Reactivity of stable trifluoroacetaldehyde hemiaminals. 1. An unexpected reaction with enolizable carbonyl compounds

Article abstract of DOI:10.1021/jo015587u

In the presence of enolizable carbonyl compounds, hemiaminals of fluoral and related polyfluoroaldehydes behave as equivalents of fluoroalkyl iminium compounds and provide β-polyfluoroalkyl β-dialkylamino ketones, which are easily transformed, under acidic conditions, into β-polyfluoroalkylenones.

Full text of DOI:10.1021/jo015587u

4826
J . Org. Chem. 2001, 66, 4826-4830
Rea ctivity of Sta ble Tr iflu or oa ceta ld eh yd e Hem ia m in a ls. 1. An
Un exp ected Rea ction w ith En oliza ble Ca r bon yl Com p ou n d s
Gae¨lle Blond, Thierry Billard,* and Bernard R. Langlois*
Laboratoire SERCOF (UMR CNRS 5622), Universite´ Claude Bernard - Lyon1,
43 Bd du 11 novembre 1918, 69622 Villeurbanne, France
Bernard.Langlois@univ-lyon1.fr
Received February 22, 2001
In the presence of enolizable carbonyl compounds, hemiaminals of fluoral and related polyfluoro-
aldehydes behave as equivalents of fluoroalkyl iminium compounds and provide â-polyfluoroalkyl
â-dialkylamino ketones, which are easily transformed, under acidic conditions, into â-polyfluoro-
alkylenones.
Sch em e 1. Hem ia m in a ls of F lu or a l
In tr od u ction
The unique behavior of fluorinated compounds makes
them suitable for various applications.1 In particular, the
lipophilicity attached to the trifluoromethyl moiety makes
trifluoromethylated molecules of great interest, especially
for biological purposes. However, their preparation by
direct introduction of a nucleophilic CF3 moiety on
organic substrates, which is the most promising general
route, is still suffering from a limited number of efficient
Ta ble 1. Rea ction of 1-3 w ith Acetop h en on e
-
reagents, able to stabilize the very unstable CF3 anion.2
Our investigations for designing new nucleophilic tri-
fluoromethylating reagents, recently led us to describe
a family of suitable reagents, namely the hemiaminals
of fluoral,3 which effectively transform nonenolizable car-
bonyl compounds into trifluoromethyl carbinols.3a-c In the
present paper, we describe their reaction toward enoliz-
able substrates, which follows a completely different
pathway.
recovd
1, 2, or
3b (%)
1, 2 or 3
entry (equiv)
conditions
THF/rt
t-BuOK (2 equiv)/24 h
THF/rt
t-BuOK (0.1 equiv)
THF/80 °C
4a or ba (%)
1
2
3
4
5
6
7
8
9
1 (2)
1 (2)
1 (2)
1 (1)
2 (1)
2 (2)
2 (2)
3 (1)
3 (2)
0
0
Resu lts a n d Discu ssion
24 h: 4a (14)
4 days: 4a (48)
7 h: 4a (68)
24 h: 4a (76)
4a (30)
During our studies of the behavior of fluoral hemiami-
nals 1-3 as nucleophilic trifluoromethylating reagents
(Scheme 1), we were driven to put them in reaction with
acetophenone. Surprisingly, when the procedures previ-
ously used to trifluoromethylate benzophenone were
applied to this substrate, the expected trifluoromethyl
carbinol was never obtained.
45
40
20
t-BuOK (0.1 equiv)
THF/80 °C/24 h
t-BuOK (0.1 equiv)
DMEc/80 °C/5 h
cat. CsF
4a (50)
4a (90)
30
47
DMEc/80 °C/5 h
cat. CsF
For example, when 1 (2 equiv) was reacted with
potassium tert-butoxide (2 equiv) and acetophenone (1
DMEc/rt
5 h: 4a (7)
24 h: 4a (23)
4b (50)
cat. CsF
DMEc/80 °C/5 h
cat. CsF
30
40
(1) (a) Gross, U.; Ru¨diger, ST. In Organo-Fluorine Compounds;
Baasner, B., Hagemann, H., Tatlow, J . C., Eds; Houben-Weyl: Methods
of Organic Chemistry; Thieme: Stuttgart, 1999; Vol. E10a, pp 18-26.
