Hidden Bronsted acid catalysis: Pathways of accidental or deliberate generation of triflic acid from metal triflates

Article abstract of DOI:10.1021/jo201631x

The generation of a hidden Bronsted acid as a true catalytic species in hydroalkoxylation reactions from metal precatalysts has been clarified in case studies. The mechanism of triflic acid (CF₃SO₃H or HOTf) generation starting either from AgOTf in 1,2-dichloroethane (DCE) or from a Cp*RuCl₂/AgOTf/phosphane combination in toluene has been elucidated. The deliberate and controlled generation of HOTf from AgOTf and cocatalytic amounts of tert-butyl chloride in the cold or from AgOTf in DCE at elevated temperatures results in a hidden Bronsted acid catalyst useful for mechanistic control experiments or for synthetic applications.

Full text of DOI:10.1021/jo201631x

Article
pubs.acs.org/joc
Hidden Brønsted Acid Catalysis: Pathways of Accidental or
Deliberate Generation of Triflic Acid from Metal Triflates
Tuan Thanh Dang, Florian Boeck, and Lukas Hintermann*
Department Chemie, Technische Universitat Munchen, Lichtenbergstrasse 4, 85748 Garching b. Munchen, Germany
̈
̈
̈
S
* Supporting Information
ABSTRACT: The generation of a hidden Brønsted acid as a true
catalytic species in hydroalkoxylation reactions from metal precatalysts
has been clarified in case studies. The mechanism of triflic acid
(CF3SO3H or HOTf) generation starting either from AgOTf in 1,2-
dichloroethane (DCE) or from a Cp*RuCl2/AgOTf/phosphane
combination in toluene has been elucidated. The deliberate and
controlled generation of HOTf from AgOTf and cocatalytic amounts
of tert-butyl chloride in the cold or from AgOTf in DCE at elevated
temperatures results in a hidden Brønsted acid catalyst useful for mechanistic control experiments or for synthetic applications.
hydroalkoxylations.1 In a recent example, Cu(OTf)2 was shown
INTRODUCTION
■
to be a precursor for HOTf via reduction to CuOTf under the
conditions of a hydroalkoxylation reaction.41 In the current
work, we have specifically investigated the role of silver triflate
(AgOTf), which is used either as a catalyst itself or as a catalyst
activator in combination with transition-metal chlorides. The
key findings are that HOTf is released according to several
specific mechanistic routes and acts as a Brønsted acid catalyst
under conditions that have earlier been proposed to involve
metal catalysis. We propose the term “hidden Brønsted acid
catalysis” to denote cases in which Brønsted acid released from
any kind of precursor is responsible for an observed catalytic
activity, rather than the precursor itself. We have established
conditions to deliberately and reliably generate HOTf in trace
amounts from AgOTf under mild conditions. Paradoxically,
such a hidden Brønsted acid catalyst may have specific
advantages over the use of pure HOTf as catalyst.
The development of metal-catalyzed additions of oxygen
nucleophiles to nonactivated alkenes1−4 had seen limited
success prior to 2002,1,4−7 but the number of reports on such
catalytic hydroalkoxylations (including hydroaryloxylations, if R
= aryl; Scheme 1), or related hydroacyloxylations, dramatically
Scheme 1. Hydroalkoxylation Reactions
increased after 2004.1,8−33 Many of the new catalysts are
electrophilic metal trifluoromethanesulfonate (triflate) salts that
display a similar substrate range and product selectivity
profile.34
RESULTS
■
Several literature reports suggest the use of AgOTf as a catalyst
in halogenated organic solvents including 1,2-dichloroethane
(hereafter abbreviated DCE), often at elevated temper-
atures.23,42−46 This is potentially problematic in view of the
halide abstracting properties of the silver cation.47 To
investigate the stability of silver triflate in DCE, we heated a
sample of the compound at reflux (84 °C) for 3 h. The
resulting turbid suspension was filtered to leave an almost
quantitative yield (99%) of AgCl on the filter (for a
photograph; see Figure S1, Supporting Information).48 The
filtrate displayed two peaks in the 19F NMR spectrum, one for
HOTf (δ = −77.6 ppm in C6D6; δ = −78.0 ppm in CDCl3; see
the discussion for a comment on the chemical shifts of HOTf
samples recorded in air and moist solvents) and a second peak
Bearing in mind Spencer’s hypothesis that many hetero-
Michael additions are catalyzed by Brønsted acidic metal aquo
complexes rather than the precursor Lewis acids introduced,35
some ambiguity remains for most of the reported examples with
respect to the actual catalytic species. Indeed, the reactions
shown in Scheme 1 have also been catalyzed by strong
Brønsted acids,36,37 and Hartwig and co-workers have shown
that traces of triflic acid (CF3SO3H/HOTf) are, in many cases,
sufficient to reproduce experimental results previously obtained
with metal triflate catalysts.38,39 However, the approach to show
that the Brønsted acid will reproduce a certain set of results
neither rules out competitive or exclusive metal catalysis nor
does it explicitly prove that Brønsted acid catalysis is
occurring.40 This may be the reason why the implications of
the suggestive results of the Spencer and Hartwig groups have
not always been appreciated in work on Lewis-acid-catalyzed
Received: August 4, 2011
Published: October 19, 2011
© 2011 American Chemical Society
9353
dx.doi.org/10.1021/jo201631x|J. Org. Chem. 2011, 76, 9353−9361

The Journal of Organic Chemistry
Article
due to 2-chloroethyltriflate (1; δ = −74.9 ppm in C6D6; δ =
−74.5 ppm in CDCl3), which also gave rise to a characteristic
triplet (δ{1H}= 3.62 ppm, J = 5.4 Hz in C6D6; δ = 4.64 ppm, J
catalyze the reaction (entries 2−4). In agreement with the
control experiment in an earlier report,43 the catalytic activity of
HOTf is inferior to that of AgOTf at catalyst loadings of 5 mol
% (entries 1, 2).42 However, the low yield obtained in the
experiment with HOTf (Table 1, entry 2) is misleading because
the reaction is very fast,55 and the product may undergo
cationic polymerization.38 When the reaction was performed
with a lower quantity of HOTf (entry 3), the product yield was
higher and similar to that earlier obtained with AgOTf (entry
1).42 Several other metal triflates among a range tested induced
the isoprene−phenol condensation, which weakens the claim
for a specific silver ion catalysis (entries 5−12).14,56−60
48,49
1
= 5.7 Hz in CDCl3) in the H NMR spectrum.
This peak
was more intense if the reaction of AgOTf with DCE was
carried out at 80 °C for 2 h but disappeared with a concomitant
increase of the HOTf peak after prolonged heating. The solvent
therefore reacts with AgOTf via an initial substitution to 1,
followed by release of HOTf, either by hydrolysis with trace
water50 or by an E1-type elimination, which would presumably
lead to vinyl chloride oligomers (Scheme 2).
A potent hidden acid catalyst was obtained by preheating
AgOTf in refluxing DCE (84 °C) and cooling to RT prior to
addition of the substrates; with only 1 mol % AgOTf, a fast
reaction gave high yields of chromane 4a (Table 1, entry 14).
Admitting that the role of the silver salt is to abstract chloride
from the chlorinated solvent, followed by elimination of HOTf
as the true catalyst, it follows that a hidden acid catalyst should
be obtained under milder conditions by addition of an alkyl
halide that is prone to E1 elimination to the reaction mixture.
