Conversion between difluorocarbene and difluoromethylene ylide

Article abstract of DOI:10.1002/chem.201303248

The interconversion between difluoromethylene ylide and difluorocarbene is described. The difluoromethylene ylide precursor, Ph₃P ⁺CF₂CO₂^-, could be turned into an efficient difluorocarbene reagent, w

Full text of DOI:10.1002/chem.201303248

FULL PAPER
DOI: 10.1002/chem.201303248
Conversion between Difluorocarbene and Difluoromethylene Ylide
Jian Zheng, Jin-Hong Lin, Ji Cai, and Ji-Chang Xiao*[a]
Dedicated to Professor Jeanꢀne M. Shreeve on the occasion of her 80th birthday
Abstract: The interconversion between difluoromethylene ylide and difluorocar-
À
bene is described. The difluoromethylene ylide precursor, Ph3P+CF2CO2 , could
be turned into an efficient difluorocarbene reagent, whereas the classical difluoro-
Keywords: carbenes
clopropanation difluoro-olefina-
tion · fluorine · ylides
· difluorocy-
carbene reagents, HCF2Cl and FSO2CF2CO2TMS, could generate highly reactive
difluoromethylene ylide. Thus the Wittig difluoro-olefination and difluorocyclopro-
panation could be selectively realized by using the same reagent. In addition, the
ylides obtained from different carbene sources showed different reactivity in
Wittig reactions.
·
Introduction
The capture of difluorocarbene from HCF2Cl by triphe-
nylphosphine to form the difluoromethylene phosphonium
ylide (Ph3P+CF2 ) was first described by Franzen,[6] but
À
Difluorocarbene and difluoromethylene phosphonium ylide
are highly reactive intermediates in the construction of fluo-
rine-containing organic molecules.[1] As a singlet carbene,[2]
difluorocarbene (DCF2) is stabilized by two fluorine atoms
and reacts readily with electron-rich substrates.[3] Triphenyl-
phosphine can trap difluorocarbene to form phophonium
a later study claimed that the reaction could not be repeat-
ed.[7] Other attempts to trap the difluorocarbene from
CF2Br2 or (CF3)2Cd with PPh3 were not successful.[7,8] The
generation of the ylide from difluorocarbene was also pro-
posed by Fuqua et al. in the Wittig reaction of aldehyde
with ClCF2CO2Na/PPh3.[9] However, Burton et al. proposed
À
À
ylide (Ph3P+CF2 ) in situ.[4] It seems that Ph3P+CF2 might
be more stable than difluorocarbene. However, computa-
tional examination of the binding and structure of difluoro-
methylene phosphonium ylide showed some discrepancy.[5]
Using ab initio molecular orbital energy at the SCF level,
Dixon and Smart found that the bond length between
À
À
that Ph3P+CF2 was generated from Ph3P+CF2CO2 , not
from the combination of difluorocarbene and PPh3.[10] Re-
cently, we successfully synthesized the intermediate
À
Ph3P+CF2CO2 and confirmed the conjecture made by
Burton et al.[11] The association of difluorocarbene with
PPh3 and the successful application in subsequent Wittig re-
action was reported by Robins and Nowak, but a toxic re-
agent, (CF3)2Hg, was employed as the difluorocarbene pre-
cursor.[4a] Dolbier et al. employed FSO2CF2CO2CH3 as the
À
carbon and phosphorus in H3P+CF2 was 3.54 ꢀ, which
means that DCF2 and PH3 could be regarded as two separate
species with little interaction.[5a] However, a later calculation
of H3P+CF2 at the HF/3-21G* level by Francl et al. showed
À
difluorocarbene source to generate the ylide (Ph3P+CF2 )
À
À
that the C P bond length was 1.635 ꢀ, which could be inter-
preted as a double bond between carbon and phosphrous.[5b]
and successfully realized the Wittig difluoro-olefination.[4b]
However, starting from common and easily available di-
fluorocarbene reagents such as HCF2Cl, attempts to form
À
Recently, Dolbier found that the calculated C P bond
À
length in Ph3P+CF2 at the M06-2X/6-311+G (2df, 2p) level
À
À
is 1.815 ꢀ.[4b] The dissociation of Ph3P+CF2 to difluorocar-
Ph3P+CF2 were not successful.[7] Therefore, to make this
bene and PPh3 in this case would only have to cross
a 9.27 kcalmolÀ1 energy barrier, which suggested the possi-
ylide formation procedure widely applicable for difluoro-
olefination, it would be worthwhile to investigate the con-
version from difluorocarbene to difluoromethylene ylide.