(b) Filler, R.; Kobayashi, Y.; Yagulpolskii, Y. L. Organofluorine
Compounds in Medicinal Chemistry and Biomedical Applications;
Elsevier: Amsterdam, 1993. (c) Banks, R. E.; Smart, B. E.; Tatlow, J .
C. Organofluorine Chemistry: Principles and Commercial Applications;
Plenum Press: New York, 1994. (d) Hudlicky, M.; Pavlath, A. E.
Chemistry of Organic Fluorine Compounds II. A critical Review; ACS
Monograph 187; American Chemical Society: Washington, DC, 1995.
(2) (a) McClinton M. A.; McClinton D. A. Tetrahedron 1992, 48,
6555-6666. (b) Prakash, G. K. S.; Yudin, A. K. Chem. Rev. 1997, 97,
757-786.
DMEc/80 °C/5 h
cat. CsF
4b (80)
Crude yields vs acetophenone determined by 19F NMR with
internal standard (PhOCF3). Yield vs 1, 2, or 3 determined by
19F NMR with internal standard (PhOCF3). c DME: 1,2-dimeth-
oxyethane.
a
b
equiv), self-aldolization of the latter compound exten-
sively occurred (Table 1, entry 1). However, when 2 or 3
(2 equiv) was opposed to acetophenone (1 equiv) in the
presence of a catalytic amount of fluoride, 3-trifluoro-
methyl-3-dialkylaminopropiophenone 4 was produced
(Table 1, entries 5-9). Moreover, the same product was
(3) (a) Billard, T.; Bruns, S.; Langlois, B. R. Org. Lett. 2000, 2, 2101-
2103. (b) Billard, T.; Langlois, B. R.; Blond, G. Tetrahedron Lett. 2000,
41, 8777-8780. (c) Billard, T.; Langlois, B. R.; Blond, G. Eur. J . Org.
Chem. 2001, 1467-1471. (d) Blond, G.; Billard, T.; Langlois, B. R.
Tetrahedron Lett. 2001, 42, 2473-2475.
10.1021/jo015587u CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/16/2001

Reactivity of Stable Trifluoroacetaldehyde Hemiaminals
J . Org. Chem., Vol. 66, No. 14, 2001 4827
Sch em e 2. Retr osyn th esis of 4
reaction between a trifluoromethyl iminium deriving
from 1 and the enolate of acetophenone (Scheme 2). In
such a hypothesis, the role of catalytic t-BuOK in the
formation of these intermediates must be elucidated.
To ascertain that the catalytic base was necessary and
that 1 was not already in equilibrium with the iminium
hydroxide 1bis, the same reaction was carried out
without base. As expected, no reaction took place (Scheme
3). Nevertheless, the absence of detectable product may
only reflect the low concentration of enolate under these
conditions. To determine if 1bis was really formed, even
in low quantities, two complementary experiments were
carried out: one in the presence of methanol and another
in the presence of 18O-enriched water. In both cases, no
transformation of 1 was observed at all, and no methoxy-
or 18O-containing product was formed. This allows us to
assert that 1 is not in equilibrium with the iminium
hydroxide 1bis.
Sch em e 3. Hyp oth esis on th e F or m a tion of 1bis
Since t-BuOK was proved to be essential in the for-
mation of 4a , we determined which reactant (acetophe-
none or 1) was deprotonated first. For this purpose, the
reaction was carried out in two steps, the first one being
the deprotonation of one of the two reactants (acetophe-
none or 1) with 0.1 equiv of t-BuOK before adding the
second partner. In both cases, the same yield of 4a was
obtained. This result demonstrated that both 1 and
acetophenone were in equilibrium with their conjugated
bases which, consequently, must be considered as inter-
mediates in the formation of 4a (Scheme 4).
obtained from acetophenone, 1 (2 equiv), and a catalytic
amount of potassium tert-butoxide (0.1 equiv) (Table 1,
entries 2-4).
