Reactions of 1,1,2,2-tetrafluoroethyl-N,N-dimethylamine with linear and cyclic 1,3-diketones

Article abstract of DOI:10.1016/j.jfluchem.2011.06.011

Ketones are known to be unreactive toward α-fluoroamines such as Ishikawa's Reagent or 1,1,2,2-tetrafluoroethyl-N,N-dimethylamine (TFEDMA). On the other hand, 1,3-diketones were found to undergo fluorination with TFEDMA. In the case of linear 1,3-diketone

Full text of DOI:10.1016/j.jfluchem.2011.06.011

Journal of Fluorine Chemistry 132 (2011) 1198–1206
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Reactions of 1,1,2,2-tetrafluoroethyl-N,N-dimethylamine with linear and cyclic
1,3-diketones
a
b
Liane M. Grieco a, Gary A. Halliday a, Christopher P. Junk a, , Steven R. Lustig , William J. Marshall ,
*
Viacheslav A. Petrov a
a DuPont Central Research and Development,1 Experimental Station, PO Box 80500, Wilmington, DE 19880-0500, United States
b DuPont Corporate Center for Analytical Sciences, Experimental Station, PO Box 80500, Wilmington, DE 19880-0500, United States
A R T I C L E I N F O
A B S T R A C T
Article history:
Ketones are known to be unreactive toward
a-fluoroamines such as Ishikawa’s Reagent or 1,1,2,2-
Received 22 April 2011
Received in revised form 8 June 2011
Accepted 9 June 2011
tetrafluoroethyl-N,N-dimethylamine (TFEDMA). On the other hand, 1,3-diketones were found to
undergo fluorination with TFEDMA. In the case of linear 1,3-diketones, the proposed mechanism
involves the formation of
b-fluoro-a,b-unsaturated ketones followed by the addition of HF to selectively
Available online 21 June 2011
give the product -difluoroketone. Interestingly, when the 1,3-diketone is cyclic (i.e. 1,3-
b,b
cyclohexadione) the outcome of the reaction is different and results in the formation of a product
with a 2,2-difluoroacetyl group on the 2-carbon.
Keywords:
1,1,2,2-Tetrafluoroethyl-N,N-
dimethylamine
TFEDMA
ß 2011 Elsevier B.V. All rights reserved.
Fluorination
1,3-Diketone
1,3-Cyclohexanedione
b
b
,
b
-Difluoroketone
-Fluoro- -unsaturated ketones
a,b
Ishikawa’s Reagent
1. Introduction
reaction involves (see Scheme 1) a nucleophilic attack of the
hydroxyl group of enol tautomer 1a on the electropositive
The
(TFEDMA) as a selective fluorinating reagent has been disclosed
previously [1] and the reactions of -fluoroamino reagents (FAR)
were recently reviewed [2]. TFEDMA, like other FAR, is an effective
reagent for the conversion of alcohols into the corresponding
fluoride (R–F). Efforts were recently undertaken to explore
additional reactions of TFEDMA. The fluorination of ketones was
of interest, but since TFEDMA is not capable of fluorinating
unactivated ketones, 1,3-diketones are the subject of this study.
use
of
1,1,2,2-tetrafluoroethyl-N,N-dimethylamine
carbon of TFEDMA (resonance form 2a). Fluoride ion then adds to
the intermediate enone (2b) via Michael addition, to form
intermediate 2c which eliminates the better leaving group: 2,2-
difluoro-N,N-dimethylacetamide (4) to form enone (2d). The
a
addition of HF across the double bond of 2d gives
b,b-
difluoroketone product (3). Although free HF is drawn in the
mechanism, it likely exists in solution complexed with the amide
byproduct (4).
This reaction was repeated with the 7- and 11-carbon diketone
analogs (3,5-heptadione, 5 and 5,7-undecadione, 7 [3], respective-
ly) to also afford the corresponding
after stirring neat under mild conditions (Scheme 2).
b,b-difluoroketones 6 and 8
2. Results and discussion
The reaction of TFEDMA with 3,5-heptanedione (5) was studied
more closely by NMR of the reaction carried out at lower
temperature. After 2 h at 16 8C, the only products detected by
19F NMR in the reaction mixture were amide 4 and a 60/40 mixture
A single reported example is the reaction of 2,4-pentanedione
(1) with TFEDMA (2) which gave 4,4-difluoro-2-pentanone (3) in
42% yield [1]. In this paper, we propose the mechanism of this
of
b-fluoro-a,b-unsaturated ketones (Z-isomer 9a, E-isomer 9b,
respectively) as shown in Scheme 3. After 48 h at ambient
temperature, intermediate compounds 9a and 9b were gone and
only product 6 and compound 4 were detected. The 19F NMR
* Corresponding author. Tel.: +1302 695 4550; fax: +1 302 695 2401.
E-mail address: chrisopther.p.junk@usa.dupont.com (C.P. Junk).
1
Publication no. 2011-332.
0022-1139/$ – see front matter ß 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2011.06.011

L.M. Grieco et al. / Journal of Fluorine Chemistry 132 (2011) 1198–1206
1199
HCF2CF2N(CH3)2
O
O
O
F
F
2
50-80 oC, 6 h
3
1
F
F
O
O
O
OH
F F
H
F
H
F
N
N
F
2a
F
2
1a
F
F
H
N
H
F
N
+
O
4
F
F
F
N
H
F
O
O
+
F
F
O
F
H
O
O
H
F
2b
2d
+
H
F
2c
O
F
O
F
F
F
2e
3
+
H
Scheme 1.
assignments of intermediates 9a and 9b are based on reported
NMR data for closely related compounds [4].
reactions 2, 5 = 11 kcal/mol, reactions 3, 6 = 9 kcal/mol. Since the
presence of two fluorines on compound 4 appeared to make no
difference in the site of protonation, it was reasonable to compute
A
recent paper from Hara et al. [5] reported a similar
fluorination of compound 7 using another FAR (DFMBA, see
Scheme 4), but in this case only the -fluoro- -unsaturated
ketones (10a and 10b) were isolated exclusively and no evidence
was reported of further formation of -difluoroketones.
and compare
D
G for the reactions of interest (5 and 6).