The addition of tert-butyl chloride (5) to a stirred solution of
AgOTf at ambient temperature in DCE or chloroform-d1
produced a white precipitate of AgCl immediately (for a
photograph, see Figure S6, Supporting Information), and a
major peak for HOTf in the 19F NMR indicated that E1
eliminiation had occurred (Scheme 3, route a).61−64 When the
Scheme 2. Generation of HOTf from AgOTf in DCE
The generation of organic triflates from haloalkanes and
AgOTf has ample precedence.51−53 Our experiment implies
that the generation of HOTf from solutions of AgOTf in DCE
must be generally expected and is probably inevitable at higher
temperatures or with prolonged reaction times. A reported
condensation of phenol (2a) with isoprene (3) to 2,2-
dimethylchromane (4a; Table 1, entry 1), mediated by catalytic
Scheme 3. HOTf Generation from AgOTf and t-BuCl
Table 1. Catalytic Phenol−Isoprene Cyclization to 2,2-
a
Dimethylchromane
time
b
entry
catalyst
AgOTf
(h)
yield (%)
1
48
2
65
21
63
63
c
2
TfOH
3
TfOH (0.5 mol %)
p-TsOH
2
4
48
48
2
5
Al(OTf)3
Sn(OTf)2
Zn(OTf)2
Bi(OTf)3
Cu(OTf)2
Fe(OTf)3
Ce(OTf)3
LaOTf
6
89
51
15
48
50
29
c
7
48
48
48
2
experiment was performed in the solvent benzene-d6, the
generation of HOTf was also observed (19F NMR), but signals
for tert-butyl benzene 6 (1H, 13C NMR) show that acid had
been generated via a Friedel−Crafts alkylation of the aromatic
solvent as another distinct mechanistic route (Scheme 3, route
b).
8
9
10
11
12
13
14
2
48
48
0.3
d
AgOTf + Et3N
c
In situ hidden Brønsted acid catalysts prepared in the manner
discussed above present distinct advantages over the direct use
of HOTf, particularly in small-scale experiments, because HOTf
does not readily dissolve homogeneously in apolar reaction
mixtures. Either the “hot” (AgOTf + DCE at reflux) or “cold”
(AgOTf + t-BuCl) protocol gave highly active catalysts for the
condensation of isoprene and phenols to chromans (Table 2).
The examples discussed so far have been concerned with acid
released from combinations of silver salt and a chlorinated
solvent. On the other hand, many other examples of metal
triflate catalyses in the literature involve nonchlorinated
solvents but must still be considered suspicious with respect
to the occurrence of hidden acid catalysis. Other mechanistic
e
AgOTf (1 mol %)
79
a
Reaction conditions: the catalyst (5 mol %) was added into a mixture
of phenol (1 mmol) and isoprene (1.5 mmol) in DCE (5 mL), and the
b
c
reaction was stirred at RT. Yield of isolated product. No product
detected. Added in large excess. Catalyst prepared by refluxing
AgOTf in DCE (5 mL).
d
e
AgOTf in DCE at room temperature or 40 °C over 1−3 days,
might well represent an example of hidden acid catalysis rather
than a specific silver ion catalysis.43 It is known that fluorinated
Brønsted acids are particularly efficient catalysts for this type of
reaction, which involves a Friedel−Crafts alkylation followed by
a hydroalkoxylation.54 We find that Brønsted acids do indeed
9354
dx.doi.org/10.1021/jo201631x|J. Org. Chem. 2011, 76, 9353−9361

The Journal of Organic Chemistry
Article
Table 2. Synthetic Applications of Deliberate Hidden
Brønsted Acid Catalysts
Table 3. Catalytic Cyclization of 2-Allylphenol (8) to 2-
Methylcoumaran (9)
a
T
time
(h)
yield
c
entry
catalyst
mol % (°C)
(%)
a
1
p-TsOH
TfOH
5
5
5
2
2
60
60
50
90
50
24
2
44
63
54
81
14
a
2
a
3
AgOTf
48
48
96
b
4
Cp*RuCl2/AgOTf/PPh3
Cp*RuCl2/AgOTf/(R)-
BINAP
b
5
b
6
Cp*RuCl2/AgOTf/(R)-
BINAP
2
2
2
2
110
50
48
96
48
48
45
11
39
3
b
7
Cp*RuCl2/AgOTf/
(R)-TolBINAP
b
8
Cp*RuCl2/AgOTf/
(R)-TolBINAP
110
110
b
9
Cp*RuCl2/AgOTf/(R,SP)-
JOSIPHOS
a
General conditions: catalyst and 8 (1 mmol) were stirred in DCE (3
b
mL) at the indicated temperature. General conditions: catalyst
prepared from Cp*RuCl2 (0.02 mmol) and AgOTf (0.04 mmol) in
PhMe (3 mL) at 90 °C for 3 h, followed by cooling and addition of a
ligand (0.02 mmol) and substrate 8 (1 mmol). Coumaran 9 was
isolated by chromatography and analyzed by NMR, GC/MS, and
a
All reactions were performed on an 1 mmol scale with AgOTf (1 mol
%) as the catalyst precursor at RT with 1.5 equiv of isoprene.
chiral HPLC.48 Yield of product after chromatography.
c
b
Conditions: Procedure A: AgOTf was stirred for 10 min with t-BuCl
(4 mol %) prior to addition of the reactants. Procedure B: AgOTf was
refluxed in DCE (10 mL) for 3 h prior to addition of the reactants at
RT. The loading of AgOTf was 2 mol %.
Notably, the conditions of this metal-catalyzed reaction
(Table 3, entry 4) appear to be harsher than those of the acid-
catalyzed reactions (entries 1−3), considering the temperature
and reaction times. The in situ catalyst preparation asks for
heating of Cp*RuCl2 and AgOTf in toluene for 3 h, followed
by cooling to ambient temperature and addition of the
phosphane ligand. In order to study the chemical processes
taking place, we first performed the procedure without addition
of a phosphane. After evaporation of the solvent, NMR spectra
of the in situ catalyst from Cp*RuCl2 and AgOTf in toluene
were recorded. The 1H NMR spectrum showed the presence of
a single major new organometallic Cp* species, which was
isolated by chromatography and turned out to be the known
complex [Cp*Ru(η 6-toluene)]OTf (10) (Scheme 4).66,67
c
pathways of HOTf generation might be operating. For example,
it has been shown that a reduction of Cu(OTf)2 to Cu(OTf)
and HOTf takes place in the presence of olefins and alcohols.41
Another intriguing organometallic catalyst is generated in situ
from Cp*RuCl2 (7),65 phosphanes, and AgOTf (1:1:2 ratio) as
the activator.8−12 It was postulated that a [Cp*Ru(PR3)n]2+
fragment (n = 1, 2) forms and activates olefins toward
nucleophlic additions of oxygen nucleophiles to bring about a
redox-neutral hydroalkoxylation (and hydroacyloxylation12)
reaction.10 Any such catalytic reaction proceeding via organo-
metallic intermediates is of major interest for synthesis because
it has the potential to be carried out enantioselectively. It must
be stressed that evidence for the realization of such a process in
the literature is scarce and conflicting.1,2,35,38 A most
remarkable aspect of the Cp*Ru(III) catalyst is a reported
asymmetric cyclization of 2-allylphenol (8) to 2-methylcoumar-
an (9; coumaran is a trivial name for the 2,3-dihydrobenzofur-
ane; Table 3) when the metal fragment is complexed to
chelating chiral diphosphanes.10 The realization of an
asymmetric catalytic reaction is probably the strongest
argument for the involvement of a chiral phosphane complex
rather than a hidden acid as the catalyst. Therefore, we decided
to take a closer look at this particular in situ catalyst and to
collect evidence for the nature of the catalytically active species
formed under reaction conditions. Initially, it was established
that the reaction 8 → 9 is catalyzed by pure Brønsted acids
(Table 3, entry 1,2)36 and also very efficiently by our hidden
acid catalyst from AgOTf in DCE (entry 3). Next, the
reported10 ruthenium-catalyzed cyclization was successfully
repeated using PPh3 as the phosphane ligand (entry 4).
