Benzimidazole-pyrrolidine/H+ (BIP/H+), a highly reactive organocatalyst for asymmetric processes

Article abstract of DOI:10.1002/ejoc.200600664

A new chiral benzimidazole-pyrrolidine has been devised, which exhibits excellent activities in aminocatalyzed aldol reactions, leading to aldol products in high yields and enantioselectivities in the presence of an equimolar amount of a Broensted acid. This organocatalyst has demonstrated remarkable reactivities in aldol processes even with equimolar amounts of aldehyde and ketone in THF. A discussion of the role of the Broensted acid as a co-catalyst is provided along with some applications of this new class of organocatalyst in Robinson annelation and α-amination processes. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejoc.200600664

FULL PAPER
DOI: 10.1002/ejoc.200600664
Benzimidazole-pyrrolidine/H+ (BIP/H+), a Highly Reactive Organocatalyst for
Asymmetric Processes
Eric Lacoste,[a] Emilie Vaique,[a] Muriel Berlande,[a] Isabelle Pianet,[a]
Jean-Marc Vincent,*[a] and Yannick Landais*[a]
Keywords: Organocatalysis / Aldol reactions / Amination / Robinson annelation / Proline / 1-Silacyclohexan-4-one
A new chiral benzimidazole-pyrrolidine has been devised,
which exhibits excellent activities in aminocatalyzed aldol
reactions, leading to aldol products in high yields and
enantioselectivities in the presence of an equimolar amount
amounts of aldehyde and ketone in THF. A discussion of the
role of the Brönsted acid as a co-catalyst is provided along
with some applications of this new class of organocatalyst in
Robinson annelation and α-amination processes.
of a Brönsted acid. This organocatalyst has demonstrated re- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
markable reactivities in aldol processes even with equimolar
Germany, 2007)
the transition state between the acidic moiety of the catalyst
and the developing alkoxide plays a crucial role in the stabi-
lisation of the transition state.[5] Consequently, proline-
based catalysts with more acidic groups or capable of mul-
tiple hydrogen-bond interactions with the aldehyde have
been prepared and were shown to exhibit significant rate
accelerations and improved enantioselectivities.[6] However,
these approaches, which mainly focus on acceptor acti-
vation, did not succeed in reducing the amount of ketone
used in the aldol process. Thus, other strategies should be
envisioned that will not only focus on aldehyde activation
but on the ketone activation as well to favour the formation
of the iminium/enamine intermediates.
Introduction
Organocatalyzed reactions have been the subject of re-
newed interest[1] over the last six years and are now part of
the organic chemistry armoury, as demonstrated by some
recent applications in the total synthesis of natural prod-
ucts.[2] The aldol process was at the origin of this revival
and is still the subject of intense research, as it represents an
effective methodology for the formation of carbon–carbon
bonds and a useful test reaction for the design of new cata-
lysts. Proline was soon recognized as an attractive catalyst
in this context and may be used in a number of cases.[3]
However, the tendency of proline to form unreactive oxazo-
lidinone[4] in significant amounts by reaction with the alde-
hyde or the ketone often led to the utilization of substoichi-
ometric amounts of catalyst (20–30 mol-%). Moreover, due
to slow reaction rates, a large amount of ketone is generally
used (20–27 equiv. or neat), which renders the method un-
suitable for an extension of the methodology to function-
alized and/or costly ketones. Finally, although proline pro-
vides satisfactory results in terms of regio- and enantio-
selectivity in a number of processes including aldol and
Michael addition reactions, there are still transformations
and substrates where it leads to disappointing results. This
suggested the need for more efficient catalysts possessing a
tuneable structure. Most of the strategies developed so far
to improve the catalytic activity of proline derivatives were
based on the assumption that the hydrogen bond formed at
We recently reported on the design, synthesis and appli-
cation of a new and simple proline surrogate, the benzimid-
azole-pyrrolidine (= BIP) (1) in asymmetric, organocata-
lyzed, aldol reactions (Scheme 1).[7] Preliminary results
showed that 1, when activated by one equiv. of a Brönsted
[a] University Bordeaux-I, Laboratoire de Chimie Organique et
Organométallique,
351, Cours de la Libération, 33405 Talence Cedex, France
Fax: +33-5-40006286
E-mail: jm.vincent@lcoo.u-bordeaux1.fr
y.landais@lcoo.u-bordeaux1.fr
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Scheme 1.
Eur. J. Org. Chem. 2007, 167–177
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
167

J.-M. Vincent, Y. Landais et al.
FULL PAPER
acid such as trifluoroacetic acid (TFA), affords aldol prod- source (AcOH, Table 1, Entry 3) effectively enhanced the
ucts in high enantioselectivity under real catalytic condi- rate of the reaction, in good agreement with recent studies
tions (2–5 mol-%), with a stoichiometric amount of the by Barbas,[8a] who emphasized the role of an added
ketone.[8] We provide here a full account of this work, in- Brönsted acid to accelerate the formation of the enamine in
cluding further examples of applications of 1/H+ as a cata- pyrrolidine-catalyzed aldol processes between aldehydes.[8]
lyst in the aldol reaction and Robinson annelation as well as Interestingly, in our case, a significant improvement of the
mechanistic insights accounting for the increased reactivity enantioselectivity was also observed. We soon found that
observed.
the strength of the acid had a major impact on both the
yield and enantioselectivity, as shown by the results ob-
served with TFA, triflic acid and heptadecafluorononanoic
acid (Table 1, Entries 4–6). In neat acetone, the aldol prod-
uct 4, possessing the (R) configuration similar to that ob-
tained with -proline,[3a] was obtained in 87% yield and
82% ee in 24 h with a catalyst loading of only 2 mol-%
(Table 1, Entry 7). More importantly, under optimized con-
ditions, the reaction was carried out in THF at 10 °C (to
avoid the formation of the corresponding α,β-unsaturated
ketone) with only one equiv. of acetone and 5 mol-% of
the catalyst to afford 4 in good yield and enantioselectivity
(Table 1, Entry 8). This result clearly demonstrates the high
reactivity of the 1/H+ couple, which allowed a significant
decrease in the quantity of both the catalyst and ketone.
Lewis acids such as Cu(OAc)2 or Zn(OTf)2 (Table 1, Entries
9 and 10) were also tested with success, with higher enantio-
selectivity and yield observed with the more Lewis-acidic
Zn salt,[10] demonstrating that 1 might also be a useful cata-
lyst for enantioselective metal-catalyzed processes other
than aldol reactions. Finally, the use of acidic hexafluoro-
2-propanol[11] (HFIP, Table 1, Entry 11) led to a relatively
slow reaction and poor yield, indicating that such a solvent
likely disrupts hydrogen bonds necessary for the aminoca-
talysis.
Results and Discussion
Synthesis of BIP (1)
Proline, which has been used with success in a number
of cases, is a readily available catalyst whose cyclic structure
has a strong influence on the stereochemical outcome of
the aldol process.[3a–3b] It is also a bifunctional catalyst,
which acts both as a Lewis base and Brönsted acid. Thus,
we envisioned that both points should be taken into ac-
count in the design of a catalyst. BIP was prepared in 55%
yield by a simple condensation of proline with a bisaniline
2 under acidic conditions (Scheme 2). High conversion was
achieved within 6–7 d, and the condensation occurs without
racemisation. It is also worth noting that the methyl groups
probably enhance the nucleophilicity of the aromatic amino
groups and also improve the solubility of the catalyst in
organic solvents. X-ray structure determination confirmed
the structure of 1 and provided useful information.[7] For
instance, it was observed that two molecules of 1 crystallize
in the presence of one molecule of water with the electron
lone pairs of the oxygen atom linked to 1 through two hy-
drogen bonds involving pyrrolidine and benzimidazole N–
H bonds with bond lengths of 1.87 Å and 2.71 Å, respec-
tively. This suggests a possible pathway for activation of the
electrophile by the formation of hydrogen bonds.
Mechanistic Considerations
The results on aldol processes gathered in Table 1 clearly
show the prominent role of added Brönsted acid with our
organocatalyst 1. Rather unexpectedly and in marked con-
trast with 1, we have also observed that addition of a pro-
ton source (TFA) in the aldol reaction catalyzed by -pro-
line led to no aldol adduct, even after a prolonged reaction
period (Table 1, Entry 12), pointing to a specific behaviour
of the less acidic pyrrolidine-benzimidazole precatalyst. In
line with these results and those previously reported by
Barbas et al.[8a] and more recently by Peng et al.,[12] we have
proposed the following mechanism to account for the high
Scheme 2.