G at 298 K is 28 kcal/mol
b
a
,b
As shown in Scheme 5, the calculated D
different for protonation of these two amides, indicating that
compound 4 is much less basic than 11. In other words, compound
11 would bind HF more tightly, likely preventing it from adding to
b,b
The mechanism proposed by Hara et al. (Scheme 4) is quite
similar to the one depicted in Scheme 1. Obviously, other than the
use of dioxane as a solvent in one case, the only difference between
these two reactions is the structure of the FAR reagent (and thus
amide by-products). In an effort to understand why TFEDMA and
DFMBA gave different products, computations were performed to
determine the basicity of the amide byproducts 4 and 11. At the
DFT level of theory, these two amides were allowed to be
protonated by hydronium ion to form a cation and a molecule of
water.
the double bond of the
b-fluoro-a,b-unsaturated ketones, thus
leaving them as final products in the reaction. The amide from
TFEDMA (4), on the other hand, is a weaker base and it cannot
prevent further reaction of HF to generate the observed
b,b-
difluoroketone products 3, 6, and 8.
The next substrates of interest were cyclic 1,3-diketones (such
as 1,3-cyclohexanedione 12). There are numerous reports of
alkylation and acylation chemistry of 12 [8–10], and it has been
Although hydrocarbon amides are known to protonate prefer-
entially on oxygen [6], computations were done with protonation
occurring on either oxygen or nitrogen [7] to check if this was true
in the case of a fluorinated amide like compound 4. As shown in
Scheme 5, each of three amides (4, 11, and DMAc) was allowed to
be protonated by hydronium ion on either oxygen or nitrogen, and
O
F
TFEDMA
Z isomer, 9a
60 mol %
HCF2CF2N(CH3)2
F
F
O
O
+
+
N
16 oC then RT
H
O
the
DG for each reaction was computed. It can be seen that
O
protonation on oxygen is thermodynamically preferred in all three
cases, by about the same amount: reactions 1, 4 = 11 kcal/mol,
F
5
H-F
E isomer, 9b
40 mol%
48 hr. at RT
TFEDMA
O
F
F
F
F
HCF2CF2N(CH3)2
RT, 48 hr.
O
O
+
N
F
F
H
R
R
O
F
F
N
R
R
+
O
H
O
4
R = Et
R = n-Bu
6, Yied = 63%
8, Yield= 48%
R = Et
R = n-Bu
5
7
6
Scheme 2.
Scheme 3.

1200
L.M. Grieco et al. / Journal of Fluorine Chemistry 132 (2011) 1198–1206
DFMBA
H-F
O
F
O
Et
Et
Et
F
E isomer, 10a
73%
Me
Me
N
N
O
O
F
Et
+
+
O
F
11
7
dioxane
30 oC, 24 h.
Z isomer, 10b
27%
Proposed mechanism:
Et
Ph
N
F
DFMBA
Et
O
OH
O
O
+ HF
Ph
Ph
Et
Et
N
N
O
O
O
O
F
Et
Et
F
Scheme 4.
Protonation on nitrogen:
O
O
H
H3O
H
Me
H
+ H2O
N
N
Reaction 1
Me
H
H
H
H
Me
Me
Calculated
ΔG (298K) = -36 kcal/mol
O
O
H
H
H3O
Me
H
Me
+ H2O
N
N
Reaction 2
Reaction 3
Me
F
F
F
F
Me
Me
Calculated
Δ
G (298K) = -21 kcal/mol
O
O
Me
H3O
N
+ H2O
N
H
Calculated
ΔG (298K) = -51 kcal/mol
Protonation on oxygen:
O
OH
H3O
H3O
Me
H
Me
H
Me
+ H2O
N
N
Reaction 4
Reaction 5
H
H
H
H
Me
Me
Calculated
Δ
G (298K) = -47 kcal/mol
O
OH
H
Me
H
Me
N
+ H2O
N
F
F
F
F
Me
Me
Calculated
Δ
G (298K) = -32 kcal/mol
O
HO
N
Me
H3O
Reaction 6
Difference
N
+ H2O
of 28 kcal/mol
Calculated
ΔG (298K) = -60 kcal/mol
Scheme 5.

L.M. Grieco et al. / Journal of Fluorine Chemistry 132 (2011) 1198–1206
1201
shown that electrophiles can attack both carbon and oxygen of 12
depending on reaction conditions. For example, in the presence of
hydrocarbon acylating reagents, acylation of 12 often proceeds
first at oxygen, and the intermediate rearranges to give the C-
acylated product under Lewis acidic conditions [10]. The effect of
fluorinated acylating reagents on cyclic 1,3-diketones such as 12
was studied in detail by Khlebnicova et al. [11]. Interestingly, in
this case, direct C-acylation was observed (see Scheme 6). The O-
acylated product was assumed to be hydrolytically unstable [11],
and no success was achieved to give C-acylated materials using
other fluorinated electrophiles.
When TFEDMA was allowed to react with 12 at ambient
temperature (either neat or in methylene chloride) the only
product detected was compound 13 derived from acylation, rather
than fluorination (Scheme 7). The structure of isolated product 13
was verified by X-ray crystallography (Fig. 1).