Scheme 4. Reaction Chemistry of Cp*RuCl2 (7) in Toluene
Related complexes have previously been prepared from either
HCp*/RuCl3 or Cp*RuCl2 and arenes in refluxing alcoholic
solution68 or alternatively from Ru(II) precursors like [Cp*Ru-
(MeCN)3]OTf by ligand exchange with arenes.67,69
9355
dx.doi.org/10.1021/jo201631x|J. Org. Chem. 2011, 76, 9353−9361

The Journal of Organic Chemistry
Article
We independently synthesized the complexes [Cp*Ru(η 6-
toluene)]Cl (11)70 and [Cp*Ru(η6-toluene)]OTf (10) from
Cp*RuCl2 by heating in toluene/isopropanol, in the absence or
presence of AgOTf, respectively (Scheme 4). The ruthenium-
(II) cation Cp*Ru+ is a very strong arenophile in organo-
metallic coordination chemistry69 that might not readily
exchange toluene for a phosphane ligand in toluene solution,
particularly if the phosphane can complex to AgCl, which is
present in the mixture in stoichiometric amounts. An in situ
catalyst was prepared by heating Cp*RuCl2 and AgOTf in
toluene, followed by cooling and addition of PPh3.10 The only
detectable Cp* species by 1H NMR in CDCl3 was again due to
[Cp*Ru(toluene)]OTf (10), with no other specific complex
signals in the 31P NMR spectrum apart from broad resonances
at δ(31P) = 10−15 ppm, which are ascribe to dynamic
(PPh3)nAgX species.71 Because we had initially removed the
toluene solvent from the in situ catalysts in vacuo before
spectroscopic investigation, any HOTf that might have been
formed was no longer present. A key experiment involved the
generation of an in situ catalyst from Cp*RuCl2 and AgOTf in
hot toluene, followed by addition of (±)-BINAP (2,2′-bis-
diphenylphosphino-1,1′-binaphthyl) as a potentially chelating
diphosphane, which should increase chances of generating and
observing a [Cp*RuII(BINAP)] complex. A filtered sample of
this in situ catalyst was analyzed by NMR spectroscopy in
(1)
Triflic acid should induce cyclization of 8 to racemic 9 under
conditions of the catalytic reaction, whereas neither the
sandwich complex 10 nor the [AgnCln(BINAP)m] species
would be expected to show catalytic activity for hydro-
alkoxylation.11 In any case, a range of cyclizations performed
with the Cp*RuCl2/AgOTf/phosphane system using the
enantiopure ligands (R)-BINAP, (R)-TolBINAP, and (1R,SP)-
Josiphos gave exclusively racemic product in our hands (Table
3, entries 5−9). Because the in situ Cp*RuCl2-AgOTf-PR3
catalyst is evidently of the hidden acid type, it remained to be
shown whether some of its previously reported reactions might
be performed with a simple deliberate hidden acid catalyst for
preparative purposes. The results in Table 4 show that this is
Table 4. Application of Deliberate Hidden Acid Catalysts in
a
Catalytic Hydroalkoxylation Reactions
1
CDCl3 solution. The H NMR spectrum revealed again the
presence of [Cp*Ru(toluene)]OTf (10) as a single major Cp*-
containing substance. The 19F NMR spectrum displayed signals
for both the OTf− counterion (δ = −77.84 ppm) and HOTf (δ
= −78.55 ppm), though this may be due to the presence of a
heterogeneous HOTf droplet phase; see the Discussion. The
a
All reactions were performed for 1 h at 90 °C with 1 mol % AgOTf,
with allylphenol (1 mmol) or a mixture of alcohol (1 mmol) and
alkene (1.5 mmol) in 10 mL DCE (10 mL). Catalyst prepared from
Figure 1. 31P NMR spectrum of an in situ catalyst mixture prepared
from Cp*RuCl2, AgOTf, and (±)-BINAP.
b
AgOTf (1 mol %) and t-BuCl (4 mol %) in DCE at RT for 10 min.
c
Reaction performed in CH2Cl2 at 50 °C for 96 h.
31P NMR spectrum (Figure 1) displayed multiple signals, which
could be interpreted as belonging to the complex species
[Ag((±)-BINAP)2]Cl (δ = 15.59 ppm) and [{AgCl((±)-
BINAP)}2] (δ = 11.24, 10.25 ppm), which appear each as
characteristic combination doublets with couplings
indeed the case; catalytic hydroalkoxylations of allylphenols8 as
well as intermolecular hydroalkoxylations of norbornene9 were
performed with a hidden acid catalyst generated from AgOTf
and either DCE or t-BuCl. As compared to the original
procedures,8,9 which had used a loading of 2 mol % [Ru] and 4
mol % AgOTf at 70 or 85 °C over 48 or 18 h, respectively, the
reactions with the deliberate hidden acid catalysts were all
performed with a loading of 1 mol % AgOTf at 90 °C within 1
h.
1
1J(109Ag,31P) and J(107Ag,31P) to silver isotopic nuclei (Figure
1 and Figure S7, Supporting Information).72
The conclusion is that an in situ catalyst prepared from
Cp*RuCl2, AgOTf, and PR3 in hot toluene solution contains
HOTf, [Cp*Ru(η 6-toluene)]OTf (10), and several
[AgmClm(PR3)n] complexes. The HOTf must result from a
redox reaction involving a Cp*RuIII species as the oxidant (eq
1). Our spectral observations give no hint as to the identity of
the reductant (RH), but judging from color changes during the
in situ catalyst preparation, the redox process occurs prior to
addition of the phosphane and might thus involve the solvent
toluene as a reducing agent.
DISCUSSION
■
Numerous reports on metal-catalyzed hydroalkoxylation
reactions have appeared in recent years.1,5−32 The substrate
scope and selectivity observed with those catalysts remain in a
relatively narrow range and are hardly discernible from results
obtained with Brønsted acid catalysts. This could, in principle,
9356
dx.doi.org/10.1021/jo201631x|J. Org. Chem. 2011, 76, 9353−9361

The Journal of Organic Chemistry
Article
be due to a common mechanistic scheme involving Lewis-acid-
assisted Brønsted acid catalysis.1,15,29,73 On the other hand, it
has been implied that traces of acid from the hydrolysis of a
metal salt precursor might be responsible for the observed acid-
like catalytic activity of transition-metal salts.35,37,38,40,41 In
support of this interpretation, some groups have shown that the
results of apparent metal catalysts can be reproduced with
respect to both activity and selectivity by use of small amounts
of strong Brønsted acid. This approach does not necessarily
prove that Brønsted acid catalysis is operating under the
conditions of the catalytic reaction. In the present work, we
have chosen the alternative approach of proving that HOTf is
generated from metal triflate precursors under reaction
conditions according to specific mechanistic pathways and
acts as a hidden Brønsted acid catalyst for the catalytic reaction
in question. The method used for detecting HOTf was 19F
NMR spectroscopy. This technique was also successful for
mixtures containing HOTf and soluble organometallic complex
salts with OTf− as the counterion, which remarkably gave rise
to separate signals in apolar solvents, which may however be a
consequence of the “wet” conditions of analysis.74 In fact, the
NMR analysis of HOTf solutions under regular laboratory
conditions in air with moist solvent (a representative sample of
CDCl3 contained 274 ppm H2O according to Karl Fischer
titration) shows peculiarities; at concentrations of ∼1 μL HOTf
per mL of CDCl3, a sample will give rise to at least two peaks in
the 19F NMR spectrum, one relatively sharp peak at δ = −76.3
± 0.5 ppm corresponding to homogeneously dissolved HOTf
in the organic phase and at least one second, broadened peak at
δ = −79.5 ± 0.5 ppm, which is sometimes accompanied by
smaller, similarly broadened peaks (Figure 2). Visual
be a secure way to prove the presence of HOTf by 19F NMR
spectroscopy also in the presence of ionic triflates.74 The
observation of the droplet HOTf phase of dilute solutions of
HOTf in moist solvents coincides with our difficulties in
preparing dilute stock solutions of HOTf in DCE or similar
nonprotic solvents for the sake of performing control
experiments at low loadings of HOTf as the catalyst. Such
stock solutions readily separate the acid in the form of small
droplets at the bottom and walls of the vessel. Use of the
supernatant then leads to erroneous experimental results and
may be a cause for false negative control experiments in studies
of metal versus acid catalysis.