Aldol Reactions Catalyzed by BIP
Preliminary investigations were carried out with acetone reactivity of 1/H+ (Figure 1). We also believe that according
as a donor and p-nitrobenzaldehyde as the acceptor. The to a wealth of literature data,[3,5] this mechanism is likely to
aldol reaction with 30 mol-% of 1 in DMSO led to a modest be applicable to other aminocatalyzed processes (such as α-
enantioselectivity as compared to that obtained with pro- amination, vide infra).
line under similar conditions (Table 1, Entries 1–2).[3a] We
Addition of a Brönsted acid (AcOH, TFA or triflic acid)
attributed this poor result to the absence of an acidic pro- to precatalyst 1 should lead to the protonation of the more
ton in 1. Indeed, while 1 can be considered a nitrogenated basic pyrrolidine ring (pKa ≈ 11–12)[13] to form intermediate
1
analogue of -proline, a major difference is that the benz- I (Figure 1). The H NMR spectrum (see the Supporting
imidazole proton (pKa ≈ 12.3)[9] is much less acidic than Information) of a mixture of 1 and TFA clearly exhibits a
that of the carboxylic acid of proline (pKa ≈ 4.75), leading pronounced deshielding effect of 0.75 ppm for proton H2
to a less stabilized transition state. The addition of a proton of the pyrrolidine ring. In contrast, benzimidazole protons
168
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 167–177

(BIP/H+), a Highly Reactive Organocatalyst
FULL PAPER
Table 1. Optimizing the reaction conditions with 1/acid.
Entry
Cat./acid
mol-%
Solvent
Acetone [equiv.]
Time [h]
T [°C]
Yield [%][a]
ee [%]
1
2
3
4
5
6
7
8
-proline/–
1/–
30
30
20
20
20
20
2
DMSO
DMSO
acetone
acetone
acetone
acetone
acetone
THF
25
25
–
–
–
–
–
1.1
25
–
1.1
25
4
8
4
20
20
–30
–35
–20
–20
–5
10
20
20
15
58
40
78
91
84
97
87
85
24
87
21
–
76[b]
44[b]
62[b]
78[b]
80[b]
76[b]
82[b]
78[c]
16[b]
74[b]
15[c]
–
1/AcOH
1/TFA
24
24
3
24
24
24
17
48
24
1/CF3SO3H
1/C8F17CO2H
1/TFA
1/TFA
5
9
1/Cu(OAc)2
1/Zn(OTf)2
1/TFA
30
20
5
DMSO
acetone
HFIP
10
11
12
-proline/TFA
20
DMSO
20
1
[a] Isolated yield of 4. [b] Enantiomeric excess estimated from H NMR of the corresponding Mosher’s ester of 4. [c] ee determined by
HPLC with a Chiralcel AD-H® chiral column.
Figure 1. Proposed mechanism for the 1/acid-catalyzed aldol reaction.
are little affected by the addition of a proton source, sup- proton source (protonated BIP I) activates the ketone to-
porting the protonation of the pyrrolidine ring to generate ward addition of the secondary amine. In order to rule out
I. The second step is the formation of the iminium II the possibility of formation of the enamine on the benz-
through nucleophilic addition of the pyrrolidine (or the imidazole ring, we carried out the condensation between
benzimidazole) onto the ketone, which should then lead to acetone and 1 in the presence of NaBH3CN to trap the
the key intermediate enamine III. In the second step, the iminium intermediate.[14] This led exclusively to the forma-
Eur. J. Org. Chem. 2007, 167–177
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
169

J.-M. Vincent, Y. Landais et al.
FULL PAPER
tion of the N-isopropylpyrrolidine 5 in 65% isolated yield non–equivalent, leading to the splitting of their signals. The
(Scheme 3),[7] indicating that iminium II is an intermediate complete assignment of the resonances of 6 has been
in the process and that the benzimidazole moiety is not nu- achieved through a range of NMR experiments including
cleophilic enough to generate an iminium.[15]
13C, TOCSY, HSQC and HMBC (see the Supporting Infor-
mation). Thus, the reactivity of 1 is similar to that of pro-
line, which was shown to react with ketones or aldehydes
to give the corresponding bicyclic oxazolidinones.[4] It has
been proposed that such parasitic consumption of the cata-
lyst was partially responsible for the low reactivity of pro-
line. Interestingly, a different reactivity was observed with
the more active 1/H+ species I (Figure 1). As shown in Fig-
ure 2, two species are clearly present (≈ 1:1 ratio) after
30 min of reaction, unreacted I and another species the
NMR (13C DEPT135, TOCSY, HSQC and HMBC) and
Scheme 3.
Interestingly, the proton generated through sequence II ES-MS data (see the Supporting Information) of which are
Ǟ III may then protonate either the enamine or the benz- in agreement with the structure of iminium II. From the 1H
imidazole ring to give the corresponding enammonium IIIЈ NMR spectrum, it is clear that the second species is not the
(pKa ≈ 4) or benzimidazolium III (pKa ≈ 5.4), which should benzimidazolidine 6 (only one resonance is observed for the
be in equilibrium. However, we propose that the active in- benzimidazole protons), while the enamine structure can be
termediate is the benzimidazolium III and not the enammo- ruled out, as the characteristic 13C resonances of the ethyl-
nium IIIЈ, which should not be nucleophilic enough to ac- enic group are not observed in the 13C NMR spectrum (13C
count for a high reactivity. Additionally, the benzimidazol- and DEPT135 spectra are given in the Supporting Infor-
ium ring in III can form relatively strong hydrogen bonds mation). Instead, the highly deshielded proton resonance at
with the aldehyde,[16] leading to a more stabilized chair-like δ = 6.24 ppm was assigned through HSQC and TOCSY
transition state IV and, consequently, to higher enantio- experiments to the –CH– of the pyrrolidine ring in II.
selectivity. Nucleophilic attack of the enamine on the re- Moreover, the quaternary carbon of iminium II was ob-
face of the aldehyde (with the largest aldehyde substituent served at δ = 160.9 ppm in the 13C spectrum, giving a broad
in a pseudo-equatorial arrangement),[5] followed by proton signal due to the couplings with the deuterium and nitrogen
transfer from the benzimidazolium ring would afford the atoms (see the Supporting Information). Importantly, the
iminium V, which upon in situ hydrolysis, would lead to ESI-MS spectrum of an acetone/1/H+ solution displayed an
an aldol having predominantly the (R) configuration. The intense peak at m/z = 256.2 (100%), which corresponds to
strength of the acid added at the beginning of the process the molecular weight of II (see the Supporting Infor-
is clearly central to this aminocatalysis. The addition of mation). Overall, this preliminary study clearly shows that
strongly acidic TFA (pKa ≈ –0.3) should ensure an efficient the addition of one equiv. of protons to 1 favours the for-
protonation of the benzimidazole ring in intermediate III. mation of reactive intermediates such as the iminium salt
On the contrary, the addition of weak acid such as AcOH II.
(pKa ≈ 4.8) provides only a partial protonation of the benz-
Having demonstrated the prominent role of acids as co-
imidazole ring, leading to a less efficient catalysis, as indi- catalysts, we then investigated the stereochemical issue of
cated by lower enantioselectivity. It is also interesting to an aldol process in which the acid would also be chiral. A
note that the addition of more than one equiv. of TFA was series of commercially available (R)- and (S)-configured chi-
detrimental to the reaction, probably due to the subsequent ral acids (1 equiv.) were thus associated with 1 in order to
protonation of enamine nitrogen site of III (as in IIIЈ), test a putative match and mismatch effect. As summarized
which slows down the process. The formation of an inactive in Table 2, the enantioselectivity is slightly improved in the
enammonium intermediate could also explain the inhibition aldol reaction with camphorsulfonic acid (CSA) as com-
of the proline activity observed upon addition of one equiv. pared to that with TFA, but there is no visible match-mis-
of TFA (Table 1, Entry 12). With proline, the enamine is match effect in this case (Entries 3 and 4, Table 2). A small
the only site available for the “extra” proton. In summary, effect is observed with binapthol phosphoric and Mosher’s
1 incorporates two basic/nucleophilic sites that are crucial acids (∆ee: 7% for Table 2, Entries 5 and 6 and ∆ee: 6% for
for the catalysis, and works synergistically in the presence Table 2, Entries 1 and 2), while a more important one is
of one equiv. of a proton source.
detected with tartaric acid (∆ee: 25% for Table 2, Entry 8),
We also initiated an NMR study to get a better under- indicating that the counter-anion is probably intimately as-
standing of the role of the added proton on the catalyst sociated with the benzimidazolium transition state IV, as
1
activation (see the H NMR spectra of 1 and 1/H+ in deu- proposed in the reaction mechanism (Figure 1).[17] In this
terated acetone, used both as solvent and reactant in Fig- case, the enantioselectivity is low so that the effect is more
ure 2). 1 alone reacts rapidly with acetone to afford, after easily detected. Therefore, although the chiral acid effect
30 min, 60% of the benzimidazolidine 6,[4] while unreacted is weak, double stereodifferentiation with the chiral ligand/
1 constitutes the other 40%. Notably, in 6, the two protons chiral acid pair has been demonstrated, which may provide
and methyl groups on the phenyl ring become magnetically an entry for further improvements. Moreover, the strength
170
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 167–177

(BIP/H+), a Highly Reactive Organocatalyst
FULL PAPER
1
1
Figure 2. H NMR analyses of aldol reactions. Upper spectrum: H NMR spectrum of 1 (25 mg, 4 Å molecular sieves) in [D6]acetone
1
(0.5 mL); Lower spectrum: H NMR spectrum of 1 (25 mg, 4 Å molecular sieves) in [D6]acetone (0.5 mL)/TFA (1 equiv., 9 µL).
of the acid is crucial for a high enantioselectivity, as shown Table 2. Aldolisation reaction with 1/chiral acids.
by the better ee obtained with the stronger CSA (Table 2,
Entries 3 and 4).