It should be pointed out that in this process TFEDMA acts as a
2,2-difluoroacetyl transfer reagent. The proposed mechanism for
this reaction is shown in Scheme 8.
In order to generate the product 13, TFEDMA is attacked by the
central carbon of enol 12a in order to give intermediate compound
12b. Unlike the mechanism for the linear analogs (Scheme 1), there
is no easy elimination of an amide due to the new carbon–carbon
bond that was formed. It is likely compound 12b is present in the
reaction mixture until aqueous workup hydrolyzes iminium cation
12c to give the observed product 13.
This process is quite general and we demonstrated that the
reactions between TFEDMA and methyl substituted 1,3-cyclohex-
anediones 14, 16, and 18 proceed similarly, giving C-acylated
products 15, 17 and 19 (Scheme 9).
Fig. 1. ORTEP drawing of compound 13 with thermal ellipsoids drawn to the 50%
probability level.
When the 2-position was blocked by a methyl group as in
compound 20 (Scheme 10), no reaction with TFEDMA took place
under similar conditions, and starting material was recovered
upon workup. This lack of reaction after steric blocking of the
central carbon is further evidence for a mechanism that requires
this carbon to be nucleophilic.
This TFEDMA acylation reaction also proceeded smoothly with
the closely related heterocyclic compound 3H-chromene-2,4-
dione (21) to give compound 22 as the only product (Scheme
10). X-ray crystallography of 22 showed that the preferred
tautomer is the one in which the OH group can be part of a 6-
membered ring intramolecular hydrogen bond (Fig. 2).
The C-acylation reaction was also shown to occur with
Ishikawa’s Reagent (N,N-diethyl-(1,1,2,3,3,3-hexafluoropropyl)a-
mine, 23) the 3-carbon analog of TFEDMA (Scheme 11).
3. Experimental
1H and 19F NMR spectra were recorded on Bruker DRX-500
(499.87 MHz) and DRX-400 (376.8485 MHz) instruments, respec-
tively using CFCl3 or TMS as an internal standards and CDCl3 as a
lock solvent, unless noted otherwise. The purity of isolated
materials was established using NMR and elemental analysis.
GC and GC/MS analysis were carried out on either Instrument A: an
HP-6890 instrument, using HP FFAP capillary column and either
TCD (GC) or mass selective (GS/MS) detectors, respectively, or
OH
O
F
F
F
F
H
H
O
H
F
H
F
H
N
N
F
F
O
12
12a
2a
2
O
O
F
O
O
O
O
Rf
O
O
Rf
O
N
O
N
O
H
Rf
N
F
H
Rf
OH
N
CF2H
CF2H
O
CHCl3
O
O
12
O
12
Rf = CF3, C2F5, C3F7
Rf
ONa
12c
12b
Aq. workup
Scheme 6.
OH
O
O
O
O
H
F
HCF2CF2N(CH3)2
- 2HF
F
F
CF2H
O
F
N
+
neat,
H
Aq. workup
O
OH
RT, 24 h
O
13
12
13, Yield = 63%
Scheme 8.
Scheme 7.

1202
L.M. Grieco et al. / Journal of Fluorine Chemistry 132 (2011) 1198–1206
O
OH
O
O
H
F
HCF2CF2N(CH3)2
- 2HF
MP = 45.9 oC
Yield = 59%
F
RT, 24-48 h
Aq. workup
Me
O
Me
14
15
O
OH
O
H
HCF2CF2N(CH3)2
RT, 24-48 h
- 2HF
MP = 30.0 oC
Yield = 46%
F
Me
Me
Me
Me
F
Aq. workup
O
O
16
17
O
OH
O
H
HCF2CF2N(CH3)2
- 2HF
F
MP = 63.1 oC
Yield = 73%
F
RT, 24-48 h
Aq. workup
O
O
Me
Me
Me
Me
18
19
Scheme 9.
O
HCF2CF2N(CH3)2
- 2HF
Me
O
No reaction -- recovered starting material
RT, 24-48 h
Aq. workup
20
O
OH
O
O
H
F
HCF2CF2N(CH3)2
RT, 24-48 h
- 2HF
MP = 142 oC
Yield = 80%
F
Aq. workup
O
O
O
3H-chromene-2,4-dione
21
22
Scheme 10.
Fig. 2. ORTEP drawing of compound 22 with thermal ellipsoids drawn to the 50% probability level.

L.M. Grieco et al. / Journal of Fluorine Chemistry 132 (2011) 1198–1206
1203
O
dione (Aldrich, 97%), pentanal (Aldrich, 97%), 2-hexanone (Aldrich,
98%), acetylacetone (Aldrich, 98%), n-butyllithium in hexanes
(Aldrich, 1.6 mol/L in hexanes), 3H-chromene-2,4-dione (Aldrich,
99%), Dess–Martin periodinane (Aldrich, 0.3 M in CH2Cl2). 1,1,2,2-
tetrafluoroethyl-N,N-dimethylamine (TFEDMA, DuPont, 80%, con-
taining 20% of 4) was stored at room temperature in a tightly closed
Teflon1 PFA bottle and handled only inside of a fume hood.
Crystallography: X-ray data for all compounds were collected at
À100 8C using a Bruker 1K CCD system equipped with a sealed tube
molybdenum source and a graphite monochromator. The struc-
tures were solved and refined using the Shelxtl [18] software
package, refinement by full-matrix least squares on F2, scattering
factors from Int. Tab. Vol C Tables 4.2.6.8 and 6.1.1.4. Crystallo-
graphic data (excluding structure factors) for the structures in this
paper have been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication no. CCDC #753415.