When only 1 equivalent of reagents 1, 2, or 3 was used,
the yield of 4 did not exceed 50% and approximately
30-40% of 1, 2, or 3 was recovered (Table 1, entries 4, 5,
and 8). With 2 equiv of these reagents, the yield of 4
increased up to 80% and, again, around 1 equiv of 1, 2,
or 3 was recovered (Table 1, entries 3, 6, and 9).
Therefore, the optimal yield of 4 was obtained at 80
°C with 2 equiv of 1, 2, or 3 and catalytic amounts of
CsF or t-BuOK (ca. 10%).
The surprising nature of product 4 and the fact that 2
equiv of 1, 2, or 3 was needed to get good results, whereas
only 1 equiv seemed to react, prompted us to investigate
the mechanism of this reaction. As these three reagents
led to almost the same result under similar conditions,
we focused our attention on 1.
As 2 equiv of 1 was required to get a good yield of 4a
though 1 equiv of 1 was recovered, it can be assumed
that the real active species are a hydrogen-bonded
dimeric form of 1 and/or a 1:1 complex between 1 and its
conjugated base. This latter hypothesis is in accordance
with our previous work,3c which demonstrated that, in
THF, the potassium salt of 1 is a dimer in which the K+
cation of one molecule is complexed by the nitrogen atoms
of the second one. These elements are taken into account
Because of the structure of 4, it could be anticipated
that this compound resulted, at least formally, from the
Sch em e 4. Mech a n ism of th e Rea ction

4828 J . Org. Chem., Vol. 66, No. 14, 2001
Ta ble 2. Syn th esis of â-Tr iflu or om eth yl, â-(N′-Ben zyl)p ip er a zin o Ca r bon yl Com p ou n d s
Blond et al.
a
b
Isolated yield. In parentheses: crude yield determined by 19F NMR with internal standard (PhOCF3). With 4 equiv of 1.
in the mechanism we propose to explain the formation
of an iminium salt from 1 (Scheme 4).
with an enolizable ketone has been reported one time
only, but the resulting compound has not been isolated.7
We, too, mentioned the formation of 3-trifluoromethyl-
3-dimethylaminopropiophenone from fluoroform and
2-benzoyl-1-dimethylaminoethylene.8 Neverteless, in all
these works, the access to reagents and/or substrates is
not always easy. In contrast, stable reagents 1-3 are
readily available, even on a large scale,3 and are able to
react directly, in an effective way, with carbonyl com-
pounds.
Apart from their potential biological activity, the main
interest of compounds 4 lies in their sensitivity to acids
and their ability to eliminate, under such conditions,
morpholine or N-benzylpiperazine to deliver trans-â-
trifluoromethyl enones 5 (Table 3).
This mechanism implies that potassium hydroxide is
formed and regenerates the conjugated base of acetophe-
none or that of 1, justifying the fact that t-BuOK must
be used in a catalytic amount. This hypothesis was
confirmed by replacing t-BuOK by KOH (50% in water):
under these conditions, the same result was obtained.
A similar mechanism can be postulated from 2 or 3.
In this case, CsF acts as a base to enolize acetophenone4
or generate the conjugate base of 1 (from 2) or that of its
morpholino analogue (from 3), whereas the trimethylsi-
lanolate anion plays the role of the hydroxide anion in
the catalytic cycle.
On a synthetic point of view, this reaction allows the
easy preparation of â-trifluoromethyl-â-dialkylamino-
acetophenone in one step from the parent carbonylated
substrate. It has been extended not only to other alkyl
aryl ketones, but also to alkyl heteroaryl ketones and
alicyclic ketones or malonic esters (Table 2).