Having shown that our catalyst was efficient with ace-
tone, we studied its behaviour toward cyclohexanone, cyclo-
pentanone and diethyl ketone (7a–c).[8d,18] As summarized
in Table 3, catalyst 1 is very efficient with such ketones and
Entry Cat./acid[a]
Time [h] T [°C]
Yield [%][b] ee [%][c]
provides aldol products 8 and 9 with excellent enantio-
selectivities, albeit with modest diastereocontrol. Compari-
son of our results with those obtained with proline, 5,5-
1
2
3
4
5
6
7
8
1/(R)-MTPA
1/(S)-MTPA
1/(+)-CSA
1/(–)-CSA
1/(R)-BINOLPO2H
1/(S)-BINOLPO2H
1/-tartaric acid
1/-tartaric acid
5
5
15
15
0
87
83
87
63
56
46
81
81
70
64
83
82
75
68
27
52
dimethylthiazolidinium-4-carboxylate (DMTC),[18] or
a
24
24
24
24
24
24
0
pyrrolidinyl-proline derivative showed that improved de as
well as ee were observed.[8d] These results are among the
best obtained to date with these substrates. Interestingly,
only 2 mol-% of organocatalyst 1 was necessary to mediate
the aldol reaction between cycloalkanones 7a and b with
only 1.1 equiv. of ketones (Table 3, Entries 1–4), in contrast
to the 22–27 equiv. of ketone and 10–20 mol-% of catalysts
previously needed.[18] The reaction was faster and more ef-
ficient in terms of yield and enantioselectivity with cyclo-
hexanone as compared to cyclopentanone (Entry 4 vs. En-
10
10
15
15
[a] Reaction carried in THF with 1.1 equiv. of acetone and 5 mol-
% of 1/chiral acid. [b] Isolated yields. [c] ee determined by HPLC
with a Chiralcel AD-H® chiral column.
tries 1–3, Table 3). Finally, the reaction was also very ef- 9c in 98% and Ͼ99% ee, respectively. Interestingly, using
ficient with the less reactive diethyl ketone 7c, and provided the (R) enantiomer of the Mosher’s acid instead of TFA
a 2.5:1 ratio of diastereomeric anti aldol 8c and syn aldol provided significantly improved results.
Eur. J. Org. Chem. 2007, 167–177
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
171

J.-M. Vincent, Y. Landais et al.
FULL PAPER
Table 3. Aldolisation reaction with 1/acids.
Entry
Ketone [equiv.]
1/acid [mol-%]
Time [h]
T [°C]
Yield [%][a]
anti/syn[d]
ee (anti) [%][e]
ee (syn) [%][e]
1
2
3
4
5
6
7a (1.1)
7a (1.1)
7a (1.1)
7b (1.1)
7c (9)
1/TFA (10)
1/TFA (2)
4
20
20
40
20
20
20
78
43[b]
46[c]
83
91
88
1.5:1
1.2:1
1.6:1
3.2:1
2.2:1
2.5:1
79
75
66
99
92
98
93
92
54
80
Ͼ99
Ͼ99
48
48
2
48
48
1/TFA (2)
1/TFA (2)
1/TFA (10)
1/(R)-MTPA (10)
7c (9)
[a] Isolated yield. [b] 48% of recovered aldehyde. [c] 52% of recovered aldehyde. [d] Ratio estimated from 1H NMR of the crude reaction
mixture. [e] ee measured by HPLC with a Chiralcel AD-H® column.
Although the diastereomeric ratio observed in the aldol here for the first time in an aldol reaction, possesses reactiv-
reaction with diethyl ketone 7c is modest, this transforma- ity analogous to that of cyclopentanone and cyclohexanone
tion is noteworthy, as it offers a rapid entry to polypropion- and leads to similar selectivities.
ate fragments with excellent enantioselectivities. In a similar
fashion, we tested our catalyst in aldol processes involving
1-silacyclohexan-4-one 12,[19] which can also be regarded as
a potential precursor of polypropionate fragments. Oxi-
dation of the C–Si bond of 11 using the well-known Tamao-
Fleming reaction[20] should effectively produce, after the
aldol reaction, polypropionate-type fragments such as 10
(Scheme 4).
Scheme 5.
Finally, several aldehydes have been tested in these aldol
reactions (Scheme 6). Benzaldehyde (14a) and 4-pyridine
Scheme 4. 1,1-Diphenyl-1-silacyclohexan-4-one (12) as a precursor
carbaldehyde (14b) were found suitable, leading to aldol
of polypropionate fragments.
products 15a and b in reasonable yield and enantio-
The aldol reaction was thus carried out starting with 1.1
equiv. of ketone 12 and 1 equiv. of our test aldehyde 3
(Scheme 5). A mixture of anti and syn aldols 13a and 13b
was obtained with modest de but good enantioselectivities.
Better enantioselectivities were observed when 1 was associ-
ated with (+)-CSA, demonstrating again the match pairing
of these chiral reagents. Interestingly, 30 mol-% of proline
in DMSO with 2.4 equiv. of 12 led to no reaction, even after
48 h of stirring at room temperature (20 °C), demonstrating
the unique reactivity of 1/acid reagents. The assignment of
the relative configurations of 13a and 13b was realized by
analogy with aldols prepared from cyclohexanone and cy-
clopentanone.[18] Finally, it can be concluded that 12, used Scheme 6.
172
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 167–177

(BIP/H+), a Highly Reactive Organocatalyst
FULL PAPER
Table 4. Robinson annelation of triones 16a and b catalyzed by 1.
Entry
Cat./acid
mol-%
Solvent
Product
Time [h]
T [°C]
Yield [%][b]
ee [%][c]
1
2
3
4
5
6
-proline/–
1/TFA
47[a]
10
CH3CN
THF
THF
CH3CN
THF
THF
17a
17a
17a
17b
17b
17b
24
48
48
24
48
72
80
0
15
80
0
87
quant.
64
83
82
84
86
88
71
68
64
1/(+)-CSA
-proline/–
1/TFA
10
47[a]
10
1/(+)-CSA
10
20
90
[a] 1  HClO4 was present in the medium. [b] Isolated yield. [c] ee measured by HPLC with a Chiralcel AD-H® column.
selectivities. Equimolar amounts of 14b and acetone were (87% ee). A single recrystallization in diethyl ether led to
used to provide 15b, while benzaldehyde (14a) was treated 21 with an improved 93% ee value. It is noteworthy that
with acetone in excess to minimize the amount of elimi- when the reaction was performed in DMF instead of THF,
nation product formed during the aldol process.[21]
21 was formed along with the aldol product in a 1:1 ratio.