The mechanistic pathways leading to HOTf release are not
hydrolyses of metal aquo complexes35 but irreversible reactions
including eliminations (Schemes 2 and 3), Friedel−Crafts
alkylations (Scheme 3), and redox processes including the
solvent or substrates as reductants (eq 1). The oxidizing power
and the halophilicity of the silver cation is responsible for the
ease of all of those processes. Interestingly, similar arguments
apply to cationic gold(I) complexes, which are readily reduced
(thereby acting as a potential source of protic acid, similarly as
illustrated for Ru(III) in eq 1) and whose halophilicity might
even be superior to that of silver(I), if the solubility product of
AgCl (Ksp = 1.77 × 10−10) versus AuCl (Ksp = 2.0 × 10−13 75 is
)
taken as an indication of the strength of the metal/halogen
interaction. Cases of gold(I)-catalyzed hydroalkoxylations of
nonactivated alkenes are certainly suspicious of accompanying
hidden Brønsted acid catalysis. In several past studies, the
interference of Brønsted acid catalysis has been excluded by
performing control experiments, in which the Brønsted acid
gave qualitatively or quantitatively different results from the
metal catalyst in question. According to Hartwig and co-
workers, the use of Brønsted acid at loadings equivalent to the
metal catalyst in such control experiments induces polymer-
ization of the reaction product and can create the false
impression that Brønsted acid is a less efficient catalyst.38 We
would also argue that control experiments with Brønsted acids
often show inferior catalytic performance because the control
experiment hardly ever receives the amount of optimization
that the key metal triflate experiment receives. We propose that
such control experiments are more readily performed with
deliberate hidden acid catalysts under the conditions of the
original catalysis. Two simple protocols have been introduced
to generate such a hidden acid catalyst from AgOTf; the first
protocol (AgOTf + t-BuCl) produces HOTf in controlled
amounts at ambient temperature or below, whereas the second
protocol (AgOTf + DCE) slowly eliminates HOTf at elevated
temperatures (>60 °C). Both procedures show advantages in
synthetic applications because the hidden acid catalyst based on
AgOTf is easier to handle in minute amounts than pure HOTf.
Dilute solutions of HOTf in moist laboratory solvents may be
inhomogeneous and difficult to dose, whereas in situ generation
provides a finely dispersed form of HOTf.
Figure 2. 19F NMR spectra of HOTf in moist CDCl3.
observation indicates that the additional signals are due to a
separate droplet phase of HOTf, which also causes the turbid
appearance of the NMR sample. The use of SiMe4 as an
internal standard further complicates the analysis because
reaction of HOTf with SiMe4 forms TMSOTf (δ{1H} = 0.52
ppm, δ{19F} = −76.9 ppm) and methane (δ{1H} = 0.23 ppm).
At higher concentrations of HOTf (>20 μL HOTf per mL of
CDCl3), the HOTf droplets collect to form an acid phase at the
bottom of the NMR tube that escapes detection. The clear
supernatant displays a signal for dissolved HOTf (−76.3 ppm).
In more diluted samples (<0.5 μL per mL of CDCl3), the signal
for dissolved HOTf becomes smaller and shifts to lower
frequency, until a single peak for a water-rich HOTf droplet
phase at δ(19F) = −78.5 ± 0.5 ppm remains (0.2 μL HOTf/mL
CDCl3). The use of a solvent like CDCl3 or C6D6 in
combination with sufficiently high concentrations appears to
CONCLUSION
■
Examples of assumed transition-metal-catalyzed hydroalkox-
ylation reactions have been investigated and shown to be
catalyzed by hidden Brønsted acid. The mechanistic pathway of
HOTf generation under reaction conditions was elucidated by
spectroscopic analysis and by isolation of side products formed
in reactions of the metal precursor. An earlier claim of an
asymmetric catalytic hydroalkoxylation reaction of a non-
activated alkene has been refuted. 19F NMR spectroscopy is a
9357
dx.doi.org/10.1021/jo201631x|J. Org. Chem. 2011, 76, 9353−9361

The Journal of Organic Chemistry
Article
by NMR without further purification. See Figure S5 (Supporting
Information) for NMR spectra.
suitable tool for studying the stability of metal triflate catalysts
under the conditions of the catalytic reactions. This technique
may prove or disprove the generation of HOTf, if peculiarities
of the 19F NMR analysis of HOTf samples in moist air and
solvents are taken into account. If the catalytic activity of
Brønsted acids in hydroalkoxylation reactions is evaluated
against the activity of metal catalysts, control experiments can
be advantageously performed with a deliberate hidden acid
catalyst rather than with pure Brønsted acids to circumvent
problems arising from the handling of small amounts of strong
acids.
(c). Reaction of DCE with AgOTf. In a dry Schlenk tube under an
Ar atmosphere, AgOTf (100 mg, 389 μmol) was suspended in 0.5 mL
of DCE and heated to 85 °C for 120 min. After cooling to RT, a
sample (∼50 μL) of the reaction was filtered and diluted with CDCl3
for NMR analysis. A second sample was treated the same way with the
exception that it was mixed with a sample of freshly prepared TFA in
CDCl3. Compare Figures S2 and S3 (Supporting Information) for
spectra.
Synthetic Applications of Deliberate Hidden Acid Cata-
lysts. (a). Generation of a Hidden Acid Catalyst from AgOTf and
t-BuCl (General Procedure A). By means of a microsyringe, t-BuCl (4
mol %) was added to a stirred solution/suspension of AgOTf
(typically 1−2 mol %) in DCE (10 mL/mmol) or any other desired
solvent (CH2Cl2, benzene, toluene). The mixture was stirred for 10
min at RT before further use. See Figure S6 (Supporting Information)
for a visual impression.
(b). Generation of a Hidden Acid Catalyst from AgOTf in DCE
(General Procedure B). For a 1 mmol scale reaction, AgOTf (2.6 mg,
1 mol %) was added to DCE (10 mL/mmol) and heated with stirring
to 90 °C (oil bath temperature) in a pressure tube (or alternatively, to
reflux in an open system with refluxing condenser) for 3 h.
Phenol−Isoprene Condensations Catalyzed by Deliberate
Hidden Acid Catalysts (Table 2). These reactions employ excess
isoprene (1.5 equivs), which is added slowly to the reaction mixture.
Batch addition may lead to isoprene polymerization, while overly slow
addition can result in isoprenylation at C-6 of the chromane products.
Individual optimization for each new phenol is necessary for optimal
results.
EXPERIMENTAL SECTION
■
General. Reaction setup was carried out in air, but catalyses were
performed in Schlenk vessels under nitrogen. Preparative chromatog-
raphy (CC, column chromatography) was performed on silica gel 60
(0.040−0.063 mm) using pressurized air (0.1−0.3 bar) for flash
elution. 1H NMR and 13C NMR data are referenced to
tetramethylsilane as the internal standard. Unless otherwise
mentioned, chemicals were obtained from commercial suppliers and
used as received. Solvents were stored over molecular sieves in air.
Isolation of AgCl from the Reaction of AgOTf with 1,2-
Dichloroethane (DCE). A suspension of AgOTf (128.5 mg, 0.50
mmol) in DCE (5 mL) was heated in a pressure tube at 90 °C with
stirring for 3 h. The precipitate was filtered to afford AgCl as a white
powder (71 mg, 99%). See Figure S1 (Supporting Information) for a
photograph. Identification of AgCl: the colorless solid was insoluble in
water or dilute HNO3 but soluble in aqueous ammonia. A sample was
dissolved in diluted aqueous ammonia and warmed to 40 °C. Addition
of a few drops of aqueous formaldehyde induced a black precipitate of
metallic silver.
