS
R
R
and insertion of the Ru atom are expected to be favored
at the C-2 carbon of indene. Previously, we also observed
the formation free benzene molecule in the reactions
catalyzed by 1 (12, 16).
Fig. 4. Proposed mechanism of the alkyl-
R'CH2OH
R
2
ation of an alkene (CH2=CHR) with an al-
cohol (R′CH2OH). rds, rate-determining step.
R = R′ = alkyl, aryl.
[Ru]
HO
1
[Ru]
22
-C6H6, -PCy3
H2O
R
R'
R
19. R. M. Roberts, A. A. Khalaf, Friedel-Crafts Alkylation
Chemistry: A Century of Discovery (Dekker, New York,
1984).
[Ru] = Ru(CO)L
[Ru] = Ru(CO)Ln
rds
L = alkene, alconhol
L = alkene, alcohol
R'
20. G. A. Olah, R. Krishnamurti, G. K. S. Prakash, in
Comprehensive Organic Synthesis, B. M. Trost, I. Fleming,
Eds. (Pergamon, New York, 1991), vol. 3, chap. 1.8.
21. Z.-Q. Liu et al., Org. Lett. 13, 2208 (2011).
22. D. E. Frantz, D. A. Singleton, J. P. Snyder, J. Am. Chem.
Soc. 119, 3383 (1997).
OH
[Ru]
OH
[Ru]
23
H
R
R
23. R. Gao, C. S. Yi, J. Org. Chem. 75, 3144 (2010).
24. F. Kakiuchi, S. Murai, Acc. Chem. Res. 35, 826 (2002).
25. Transition metal–vinyl complexes have been proposed to
be a key intermediate in C–H bond activation chemistry.
For representative examples, see (32, 33).
26. The formation of a free benzene molecule (2%) is
detected after heating the reaction mixture at 70°C for
15 min using 5 mol % of 1. Previously, we observed
similar results in the reactions catalyzed by 1 (12, 16).
27. Y. Feng et al., J. Am. Chem. Soc. 127, 14174 (2005).
28. T. Matsumoto, Y. Nakaya, N. Itakura, K. Tatsumi,
J. Am. Chem. Soc. 130, 2458 (2008).
2. R. F. Heck, Acc. Chem. Res. 12, 146 (1979).
3. J.-Q. Yu, Z. Shi, Eds., C–H Activation (Springer, New York,
2010).
for the alkylation reaction. We (23) and others
(24) observed similar carbon isotope effects in
ruthenium-catalyzed C–H insertion reactions,
where C–C bond formation has been proposed
as the rate-limiting step.
4. S. Murai et al., Nature 366, 529 (1993).
5. D. A. Colby, R. G. Bergman, J. A. Ellman, Chem. Rev.
110, 624 (2010).
6. C. S. Yeung, V. M. Dong, Chem. Rev. 111, 1215
(2011).
7. C.-J. Li, Acc. Chem. Res. 42, 335 (2009).
8. D. R. Stuart, K. Fagnou, Science 316, 1172 (2007).
9. D.-H. Wang, K. M. Engle, B.-F. Shi, J.-Q. Yu, Science 327,
315 (2010); 10.1126/science.1182512.
10. M. B. Smith, J. March, March’s Advanced Organic
Chemistry (Wiley, New York, 2001).
11. C. S. Yi, D. W. Lee, Organometallics 28, 4266 (2009).
12. C. S. Yi, D. W. Lee, Organometallics 29, 1883 (2010).
13. G. E. Dobereiner, R. H. Crabtree, Chem. Rev. 110,
681 (2010).
14. D. V. McGrath, R. H. Grubbs, Organometallics 13,
224 (1994).