In contrast, -proline in DMSO provided only the aldol
product, indicating again that 1/TFA catalyzes not only the
aldol reaction but also the elimination process to provide, in
a single operation, the annelation product 21. The absolute
configuration of 21 was assigned based on the known con-
figuration of 17a and b above.[22b,22c]
Robinson Annelation Catalyzed by 1/H+
As a continuation of our investigation on the potential
of the 1/H+ reagent, we extended the study to the Robinson
annelation of diones 16a and b (Table 4).[22,23] The entire
aldol reaction-elimination process could be carried out in
one pot, as we observed that the dehydration step was
smoothly catalyzed by the 1/TFA. This is in contrast with
proline catalysis, for which the reaction has to be performed
at elevated temperature in the presence of 1  HClO4 to
achieve this elimination step. The results are summarized in
Table 4. A comparison with the results obtained with -pro-
line shows that a better yield of 17a[24] and similar enantio-
selectivity was obtained with 1/TFA under milder condi-
tions (Table 4, Entry 2 vs. 1). As already noticed in the aldol
reactions, only 10 mol-% of 1/TFA was needed, as com-
pared with 47 mol-% of -proline (Table 4, Entry 1). A sim-
ilar conclusion was reached with the homologous trione
16b. A better yield of bicyclic ketone 17b was obtained with
1/TFA, albeit with slightly lower enantioselectivity (Table 4,
Entry 5). Finally, the reagent 1/(+)-CSA catalyzed the an-
nelation of trione 16a with slightly better enantioselectivity
but lower yield (Entry 3), while the contrary was observed
upon annelation of 16b (Table 4, Entry 6).
Scheme 7.
1-mediated elimination was supported by an experiment
carried out on the racemic aldol product 15a. Treatment of
racemic 15a in THF in the presence of 10 mol-% of 1/TFA
led, after about 80% conversion into the corresponding
chalcone (the reaction was followed by HPLC), to an en-
antioenriched aldol product 15a [39% ee, major isomer (S)]
as a result of a kinetic resolution.
The approach was then extended to the Robinson annel-
ation of an analogue of 16a possessing a more useful allyl
substituent on the quaternary centre (Scheme 7). Such an
allyl group can be functionalized further to generate valu-
able intermediates for the synthesis of polycyclic systems.
Thus, symmetrical dione 20 was prepared in two steps from
cyclopentane-1,3-dione (18) through a palladium-catalyzed
allylation, followed by Michael addition of 19[25] onto
α-Amination of Ketones Catalyzed by 1
The α-amination of ketones is a synthetically relevant
methylvinyl ketone. Robinson annelation of 20, catalyzed transformation which has received a great deal of attention
by 1/TFA (10 mol-%) finally led to the desired bicyclic in the context of aminocatalysis.[27] We have tested our or-
ketone 21[26] in excellent yield and good enantioselectivity ganocatalytic system on this reaction and found that the
Eur. J. Org. Chem. 2007, 167–177
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
173

J.-M. Vincent, Y. Landais et al.
FULL PAPER
Table 5. α-Amination of ketones catalyzed by 1.
Entry
Cat./acid
mol-%
Solvent
Product
Time [h]
T [°C]
Yield [%][a]
ee [%][b]
1
2
3
4
5
-proline/–
1/TFA
15
10
10
20
20
CH2Cl2
CH2Cl2
THF
CH2Cl2
CH2Cl2
22
22
–
23
23
24
24
24
24
24
20
20
20
20
20
88
92
0
85
65
84
66
–
60
71
1/TFA
-proline/–
1/TFA
[a] Isolated yield. [b] ee was measured by HPLC with a Chiralcel AD-H® column.
terated chloroform as the solvent. The chemical shifts (δ) for 13C
and 1H signals are given compared to the internal reference (TMS)
and are expressed in ppm. Some mass spectra (low resolution) were
obtained with a Thermo Quest Finnigan Trace GC-MS apparatus.
Ionization was carried out by electronic impact (potential of ioniza-
tion: 70 eV). Other mass spectra (low and high resolution) were
obtained on a Micromass autospec-Q spectrometer. Ionization em-
ployed was the electronic impact mode (potential of ionization:
70 eV), and the LSIMS mode [potential of ionization: 35 keV, ma-
trix: (3-nitrophenyl)methanol]. IR spectra were obtained on a Per-
reaction occurred smoothly in CH2Cl2 leading to the α-ami-
noketones 22 and 23 in good yield and reasonable enantio-
selectivity (Table 5). 1/TFA was less efficient than -proline
in terms of enantioselectivity when cyclohexanone 7b was
used but led to better results with 1-silacyclohexan-4-one
12.[28]
Conclusions
kin–Elmer Paragon 1000 FT-IR spectrometer. The wavelengths (ν)
˜
In summary, we have developed a new organocatalyst,
which is available in one step from -proline. As demon-
strated by 1H NMR and mass spectrometry studies, this
organocatalyst exhibits an enhanced reactivity toward
ketones, accelerating significantly the formation of the key
iminium and the enamine intermediates in aminocatalyzed
aldol and amination reactions. Excellent levels of enantio-
selectivity have been attained in several cases, with low cata-
lyst loading (Ͻ5–10%) and as little as 1.1. equiv. of ketone.
Modifications of the basic core of the benzimidazole and
that of pyrrolidine should provide analogues of BIP that
might exhibit higher reactivity and selectivity. Extension of
these aldol processes to various types of ketone donors and
application of our catalyst to the synthesis of enantioen-
riched intermediates and biologically relevant targets is cur-
rently under way in our laboratory.
are expressed in cm–1.
(S)-5,6-Dimethyl-2-(pyrrolidin-2-yl)-1H-benzimidazole (1): -proline
(3.4 g, 27 mmol) was treated with 4,5-dimethyl-1,2-phenylenediam-
ine 2 (3.1 g, 23 mmol) in aqueous HCl (4 , 40 mL) at reflux for
6–7 d. The solution was then treated with aqueous NaOH (4 ,
until pH = 12), affording a brown sticky residue. After vigorous
stirring for 5 h, the precipitate that formed was filtered, washed
with water and thoroughly washed with diethyl ether to give 1 as
a beige powder (2.7 g, 55% yield). 1 can be recrystallised by slow
evaporation of a water/MeOH solution. M.p. 94 °C (beige powder).
IR (KBr): ν = 3280 (NH), 2970, 2871, 2282, 1634, 1444, 1310, 825
˜
cm–1. 1H NMR ([D6]DMSO): δ = 7.30 (s, 2 H), 4.46–4.41 (m, 1
H), 3.06–2.98 (m, 2 H), 2.34 (s, 6 H), 2.22–2.13 (m, 1 H), 2.02–1.98
(m, 1 H), 1.85–1.77 (m, 2 H) ppm. 13C NMR ([D6]DMSO): δ =
155.2, 136.9, 131.2, 115.1, 56.3, 46.5, 32.2, 25.5, 20.2 ppm. MS
(EI): m/z (%) = 215 (71) [M]+, 187 (53), 186 (37), 173 (100), 160
(75), 147 (26). C13H17N3·1.5H2O (241.1) calcd. C 64.46, H 8.26, N
17.35; found C 64.50, H 7.85, N 17.25.
General Procedure for Aldol Reactions with 1: Compound 1 (4.9 mg,
23 µmol) and TFA (1.8 µL, 23 µmol) were stirred in THF (1 mL)
with ketone (1.1 equiv.) for 10 min. p-Nitrobenzaldehyde (3,
177.6 mg, 1.17 mmol) was added in one portion. The reaction pro-
gress was monitored by TLC (hexanes/EtOAc, 7:3). The crude reac-
tion mixture was purified by flash column chromatography on sil-
ica gel (hexanes/EtOAc 7:3) to give the aldol adduct. Enantiomeric
Acknowledgments
This work was supported by the Ministère de la Recher-
che et de la Technologie through a PhD grant to E.L. We
gratefully acknowledge the Institut Universitaire de France
and the CNRS for financial support. We thank Prof. Petri
Pihko (University of Helsinki) for a generous gift of DBAB.
1
excess (ee) was determined through H-NMR spectroscopy of the
corresponding Mosher’s ester or with chiral HPLC. The anti/syn
ratio was estimated from a 1H NMR analysis of crude reaction
mixture. The enantioselectivities of the anti and syn isomers were
measured by chiral HPLC analysis.
Experimental Section
General Remarks: NMR analysis were carried out on Bruker AC-
200 FT (200 MHz for H and 50.4 MHz for 13C), Bruker AC-250 4-Hydroxy-4-(4-nitrophenyl)butan-2-one (4): This aldol product was
1
1
FT (250 MHz for H and 63 MHz for 13C) and Bruker DPX-300
prepared according to the general procedure described above. Rf
(300 MHz for H and 75.5 MHz for 13C) spectrometers with deu- (hexanes/EtOAc, 7:3) = 0.3. HPLC: Chiralcel AD-H®, hexanes/
1
174
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 167–177

(BIP/H+), a Highly Reactive Organocatalyst
FULL PAPER
MeOH, 96:4, retention time 50.3 min. (S) and 51.9 min. (R). 1H
NMR (CD2Cl2): δ = 8.18 (d, J = 9 Hz, 2 H), 7.54 (d, J = 8.85 Hz,
2 H), 5.28–5.22 (m, 1 H), 3.58 (broad s, 1 H), 2.87–2.82 (m, 2 H),
2.49–2.43 (m, 1 H), 1.05–0.95 (m, 6 H). 13C NMR (CDCl3): δ =
215.6, 146.6, 126.8, 123.5, 71.9, 51.4, 35.1, 9.9, 7.4 ppm. MS
(FAB+): m/z (%) = 260 (100) [M + Na]+, 238 (20) [M + H]+, 212
2.19 (s, 3 H) ppm. 13C NMR (CD2Cl2): δ = 208.8, 150.8, 147.7, (21). HRMS: calcd. for C12H15NO4Na [M + Na]+ 260.089878;
129.4, 124.0, 69.3, 51.8, 30.8 ppm. MS (FAB+): m/z (%) = 232
(100) [M + Na]+, 212 (40), 210 (35) [M + H]+. HRMS: calcd. for
C10H11NO4Na [M + Na]+ 232.058578; found 232.053874.
found 260.090584.