Copies of the data can be obtained, free of charge, on application to
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: +44 1223
336033 or e-mail: deposit@ccdc.cam.ac.uk).
CF3CFHCF2NEt2
OH
O
F
F
F
23
- 2HF
H
RT, 24 h
F
Aq. workup
O
O
12
24, Yield = 63%
Scheme 11.
Instrument B: an Agilent 6890N gas chromatograph (GC) equipped
with an Agilent 5973N mass selective detector (MS) operating in
positive chemical ionization (CI) mode using methane as the
reagent gas. An Agilent 122-5532G DB-5MS column (30 m long,
0.25 mm i.d., 0.25
mm dF) was installed on the GC. Results from
Instrument B typically give characteristic m/e of [M+1], [M+29] and
[M+41]. Elemental analysis was performed by Micro-Analysis, Inc.,
Wilmington, DE.
All calculations were performed using the density functional
theory electronic structure program DMol3 [12,13] with graphical
displays generated with Materials Studio [14]. The Perdew–Wang
(PW91) generalized gradient approximation [15] was used for the
exchange and correlation potentials with restricted spin polariza-
3.1. Preparation of 5,5-difluoroheptan-3-one (6)
A 25 mL Teflon1 PFA flask equipped with digital thermometer,
Teflon1 stir bar, and N2 inlet was charged with 1,1,2,2-
tetrafluoroethyl-N,N-dimethylamine (TFEDMA, 5.2 g, 35.9 mmol,
2.0 equiv.) and cooled to 16 8C using an ice bath. 3,5-heptanedione
(2.3 g, 17.9 mmol, 1.0 equiv.) was then added dropwise via syringe
at a rate to prevent the temperature from rising. The 19F NMR
spectrum (CDCl3) of a sample taken after 2 h at 16 8C showed a
˚
tion and fine DNP numerical basis sets with 4 A cut-off. Atomic
cores are described with the all-electron treatment. SCF iterations
were considered converged with 10À6 Ha tolerance and no thermal
smearing was used in the orbital occupancy. Molecular geometries
were refined via energy optimizations in which convergence
tolerance criteria included energy changes less than 10À5 Ha,
˚
maximum forces less than 0.002 Ha/A and maximum displace-
mixture of E/Z
no desired product.
b
-fluoro-
a
,b
-unsaturated ketones, byproduct 4, but
˚
ments less than 0.005 A. Thermodynamic properties such as
entropy, enthalpy and Gibbs energy are computed at finite
temperatures after fine geometric optimization and vibrational
analysis or Hessian evaluation. Vibrational, rotational and transla-
tional contributions to the molecular partition function are
computed according to standard statistical mechanics in the ideal
gas approximation [16]. Fukui functions [17] are used to assess
reactivity from a frontier orbital analysis. The function f+ measures
the change in the local electron density when the molecule gains
electrons and, hence, corresponds to reactivity with respect to
nucleophilic attack. Conversely, the function fÀ measures the
change in the local electron density when the molecule loses
electrons, and hence, indicates the reactivity with respect to
electrophilic attack. For the ensuing discussion these functions are
calculated from differences in atom-centered charges. These
charges are calculated from Mulliken population analysis before
and after 0.1 electrons are removed (or added) for fÀ (or f+),
respectively. A larger difference between the molecule’s neutral
and partial ion states in these atom-centered charges indicates a
higher kinetic reactivity. Mechanistic pathways are followed by
constraining the distance between two atoms. Starting from a
confirmed vibrational ground state all other degrees of freedom are
varied to minimize the total energy and then reducing the biatomic
approach distance by as little as 0.01 A. The pathway is followed as
the system energy transcends a first maximum followed by a
minimum.
3
E
isomer: 19F NMR: À78.3 (dt, JFH = 20 Hz (vinyl C–H),
3JFH = 25 Hz (CH2)).
1H NMR: 5.9 ppm (3JFH = 20 Hz, vinyl C–H).
3
Z
isomer: 19F NMR: À79.6 (dt, JFH = 39 Hz (vinyl C–H),
3JFH = 14 Hz (CH2)).
1H NMR: 5.3 ppm (3JFH = 39 Hz, vinyl C–H).
The 19F NMR spectrum (CDCl3) of a sample taken after 22 h at
room temperature showed a mixture of E isomer/Z isomer/product
of 14/22/64 mol%, respectively. The mixture was allowed to stir for
an additional 24 h at ambient temperature. It was then poured over
crushed ice (40 g) which was allowed to melt. The mixture was
extracted with hexanes (4Â 20 mL), organic layers were combined
and dried over MgSO4, filtered, and concentrated in vacuo. The
product was isolated by distillation through a short path glass
column with cold water condenser to give 1.7 g of a clear liquid
(63% isolated yield, 90% purity). BP = 29 8C (75 Torr). GCMS
(Instrument B: m/e 191 [M+41], 179 [M+29], 151 [M+1], 131
[M-19], 57 [Base, C3H5O].
D
B
C
A
E
F
F
O
Caution: TFEDMA reacts violently with water producing HF, can
cause severe chemical burns, and has to be handled with great
caution by trained personnel. Due to the high reactivity of TFEDMA
toward water, it is important that all substrates are dry.
The following chemicals were obtained from commercial
sources and used without further purification: 2,4-pentanedione
(Aldrich, 99%), 3,5-heptanedione (97%), 1,3-cyclohexanedione
(Aldrich, 97%), 5-methyl-1,3-cyclohexanedione (Aldrich, 98%),
5,5-dimethyl-1,3-cyclohexanedione (Aldrich, 95%), 4,4-dimethyl-
1,3-cyclohexanedione (Aldrich, 98%), 2-methyl-1,3-cyclohexane-
19F NMR (CDCl3) À95.7 (tt, JHbF = 17.2 Hz, JHcF = 16 Hz, 2F).