15. F. C. Courchay, J. C. Sworen, I. Ghiviriga, K. A. Abboud,
K. B. Wagener, Organometallics 25, 6074 (2006).
16. D. W. Lee, C. S. Yi, Organometallics 29, 3413 (2010).
17. M. Karplus, J. Chem. Phys. 30, 11 (1959).
18. The regioselective deuterium exchange on the C-2 carbon of
indene is most consistent with the vinyl C−H activation
mechanism. The selective deuterium incorporation on the
C-1 and C-3 carbons of indan can readily be explained
from regioselective insertion of indene to Ru−D, followed
by reductive elimination with the hydride formed from
the vinyl C−H activation of another indene molecule. Due
to steric hindrance from the arene hydrogen, metalation
On the basis of these results, we propose a
mechanism involving a cationic Ru-alkenyl spe-
cies 22, which is initially formed from the re-
action of 1 with two equivalents of the olefin
substrate via the vinylic C–H activation and an
alkane elimination step (Fig. 4) (25). As support-
ing evidence for the Ru-alkenyl species, we have
been able to detect the formation of indan and
free benzene molecules from the crude reaction
mixture of indene with 4-methoxybenzyl alcohol
(26). Though the exact mechanism of the C–O
cleavage step is not clear at the present time, one
possible pathway involves the oxidative addition
of the C–O bond to form a cationic Ru(IV)-
alkenyl-alkyl species, from which reductive elim-
ination proceeds to yield the alkylation product
and a cationic Ru-hydroxo complex 23. Notably,
Ru-hydroxo complexes have been shown to ex-
hibit high activity for both C–H and H–H bond
activation reactions (27, 28). The carbon iso-
tope effect study provides strong support for a
turnover-limiting C–C bond formation step. A
few electrophilic ruthenium complexes have been
found to mediate the C–O bond cleavage re-
action of alcohols and acetals (29–31). The
subsequent vinylic C–H activation and water
elimination is envisaged for the regeneration of
the Ru-alkenyl species 22. Extensive hydrogen-
deuterium exchange between indene and 2-
propanol-d8 substrates suggests that the vinylic
C–H bond activation step is facile and reversible
under the catalytic conditions.
Our catalytic alkylation method exhibits a
broad substrate scope, as well as a high degree of
regio- and chemoselectivity without requiring any
co-catalysts or additives, and generates no wasteful
by-products other than water. Whereas chlorinated
solvents such as CH2Cl2 and C6H5Cl generally
give the best results, nonchlorinated solvents such
as toluene and tetrahydrofuran can be used in most
cases without sacrificing product yields or selec-
tivity (though they require a longer reaction time).
We anticipate that the catalytic method will con-
tribute substantially to the development of next-
generation “green” catalytic processes.
29. D. Rondon, B. Chaudret, X. D. He, D. Labroue,
J. Am. Chem. Soc. 113, 5671 (1991).
30. D. B. Grotjahn, H. C. Lo, Organometallics 15, 2860 (1996).
31. S. Ueno, E. Mizushima, N. Chatani, F. Kakiuchi,
J. Am. Chem. Soc. 128, 16516 (2006).
32. P. O. Stoutland, R. G. Bergman, J. Am. Chem. Soc. 107,
4581 (1985).
33. K. B. Renkema, Y. V. Kissin, A. S. Goldman, J. Am.
Chem. Soc. 125, 7770 (2003).
Acknowledgments: We thank the NSF for financial support
(grant CHE-1011891).
Supporting Online Material
www.sciencemag.org/cgi/content/full/333/6049/1613/DC1
SOM Text
Figs. S1 to S3
Tables S1 to S3
References (34–38)
24 May 2011; accepted 3 August 2011
10.1126/science.1208839
Soil Nitrite as a Source of
Atmospheric HONO and OH Radicals
Hang Su,1*† Yafang Cheng,2,3*† Robert Oswald,1 Thomas Behrendt,1 Ivonne Trebs,1
Franz X. Meixner,1,4 Meinrat O. Andreae,1 Peng Cheng,2 Yuanhang Zhang,2 Ulrich Pöschl1†
Hydroxyl radicals (OH) are a key species in atmospheric photochemistry. In the lower atmosphere,
up to ~30% of the primary OH radical production is attributed to the photolysis of nitrous acid
(HONO), and field observations suggest a large missing source of HONO. We show that soil nitrite
can release HONO and explain the reported strength and diurnal variation of the missing source.
Fertilized soils with low pH appear to be particularly strong sources of HONO and OH. Thus, agricultural
activities and land-use changes may strongly influence the oxidizing capacity of the atmosphere.
Because of the widespread occurrence of nitrite-producing microbes, the release of HONO from soil
may also be important in natural environments, including forests and boreal regions.
eterogeneous reactions of NO2 have NO2, however, atmospheric aerosols have too
been suggested to be a missing source little surface area to account for the measured
of nitrous acid (HONO) to the atmo- HONO production rates (7, 8). Thus, recent studies
H
sphere (1–4), and photoenhanced reactions on suggested that the reduction of N(IV) in NO2 to
the surface of soot and other aerosol particles N(III) in HONO may proceed on ground surfaces
have been proposed to explain the observed en- (7, 8), and field data indicate that HONO is in-
hancement of HONO production rates during the deed released from the ground (10, 11).
The dominant sources of N(III) in soil, how-
(5–9). Because of low uptake coefficients for ever, are biological nitrification and denitrifica-
References and Notes
1. G. A. Olah, A. Molnár, Hydrocarbon Chemistry
(Wiley, New York, 2003).
daytime [up to 5 parts per billion (ppb) hour–1]
1616
16 SEPTEMBER 2011 VOL 333 SCIENCE www.sciencemag.org