As for the reverse process, Burton and co-workers ob-
served the dissociation of the difluoromethylene ylide into
difluorocarbene in the reaction of (bromodifluoromethyl)tri-
phenylphosphonium bromide with tertiary phosphine and
trialkyl phosphite.[12] A similar process was proposed by
Jaiswal et al. during the preparation of hydrofluorocar-
bons.[13] However, this procedure was not applied in typical
reactions involving difluorocarbene. Thus the conversion
from the difluoromethylene ylide to difluorocarbene re-
mains to be further investigated.
bility of interconversion between the ylide (Ph3P+CF2 ) and
À
carbene (DCF2).
[a] J. Zheng, Dr. J.-H. Lin, Dr. J. Cai, Prof. Dr. J.-C. Xiao
Key Laboratory of Organofluorine Chemistry
Shanghai Institute of Organic Chemistry
Chinese Academy of Sciences
345 Lingling Road, Shanghai 200032 (P.R. China)
Fax : (+86)21-6416-6128
E-mail: jchxiao@sioc.ac.cn
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201303248.
Chem. Eur. J. 2013, 00, 0 – 0
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Table 2. Reactions of difluorocarbene generated from the ylide
(Ph3P+CF2À).[a]
Results and Discussion
We have previously demonstrated that the phosphobetaine
À
intermediate (Ph3P+CF2CO2 , 1) is an efficient and reliable
precursor of difluoromethylene ylide.[11] The simple decar-
À
boxylation of Ph3P+CF2CO2 at temperatures below 808C
can give the difluoromethylene ylide, which would make it
feasible and convenient to explore the conversion from di-
fluoromethylene ylide to difluorocarbene. 4-Vinyl-1,1’-bi-
phenyl (2a) was chosen as a model substrate to detect the
generation of difluorocarbene. It was found that the desired
reaction did not happen in polar solvents (Table 1, entries 1
Entry
1
Reactant
Product
Yield [%][b]
80
2
3
73
92
Table 1. Conditions
for
the
difluorocyclopropanation
with
À
[a]
Ph3P+CF2CO2
.
4
50
Entry
Solvent
T [8C]
1:2a[b]
t [h]
Yield [%][c]
5
6
77
62
1
2
3
4
5
6
7
8
9
DMF
DG
methyl benzoate
1,4-dioxane
THF
p-xylene
cyclohexane
p-xylene
p-xylene
p-xylene
p-xylene
p-xylene
80
80
90
80
80
80
80
80
80
60
90
90
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
1.5:1
2:1
2:1
2:1
8
8
2
8
8
8
8
4
4
8
8
8
ND[d]
ND[d]
24
45
30
77
73
75
54
7[c]
8[c]
70
87
10
11
12[e]
22
80
80
9[c]
81
[a] 0.1 mmol of 2a in 1 mL of solvent. [b] Molar ratio. [c] Determined by
19F NMR spectroscopy with the quantitative addition of trifluoromethyl-
benzene as the internal standard. [d] Not detected by 19F NMR spectros-
copy. [e] 0.8 mmol of 2a in 1.2 mL of p-xylene. DG=diglyme.
[a] Reaction conditions:
(1.2 mL) for 8 h at 908C. [b] Yield of isolated product. [c] Performed for
1 h.