3
3
1H NMR (CDCl3) 3.0 (t, JHcF = 15.4 Hz, 2H); 2.5 (q,
3
3JHdHe = 7.3 Hz, 2H); 2.0 (tq, JHbF = 17.2 Hz, JHbHa = 7.5 Hz, 2H);
3
3
3
3
1.1 (t, JHdHe = 7.2 Hz, 3H); 1.0 (t, JHbHa = 7.5 Hz, 3H).
3.2. Preparation of 7-hydroxyundecan-5-one
This compound had been synthesized previously [19] but full
characterization was not presented.

1204
L.M. Grieco et al. / Journal of Fluorine Chemistry 132 (2011) 1198–1206
Formation of lithium diisopropylamide (LDA): an oven dried
tetrafluoroethyl-N,N-dimethylamine (TFEDMA, 1.6 g, 11.0 mmol,
2.1 equiv.) and cooled to 0 8C using an ice bath. 5,7-undecanedione
(0.97 g, 5.2 mmol, 1.0 equiv.) was then added dropwise via syringe
500 mL 3-neck round bottom flask with stir bar, N2 inlet, addition
funnel, septum and thermocouple was charged with a solution of
diisopropylamine (4.40 g, 43.4 mmol) in dry THF (50 mL). This
solution was cooled to À70 8C with stirring using a dry ice/acetone
bath. The addition funnel was charged with n-butyllithium
(15.8 mL, 44.0 mmol, 2.79 M in hexanes), which was added
dropwise over 20 min. The cold bath was removed and the flask
was allowed to warm to 0 8C to facilitate the formation of LDA.
Addition of 2-hexanone: upon reaching 0 8C the cold bath was
replaced and the flask cooled back to À70 8C. The addition funnel
was recharged with a solution of 2-hexanone (4.00 g, 40.0 mmol)
in dry THF (40 mL). This was added dropwise over 15 min. The
hexanone/LDA mixture was stirred at À70 8C for 30 min. The
addition funnel was recharged with a solution of pentanal (3.44 g,
40.0 mmol) in dry THF (40 mL). This was added drop wise over
15 min. The mixture was stirred for 6 min at À70 8C and then
transferred via cannula to a 1 L flask containing saturated aqueous
NH4Cl solution (400 mL) to quench. The quenched mixture was
stirred for 20 min at RT prior to workup. Diethyl ether (250 mL)
was then added and stirred for 5 min. The phases were separated
and organic layer was washed with saturated NH4Cl (50 mL)
solution and brine (50 mL). The aqueous layer was back-extracted
with diethyl ether (50 mL). Organic layers were combined and
dried over MgSO4, filtered and concentrated in vacuo, then vacuum
distilled at 60 8C/350 mTorr to obtain 4.95 g desired product (66%
yield).
at a rate to prevent the temperature from rising (10 min.). The 19
F
NMR spectrum (CDCl3) of a sample taken after 4 h at ambient
temperature showed a mixture of E/Z -fluoro- -unsaturated
b
a,b
ketones, byproduct 4, and desired product. After 48 h at ambient
temperature the unsaturated ketone intermediates were gone by
19F NMR and the reaction was quenched by pouring onto crushed
ice (25 g) and allowing to melt. The mixture was extracted with
hexanes (3Â 20 mL). The combined organic layers were washed
with deionized water (10 mL), dried over MgSO4, filtered, and
reduced in vacuo to give 1.2 g of a crude red oil. Purification by
column chromatography (0.75 in. Â 8 in. silica gel, CH2Cl2/hexane
75/25) gave 0.52 g of the liquid product (48% yield). GCMS
(Instrument B: m/e 206 [M+], 149 [M+ÀC4H9], 129 [C7H10FO+], 85
[base, C5H9O+], 57 [C4H9+], 43 [C3H7+], 29 [C2H5+].
H
F
E
D
B
C
G
A
I
F
F
O
3
3
19F NMR (CDCl3) À93.4 (tt, JHbF = 17.2 Hz, JHcF = 16 Hz, 2F).
1H NMR (CDCl3) 3.0 (t, 3JHeF = 15.4 Hz, 2H); 2.5 (t, 3JHfHg = 7.3 Hz,
2H); 1.9 (m, 2H); 1.6 (m, 2H); 1.5 (m, 2H); 1.3 (m, 4H); 0.9 (m, 6H).
3.5. Preparation of 2-difluoroacetyl-1,3-cyclohexadione (13)
A 100 mL Teflon1 PFA flask equipped with digital thermometer,
pressure equalized addition funnel, Teflon1 PTFE stir bar, and N2
inlet was charged with N,N-dimethylamino-1,1,2,2-tetrafluor-
oethane (TFEDMA, 8.3 g, 57 mmol, 1.6 equiv.).
In a separate flask, 1,3-cyclohexanedione (4.0 g, 36 mmol,
1.0 equiv.) was dried by combining with anhydrous magnesium
sulfate (2 g) in methylene chloride (75 mL). The solution was
agitated for 5 min. and then allowed to sit for 10 min. to allow the
MgSO4 to settle. The solution was carefully removed with a pipette,
F
E H
H
B
J
D
A
I
C
K
O
OH
G
1H NMR: 4.1 (br s, HF, 1H); 3.1 (s, Hg, 1H); 2.4–2.7 (m, HD, HE,
4H); 1.2–1.6 (m, HB, HC, HH, HI, 10H); 0.9 (t, HA, 3H); 0.9 (t, HK, 3H).