2,2-Dimethylchromane (4a). The catalyst solution was prepared
according to the procedure A from AgOTf (2.6 mg, 1 mol %) and t-
BuCl (4.5 μL, 4 mol %) in DCE (10 mL). Phenol (94 mg, 1 mmol)
was added to the suspension, followed by slow addition of a solution
of isoprene (150 μL, 1.5 mmol) in DCE (2 mL) over 10 min. The
mixture was stirred at RT for 20 min. The solvent was removed in
vacuo and the residue purified by flash column chromatography (SiO2,
n-hexane) to give product 4a (139 mg, 86%) as a colorless oil.
2,2-Dimethyl-8-isopropylchromane (4b). A catalyst solution
was prepared according to procedure A from AgOTf (5.2 mg, 2 mol
%) and t-BuCl (4.5 μL, 4 mol %) in DCE (10 mL). After addition of 2-
isopropylphenol (136 mg, 1.0 mmol), a solution of isoprene (150 μL,
1.5 mmol) in DCE (2 mL) was slowly added over 10 min to the
reaction mixture. After stirring for 30 min at RT, the solvent was
removed in vacuo and the residue purified by flash column
chromatography (SiO2, n-hexane) to give 4b (134 mg, 66%) as a
colorless oil. 1H NMR (360 MHz, CDCl3): δ = 1.20 (d, J = 6.9 Hz, 6
H), 1.32 (s, 6 H), 1.78 (t, J = 6.8 Hz, 2 H), 2.77 (t, J = 6.8 Hz, 2 H),
3.26 (sept, J = 6.9 Hz, 1 H), 6.77 (t, J = 7.5 Hz, 1 H), 6.87−6.92 (m, 1
H), 6.99−7.04 (m, 1 H) ppm. 13C NMR (90 MHz, CDCl3): δ = 22.5,
22.8, 26.9, 27.0, 32.7, 73.7, 119.1, 120.3, 123.6, 126.8, 136.5, 151.1
ppm. HRMS (EI) calcd for C14H20O+ 204.1509; found 204.1509.
8-tert-Butyl-2,2-dimethylchromane (4c). A catalyst solution
was prepared according to procedure B from AgOTf (2.6 mg, 1 mol
%) in DCE (10 mL). After cooling to RT, 2-tert-butylphenol (150 mg,
1.0 mmol) was added to the suspension. A solution of isoprene (150
μL, 1.5 mmol) in DCE (2 mL) was slowly added over 10 min and the
reaction mixture stirred at RT for 3 h. The solvent was removed in
vacuo and the residue purified by flash column chromatography (SiO2,
Generation of 2-Chloroethyl Triflate (1) and HOTf from the
Reaction of AgOTf with DCE. In a dry Schlenk tube under an Ar
atmosphere, AgOTf (100 mg, 389 μmol) was suspended in 0.5 mL of
DCE and heated to 80 °C for 2 h. After cooling to RT, a sample (∼50
μL) of the reaction was filtered and diluted with CDCl3 or C6D6 for
NMR analysis. 1H NMR (500 MHz, CDCl3): δ = 4.64 ppm, (t, J = 5.7
Hz; 1); 1H NMR (250 MHz, C6D6): δ = 3.62 ppm (t, J = 5.4 Hz; 1).
19F NMR (471 MHz, CDCl3): δ = −74.5 ppm (1), −78.0 ppm
(HOTf); 19F NMR (235 MHz, C6D6): δ = −74.9 ppm (1), −77.5
ppm (HOTf). See Figures S2 and S3 (Supporting Information) for
selected spectra.
Catalytic Phenol−Isoprene Condensation to 2,2-Dimethyl-
chromane (Table 1): General Procedure. The catalyst (usually 5
mol %) was added to a stirred solution of phenol (2a; 1 mmol) and
isoprene (3; 1.5 mmol) in DCE (5 mL). Reaction progress was
monitored by TLC analysis. After completion of the reation, the
solvent was evaporated and the residue separated by flash column
chromatography (SiO2, n-hexane) to give the product 4a as colorless
oil. See Table 1 for results.
2,2-Dimethylchromane (4a). 1H NMR (360 MHz, CDCl3): δ =
1.33 (s, 6 H), 1.79 (t, J = 6.8 Hz, 2 H), 2.77 (t, J = 6.8 Hz, 2 H), 6.76−
6.83 (m, 2 H), 7.03−7.10 (m, 2 H) ppm. 13C NMR (90 MHz,
CDCl3): δ = 22.6, 27.0, 32.9, 74.2, 117.4, 119.7, 121.0, 127.4, 129.6,
154.1 ppm. Known compound, CAS 1198-96-5.
Generation of Deliberate Hidden Acid Catalysts (NMR
Experiments). (a). Generation of HOTf in CDCl3 from t-BuCl +
AgOTf. In a dry Schlenk tube under an Ar atmosphere, AgOTf (23.3
mg, 91 μmol) was added to a solution of t-BuCl (10 μL, 91 μmol) in
CDCl3 (0.6 mL). The suspension was stirred vigorously for 10 min.
The colorless suspension was directly filtered into an NMR tube for
further analysis. See Figure S4 (Supporting Information) for the NMR
spectrum.
1
n-hexane) to give 4c (214 mg, 98%) as a colorless oil. H NMR (250
MHz, CDCl3): δ = 1.35 (s, 6 H), 1.38 (s, 9 H), 1.78 (t, J = 6.9 Hz, 2
H), 2.79 (t, J = 6.9 Hz, 2 H), 6.75 (t, J = 7.6 Hz, 1 H), 6.89−6.95 (m, 1
H), 7.10 (dd, J = 7.7, 1.6 Hz, 1 H) ppm. 13C NMR (62.5 MHz,
CDCl3): δ = 23.0, 27.0, 29.7, 32.7, 34.8, 73.8, 118.8, 120.9, 124.3,
127.5, 137.8, 152.5 ppm. HRMS (EI) calcd for C15H22O+ 218.1665;
found 218.1668.
6-Methoxy-2,2-dimethylchromane (4d). A catalyst solution
was prepared according to procedure B from AgOTf (5.2 mg, 2 mol
%) in DCE (10 mL). In the resulting suspension, 4-methoxyphenol
(124 mg, 1.0 mmol) was dissolved, and a solution of isoprene (150 μL,
(b). Generation of HOTf in C6D6 Using t-BuCl + AgOTf. In a dry
Schlenk tube under an Ar atmosphere, t-BuCl (10 μL, 91 μmol) was
dissolved in C6D6 (0.6 mL) and mixed with AgOTf (23.3 mg, 91
μmol). The suspension was stirred vigorously for 180 min while a
white precipitate appeared. The suspension was filtered and analyzed
9358
dx.doi.org/10.1021/jo201631x|J. Org. Chem. 2011, 76, 9353−9361

The Journal of Organic Chemistry
Article
1.5 mmol) in DCE (2 mL) was slowly added over 10 min. After
stirring the reaction mixture at RT for 1 h, the solvent was removed in
vacuo and the residue purified by flash column chromatography (SiO2,
degassed toluene/i-PrOH (1:1, 10 mL) was heated at 90 °C for 4 h
with stirring. The mixture was filtered and the filtrate concentrated
under reduced pressure. The residual solid was washed with Et2O and
1
1
hexanes to give a tan solid 11 (54 mg, 74%). H NMR (360 MHz,
n-hexane) to give 4d (114 mg, 59%) as a colorless oil. H NMR (360
CDCl3): δ = 2.02 (s, 15H), 2.23 (s, 3H), 5.87 (d, J = 5.8 Hz, 2H), 5.95
(br t, J = 5.5 Hz, 1H), 6.03 (br t, J = 5.6 Hz, 2H) ppm. 13C NMR
(90.6 MHz, CDCl3): δ = 11.0, 18.7, 87.0, 87.7, 88.5, 96.0, 100.0 ppm.