1
(1.6 mmol) and
2
(0.8 mmol) in p-xylene
and 2). To our delight, the difluorocyclopropanation did
occur in less-polar solvents (Table 1, entries 3–5). Better
yields were obtained in cyclohexane or para-xylene (Table 1,
entries 6 and 7). These results suggest that polar solvents
could help to stabilize the charge-separation ylide
uct in moderate to good yields (Table 2, entries 1–6). An
electron-rich alkene reacted better than an electron-poor
alkene (Table 2, entry 3 vs. 4). Moderate yield could be ob-
tained in the case of the disubstituted alkene (Table 2,
entry 6), indicating that the steric hindrance did not signifi-
cantly influence the reaction. Besides the difluorocyclopro-
panation reaction, heteroatom–hydrogen bond insertion
À
Ph3P+CF2 , whereas less-polar or non-polar solvents would
favor the dissociation of the ylide into difluorocarbene. Fur-
ther screening of the reaction temperature and molar ratio
showed that a 2:1 ratio of 1 to 2a at 908C for 8 h were the
optimal reaction conditions (Table 1, entry 11). Due to the
volatility of the product, less solvent (0.1 mmol of 2a in
1 mL of para-xylene) was used so that the reaction mixture
could be directly subject to column chromatography after
the completion of the reaction, thus reducing the inevitable
loss of the product during work-up. The reaction proceeded
equally well for a scaled-up reaction in a small amount of
para-xylene (Table 1, entry 12), indicating the high efficiency
of this reaction.
À
(X H, X=O, S, N) is another typical reaction of difluoro-
carbene.[14] However, due to the poor electrophilicity of di-
fluorocarbene, excess base was usually required to produce
the heteroatom anion, thereby promoting its reaction with
difluorocarbene.[14b–d,15] It was found that the difluorocar-
À
bene derived from Ph3P+CF2CO2 inserted smoothly into
À
the X H bond without the addition of any base, giving good
yields of the difluoromethylation product (Table 2, en-
tries 7–9). The above difluorocyclopropanation and difluoro-
methylation reactions demonstrated the successful conver-
sion from difluoromethylene ylide to difluorocarbene. The
difluorocarbene can be generated at a relatively low reac-
tion temperature without any external additive or catalyst,
which might make the difluoromethylene ylide precursor
We then investigated the scope of the reaction under the
optimal reaction conditions. In most cases, the reaction pro-
ceeded very well to give the difluorocyclopropanation prod-
&
2
&
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Conversion between Difluorocarbene and Difluoromethylene Phosphonium Ylide
FULL PAPER
À
(Ph3P+CF2CO2 , 1) a simple and convenient difluorocarbene
reagent.
that polar solvents favored the ylide formation, which is
consistent with the previous observation.[4b] Further optimi-
zation of the temperature, the amount of propylene epoxide
and the reaction time revealed that a moderate yield of the
desired product could be obtained after being heated at
1108C for 6 h in the presence of 2 equivalents of propylene
The successful conversion from difluoromethylene ylide
to difluorocarbene prompted us to investigate its reverse
process, from difluorocarbene to difluoromethylene ylide.
Although many difluorocarbene reagents have been devel-
oped, most are inefficient or difficult to obtain: difluorodia-
zirine, Me3SnCF3, and (CF3)2Hg, for instance, require sever-
al steps to prepare and involve expensive or toxic materi-
als.[16] The generation of difluorocarbene from hexafluoro-
propylene oxide (HFPO) must be carried out in an auto-
clave at high temperature.[17] Chlorodifluoromethane
(HCF2Cl) is a classical difluorocarbene reagent from which
difluorocarbene can be easily generated in the presence of
base, such as potassium tert-butoxide or hydroxide.[14a,18]
Speziale and Ratts once tried to capture the difluorocarbene
from HCF2Cl by PPh3 to form the difluoromethylene ylide
Ph3P+CF2À; however, their attempts failed.[7] It was suggest-
ed that difluorocarbene preferentially reacted with the
strong base rather than PPh3. We thought that reducing the
concentration of base might help the capture of difluorocar-
bene with PPh3. It was reported that the low concentration
of alkoxy anion produced from the ring-opening reaction of
ethylene epoxide could be used as the base for the genera-
tion of DCF2 from HCF2Cl.[19] Therefore, this procedure was
adopted for the trapping of difluorocarbene with PPh3.