GCMS (Instrument A: m/e 227 [M+41], 187 [M+1], 169 [base,
M+ÀOH], 129 [C3H13O2+], 101 [C6H13O+], 85 [C5H9O+].
and added to a syringe capped by a 0.45
mm syringe filter. The
dried solution was then pushed through the syringe filter (to
remove any trace MgSO4) into the addition funnel. The CH2Cl2
solution was slowly added into the reaction pot with stirring. The
reaction was left to magnetically stir at 25 8C for 15 h. The reaction
mixture was poured over an equal volume of 10% HCl solution.
EtOAc (ꢀ50 mL) was added, and the layers were separated. The
aqueous layer was back extracted with EtOAc (50 mL) and the
organic layers were combined, washed with deionized water (2Â
20 mL), dried over MgSO4, filtered, and concentrated in vacuo to
give a colorless oil which crystallized upon drying at 200-
250 mTorr and RT for 3 h (7.4 g, 87 wt.% product 13, 13 wt.% 4 by
19F NMR). The compound was further purified by recrystallization
from diethyl ether/hexanes to give 3.5 g (41%) product. 1H NMR
3.3. Preparation of 5,7-undecanedione[3] (7)
7-Hydroxyundecan-5-one (0.50 g, 2.68 mmol, 1.0 equiv.) was
dissolved in CH2Cl2 (5 mL) at RT and the flask was wrapped with
aluminum foil to shield from light. Dess–Martin periodinane
solution (DMP, 10.79 mL of 0.3 M in CH2Cl2, 3.22 mol, 1.2 equiv.)
was added via syringe and stirred for 30 min while shielded from
light. Reaction mixture was then diluted with ether (25 mL via
syringe) and stirred for 15 min. An aqueous solution of Na2S2O3
(2.14 g) in saturated bicarbonate (10 mL) was then added (to
consume DMP byproduct) and the solution was stirred for another
15 min. The flask was unwrapped and the phases separated. The
organics layer was washed with saturated aqueous sodium
bicarbonate (20 mL), brine (20 mL) and dried over MgSO4.
Concentrating in vacuo gave an impure oil that crystallized upon
standing. Purification by column chromatography (0.75 in. Â 8 in.
silica gel, CH2Cl2/hexane 50–90%) gave 0.45 g product (90% yield).
GCMS (Instrument B: m/e 225 [M+41], 213 [M+29], 199 [M+15],
185 [M+1], 85 [C5H9O].
2
(CDCl3) 6.9 (t, JFH = 53.6 Hz, 1H); 2.77 (t, J = 6.4 Hz, 2H); 2.53 (t,
J = 6.4 Hz, 2H); 2.06 (tt, J = 6.4 Hz, J = 6.4 Hz, 2H). 19F NMR (CDCl3)
À130.4 (d, JFH = 53.6 Hz, 2F). MP by DSC: 57.7 8C. Anal. Calc. for
C8H8F2O3: C, 50.53: H, 4.24. Found: C, 50.24: H, 4.26. GCMS
(Instrument A: m/e 190 [M+], 170 [M+ÀHF], 162 [M+ÀC2H4], 139
[base, M+ÀCF2H], 111 [M+ÀC2HF2O], 69 [C4H5O+], 55 [C3H3O+], 42
[C2H2O+].
1H NMR (CDCl3) 3.6 (s, 2H); 2.3 (t, J = 7.5 Hz, 4H); 1.6 (m, 4H);
1.4 (m, 4H); 0.9 (t, J = 7.3 Hz, 6H).
3.6. Preparation of 2-difluoroacetyl-5-methyl-1,3-cyclohexadione
(15)
3.4. Preparation of 7,7-difluoroundecan-5-one (8)
A 50 mL Teflon1 PFA flask equipped with digital thermometer,
pressure equalized addition funnel, Teflon1 PTFE stir bar, and N2
inlet was charged with N,N-dimethylamino-1,1,2,2-tetrafluor-
oethane (TFEDMA, 1.9 g, 13.1 mmol, 1.6 equiv.).
A 25 mL Teflon1 PFA flask equipped with digital thermometer,
Teflon1 stir bar, and N2 inlet was charged with 1,1,2,2-

L.M. Grieco et al. / Journal of Fluorine Chemistry 132 (2011) 1198–1206
1205
In a separate flask, 5-methyl-1,3-cyclohexanedione (1.0 g,
7.9 mmol, 1.0 equiv.) was dried by combining with anhydrous
magnesium sulfate (1 g) in methylene chloride (50 mL). The
solution was agitated for 5 min. and then allowed to sit for 10 min.