3-[Hydroxy(4-nitrophenyl)methyl]-1,1-diphenylsilinan-4-one (13a
and 13b): Rf (hexanes/EtOAc, 8:2, syn) = 0.4, Rf (hexanes/EtOAc,
8:2, anti) = 0.35. HPLC: Chiralcel AD-H®, hexanes/MeOH, 96:4,
retention time for syn-13b 34.4 min and 36.7 min; anti-13a 71.1 min
and 96.3 min. syn-13b: 1H NMR (CDCl3): δ = 8.15 (d, J = 8.7 Hz,
2 H), 7.39 (d, J = 8.7 Hz, 2 H), 7.37–7.29 (m, 10 H), 5.35 (s, 1 H),
3.47 (d, J = 3 Hz, 1 H), 2.88–2.84 (m, 1 H), 2.82–2.64 (m, 2 H),
1.47–1.35 (m, 4 H) ppm. 13C NMR (CDCl3): δ = 216.6, 148.9,
147.1, 134.7, 130.7, 128.5, 128.4, 127.1, 123.8, 71.8, 52.8, 39.0, 10.4,
8.2 ppm. anti-13a: 1H NMR (CDCl3): δ = 8.15 (d, J = 8.7 Hz, 2
H), 7.39 (d, J = 8.7 Hz, 2 H), 7.37–7.29 (m, 10 H), 4.96 (d, J =
8 Hz, 1 H), 3.7 (br., 1 H), 2.96–2.92 (m, 1 H), 2.75–2.69 (m, 2 H),
1.67–1.58 (m 4 H) ppm. 13C NMR (CDCl3): δ = 216.6, 148.9,
147.1, 134.7, 130.7, 128.5, 128.4, 127.1, 123.8, 76.3, 53.3, 39.4, 14.3,
10.8 ppm.
(S)-2-(1-Isopropylpyrrolidin-2-yl)-5,6-dimethyl-1H-benzimidazole
(5): 1 (50 mg, 0.23 mmol) and TFA (18 µL, 0.23 mmol) were stirred
in acetone for 10 min, and NaBH3CN (30 mg, 0.5 mmol) was
added in portions. After 24 h at room temperature, the resulting
solution was treated with water and Na2CO3. The reaction mixture
was extracted with EtOAc, and the resulting organic layer was
dried with MgSO4. The solvent was evaporated in vacuo to afford
5 as a white powder (39 mg, 65 % yield). M.p. 160–178 °C. IR
(KBr): ν = 2966, 1635, 1417, 1309 cm–1. 1H NMR (CDCl3): δ =
˜
7.25 (s, 2 H), 4.11–4.08 (m, 1 H), 3.15–3.05 (m, 1 H), 2.85–2.75 (m,
1 H), 2.60–2.40 (m, 1 H), 2.28 (s, 6 H), 2.25–2.15 (m, 1 H), 1.95–
1.75 (m, 1 H), 1.73–1.65 (m, 2 H), 0.99 (d, J = 6.4 Hz, 3 H), 0.95
(d, J = 6.4 Hz, 3 H) ppm. 13C NMR ([D6]DMSO): δ = 159.2, 131.3,
115.4, 59.7, 53, 50.3, 33.8, 24.7, 22.2, 20.7, 19.4 ppm. MS (FAB+):
m/z (%) = 280 [M + Na]+, 258 (100) [M + H]+, 257 (26), 256 (80),
214 (65), 212 (22). HRMS: calcd. for C16H24N3 [M + H]+
258.196543; found 258.197023.
4-Hydroxy-4-(pyridin-4-yl)butan-2-one (15b):[29] Rf (CH2Cl2/diethyl
ether, 8:2) = 0.5. HPLC: Chiralcel AD-H®, hexanes/MeOH, 90:10,
retention time 15.5 min. (S) and 16.7 min. (R). 1H NMR (CD2Cl2):
δ = 8.49 (d, J = 6 Hz, 2 H), 7.22 (d, J = 6 Hz, 2 H), 5.08 (t, J =
6.4 Hz, 1 H), 3.65 (br., 1 H), 2.76 (d, J = 5.3 Hz, 2 H), 2.15 (s, 3
H) ppm. 13C NMR (CDCl3): δ = 208.4, 151.6, 149.9, 120.5, 68.4,
51.2, 30.7 ppm. MS (EI): m/z (%) = 132 (20), 107,(100), 106,(37),
78(55).
2-[Hydroxy-(4-nitrophenyl)methyl]cyclopentanone (8a and 9a):[18b]
Rf (hexanes/EtOAc, 7:3) = 0.4. HPLC: Chiralcel AD-H®, hexanes/
MeOH, 95:5, retention time for syn-9a 26.7 min and 42 min; anti-
8a 52.6 min and 55.3 min. anti-8a: 1H NMR (CD2Cl2): δ = 8.18 (d,
J = 10.6 Hz, 2 H), 7.53 (d, J = 10.2 Hz, 2 H), 4.82 (d, J = 9.0 Hz,
1 H), 4.67 (s, 1 H), 2.30–1.50 (m, 7 H) ppm. 13C NMR (CDCl3): δ
= 240.1, 150.8, 148.0, 126.8, 123.9, 74.8, 55.4, 39.0, 27.2, 20.7 ppm.
2-Methyl-2-(3-oxobutyl)cyclopentane-1,3-dione (16a):[24] 2-Methyl-
cyclopentane-1,3-dione (2 mmol) was placed in CH3CN (3 mL),
and triethylamine (1 mL) was added, followed by methylvinyl
ketone (200 µL, 2.4 mmol). The reaction mixture was stirred at
room temperature. The reaction progress was monitored by TLC
(hexanes/EtOAc, 1:1). The solvents were removed under vacuum,
and the residue was purified through chromatography over silica
gel to give 16a as an oil (100% yield); Rf (hexanes/EtOAc, 1:1) =
1
syn-9a: H NMR (CDCl3): δ = 8.18 (d, J = 7.2 Hz, 2 H), 7.53 (d,
J = 7.5 Hz, 2 H), 5.38 (m, 1 H), 2.59 (d, J = 4.9 Hz, 1 H), 2.30–
1.50 (m, 7 H) ppm. 13C NMR (CDCl3): δ = 220.1, 149.4, 127.8,
123.4, 70.9, 56.3, 39.2, 22.8, 20.7 ppm. MS (FAB+): m/z (%) = 258
(100) [M + Na]+ , 218 (37), 212 (42). HRMS: calcd. for
C12H13NO4Na [M + Na]+ 258.074228; found 258.074435.
1
0.4. H NMR (CDCl3): δ = 2.89–2.73 (m, 4 H), 2.47 (t, J = 7 Hz,
2-[Hydroxy(4-nitrophenyl)methyl]cyclohexanone (8b and 9b): Rf
(hexanes/EtOAc, 7:3, syn) = 0.25; Rf (hexanes/EtOAc, 7:3, anti) =
0.2. HPLC: Chiralcel AD-H®, hexanes/MeOH, 90:10, retention
time syn-9b 14.6 min and 16.1 min; anti-8b 17.4 min and 24.2 min.
2 H), 2.12 (s, 3 H), 1.91 (t, J = 5 Hz, 2 H), 1.12 (s, 3 H) ppm. 13C
NMR (CDCl3): δ = 215.7, 207.8, 55.1, 37.4, 34.7, 30.0, 27.8, 19.1
ppm. MS (FAB+): m/z (%) = 183 (100) [M + H]+. HRMS: calcd.
for C10H15O3 [M + H]+ 183.102120; found 183.10190.