In the presence of propylene oxide and tetra-n-butylam-
monium chloride, the reaction of 4-bromobenzaldehyde
with HCF2Cl and PPh3 gave the difluoro-olefination prod-
uct, indicating the formation of difluoromethylene ylide
oxide (Table 3, entry 8). Water was known to be detrimental
À [4b,14d]
to the reaction of DCF2 or Ph3P+CF2 .
Therefore, 4 ꢀ
MS were employed to remove trace amounts of water in the
reaction. It was found that the yield was significantly in-
creased with the addition of 4 ꢀ MS (Table 3, entry 9).
Under these conditions, we tried to reduce the amount of
nBu4NCl. The reaction proceeded smoothly, even when only
a catalytic amount of nBu4NCl was used (Table 3, entries 10
and 11). To our surprise, the difluoro-olefination reaction
took place even without the presence of nBu4NCl (Table 3,
entry 12), suggesting that 4 ꢀ MS acted both as a drying
agent and as a Lewis acid to promote the ring opening of
propylene epoxide. In the absence of propylene oxide, no
desired product was detected (Table 3, entry 13), which
means that the alkoxy anion from propylene epoxide initiat-
ed the reaction of HCF2Cl.
The difluoromethylene ylide derived from HCF2Cl could
be applied to the difluoro-olefination of a variety of aryl al-
dehydes, giving the corresponding gem-difluoroalkenes in
good to excellent yields (Table 4, entries 1–5). The relatively
lower yield of meta-trifluoromethyl benzaldehyde obtained
is due to the high volatility of the product; the yield deter-
mined by 19F NMR analysis was 81% (Table 4, entry 2).
Heteroaryl aldehydes are also suitable substrates for this re-
action (Table 4, entries 6 and 7). The reaction proceeded
quite well for the a,b-unsaturated aldehyde or enolizable al-
dehyde (Table 4, entries 8 and 9). However, even in the case
of an activated ketone, only a moderate yield of difluoro-
olefinated product could be obtained (Table 4, entry 10).
This reaction occurred through the following process
(Scheme 1). The ring-opening reaction of propylene oxide
Ph3P+CF2 in the reaction (Table 3, entry 1). Among the sol-
vents tested, DMF was shown to be the most suitable for
the difluoro-olefination (Table 3, entries 1–4). This indicated
À
Table 3. Optimization of reaction conditions for difluoro-olefination with
HCF2Cl.[a]
Entry
Solvent
nBu4NCl
[equiv]
T
[8C]
PO[b]
[equiv]
t
Yield[c]
[%]
G
N
[h]
1
2
3
4
5
6
7
8
DMF
DG
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.2
0.1
–
80
80
80
80
90
110
110
110
110
110
110
110
110
1.0
1.0
1.0
1.0
1.0
1.0
1.0
2.0
2.0
2.0
2.0
2.0
–
4
4
4
4
4
4
6
6
6
6
6
6
6
15
<1
6
MB[d]
p-xylene
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
3
24
42
51
56
80
83
75
42
ND[f]
Scheme 1. Formation of difluorocarbene and difluoromethylene ylide
from HCF2Cl.
9[e]
10[e]
11[e]
12[e]
13[e]
by chloride afforded the alkoxy anion. This then abstracted
a proton from HCF2Cl to form the chlorodifluoromethyl
anion, which underwent decomposition to give difluorocar-
bene and a chloride anion. The chloride anion entered into
the next reaction cycle. Therefore, only a catalytic amount
of nBu4NCl was needed for the reaction. If water was pres-
ent in the reaction system, the alkoxy anion would be con-
–
[a] Performed on a 0.1 mmol scale based on 4a. [b] Propylene oxide.
[c] Determined by 19F NMR spectroscopy with the quantitative addition
of trifluoromethylbenzene as the internal standard. [d] Methylbenzoate.
[e] 50 mg 4 ꢀ MS was added. [f] Not detected by 19F NMR spectroscopy.