to allow the MgSO4 to settle. The solution was carefully removed
deionized water (2Â 20 mL), dried over MgSO4, and concentrated
in vacuo for 12 h to give a flaky yellow solid (4.1 g) that still
contained some EtOAc. The crude product was recrystallized from
a warm CH2Cl2/hexanes mixture to give a white solid (2.82 g, 73%
yield). Due to lack of symmetry, the two enol isomers of this
compound show up separately in NMR as a 60/40 mol% mixture:
with a pipette, and added to a syringe capped by a 0.45
mm syringe
filter. The dried solution was then pushed through the syringe filter
(to remove any trace MgSO4) into the addition funnel. The CH2Cl2
solution was slowly added into the reaction pot at ambient
temperature with stirring. The reaction was left to magnetically
stir at ambient temperature for 24 h, then poured over an equal
volume of 10% HCl solution. EtOAc (ꢀ50 mL) was added, and the
layers were separated. The aqueous layer was back extracted with
EtOAc (50 mL) and the organic layers were combined, washed with
deionized water (2Â 20 mL), dried over MgSO4, filtered, and
concentrated in vacuo to give a white solid (1.1 g, 87 wt.% product,
13 wt.% 4 by 19F NMR). 1H NMR (CDCl3) 6.9 (t, 2JFH = 53.4 Hz, 1H);
F
F
H
H
O
O
F
F
HO
O
O
OH
CH3
CH3
CH3
CH3
1H NMR: 6.90 (t, JFH = 53.6 Hz, 1H); 2.78 (t, JHH = 6.5 Hz, 2H);
1.87 (t, JHH = 6.5 Hz, 2H); 1.18 (s, 6H). 19F NMR: À130.2 (d,
2
2.78, 2.46 (subsplit ABq, JHH = 18.5 Hz, 2H); 2.58, 2.19 (subsplit
JFH = 53.5 Hz, 2F).
2
3
ABq, JHH = 16.1 Hz, 2H); 2.28 (m, 1H); 1.11 (d, JHH = 6.5 Hz, 1H).
19F NMR (CDCl3) À130.0, À130.8 (dABq, JFH = 53.2 Hz, JFF = 307 Hz,
2F). An analytical sample was obtained by recrystallization from
hexanes (0.175 g). MP by DSC: 45.9 8C. Anal. Calc. for C9H10F2O3: C,
52.94: H, 4.94. Found: C, 52.83: H, 5.04.
1H NMR: 6.93 (t, JFH = 53.6 Hz, 1H); 2.56 (t, JHH = 6.7 Hz, 2H);
1.89 (t, JHH = 6.7 Hz, 2H); 1.36 (s, 6H). 19F NMR: À130.0 (d,
JFH = 53.5 Hz, 2F). MP by DSC: 63.1 8C. Anal. Calc. for C10H12F2O3: C,
55.05: H, 5.54: F, 17.41. Found: C, 54.25: H, 5.57: F, 17.95.
3.9. Preparation of 3-difluoroacetylchromene-2,4-dione (22)
3.7. Preparation of 2-difluoroacetyl-5,5-dimethyl-1,3-cyclohexadione
(17)
To a solution of 3H-chromene-2,4-dione (3.2 g, 20 mmol,
1.0 equiv.) in CH2Cl2 (30 mL) at ambient temperature was added
A 50 mL Teflon1 PFA flask equipped with digital thermometer,
1,1,2,2-tetrafluoroethyl-N,N-dimethylamine
(TFEDMA,
5.0 g,
Teflon1 PTFE stir bar, and N2 inlet was charged with N,N-
34 mmol, 1.7 equiv.) dropwise over 15 min. The reaction mixture
was magnetically stirred at ambient temperature for 16 h, washed
with water (3Â 100 mL), dried over MgSO4, concentrated in vacuo,
and the solid residue was kept under dynamic vacuum
(0.1 mm Hg) for 10 h. 3.8 g (80%) of crude 3-(2,2-difluoroacetyl)-
3H-chromene-2,4-dione was isolated, purity 98% (NMR). Analyti-
cally pure sample was prepared by crystallization from chloroform
to afford white needles. MP 142–143 8C.
dimethylamino-1,1,2,2-tetrafluoroethane
(TFEDMA,
2.5 g,
17 mmol, 1.6 equiv.). At ambient temperature, 5,5-dimethyl-1,3-
cyclohexanedione (1.5 g, 10.7 mmol, 1.0 equiv.) was added to the
stirring TFEDMA. No exotherm was observed upon addition. The
reaction mixture was left to magnetically stir at ambient tempera-
ture for 15 h. The crude mixture was a viscous yellow oil. EtOAc
(20 mL) was added, and the reaction mixture was poured over an
equal volume of 10% HCl solution. The aqueous layer was back
extracted with EtOAc (10 mL) and all organic layers were combined.
The organic layer was washed with additional 10% HCl (1Â 20 mL),
deionized water (2Â 20 mL), dried over MgSO4 and concentrated in
vacuo at 200–250 mTorr at RT for 12 h. The material crystallized to
1H NMR (CDCl3): 15.43 (s, 1H); 8.03 (dd, J = 8.0 Hz, J = 1.8 Hz,
1H), 7.71 (ddd, J = 8.9 Hz, J = 7.0 Hz, J = 1.1 Hz, 1H); 7.34 (ddd,
J = 8.3 Hz, J = 7.1, J = 1.1, 1H); 6.99 (t, J = 53.2 Hz, 1H), 2.63 (s, 2H).
19F NMR: À130.68 (d, J = 53.2 Hz, 2F). Anal. Calc. for C11H8F2O2: C,
54.55: H, 3.33, F, 15.69. Found: C, 54.14: H, 2.81, F, 15.69.
give 1.72 g of dark yellow solid (87 wt.% product, 13 wt.% 4 by 19
F
NMR). Byproduct 4 was removed by recrystallization from hexanes
to afford a white crystalline solid (1.08 g, 46% yield).
1H NMR (CDCl3) 6.94 (t, JFH = 53.5 Hz, 1H); 2.63 (s, 2H); 2.40 (s,
2H); 1.12 (s, 6H).
19F NMR (CDCl3) À130.4 (d, JFH = 53.5 Hz, 2F). MP by DSC:
30.0 8C. Anal. Calc. for C10H12F2O3: C, 55.05: H, 5.54. Found: C,
54.55: H, 5.61.