MS (FAB): m/z (%) = +329 (100) [Cp*Ru(C7H8)]+. Analysis calcd
for C17H23ClRu: C 56.11, H 6.37; found: C 56.39, H 6.64. Data agree
with literature values.70
Deliberate Hidden Acid Catalysis of Hydroalkoxylation
Reactions (Table 4): 2-Methylcoumaran (9). A catalyst solution
was prepared according to general procedure B from AgOTf (2.6 mg,
1 mol %) in DCE (10 mL). After addition of 2-allylphenol (134 mg,
1.0 mmol), the mixture was heated at 90 °C for 1 h. After cooling, the
solvent was removed in vacuo. The residue was purified by flash
column chromatography (SiO2, n-hexane) to give product 9 (119 mg,
89%) as a colorless oil.
MHz, CDCl3): δ = 1.31 (s, 6H), 1.78 (t, J = 6.8 Hz, 2H), 2.75 (t, J =
6.8 Hz, 2H), 3.74 (s, 3H), 6.59−6.62 (m, 1H), 6.65−6.72 (m, 2H)
ppm. 13C NMR (360 MHz, CDCl3): δ = 22.8, 26.7, 32.8, 55.7, 73.7,
113.3, 113.9, 117.7, 121.4, 147.9, 152.8 ppm. Known compound, CAS
69888-41-1.
Catalytic Cyclization of 2-Allylphenol to 2-Methylcoumaran
(Table 3). (a). Cyclization with Various Catalysts (Table 3, entries
1−3). A solution of the catalyst (5 mol %) and 8 (134 mg, 1.0 mmol)
in DCE (3 mL) was stirred under the conditions indicated in Table 3.
The solvent was evaporated and the product isolated by flash column
chromatography (SiO2, n-hexane).
2-Methylcoumaran (9). 1H NMR (360 MHz, CDCl3): δ = 1.46
(d, J = 6.3 Hz, 3H), 2.80 (dd, J = 15.4, 7.7 Hz, 1H), 3.29 (dd, J = 15.4,
8.8 Hz, 1H), 4.84−4.96 (m, 1H), 6.73−6.84 (m, 2H), 7.05−7.17 (m,
2H) ppm. 13C NMR (360 MHz, CDCl3): δ = 21.7, 37.1, 79.4, 109.3,
120.1, 124.9, 127.0, 127.9, 159.5 ppm. Known compound, CAS 1746-
11-8.
5,7-Di-tert-butyl-2-methylcoumaran (12). (a) A catalyst sol-
ution was prepared according to general procedure B from AgOTf (2.6
mg, 1 mol %) in DCE (10 mL). After addition of 2-allyl-4,6-di-tert-
butylphenol (1 mmol, 246 mg), the mixture was heated to 90 °C for 1
h. After cooling, the solvent was removed in vacuo, and the residue
was purified by flash column chromatography (SiO2, n-hexane) to give
the product (161 mg, 65%) as a colorless oil. (b) The reaction was
performed at a 1.03 mmol scale as under (a) but in CH2Cl2 as the
solvent with heating to 50 °C (closed vessel) over 4 days to give a
(b). Cyclization with In Situ Catalysts Derived from Cp*RuCl2-
AgOTf-phosphane (Table 3). In a dry Schlenk tube under an Ar
atmosphere, Cp*RuCl2 (6.1 mg, 0.02 mmol; 2 mol %) and AgOTf
(10.3 mg, 0.040 mmol; 4 mol %) were added subsequently into freshly
degassed toluene (3 mL). The resulting suspension was heated to 110
°C for 3 h. After cooling to RT, the ligand (PPh3, (R)-BINAP, (R)-
Tol-BINAP, or (1R,SP)-Josiphos; 2 mol %) and 2-allylphenol (134 mg,
1.0 mmol) were added. The mixture was heated at the temperature
and time indicated in Table 3. The reaction mixture was directly
chromatographed on SiO2 (pentane/Et2O = 50:1) to isolate pure
coumaran 9, which was identified by NMR and GC/MS. The
enantiomeric excess was analyzed by HPLC with UV detection
(Chiralcel OJ, hexane/i-propanol = 98:2, flow = 0.8 mL/min, T = 15
°C; UV detection at λ = 280 nm). No enantiomeric excess was
observed within the accuracy of the analytical method (≤1% ee). See
Table 3, entries 5−9, and the Supporting Information for results.
NMR Analysis of the Cp*RuCl2-AgOTf-BINAP In Situ
Catalyst. In a dry Schlenk tube under an Ar atmosphere, Cp*RuCl2
(13.3 mg, 43 μmol) and AgOTf (20.5 mg, 80 μmol) were added
subsequently into freshly degassed toluene (3 mL), and the mixture
was heated to 110 °C for 210 min. After cooling to RT, (±)-BINAP
(25 mg, 40 μmol) was added and the mixture stirred for 30 min at RT.
A sample of the mixture was filtered, diluted with CDCl3, and analyzed
by 1H, 19F, and 31P NMR. See Figure S7 (Supporting Information) for
NMR spectra.
1
colorless oil (203 mg, 80%). H NMR (250 MHz, CDCl3): δ = 1.29
(s, 9H), 1.36 (s, 9H), 1.43 (d, J = 6.2 Hz, 3H), 2.75 (ddt, J = 15.2, 7.5,
0.9 Hz, 1H), 3.24 (ddt, J = 15.2, 8.8, 0.9, 1H), 4.87 (qdd, 8.8, 7.5, 6.2
Hz, 1H), 7.04 (dt, J = 2.1, 1.0 Hz, 1H), 7.11 (dt, J = 2.1, 0.8 Hz, 1H).
13C NMR (63 MHz, CDCl3): δ = 21.9, 29.3, 31.9, 34.3, 34.4, 37.2,
78.6, 119.4, 121.5, 126.7, 131.7, 142.5, 155.2 ppm. HRMS (EI): calcd
for C17H26O+ 246.1978; found 246.1972.
2-exo-Phenyloxybicyclo[2.2.1]heptane (13). A catalyst solu-
tion was prepared according to general procedure B from AgOTf (2.6
mg, 1 mol %) in DCE (10 mL). After addition of phenol (94 mg, 1.0
mmol) and norbornene (141 mg, 1.5 mmol), the reaction mixture was
heated at 90 °C for 1 h. The solvent was removed in vacuo and the
residue purified by flash column chromatography (SiO2, n-hexane) to
1
give the product (97%, 0.182 g) isolated as a colorless oil. H NMR
(360 MHz, CDCl3): δ = 1.08−1.22 (m, 3 H), 1.41−1.79 (m, 5 H),
2.31 (t, J = 4.1 Hz, 1 H), 2.45 (d, J = 4.2, 1 H), 4.15 (d, J = 6.7, 1 H),
6.82−6.93 (m, 3 H), 7.21−7.27 (m, 2 H). 13C NMR (91 MHz,
CDCl3): δ = 24.3, 28.5, 35.2, 35.4, 40.0, 41.1, 79.9, 115.4, 120.1, 129.3,
157.7 ppm. Data are consistent with the exo-stereoisomer.76 Known
compound, CAS 50414-48-7.
Synthesis of Ruthenium(II) Complexes from Cp*RuCl2: η5-
Pentamethylcyclopentadienyl(η 6-toluene)ruthenium(II) Tri-
fluoromethanesulfonate (10). (a). Direct Synthesis from 7. A
mixture of Cp*RuCl2 (7; 18 mg, 0.059 mmol) and AgOTf (30.9 mg,
0.12 mmol) in degassed toluene/i-PrOH (1:1, 4 mL) was heated at 90
°C for 4 h with stirring. The mixture was filtered to remove the
precipitate (AgCl), and the filtrate was concentrated under reduced
pressure. The residual solid was washed with Et2O and hexanes to give
19 mg of tan solid 10 (68% yield).