Chem. Eur. J. 2013, 00, 0 – 0
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

J.-C. Xiao et al.
Table 4. Wittig difluoro-olefination with HCF2Cl.[a]
It has been reported that FSO2CF2CO2TMS (TFDA)
could smoothly generate difluorocarbene in methylbenzoa-
te.[20a,c] The first attempt at Wittig difluoro-olefination of 4-
bromobenzaldehyde with TFDA in the presence of PPh3
was made in this solvent. Indeed, the reaction gave the de-
sired product, indicating the successful conversion from di-
Entry
1
Reactant
Product
Yield [%][b]
77
À
fluorocarbene to difluoromethylene ylide Ph3P+CF2 , but
only a low yield was obtained (Table 5, entry 1). Some other
2
3
49, 81[c]
92
Table 5. Optimization of reaction conditions for Wittig difluoro-olefina-
tion of FSO2CF2CO2TMS.
4
5
78
94
Entry
Solvent
T
[8C]
t
Molar ratio
6:7:4a
Yield
[%][a]
[h]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
MB[b]
DMF
p-xylene
benzonitrile
CH3CN
EA
EA
EA
EA
EA
EA
EA
EA
EA
EA
EA
120
120
120
120
120
120
70
80
90
100
90
90
2
2
2
2
2
2
2
2
2
2
2
2
2
2
0.5
1
2:2:1
2:2:1
2:2:1
2:2:1
2:2:1
2:2:1
2:2:1
2:2:1
2:2:1
2:2:1
1:1:1
1.5:1.5:1
2.5:2.5:1
3:3:1
2.5:2.5:1
2.5:2.5:1
26
12
27
19
53
76
72
83
86
85
46
80
94
91
70
95
6
7
89
74
8
9
90
61
90
90
90
90
10
43
[a] Determined by 19F NMR spectroscopy with the quantitative addition
of trifluoromethylbenzene as the internal standard. [b] Methylbenzoate.
[a] Reaction conditions: Ph3P (1.6 mmol), HCF2Cl (3.5 mmol), 4 ꢀ MS
(75 mg), nBu4NCl (0.16 mmol), propylene epoxide (1.6 mmol), and 4
(0.8 mmol) in DMF (1.5 mL) for 6 h at 1108C. [b] Yield of isolated prod-
uct. [c] Determined by 19F NMR spectroscopy with trifluoroacetic acid as
the internal standard.
solvents were then investigated (Table 5, entries 2–6) and it
was found that solvent had significant impact on the reac-
tion. In ethyl acetate (EA), the result was much better than
that in other solvents (Table 5, entry 6). After further
screening of the reaction temperature, reaction time, and
the molar ratio of Ph3P/FSO2CF2CO2TMS/substrate
(Table 5, entries 7–16), we decided that EA as solvent, a re-
action temperature of 908C, and a molar ratio of 2.5:2.5:1
were the optimal conditions (Table 5, entry 16).
The reaction could be applied to a variety of carbonyl
compounds including aromatic, heteroaromatic, and aliphat-
ic aldehydes and activated ketones (Table 6). More impor-
tantly, the reaction worked very well in the case of electron-
deficient aldehydes (Table 6, entries 3 and 9), demonstrating
a higher reactivity of the difluoromethylene ylide obtained
from TFDA in comparison with other reported meth-
ods.[4b,9c] In the case of 4-nitrobenzaldehyde (Table 6,
entry 3), difluoro-olefination with difluoromethylene ylide
usually gave very low yields of the expected difluoro-olefi-
nated product; only 2.1% yield was obtained by Fuquaꢂs
method,[9c] and only 15% yield by Dolbierꢂs method.[4b] This
indicated that ylides obtained from different carbene sour-
ces have different reactivity.
sumed. Thus molecular sieves were used to remove it. Be-
cause the alkoxy anion was generated by ring-opening of
the epoxide, it might always exist in very low concentrations
and not impede the capture of difluorocarbene with PPh3,
which would help the formation of difluoromethylene ylide.