1
anti-8b: H NMR (CDCl3) δ = 8.19 (d, J = 10.3 Hz, 2 H), 7.50 (d,
2-Methyl-2-(3-oxobutyl)cyclohexane-1,3-dione (16b):[24] Compound
16b was prepared following the same procedure as described for
16a above. Rf (hexanes/EtOAc, 1:1) = 0.4. H NMR (CDCl3): δ =
2.79–2.59 (m, 4 H), 2.36 (t, J = 7 Hz, 2 H), 2.13 (s, 3 H), 2.09–2.01
(m, 2 H), 1.99–1.85 (m, 2 H), 1.26 (s, 3 H) ppm. 13C NMR
(CDCl3): δ = 210.0, 207.5, 64.3, 38.4, 37.8, 29.9, 29.6, 20.0, 17.6
ppm. MS (FAB+): m/z (%) = 197 (100) [M + H]+. HRMS: cacld.
for C11H17O3 [M + H]+ 197.117770; found 197.118217.
J = 10.2 Hz, 2 H), 4.88 (dd, J = 6.2 and 3.7 Hz, 1 H), 4.09 (d, J =
4 Hz, 1 H), 2.64–2.28 (m, 3 H), 2.14–2.06 (m, 1 H), 1.83–1.30 (m,
5 H) ppm. 13C NMR (CDCl3): δ = 214.7, 148.4, 127.9, 123.6, 74.0,
57.2, 42.7, 30.7, 27.6, 24.7 ppm. syn-9b: 1H NMR (CDCl3): δ =
8.21 (d, J = 7.5 Hz, 2 H), 7.49 (d, J = 7.5 Hz, 2 H), 5.49 (s, 1 H),
3.20 (d, J = 3 Hz, 1 H), 2.67–2.53 (m, 3 H), 2.16–2.07 (m, 1 H),
1.89–1.45 (m, 5 H) ppm. 13C NMR (CDCl3): δ = 213.9, 147.6,
126.6, 123.4, 70.1, 56.8, 42.6, 27.8, 25.9, 24.8 ppm. MS (FAB+):
m/z (%) = 272 (100) [M + Na]+, 232 (30), 212 (49). HRMS: calcd.
for C13H15NO4Na [M + Na]+ 272.089878; found 272.090203.
1
7α-Methyl-2,3,7,7α-tetrahydro-6H-indene-1,5-dione (17a):[22c] In a
2 mL flask, 1 (11 mg, 48 µmol), TFA (3.9 µL, 50 µmol) and 16a
(98 mg, 5 mmol) were mixed in THF (1 mL), and the reaction mix-
1-Hydroxy-2-methyl-1-(4-nitrophenyl)pentane-3-one (8c and 9c):[18c]
Rf (hexanes/EtOAc, 7:3) = 0.3. HPLC: Chiralcel OD®, hexanes/
ture was stirred at room temperature. The reaction progress was
1
MeOH, 92:8, retention time for syn-9c 14.9 min and 15.4 min; anti- monitored by H NMR analysis. The crude reaction mixture was
1
8c 13.8 min and 16.4 min. anti-8c: H NMR (CDCl3): δ = 8.22 (d,
purified by flash column chromatography on silica gel (hexanes/
J = 8.8 Hz, 2 H), 7.51 (d, J = 8.5 Hz, 2 H), 4.88 (d, J = 7.5 Hz, 1 EtOAc, 9:1) to give ketone 17a. HPLC: Chiralcel OD®, hexanes/
H), 2.95–2.91 (m, 1 H), 2.55–2.51 (m, 1 H), 2.49–2.43 (m, 1 H), MeOH, 96:4, retention time 31.9 min and 36.7 min. 1H NMR
1.05–0.95 (m, 6 H) ppm. 13C NMR (CDCl3): δ = 215.6, 149.6, (CDCl3): δ = 5.98 (s, 1 H), 3.05–2.92 (m, 1 H), 2.85–2.77 (m, 2 H),
127.4, 123.7, 75.6, 52.2, 36.4, 14.5, 7.5 ppm. syn-9c: 1H NMR
2.58–2.47 (m, 2 H), 2.17–2.04 (m, 2 H), 1.98–1.84 (m, 1 H), 1.33
(CDCl3): δ = 8.22 (d, J = 8.8 Hz, 2 H), 7.51 (d, J = 8.5 Hz, 2 H), (s, 3 H) ppm. 13C NMR (CDCl3): δ = 214.3, 196.0, 167.6, 121.8,
5.23 (d, J = 3 Hz, 1 H), 2.85–2.75 (m, 1 H), 2.55–2.51 (m, 1 H), 94.4, 32.6, 27.6, 24.7, 18.4 ppm. MS (FAB+): m/z (%) = 165 (100)
Eur. J. Org. Chem. 2007, 167–177
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
175

J.-M. Vincent, Y. Landais et al.
FULL PAPER
[M + H]+, 147 (31), 141 (54), 135 (33). HRMS: calcd. for C10H13O2
Dibenzyl 1-(2-Oxocyclohexyl)hydrazine-1,2-dicarboxylate (22): 1
(12 mg, 48 µmol) and TFA (4 µL, 52 µmol) were stirred in CH2Cl2
(2 mL), and cyclohexanone 7b (80 µL, 773 µmol) and DBAB
(150 mg, 504 µmol) were added. The reaction mixture was stirred
at room temperature, and the reaction progress was monitored by
TLC (hexanes/EtOAc, 7:3). The crude reaction mixture was puri-
fied by flash column chromatography on silica gel (hexanes/EtOAc,
8:2) to give 22. Rf (CH2Cl2/diethyl ether, 8:2) = 0.25. HPLC: Chi-
[M + H]+ 165.091555; found 165.091101.
8α-Methyl-3,4,8,8α-tetrahydronaphthalene-1,6(2H,7H)-dione
(17b):[22c] 17b was prepared following the same procedure as de-
scribed for 17a above. HPLC: Chiralcel OD®, hexanes/MeOH,
95:5, retention time 21.9 min and 24.7 min. 1H NMR (CDCl3): δ
= 5.85 (s, 1 H), 2.85–2.70 (m, 2 H), 2.65–2.40 (m, 4 H), 2.25–2.05
(m, 3 H), 1.85–1.65 (m, 1 H), 1.45 (s, 3 H) ppm. 13C NMR
(CDCl3): δ = 165.8, 125.9, 50.6, 37.7, 33.6, 31.8, 29.7, 23.3, 23.0
ppm. MS (FAB+): m/z (%) = 179 (100) [M + H]+. HRMS: calcd.
for C11H15O2 [M + H]+ 179.107205; found 179.107743.
ralcel OD®, hexanes/MeOH, 80:20, retention time 34.8 min and
1
41.7 min. IR (NaCl): ν = 3387, 1756, 1727, 1700 cm–1. H NMR
˜
(CDCl3, 55 °C): δ = 7.38–7.33 (m, 10 H), 6.89 (br., 1 H), 5.24–5.17
(m, 4 H), 5.00–4.82 (br., 1 H), 2.54–2.42 (m, 3 H), 1.94–1.59 (m, 5
H) ppm. 13C NMR (CDCl3, 55 °C): δ = 206.9, 156.2, 135.7, 135.6,
128.4, 127.6, 68.5, 67.7, 67.5, 41.1, 30.6, 26.6, 24.2 ppm. MS (EI):
m/z (%) = 396 (0.2), 91 (100). HRMS: calcd. for C22H24N2O5
396.168522; found 396.167668.
2-Allylcyclopentane-1,3-dione (19):[25] In a 100 mL flask with a con-
denser, allyl palladium chloride dimer (68 mg, 0.18 mmol) and
dppe (199 mg, 0.50 mmol) were introduced and degassed with N2.
Anhydrous THF (30 mL) was added, followed by allyl acetate
(0.53 mL, 4.9 mmol). To the solution were added sequentially 1,3-
cyclopentanedione (731 g, 7.45 mmol), BSA (1.83 mL, 7.43 mmol)
and sodium acetate (24 mg, 0.29 mmol). The reaction mixture was
heated to 70 °C, and the reaction progress was monitored by TLC
(EtOAc/MeOH, 97:3). The crude reaction mixture was filtered
through a celite plug and washed with MeOH (3ϫ10 mL). The
solvents were removed under vacuum, and the product was purified
through column chromatography on silica gel (EtOAc/MeOH,
97:3) to afford a white powder (530 mg, 78% yield). Rf (CH2Cl2/
Dibenzyl 1-(4-Oxo-1,1-diphenylsilinan-3-yl)hydrazine-1,2-dicarbox-
ylate (23): 23 was prepared following the same procedure as de-
scribed for 22 above. Rf (CH2Cl2/diethyl ether, 8:2) = 0.25. HPLC:
Chiralcel OD®, hexanes/iPrOH, 70:30, retention time 12.5 min and
1
15.8 min. IR (NaCl): ν = 3386, 1775, 1727, 1712 cm–1. H NMR
˜
(CDCl3, 55 °C): δ = 7.69–7.30 (m, 20 H), 6.89 (br., 1 H), 5.17–4.98
(m, 5 H), 2.75–2.61 (m, 2 H), 1.77–1.05 (m, 4 H) ppm. 13C NMR
(CDCl3): δ = 210.5, 156.7, 134.9, 130.8, 130.6, 129.0, 128.9, 128.7,
128.6, 128.5, 127.9, 68.6, 67.9, 64.2, 37.4, 14.9, 9.8 ppm. MS
(FAB+): m/z (%) = [M + Na]+ 587.3 (100). HRMS: calcd. for
C33H32N2O5SiNa 587.197739; found 587.197821.