2-exo-Benzyloxybicyclo[2.2.1]heptane (14). A catalyst solu-
tion was prepared according to procedure A from AgOTf (2.6 mg, 1
mol %) and tert-butylchloride in DCE (10 mL). Benzyl alcohol (108
mg, 1 mmol) and norbornene (141 mg, 1.50 mmol) were added with
stirring, and the mixture was heated to 90 °C for 1 h. The solvent was
removed in vacuo and the residue purified by flash column
chromatography (SiO2, n-hexane) to give product 2d (117 mg,
1
58%) as a colorless oil. H NMR (360 MHz, CDCl3): δ = 0.91−1.16
(b). Synthesis of 10 by Chloride Abstraction from 11. A mixture
of 11 (30 mg, 0.082 mmol) and AgOTf (26 mg, 0.10 mmol) in
degassed toluene/i-PrOH (1:1, 6 mL) was stirred at RT for 4 h. The
mixture was filtered to remove the precipitate (AgCl), and the filtrate
was concentrated under reduced pressure. The residual solid was
washed with Et2O and hexanes to give tan solid 10 (38 mg, 98% yield).
1H NMR (360 MHz, CDCl3): δ = 1.97 (s, 15H), 2.19 (s, 3H), 5.73 (d,
(m, 3 H), 1.36−1.67 (m, 5 H), 2.24 (br. s, 1 H), 2.37 (d, J = 4.6 Hz, 1
H), 3.44 (dt, J = 6.7, 1.1 Hz, 1 H), 4.43 (d, J = 11.9 Hz, 1 H), 4.48 (d,
J = 11.9 Hz, 1 H), 7.20−7.35 (m, 5 H) ppm. 13C NMR (90 MHz,
CDCl3): δ = 24.6, 28.6, 34.9, 35.2, 39.6, 40.4, 70.1, 82.1, 127.3, 127.5,
128.3, 139.1 ppm. Known compound.17
ASSOCIATED CONTENT
J = 6.0 Hz, 2H), 5.75−5.87 (m, 3H) ppm. 13C NMR (90.6 MHz,
■
CDCl3): δ = 10.6, 18.5, 86.6, 87.3, 88.2, 96.1, 100.0, 120.9 (q, JFC
=
S
* Supporting Information
320.6 Hz) ppm. MS (FAB): m/z (%) = +329 (100) [Cp*Ru(C7H8)]+,
−149 (100) OTf−. Analysis calcd for C18H23F3O3RuS: C 45.28, H
4.85, S 6.72; found: C 45.19, H 4.77, S 6.95. Spectral data in
agreement with the literature values.66,67
NMR spectra and photographs of key experiments, H and 13
C
1
NMR spectra of pure reaction products, and additional data
including HPLC chromatograms for attempted asymmetric
experiments. This material is available free of charge via the
Internet at http://pubs.acs.org.
η 5-Pentamethylcyclopentadienyl(η 6-toluene)ruthenium(II)
Chloride (11). A suspension of Cp*RuCl2 (7; 62 mg, 0.20 mmol) in
9359
dx.doi.org/10.1021/jo201631x|J. Org. Chem. 2011, 76, 9353−9361

The Journal of Organic Chemistry
Article
(35) Wabnitz, T. C.; Yu, J. Q.; Spencer, J. B. Chem.Eur. J. 2004, 10,
AUTHOR INFORMATION
Corresponding Author
*E-mail: lukas.hintermann@tum.de.
■
484.
(36) Coulombel, L.; Dunach, E. Green. Chem. 2004, 6, 499.
̃
(37) Li, Z.; Zhang, J.; Brouwer, C.; Yang, C. G.; Reich, N. W.; He, C.
Org. Lett. 2006, 8, 4175.
(38) Rosenfeld, D. C.; Shekhar, S.; Takemiya, A.; Utsunomiya, M.;
Hartwig, J. F. Org. Lett. 2006, 8, 4179.
(39) Compare: Taylor, J. G.; Adrio, L. A.; Hii, K. M. Dalton Trans.
2010, 39, 1171.
ACKNOWLEDGMENTS
This work was supported through an Alexander von Humboldt
fellowship to TTD.
■
(40) For additional studies addressing the question of metal versus
acid catalysis, see: McKinney Brooner, R. E.; Widenhoefer, R. A.
Chem.Eur. J. 2011, 17, 6170.
REFERENCES
■
(1) Hintermann, L. Top. Organomet. Chem. 2010, 31, 123 and cited
literature.
(2) Beller, M.; Seayad, J.; Tillack, A.; Jiao, H. Angew. Chem., Int. Ed.
2004, 43, 3368.
(41) Tschan, M. J. L.; Thomas, C. M.; Strub, H.; Carpentier, J. F.
Adv. Synth. Catal. 2009, 351, 2496.
(42) Bandini, M.; Eichholzer, A.; Kotrusz, P.; Tragni, M.; Troisi, S.;
Umani-Ronchi, A. Adv. Synth. Catal. 2009, 351, 319.
(43) Youn, S. W.; Eom, J. I. J. Org. Chem. 2006, 71, 6705.
(44) Harrison, T. J.; Dake, G. R. Org. Lett. 2004, 6, 5023.
(45) Deng, C.-L.; Zou, T.; Wang, Z.-Q.; Song, R.-J.; Li, J.-H. J. Org.
Chem. 2009, 74, 412.
(46) Arcadi, A.; Alfonsi, M.; Marinelli, F. J. Organomet. Chem. 2007,
692, 5322.
(47) Silver in Organic Chemistry; Harmata, M., Ed.; Wiley: New York,
2010.
(48) See the Supporting Information for additional data and NMR
(3) Alonso, F.; Beletskaya, I. P.; Yus, M. Chem. Rev. 2004, 104, 3079.
(4) Tani, K.; Kataoka, Y. In Catalytic Heterofunctionalization; Togni,
A., Grutzmacher, H., Eds.; Wiley-VCH: Weinheim, Germany, 2001; p
̈
171.
(5) Hori, K.; Kitagawa, H.; Miyoshi, A.; Ohta, T.; Furukawa, I. Chem.
Lett. 1998, 1083.
(6) Marotta, E.; Foresti, E.; Marcelli, T.; Peri, F.; Righi, P.; Scardovi,
N.; Rosini, G. Org. Lett. 2002, 4, 4451.
(7) Loh, T.-P.; Hu, Q.-Y.; Ma, L.-T. J. Am. Chem. Soc. 2001, 123,
2450.
(8) Oe, Y.; Ohta, T.; Ito, Y. Chem. Commun. 2004, 1620.
(9) Oe, Y.; Ohta, T.; Ito, Y. Synlett 2005, 179.
(10) Ohta, T.; Kataoka, Y.; Miyoshi, A.; Oe, Y.; Furukawa, I.; Ito, Y.
J. Organomet. Chem. 2007, 692, 671.
(11) Ito, Y.; Kato, R.; Hamashima, K.; Kataoka, Y.; Oe, Y.; Ohta, T.;
Furukawa, I. J. Organomet. Chem. 2007, 692, 691.
(12) Oe, Y.; Ohta, T.; Ito, Y. Tetrahedron Lett. 2010, 51, 2806.
spectra.
(49) Falzon, C. L.; Ackermann, U.; Spratt, N.; Tochon-Danguy, H.
J.; White, J.; Howells, D.; Scott, A. M. J. Labelled Compds. Radiopharm.
2006, 49, 1089.
(50) 2-Chloroethyltriflate is described as sensitive to hydrolysis,
fuming in air: Lown, J. W.; Joshua, A. V.; McLaughlin, L. W J. Med.
Chem. 1980, 23, 798.
(13) Coulombel, L.; Favier, I.; Dunach, E. Chem. Commun. 2005,
̃
2286.
(51) Howells, R. D.; Mc Cown, J. D. Chem. Rev. 1977, 77, 69.