It should be noted that the difluoromethylene ylide de-
rived from HCF2Cl did not react well with electron-deficient
aldehydes, such as 4-nitrobenzaldehyde. This was also re-
ported by Fuqua and Dolbier and co-workers.[4b,9c] As we
know, the reactivity of difluorocarbene from different pre-
cursors varied greatly. This prompted us to consider that dif-
ferent difluorocarbene sources might make some difference
in the reactivity of the corresponding difluoromethylene
ylides. Recently, trimethylsilyl fluorosulfonyldifluoroacetate
(FSO2CF2CO2TMS; TFDA) has been found to be an effi-
cient difluorocarbene reagent that can react smoothly with
electron-deficient alkenes.[20] To explore the applicability of
this ylide formation method and the reactivity of the ylide
from different carbene sources, we investigated the genera-
tion and reaction of difluoromethylene ylide from TFDA.
&
4
&
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Conversion between Difluorocarbene and Difluoromethylene Phosphonium Ylide
Table 6. Wittig difluoro-olefination of FSO2CF2CO2TMS.[a]
FULL PAPER
Experimental Section
General methods: Solvents and reagents were purchased from commer-
cial sources and used as received, unless otherwise noted. The solvents
para-xylene, DMF, and EA were distilled over CaH2. 1H, 13C and
19F NMR spectra were detected on a 500, 400or 300 MHz NMR spec-
Entry
1
Reactant
Product
Yield [%][b]
87
1
trometer. Data for H, 13C and 19F NMR were recorded as follows: chem-
ical shift (d, ppm), multiplicity (s=singlet, d=doublet, t=triplet, m=
multiplet, q=quartet), coupling constant/s in Hz. Mass spectra were ob-
tained on a GC-MS. High-resolution mass data were recorded on a high
resolution mass spectrometer in the EI or ESI mode.
2
3
91
70
À
Dissociation of difluoromethylene ylide to carbene: Ph3P+CF2CO2 (1,
570 mg, 1.6 mmol) and 4-vinyl-1,1’-biphenyl (2a, 144 mg, 0.8 mmol) were
added to a 5 mL sealed tube. The mixture was degassed and then para-
xylene (1.2 mL) was added under N2. The reaction mixture was stirred at
908C for 8 h. After being cooled to room temperature, the solution was
subjected to flash column chromatography to give the pure product 3a.
4
5
6
84
86
81
Wittig difluoro-olefination of HCF2Cl: Ph3P (420 mg, 1.6 mmol), 4-bro-
mobenzaldehyde (4a, 148 mg, 0.8 mmol), and 4 ꢀ MS (75 mg) were
added to a 5 mL sealed tube. The mixture was degassed and then a solu-
tion of HCF2Cl in DMF (2.3m, 1.5 mL) was added under N2. The reac-
tion mixture was stirred at 1108C for 6 h. After being cooled to room
temperature, the solution was subjected to flash column chromatography
to give the pure product 5a.
7
8
87
93
Wittig difluoro-olefination of FSO2CF2CO2TMS: Ph3P (525 mg,
2.0 mmol), 4-bromobenzaldehyde (4a, 148 mg, 0.8 mmol), and NaF
(1.6 mg, 0.038 mmol) were added to a 25 mL Schlenk tube. The mixture
was degassed and then EA (1.2 mL) was added under N2. The reaction
mixture was stirred at 908C for 1 min. FSO2CF2CO2TMS (501 mg,
2.0 mmol) was added dropwise over 10 min. The resulting mixture was
stirred at the same temperature for another 1 h. After being cooled to
room temperature, it was subjected to flash column chromatography di-
rectly to give the pure product 5a.
9
96
54
10
11
65
Acknowledgements
We thank the National Natural Science Foundation (21032006,
21172240), the 973 Program of China (2012CBA01200), and the Chinese
Academy of Sciences.
[a] Reaction conditions: Ph3P (2.0 mmol), FSO2CF2CO2TMS (2.0 mmol),
NaF (0.016 mmol), and 4 (0.8 mmol) in EA (1.5 mL) for 1 h at 908C.