1
MeOH, 99.5:0.5) = 0.3. H NMR (300 MHz, CD3OD): δ = 5.85–
5.74 (m, 1 H), 4.92–4.82 (m, 3 H), 2.85 (d, J = 6 Hz, 2 H), 2.52 (s,
4 H) ppm. 13C NMR (CDCl3): δ = 136.4, 116.4, 114.9, 31.3, 25.9
ppm. MS (EI): m/z (%) = 138 (100), 123 (32), 109 (25), 95 (60), 81
(25), 67 (25), 53 (28), 39 (32), 27 (36). HRMS: calcd. for C8H10O2
138.068080; found 138.068106.
[1] a) P. I. Dalko, L. Moisan, Angew. Chem. Int. Ed. 2001, 40,
3727–3748; b) E. R. Jarvo, S. J. Miller, Tetrahedron 2002, 58,
2481–2495; c) Synlett (special issue) 2003, 1901–1939; d) Acc.
Chem. Res. (special issue) 2004, 8, 487–631; e) A. Berkessel,
H. Gröger, in: Asymmetric Organocatalysis: From Biomimetic
Concepts to Applications in Asymmetric Synthesis, John
Wiley & Sons, Wiley, New York, 2005; f) D. Enders, M. R. M.
Hüttl, C. Grondal, G. Raabe, Nature 2006, 441, 861–863; g) Y.
Huang, A. M. Walji, C. H. Larsen, D. W. C. McMillan, J. Am.
Chem. Soc. 2005, 127, 15051–15053; h) B. List, Chem. Com-
mun. 2006, 819–824; i) P. I. Dalko, L. Moisan, Angew. Chem.
Int. Ed. 2004, 43, 5138–5175; j) G. Lelais, D. W. C. McMillan,
Aldrichimica Acta 2006, 39, 79–87.
[2] a) A. C. Kinsman, M. A. Kerr, J. Am. Chem. Soc. 2003, 125,
14120–14125; b) A. B. Northrup, I. K. Mangion, F. Hettche,
D. W. C. McMillan, Angew. Chem. Int. Ed. 2004, 43, 2152–
2154; c) A. Córdova, I. Ibrahem, J. Casas, M. Engqvist, B.
Kaynak, Angew. Chem. Int. Ed. 2005, 44, 1343–1345; d) D.
Enders, C. Grondal, Angew. Chem. Int. Ed. 2005, 44, 1210–
1212; e) D. Enders, J. Palecek, C. Grondal, Chem. Commun.
2006, 655–657; f) T. Itoh, M. Yokoya, K. Miyauchi, K. Nagata,
A. Ohsawa, Org. Lett. 2006, 8, 1533–1535.
2-Allyl-2-(3-oxobutyl)cyclopentane-1,3-dione (20):[25] Into a 25 mL
flask was introduced 19 (753 mg, 5.8 mmol) and H2O (2.6 mL),
followed by methyl vinyl ketone (1.2 mL, 14.5 mmol). The reaction
mixture was stirred at room temperature, and the progress of the
reaction was monitored by TLC (CH2Cl2/MeOH, 97:3). The reac-
tion mixture was diluted with diethyl ether (3ϫ10 mL), and the
organic layer was washed with a solution of NaCl (3ϫ10 mL). The
organic layer was dried with MgSO4, and the solvent was removed
under vacuum. The crude product was purified through column
chromatography on silica gel (CH2Cl2/MeOH, 97:3) to afford 20 as
a yellow oil (649 mg, 54% yield). Rf (CH2Cl2/MeOH, 96:4) = 0.4.
IR (film): ν = 3464, 1722, 1640 cm–1. 1H NMR (CDCl ): δ = 5.72–
˜
3
5.45 (m, 1 H), 5.09–5.02 (m, 2 H), 2.85–2.55 (m, 4 H), 2.42 (t, J =
7.3 Hz, 2 H), 2.31 (d, J = 2.3 Hz, 2 H), 2.08 (s, 3 H), 1.88 (t, J =
7.3 Hz, 2 H) ppm. MS (FAB+): m/z (%) = 231.2 (100)
[M + Na]+. HRMS: calcd. for C12H16O3Na [M + Na]+ 231.099309;
found 231.099714.
[3] a) B. List, R. A. Lerner, C. F. Barbas III, J. Am. Chem. Soc.
2000, 122, 2395–2396; b) B. List, Tetrahedron 2002, 58, 5573–
5598; c) H. Gröger, J. Wilken, Angew. Chem. Int. Ed. 2001, 40,
529–532; d) N. Suzuki, Y. Hayashi, M. Shoji, W. Tsuboi, J. Am.
Chem. Soc. 2003, 125, 11208–11209; e) G. Zhong, Angew.
Chem. Int. Ed. 2003, 42, 4247–4250; f) Y. Hayashi, W. Tsuboi,
I. Ashimine, T. Urushima, M. Shoji, K. Sakai, Angew. Chem.
Int. Ed. 2003, 42, 3677–3680; g) R. O. Duthaler, Angew. Chem.
Int. Ed. 2003, 42, 975–978; h) B. List, N. Vignola, L. Hoang,
C. Pidathala, Angew. Chem. Int. Ed. 2003, 42, 2785–2788; i)
D. W. C. McMillan, A. B. Northrup, J. Am. Chem. Soc. 2002,
124, 6798–6799; j) C. F. Barbas III, W. Notz, A. Córdova, J.
Org. Chem. 2002, 67, 301–303; k) B. Anders, K. Nagaswamy,
K. A. Jørgensen, Chem. Commun. 2002, 620–621; l) C. Palomo,
M. Oiarbide, J. M. Garcia, Chem. Eur. J. 2002, 8, 36–44; m) A.
Córdova, H. Sundén, J. Casas, Tetrahedron Lett. 2004, 45,
6117–6119; n) S. Chandrasekhar, Ch. Narsihmulu, N. R.
7α-Allyl-2,3,7,7α-tetrahydro-6H-indene-1,5-dione (21):[26] In a
25 mL flask, 1,3-dione 20 (342 mg, 1.64 mmol) was stirred in THF
(5 mL), followed by the addition of 1 (37 mg, 0.16 mmol) and TFA
(12 µL, 0.16 mmol). The reaction progress was monitored by TLC
(EtOAc/pentane, 4:6). The solvent was removed under vacuum, and
the crude reaction mixture was purified through chromatography
on silica gel to furnish 21 as a yellow oil (218 mg, 70% yield, 87%
1
ee). Rf (EtOAc/pentane, 4:6) = 0.45. H NMR (300 MHz, CDCl3):
δ = 6.01 (d, J = 2 Hz, 1 H), 5.82–5.67 (m, 1 H), 5.18–5.09 (m, 2
H), 2.97–2.16 (m, 8 H), 2.82–2.69 (m, 2 H) ppm. 13C NMR
(75.47 MHz, CDCl3): δ = 198.06, 168.89, 131.69, 124.60, 119.73,
52.57, 38.95, 36.07, 32.53, 27.26 ppm. MS (FAB+): m/z (%) = 191
[M + H]+, 213 [M + Na]+. HRMS: calcd. for C12H14O2Na [M +
Na]+ 213.0891; found 213.0896.
176
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 167–177

(BIP/H+), a Highly Reactive Organocatalyst
FULL PAPER
Reddy, S. S. Sultana, Chem. Commun. 2004, 2450–2451; o) S. P.
Mathew, H. Iwamura, D. G. Blackmond, Angew. Chem. Int.
Ed. 2004, 43, 3317–3321; p) J. T. Suri, D. B. Ramachary, C. F.
Barbas III, Org. Lett. 2005, 7, 1383–1385; q) O. Tokuda, T.