(52) Chapman, R. D.; Andreshak, J. L.; Herrlinger, S. P.; Shackelford,
S. A.; Hildreth, R. A.; Smith, J. P. J. Org. Chem. 1986, 51, 3792.
(53) Chapman, R. D.; Andreshak, J. L.; Shackelford, S. A. J. Org.
Chem. 1988, 53, 3771.
(54) Bonrath, W.; Haas, A.; Hoppmann, E.; Netscher, T.; Pauling,
H.; Schager, F.; Wildermann, A. Adv. Synth. Catal. 2002, 344, 37.
(55) The addition of a small quantity of HOTf to a mixture of 4-
methoxyphenol and isoprene (1:2 ratio) dissolved in CDCl3 in an
NMR tube produced an immediate exotherm. A 1H NMR spectrum of
the sample, measured within 1 h, showed peaks of the
methoxychromane with no more remaining signals for isoprene.
(56) The reaction conditions with the metal triflates are not
optimized. The results appear to depend as much on solubility of the
precatalysts as on the nature of the cation. For other Lewis-acid-
catalyzed isoprene/phenol condensations, see refs 57−60.
(57) Youn, S. W. Synlett 2007, 3050.
(58) Adrio, L. A.; Hii, K. K. Chem. Commun. 2008, 2325.
(59) Bigi, F.; Carloni, S.; Maggi, R.; Muchetti, C.; Rastelli, M.;
Sartori, G. Synthesis 1998, 301.
(60) De Felice, V.; De Renzi, A.; Funicello, M.; Panunzi, A.; Saporito,
A. Gazz. Chim. Ital. 1985, 115, 13.
(61) The reaction of a silver salt AgY with t-BuX is an elegant
method to generate strong protic acids HY in solution, particularly in
cases where the pure substance HY is unstable; see refs 62 and 63.
(62) Jona, H.; Mandai, H.; Chavasiri, W.; Takeuchi, K.; Mukaiyama,
T. Bull. Chem. Soc. Jpn. 2002, 75, 291.
(14) Lemechko, P.; Grau, F.; Antoniotti, S.; Dunach, E. Tetrahedron
̃
Lett. 2007, 48, 5731.
(15) Coulombel, L.; Rajzmann, M.; Pons, J. M.; Olivero, S.; Dunach,
̃
E. Chem.Eur. J. 2006, 12, 6356.
(16) Kelly, B. D.; Allen, J. M.; Tundel, R. E.; Lambert, T. H. Org.
Lett. 2009, 11, 1381.
(17) Taylor, J. G.; Whittall, N.; Hii, K. K. M. Chem. Commun. 2005,
5103.
(18) Chaminade, X.; Coulombel, L.; Olivero, S.; Dunach, E. Eur. J.
̃
Org. Chem. 2006, 3554.
(19) Yang, C. G.; He, C. J. Am. Chem. Soc. 2005, 127, 6966.
(20) Reich, N. W.; Yang, C.; Shi, Z.; He, C. Synlett 2006, 1278.
(21) Li, Z.; Zhang, J.; Brouwer, C.; Yang, C. G.; Reich, N. W.; He, C.
Org. Lett. 2006, 8, 4175.
(22) Grant, V. H.; Liu, B. Tetrahedron Lett. 2005, 46, 1237.
(23) Yang, C. G.; Reich, N. W.; Shi, Z.; He, C. Org. Lett. 2005, 7,
4553.
(24) Qian, H.; Han, X.; Widenhoefer, R. A. J. Am. Chem. Soc. 2004,
126, 9536.
(25) Liu, C.; Bender, C. F.; Han, X.; Widenhoefer, R. A. Chem.
Commun. 2007, 3607.
(26) Komeyama, K.; Morimoto, T.; Nakayama, Y.; Takaki, K.
Tetrahedron Lett. 2007, 48, 3259.
(27) Choi, J. C.; Kohno, K.; Masuda, D.; Yasuda, H.; Sakakura, T.
Chem. Commun. 2008, 777.
(28) Dzudza, A.; Marks, T. J. Org. Lett. 2009, 11, 1523.
(29) Dzudza, A.; Marks, T. J. Chem.Eur. J. 2010, 16, 3403.
(30) Zhang, X.; Corma, A. Chem. Commun. 2007, 3080.
(31) Vece, V.; Ricci, J.; Poulain-Martini, S.; Nava, P.; Carissan, Y.;
(63) Kato, T.; Tanaka, M.; Hoshikawa, M.; Yagi, M. Tetrahedron Lett.
1998, 39, 7553.
(64) Claims regarding the use of in situ prepared t-BuOTf can be
found in the literature, but this compound has only been characterized
by NMR spectroscopy at −60 °C and decomposes above −30 °C with
generation of polyisobutene: Hommeltoft, S. I.; Ekelund, O.; Zavilla,
J. Ind. Eng. Chem. Res. 1997, 36, 3491.
Humbel, S.; Dunach, E. Eur. J. Org. Chem. 2010, 6239.
(32) Adrio, L. A.; Quek, L. S.; Taylor, J. G.; Hii, K. K. M. Tetrahedron
2009, 65, 10334.
(33) Hirai, T.; Hamasaki, A.; Nakamura, A.; Tokunaga, M. Org. Lett.
2009, 11, 5510.
̃
(65) The monomeric formula Cp*RuCl2 is used to avoid confusion
regarding stoichiometric ratios. Complex 7 is a dimer in the solid state:
(34) Antoniotti, S.; Poulain-Martini, S.; Dunach, E. Synlett 2010,
2973.
̃
9360
dx.doi.org/10.1021/jo201631x|J. Org. Chem. 2011, 76, 9353−9361

The Journal of Organic Chemistry
Article
Kolle, U.; Kossakowski, J.; Klaff, N.; Wesemann, L.; Englert, U.;
̈
Heberich, G. E. Angew. Chem., Int. Ed. Engl. 1991, 30, 690.
(66) Masuda, K.; Ohkita, H.; Kurumatani, S.; Itoh, K. Organometallics
1993, 12, 2221.
(67) Fagan, P. J.; Ward, M. D.; Calabrese, J. C. J. Am. Chem. Soc.
1989, 111, 1698.
(68) (a) Kaganovich, V. S.; Kudinov, A. R.; Rybinskaya, M. I. Russ.
Chem. Bull. 1986, 35, 456. (b) Kaganovich, V. S.; Kudinov, A. R.;
Rybinskaya, M. I. Izv. Akad. Nauk SSSR, Ser. Khim. 1986, 2, 492.
(69) Wheeler, D. E.; Hill, S. T.; Williams, J. D.; Wedyck, E. J. Inorg.
Chim. Acta 1995, 231, 133.
(70) Suzuki, H.; Omori, H.; Lee, D. H.; Yoshida, Y.; Morooka, Y.
Organometallics 1988, 7, 2243.
́
(71) Bardajı, M.; Crespo, O.; Laguna, A.; Fischer, A. K. Inorg. Chim.
Acta 2000, 304, 7.
(72) The chemical shifts and J(107/109Ag,31P) coupling constants are
characteristic for complexes of the indicated stoichiometry; the
[{AgCl(BINAP)}2] complex is presumably occurring in meso- and
rac-forms. For reference data in a AgPF6/BINAP system, see:
Ohkouchi, M.; Masui, D.; Yamaguchi, M.; Yamagishi, T. J. Mol.
Catal. A: Chem. 2001, 170, 1.
(73) Yamamoto, H.; Futatsugi, K. Angew. Chem., Int. Ed. 2005, 44,
1924.
(74) Performing a vacuum transfer of volatiles and recording of a 19
F
NMR spectrum of the distillate is another method to prove the
presence of HOTf in a mixture with ionic triflate.29
(75) Speight, J. G. Lange’s Handbook of Chemistry, 16th ed.; McGraw-
Hill: New York, 2005.
(76) Lajunen, M.; Himottu, M. Acta Chem. Scand. 1989, 43, 957.
9361
dx.doi.org/10.1021/jo201631x|J. Org. Chem. 2011, 76, 9353−9361