[b] Yield of isolated product.
[1] For reviews, see: a) D. L. S. Brahms, W. P. Dailey, Chem. Rev. 1996,
96, 1585–1632; b) D. Burton, Z.-Y. Yang, W. Qiu, Chem. Rev. 1996,
96, 1641–1716; c) W. R. Dolbier, Jr., M. A. Battiste, Chem. Rev.
2003, 103, 1071–1098; for examples, see: d) R. A. Moss, L. Wang, K.
Krogh-Jespersen, J. Am. Chem. Soc. 2009, 131, 2128–2130; e) J. M.
Lenhardt, M. T. Ong, R. Choe, C. R. Evenhuis, T. J. Martinez, S. L.
Craig, Science 2010, 329, 1057–1060; f) F. Wang, T. Luo, J. Hu, Y.
Wang, H. S. Krishnan, P. V. Jog, S. K. Ganesh, G. K. S. Prakash,
G. A. Olah, Angew. Chem. 2011, 123, 7291–7295; Angew. Chem. Int.
Ed. 2011, 50, 7153–7157; g) Y. Zhao, B. Gao, J. Hu, J. Am. Chem.
Soc. 2012, 134, 5790–5793.
[2] M. S. Gordon, J. A. Pople, J. Chem. Phys. 1968, 49, 4643–4650.
[3] J. F. Harrison, J. Am. Chem. Soc. 1971, 93, 4112–4119.
[4] a) I. Nowak, M. J. Robins, Org. Lett. 2005, 7, 721–724; b) C. S. Tho-
moson, H. Martinez, W. R. Dolbier, Jr., J. Fluorine Chem. 2013, 150,
53–59.
Conclusion
The interconversion between difluoromethylene ylide and
difluorocarbene has been successfully achieved. By changing
the reaction conditions, the difluoromethylene ylide precur-
À
sor, Ph3P+CF2CO2 , was converted into an efficient difluoro-
carbene reagent, whereas the classical difluorocarbene re-
agents, HCF2Cl and FSO2CF2CO2TMS, generated the highly
reactive difluoromethylene ylide. The difluoromethylene
ylides obtained from different difluorocarbene reagents
showed different reactivity in Wittig reactions. Therefore,
the difluoro-olefination and difluorocyclopropanation could
be selectively conducted by using the same reagent. Further
research on the conversion between other fluorinated ylide
and carbene is currently underway.
[5] a) D. A. Dixon, B. E. Smart, J. Am. Chem. Soc. 1986, 108, 7172–
7177; b) M. M. Francl, R. C. Pellow, L. C. Allen, J. Am. Chem. Soc.
1988, 110, 3723–3728.
[6] V. Franzen, Angew. Chem. 1960, 72, 566.
[7] A. J. Speziale, K. W. Ratts, J. Am. Chem. Soc. 1962, 84, 854–859.
[8] L. Riesel, H. Vogt, V. Kolleck, Z. Anorg. Allg. Chem. 1989, 574,
143–152.
[9] a) S. A. Faqua, W. G. Duncan, R. M. Silverstein, Tetrahedron Lett.
1964, 5, 1461–1463; b) S. A. Fuqua, W. G. Duncan, R. M. Silver-
Chem. Eur. J. 2013, 00, 0 – 0
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

J.-C. Xiao et al.
stein, J. Org. Chem. 1965, 30, 2543–2545; c) S. A. Fuqua, W. G.
Duncan, R. M. Silverstein, J. Org. Chem. 1965, 30, 1027–1029.
[10] E. H. Frank, J. B. Donald, J. Org. Chem. 1967, 32, 1311–1318.
[11] J. Zheng, J. Cai, J.-H. Lin, Y. Guo, J.-C. Xiao, Chem. Commun.
2013, 49, 7513–7515.