Thus, (E)-â-trifluoromethyl enones can be prepared
in an efficient two-step procedure by a formal (trifluoro-
methyl)methylidenation of ketones. These electron-
withdrawing substituted enones constitute very valuable
synthetic intermediates. Moreover, compound 5g could
The synthesis of some of these compounds have been
already described in the literature from hemiaminals5 or
aminals6 of fluoral and silyl enol ethers but not from
ketones. The reaction of a trifluoromethylated iminium
(6) (a) Dolbier, W. R., J r.; Xu, Y. Tetrahedron Lett. 1998, 39, 9151-
9154. (b) Dolbier, W. R., J r.; Xu, Y. J . Org. Chem. 2000, 65, 2134-
2137. (c) Fuchigami, T.; Nakagawa, Y.; Nonaka, T. J . Org. Chem. 1987,
52, 5489-5491.
(7) Ates, C.; J anousek, Z.; Viehe, H. G. Tetrahedron Lett. 1993, 34,
5711-5714.
(4) Clark, J . H. Chem. Rev. 1980, 80, 429-452.
(5) Takaya, J . H.; Kagoshima, H.; Akiyama, T. Org. Lett. 2000, 2,
1577-1579.
(8) Large, S.; Roques, N.; Langlois, B. R. J . Org. Chem. 2000, 65,
8848-8856.

Reactivity of Stable Trifluoroacetaldehyde Hemiaminals
J . Org. Chem., Vol. 66, No. 14, 2001 4829
Ta ble 3. Syn th esis of â-Tr iflu or om eth yl En on es
Ta ble 5. (P er h a logen oa lk yl)Meth ylid en a tion of
Ca r bon yl Com p ou n d s
a
Isolated yield. In parentheses: crude yield determined by 19F
NMR with internal standard (PhOCF3).
Ta ble 4. Use of Lew is Acid to Gen er a te Im in iu m
entry
Lewis acid (equiv)
conditions
4a a (%)
a
Isolated yields vs ketones. In parentheses: crude yield deter-
1
2
3
4
5
6
7
8
9
BF3‚Et2O (1)
THF/rt
THF/rt
THF/rt
THF/rt
THF/rt
THF/rt
THF/rt
CH2Cl2/50 °C
CH2Cl2/50 °C
CH2Cl2/50 °C
30
58
59
0
58
0
b
mined by 19F NMR with internal standard (PhOCF3). With 2
equiv of 2.
BF3‚Et2O (2)
BF3‚Et2O (3)
BF3‚Et2O (0.1)
BF3‚Et2O (1.2)
ClTi(O-i-Pr)3 (1.2)
Ti(O-i-Pr)4 (1.2)
BF3‚Et2O (1)
generate the iminium intermediate and 0.2 equiv to
catalyze the enolization of acetophenone.
0
This reaction confirms the possibility to generate a
trifluoromethylated iminium from 1-3 and, therefore,
reinforce the mechanism proposed in Scheme 4. It also
offers a second route to compounds 4. Moreover, as this
new synthesis of 4 is carried out under acidic conditions,
like the transformation of 4 to 5, these two steps were
combined in a one-pot preparation of 5 from simple
carbonyl compounds.
51
70
66
BF3‚Et2O (1.2)
BF3‚Et2O (2)
10
a
Determined by 19F NMR with internal standard (PhOCF3).
be an interesting analogue of the choleretic drug Cyclo-
valone (2,6-bis-(3-methoxy-4-hydroxybenzylidene)cyclo-
hexanone).9
As silylated hemiaminals of pentafluoropropionalde-
hyde and chlorodifluoroacetaldehyde can be prepared in
the same way as 2 from the hydrates of the corresponding
aldehydes, this methodology has been successfully ex-
tended to the one-pot (perhalogenoalkyl)methylidenation
of carbonyl compounds, even elaborated ones such as
cholestenone (Table 5).