3.10. Preparation of 2-(2,3,3,3-tetrafluoropropanoyl)-1,3-
cyclohexanedione (24)
A 50 mL Teflon1 PFA flask equipped with digital thermometer,
Teflon1 PTFE stir bar, and N2 inlet was charged with N,N-
diethylamino-1,1,2,3,3,3-hexafluoropropane (Ishikawa’s Reagent,
7.5 g, 33.6 mmol, 1.5 equiv.). The flask was placed in an ice bath
until the internal temperature was 2 8C. Solid 1,3-cyclohexane-
dione (2.5 g, 22.3 mmol, 1.0 equiv.) was added slowly at a rate to
keep the temperature below 15 8C over a period of 20 min. The
mixture was allowed to stir at ambient temperature for 48 h and
then poured onto crushed ice (50 g) and allowed to melt. The
mixture was extracted with ethyl acetate (3Â 20 mL). The
combined organic fractions were washed with 10% HCl (2Â
10 mL) and then with deionized water (2Â 10 mL). The organic
layer was dried over MgSO4 and solvent was removed in vacuo. The
resulting oil (6.24 g) was purified by putting under vacuum
(200 mTorr) for 10 h with an intermittent warm water bath to aid
in removal of the amide byproduct (CF3CFHCONEt2) as an oil that
collected on the glass above the remaining product. The desired
product crystallized upon standing to give 3.35 g (63% yield).
3.8. Preparation of 2-difluoroacetyl-4,4-dimethyl-1,3-cyclohexadione
(19)
A 25 mL Teflon1 PFA flask equipped with digital thermometer,
Teflon1 PTFE stir bar, and N2 inlet was placed in an ice/water bath
and then charged with N,N-dimethylamino-1,1,2,2-tetrafluor-
oethane (TFEDMA, 4.1 g, 28.5 mmol, 1.6 equiv.). Once the internal
temperature reached 8 8C, 4,4-dimethyl-1,3-cyclohexanedione
(2.5 g, 17.8 mmol, 1.0 equiv.) was added in small portions to the
stirring TFEDMA so as to keep the temperature below 15 8C. The
reaction mixture was left to magnetically stir at ambient
temperature for 5 days. The crude mixture was a viscous yellow
oil. EtOAc (20 mL) was added, and the reaction mixture was poured
over an equal volume of 10% HCl solution. The aqueous layer was
back extracted with EtOAc and all organic layers were combined.
The organic layer was washed with additional 10% HCl (20 mL),
2
3
1H NMR (CDCl3) 6.8 (dq, JFH = 47 Hz, JFH = 6 Hz, 1H); 2.8 (m,
2H); 2.5 (m, 2H); 2.0 (m, 2H). 19F NMR (CDCl3) À74.7 (dd,
3JFH = 6 Hz, JFF = 12 Hz, 3F); À207.0 (dq, JFH = 47 Hz, JFF = 12 Hz,
3
2
3

1206
L.M. Grieco et al. / Journal of Fluorine Chemistry 132 (2011) 1198–1206
[7] The authors thank Reviewer #1 for suggesting to compute the energies for
protonation on both nitrogen and oxygen
[8] H. Smith, J. Chem. Soc. (Resumed) (1953) 803–810.
[9] D.B. Rubinov, I.L. Rubinova, A.A. Akhrem, Chem. Rev. 99 (1999) 1047–1066.
[10] A.A. Akhrem, F.A. Lakhvich, S.I. Budai, T.S. Khlebnicova, I.I. Petrusevich, Synthesis
(1978) 925–927.
1F). Anal. Calc. for C9H8F4O3: C, 45.01: H, 3.36. Found: C, 44.88: H,
3.10. MP: 32.1–33.2 8C. GCMS (Instrument A: m/e 240 [M+], 220
[M+ÀHF], 212 [M+ÀC2H4], 184 [M+ÀC3H4O], 139 [base,
M+ÀC2HF4], 101 [C2HF4+], 69 [C4H5O+], 55 [C3H3O+], 42 [C2H2O+].
[11] T.S. Khlebnicova, V.G. Isakova, A.V. Baranovsky, E.V. Borisov, F.A. Lakhvich, J.
Fluorine Chem. 127 (2006) 1564–1569.
References
[12] B. Delley, J. Chem. Phys. 92 (1990) 508.
[13] B. Delley, J. Chem. Phys. 113 (2000) 7756.
[14] Accelrys Software Inc., Accelrys, MS Modeling, Release 4.4, San Diego, 2010.
[15] J.P. Perdew, Y. Wang, Phys. Rev. B 45 (1992) 13244.
[16] T. Hirano, J.J.P. Stewart (Eds.), MOPAC Manualseventh ed., 1993.
[17] R.G. Parr, W. Yang, Density-functional Theory of Atoms and Molecules, 1989.
[18] G.M. Sheldrick, Acta Crystallogr. Sect. A: Found. Crystallogr. 64 (2007) 112–122.
[19] K. Narasaka, F.C. Pai, Tetrahedron 40 (1984) 2233–2238.
[1] V.A. Petrov, S. Swearingen, W. Hong, W. Chris Petersen, J. Fluorine Chem. 109
(2001) 25–31.
[2] V.A. Petrov, Adv. Org. Synth. 2 (2006) 269–290.
[3] D.S. Matteson, K.M. Sadhu, M.L. Peterson, J. Am. Chem. Soc. 108 (1986) 810–819.
[4] J.P. Gillet, R. Sauvetre, J.F. Normant, Synthesis (1982) 297–301.
[5] K. Sano, T. Fukuhara, S. Hara, J. Fluorine Chem. 130 (2009) 708–713.
[6] R.J. Gillespie, T. Birchall, Can. J. Chem. 41 (1963) 148–164.