Kano, W.-G. Gao, T. Ikemoto, K. Maruoka, Org. Lett. 2005,
7, 5103–5105; r) B. Rodriguez, C. Bolm, J. Org. Chem. 2006,
71, 2888–2891; s) S. Samanta, C.-G. Zhao, J. Am. Chem. Soc.
2006, 128, 7442–7443.
[13]
a) H. K. Hall, J. Am. Chem. Soc. 1957, 79, 5441–5444; b) E. J.
Stamhuis, W. Maas, H. Wynberg, J. Org. Chem. 1965, 30, 2160–
2163.
R. F. Borch, M. D. Bernstein, H. D. Durst, J. Am. Chem. Soc.
1971, 93, 2897–2904.
As an indication, pyrrolidine is twenty times more nucleophilic
than imidazole in the S2 reaction with methyl iodide, see:
R. G. Pearson, H. R. Sobel, J. J. Songstad, J. Am. Chem. Soc.
1968, 90, 319–326.
M. S. Taylor, E. N. Jacobsen, Angew. Chem. Int. Ed. 2006, 45,
1520–1543.
S. Mayer, B. List, Angew. Chem. Int. Ed. 2006, 45, 4193–4195.
a) Y.-S. Wu, W.-Y. Shao, C.-Q. Zheng, Z.-L. Huang, J. Cai,
Q.-Y. Deng, Helv. Chim. Acta 2004, 87, 1377–1383; b) C. F.
Barbas III, K. Sakthivel, W. Notz, T. Bui, J. Am. Chem. Soc.
2001, 123, 5260–5267.
J. A. Soderquist, A. Negron, J. Org. Chem. 1989, 54, 2462–
2464.
a) G. R. Jones, Y. Landais, Tetrahedron 1996, 52, 7599–7662;
b) I. Fleming, Chemtracts, Org. Chem. 1996, 9, 1–64; c) K.
Tamao, Adv. Silicon Chem. 1996, 3, 1–62.
[14]
[15]
[4] a) B. List, L. Hoang, H. J. Martin, Proc. Natl. Acad. Sci. 2004,
101, 5839–5842; b) A. Hartikka, P. I. Arvidsson, Eur. J. Org.
Chem. 2005, 4287–4295; c) D. Seebach, M. Boes, R. Naef, B.
Schweizer, J. Am. Chem. Soc. 1983, 105, 5390–5398.
[16]
[17]
[18]
[5] a) F. R. Clemente, K. N. Houk, Angew. Chem. Int. Ed. 2004,
43, 5766–5768; b) C. Allemann, R. Gordillo, F. R. Clemente,
P. H.-Y. Cheong, K. N. Houk, Acc. Chem. Res. 2004, 37, 558–
569; c) L. Hoang, S. Bahmanyar, K. N. Houk, B. List, J. Am.
Chem. Soc. 2003, 125, 16–17; d) S. Bahmanyar, K. N. Houk, J.
Am. Chem. Soc. 2001, 123, 11273–11283; e) S. Bahmanyar,
K. N. Houk, H. J. Martin, B. List, J. Am. Chem. Soc. 2003,
125, 2475–2479; f) K. N. Rankin, J. W. Gauld, R. J. Boyd, J.
Phys. Chem. A 2002, 106, 5155–5159; g) M. Arno, L. R. Do-
mingo, Theor. Chem. Acc. 2002, 108, 232–239; h) P. H.-Y.
Cheong, K. N. Houk, Synthesis 2005, 1533–1537.
[6] a) C. E. T. Mitchell, S. E. Brenner, S. V. Ley, Chem. Commun.
2005, 5346–5348; b) D. B. Ramachary, C. F. Barbas III, Org.
Lett. 2005, 7, 1577–1580; c) A. Hartikka, P. I. Arvidsson, Tet-
rahedron: Asymmetry 2004, 15, 1831–1834; d) M. Arno, R. J.
Zaragoza, L. R. Domingo, Tetrahedron: Asymmetry 2005, 16,
2764–2770; e) Z. Tang, L.-F. Cun, X. Cui, A.-Q. Mi, Y.-Z. Ji-
ang, L.-Z. Gong, Org. Lett. 2006, 8, 1263–1266; f) A. Berkessel,
B. Koch, J. Lex, Adv. Synth. Catal. 2004, 346, 1141–1146; g)
Z. Tang, F. Jiang, L.-T. Yu, X. Cui, L.-Z. Gong, A.-Q. Mi, Y.-
Z. Jiang, Y.-D. Wu, J. Am. Chem. Soc. 2003, 125, 5262–5263.
[7] E. Lacoste, Y. Landais, K. Schenk, J.-B. Verlhac, J.-M. Vincent,
Tetrahedron Lett. 2004, 45, 8035–8038.
[19]
[20]
[21]
A. Córdova, H. Sunden, Y. Xu, I. Ibrahem, W. Zou, M.
Engqvist, Chem. Eur. J. 2006, 12, 5446–5451.
[22] a) G. Sauer, U. Eder, G. Haffer, G. Neef, R. Wiechert, Angew.
Chem. Int. Ed. Engl. 1975, 14, 417–417; b) Z. G. Hajos, D. R.
Parrish, J. Org. Chem. 1974, 39, 1615–1621; c) T. Bui, C. F.
Barbas III, Tetrahedron Lett. 2000, 41, 6951–6954.
[23] a) P. H.-Y. Cheong, K. N. Houk, J. S. Warrier, S. Hanessian,
Adv. Synth. Catal. 2004, 346, 1111–1115; b) K. Inomata, M.
Barragué, L. A. Paquette, J. Org. Chem. 2005, 70, 533–539; c)
F. R. Clemente, K. N. Houk, J. Am. Chem. Soc. 2005, 127,
11294–11302.
[24] R. A. Bunce, W. G. Dauben, J. Org. Chem. 1983, 48, 4642–
4648.
[8] a) C. F. Barbas III, F. Tanaka, N. Mase, Org. Lett. 2003, 5,
4369–4372; b) C. Bolm, T. Rantanen, I. Schiffers, L. Zani, An-
gew. Chem. Int. Ed. 2005, 44, 1758–1763; c) C. Cheng, J. Sun,
C. Wang, Y. Zhang, S. Wei, F. Jiang, Y. Wu, Chem. Commun.
2006, 215–217; d) M. Nakadai, S. Saito, H. Yamamoto, Tetra-
hedron 2002, 58, 8167–8177; e) C. Cheng, J. Sun, C. Wang, Y.
Zhang, S. Wei, F. Jiang, Y. Wu, Chem. Commun. 2006, 215–
217; f) S. S. Chimni, D. Mahajan, V. V. Suresh Babu, Tetrahe-
dron Lett. 2005, 46, 5617–5619.
[25] P. K. Ruprah, J.-P. Cros, J. E. Pease, W. G. Whittingham,
J. M. J. Williams, Eur. J. Org. Chem. 2002, 3145–3152.
[26] M. E. Adler, S. A. Yumagulova, M. S. Miftakov, Z. Org. Khim.
1994, 30, 943–944.
[27] a) B. List, J. Am. Chem. Soc. 2002, 124, 5656–5657; b) W. Zhu-
ang, N. Kumaragubaran, K. Juhl, A. Bogevig, K. A.
Jørgensen, J. Am. Chem. Soc. 2002, 124, 6254–6255; c) A. Bog-
evig, K. Juhl, N. Kumaragubaran, W. Zhuang, K. A.
Jørgensen, Angew. Chem. Int. Ed. 2002, 41, 1790–1793.
[28] The absolute configuration of the major enantiomer was as-
signed the (R) configuration by analogy with the configuration
of the closely related α-amination products obtained with -
proline.[27b]
[9] G. L. Schmir, T. C. Bruice, J. Am. Chem. Soc. 1958, 80, 148–
156.
[10] a) T. Darbre, M. Machuqueiro, Chem. Commun. 2003, 1090–
1091; b) J. Kofoed, M. Machuqueiro, J. L. Reymond, T. Dar-
bre, Chem. Commun. 2004, 1540–1541; c) J. Kofoed, T. Darbre, [29] K. Maruoka, T. Kano, J. Takai, O. Tokuda, Angew. Chem. Int.
J. L. Reymond, Chem. Commun. 2006, 1482–1484.
Ed. 2005, 44, 3055–3057.
[11] D. Bonnet-Delpon, B. Crousse, J.-P. Begue, Synlett 2004, 18–
Received: July 31, 2006
Published Online: October 10, 2006
29.
[12] C. Ji, Y. Peng, C. Huang, N. Wang, Y. Jiang, Synlett 2005, 987–
Note: Minor modifications have been made to the manuscript
990.
at the authors’ request since its publication in Early View.
Eur. J. Org. Chem. 2007, 167–177
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
177