In conclusion, this study shows that the stable and
readily available hemiaminals of fluoral can, under
specific conditions, generate iminium species that react
with enolizable carbonyl compounds to deliver â-trifluo-
romethyl â-dialkylamino carbonyl compounds in an ef-
ficient way. In a second step, which can be carried out in
the same pot, trifluoromethylenones can be prepared
Coming back to compounds 4, it was important to
confirm that iminium species were involved as interme-
diates in their synthesis. For this purpose, we decided
to generate iminium salts from 2 and Lewis acids and to
react them with enolizable ketones (Table 4).
As expected, 4a was produced in this way from ac-
etophenone, 2, and BF3 etherate (Table 4). It must be
noticed that Ti(IV) derivatives were not able to catalyze
this condensation (Table 4, entries 6 and 7). Best results,
similar to those resulting from fluoride activation of 2,
were obtained with 1.2 equiv of boron trifluoride (Table
4, entries 5 and 9). Probably 1 equiv was necessary to
(9) Rumpel, W. Austrian Patent Nï. 180258, 1954.

4830 J . Org. Chem., Vol. 66, No. 14, 2001
Blond et al.
evaporation, the crude mixture was deposited at the top of a
column filled with silica gel and eluted.
from them under acidic conditions. This reaction can be
extended to hemiaminals of other perhalofluoroalde-
hydes. The synthetic potential of perhalofluoro iminiums
species is under study in our laboratory.
Typ ica l P r oced u r e for th e Syn th esis of 5 fr om 4.
Trifluoroacetic acid (0.5 mL) was added, at 0 °C, to a solution
of 4 (1 mmol) in CH2Cl2 (1 mL). The mixture was heated in a
closed vessel at 80 °C for 5 h. After cooling, it was washed
twice with a 6% aqueous solution of NaHCO3, dried over
Na2SO4, and evaporated in vacuo at room temperature. Then,
the crude residue was purified by flash chromatography.
Typ ica l P r oced u r e for th e On e-P ot Syn th esis of 5 fr om
2, 6, or 7. BF3‚Et2O (1.2 mmol) was added, at room temper-
ature, to a solution of 2, 6, or 7 (1 mmol) and enolizable
carbonyl substrate (1 mmol) in CH2Cl2 (1 mL). Then, the
mixture was heated to 50 °C for 5 h. After the mixture was
cooled to 0 °C, trifluoroacetic acid (0.5 mL) was added and the
mixture was heated again to 50 °C for 5 h. After cooling and
evaporation, the crude mixture was deposited at the top of a
column filled with silica gel and eluted.
Exp er im en ta l Section
Gen er a l Rem a r k s. Solvents were distilled prior to use.
Other reagents were used as received. Flash chromatography
was performed on silica gel 60M (0.04-0.063 mm). Reagents
1-3 were prepared according to our previous work.3a-c Pen-
tafluoroethyl (6) and chlorodifluoro analogues (7) were pre-
pared with the same techniques.
Typ ica l P r oced u r e for th e Syn th esis of 4 u n d er Ca -
ta lysis w ith t-Bu OK. A 1 M solution of t-BuOK in THF (1
mL) was added, at room temperature, to a solution of 1 (2
mmol) and enolizable carbonyl substrate (1 mmol) in THF (1
mL). Then, the reaction mixture was heated to 80 °C for 24 h.
After cooling, the crude mixture was evaporated and purified
by flash chromatography. Because of their rather limited
stability, compounds 4 were rapidly engaged in the elimination
step after NMR control.
Ack n ow led gm en t. Mrs. Marie-France Cadot and
Odile Miani are greatly acknowlegded for recording
NMR Spectra. G.B. also thanks the French Minister of
Research for a Ph.D. grant.
Typ ica l P r oced u r e for th e Syn th esis of 4 u n d er Ca -
ta lysis w ith BF 3‚Et2O. BF3‚Et2O (1.2 mmol) was added, at
room temperature, to a solution of 2, 6, or 7 (1 mmol) and
enolizable carbonyl substrate (1 mmol) in CH2Cl2 (1 mL). Then,
the mixture was heated to 50 °C for 5 h. After cooling and
Su p p or tin g In for m a tion Ava ila ble: Characterization
data of all compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
J O015587U