[12] a) D. J. Burton, D. G. Naae, R. M. Flynn, B. E. Smart, D. R. Brittelli,
J. Org. Chem. 1983, 48, 3616–3618; b) R. M. Flynn, D. J. Burton,
D. M. Wiemers, J. Fluorine Chem. 2008, 129, 583–589.
[16] a) R. A. Mitsch, E. W. Neuvar, R. J. Koshar, D. H. Dybvig, J. Heter-
ocycl. Chem. 1965, 2, 371–375; b) R. A. Mitsch, J. Am. Chem. Soc.
1965, 87, 758–761; c) D. Seyferth, H. Dertouzos, R. Suzuki, J. Y. P.
Mui, J. Org. Chem. 1967, 32, 2980–2984; d) I. Nowak, J. F. Cannon,
M. J. Robins, Org. Lett. 2004, 6, 4767–4770.
[17] P. B. Sargeant, C. G. Krespan, J. Am. Chem. Soc. 1969, 91, 415–419.
[18] a) T. G. Miller, J. W. Thanassi, J. Org. Chem. 1960, 25, 2009–2012;
b) G. C. Robinson, Tetrahedron Lett. 1965, 6, 1749–1752.
[19] a) P. Weyerstahl, D. Klamann, C. Finger, F. Nerdel, J. Buddrus,
Chem. Ber. 1967, 100, 1858–1869; b) M. Kamel, W. Kimpenhaus, J.
Buddrus, Chem. Ber. 1976, 109, 2351–2369.
[20] a) F. Tian, V. Kruger, O. Bautista, J.-X. Duan, A.-R. Li, W. R. Dol-
bier, Jr., Q.-Y. Chen, Org. Lett. 2000, 2, 563–564; b) W. Xu, Q.-Y.
Chen, J. Org. Chem. 2002, 67, 9421–9427; c) W. R. Dolbier, Jr., F.
Tian, J.-X. Duan, A.-R. Li, S. Ait-Mohand, O. Bautista, S. Buathong,
J. Marshall Baker, J. Crawford, P. Anselme, X. H. Cai, A. Modze-
lewska, H. Koroniak, M. A. Battiste, Q.-Y. Chen, J. Fluorine Chem.
2004, 125, 459–469.
[13] P. S. Bhadury, M. Palit, M. Sharma, S. K. Raza, D. K. Jaiswal, J. Flu-
orine Chem. 2002, 116, 75–80.
[14] a) J. Hine, J. J. Porter, J. Am. Chem. Soc. 1957, 79, 5493–5496; b) L.
Zhang, J. Zheng, J. Hu, J. Org. Chem. 2006, 71, 9845–9848; c) Y. Za-
frani, G. Sod-Moriah, Y. Segall, Tetrahedron 2009, 65, 5278–5283;
d) P. S. Fier, J. F. Hartwig, Angew. Chem. 2013, 125, 2146–2149;
Angew. Chem. Int. Ed. 2013, 52, 2092–2095.
[15] a) G. Liu, X. Wang, X. Lu, X.-H. Xu, E. Tokunaga, N. Shibata,
ChemistryOpen 2012, 1, 227–231; b) G. Liu, X. Wang, X.-H. Xu, X.
Lu, E. Tokunaga, S. Tsuzuki, N. Shibata, Org. Lett. 2013, 15, 1044–
1047.
Received: August 17, 2013
Published online: && &&, 0000
&
6
&
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Conversion between Difluorocarbene and Difluoromethylene Phosphonium Ylide
FULL PAPER
Ylides
J. Zheng, J.-H. Lin, J. Cai,
J.-C. Xiao* .................... &&&& — &&&&
Book of ylide: Difluoromethylene
FSO2CF2CO2TMS, generated the
highly reactive difluoromethylene
ylide. Interestingly, it was found that
the ylides obtained from different car-
bene sources showed different reactiv-
ity in Wittig reactions (see scheme).
Conversion between Difluorocarbene
and Difluoromethylene Ylide
phosphonium ylide, generated from
À
Ph3P+CF2CO2 , was turned into an
efficient difluorocarbene reagent,
whereas the classical difluorocarbene
reagents, HCF2Cl and
Chem. Eur. J. 2013, 00, 0 – 0
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!