Replacement of BF4- by PF6- makes Selectfluor greener

Article abstract of DOI:10.1016/j.jfluchem.2012.05.010

A combination of F-TEDA-PF₆ and CuBr (0.1 equiv.) provides a potent oxidant that readily oxidizes amides to provide imides at room temperature. Replacement of BF₄^-, the anion of Selectfluor (F-TEDA-BF₄), by PF₆^-, dramatically reduces CuBr loading in this oxidative reaction. A possible rationale for this dramatic counterion effect is provided.

Full text of DOI:10.1016/j.jfluchem.2012.05.010

Journal of Fluorine Chemistry 143 (2012) 226–230
Contents lists available at SciVerse ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
À
À
Replacement of BF4 by PF6 makes Selectfluor greener
Zhuang Jin a, Bo Xu a, Stephen G. DiMagno b, , Gerald B. Hammond
*
a,**
a Department of Chemistry, University of Louisville, Louisville, KY, USA
b Department of Chemistry and Center for Materials Research and Analysis, University of Nebraska–Lincoln, Lincoln, NE, USA
A R T I C L E I N F O
A B S T R A C T
Article history:
A combination of F-TEDA-PF6 and CuBr (0.1 equiv.) provides a potent oxidant that readily oxidizes
amides to provide imides at room temperature. Replacement of BF4À, the anion of Selectfluor (F-TEDA-
BF4), by PF6À, dramatically reduces CuBr loading in this oxidative reaction. A possible rationale for this
dramatic counterion effect is provided.
Received 18 April 2012
Received in revised form 8 May 2012
Accepted 10 May 2012
Available online 18 May 2012
ß 2012 Elsevier B.V. All rights reserved.
Keywords:
Imides
Amide oxidation
Selectfluor
1. Introduction
amides to imides at room temperature [6]. Although this method is
efficient, the use of a stoichiometric amount of copper makes the
Imides are core structures in many therapeutic agents and
agrochemicals [1]. Imides are readily prepared by condensation
reactions of carboxylic acid derivatives and ammonia or primary
amines [2], however a number of other protocols have been
employed for imide synthesis [3], including cross coupling [3d, e],
oxidation of oxazoles [3b], N-formylation [3c], and the metal
catalyzed nucleophilic ring opening of aziridines with malonates
[3a]. We are interested in a direct synthetic pathway to prepare
methodology unattractive for process chemistry. Thus, we
investigated conditions that would permit use of a catalytic
amount of cuprous salt. Here we are pleased to disclose an
interesting finding: a simple counterion exchange on Selectfluor
(F-TEDA-PF6 instead of F-TEDA-BF4) permits the amount of copper
catalyst to be reduced significantly (to 10 mol%). The oxidizing
power of F-TEDA-PF6, a reagent that can be easily prepared from
Selectfluor, is at least as potent as Selectfluor itself; amides are
oxidized to imides as rapidly and efficiently as we reported
previously, despite the significant reduction in CuBr loading
(Scheme 1).
unsymmetricalimides by direct oxidation of the
of amides. Several methods to accomplish this transformation have
been reported [4], however, efficient methodologies for -methy-
a-methylenegroup
a
lene oxidation are relatively rare [4e, i], and most of the proposed
oxidative methods suffer from low yields and/or limited substrate
scope[4a–d, f–h, j]. Forexample, althoughRuO4 iswidelyrecognized
as a capable oxidant for amide oxidation [4a, f, j]; it is such an active
reagent that amide oxidation is accompanied by the oxidation of
even unactivated tertiary C–H bonds [5]. Of the two most efficient
2. Result and discussions
The results of amide oxidation experiments are summarized in
Table 1, where
a direct comparison is made between the
stoichiometric (Condition A, 120 mol% (or 1.2 equiv.) CuBr) and
catalytic (Condition B, 10 mol% CuBr) processes. First, amide 1a,
CuBr (10 mol%) and F-TEDA-PF6 (2.2 equiv.) were allowed to react
in acetonitrile for 5 h. This procedure provided a yield of imide (2a)
that was nearly identical to our previous results obtained for the
stoichiometric reaction, albeit after a slightly longer reaction time.
Similarly, amides 1b (entry 2) and 1c (entry 3) were also efficiently
oxidized into the corresponding imides in good yield under
catalytic conditions. For entry 4 (amide 1d), the new protocol gave
amide oxidative methods, the first employs
a heavy metal
(chromium VI) reagent [4i]. Later, Nicolau and coworkers demon-
strated that such toxic heavy metal oxidants could be supplanted by
the more environmentally benign Dess–Martin periodinane (DMP),
although effective oxidation did require heating of the hypervalent
iodine reagent to elevated temperatures (80–85 8C) [4e].
Previously, we reported that the combination of stoichiometric
amounts of CuBr and excess Selectfluor (F-TEDA-BF4) can oxidize
an enhanced yield of imide. Table
1 entries 5–8 further
demonstrate the comparability of the catalytic and stoichiometric
protocols. Ester functionality can be tolerated in the new oxidative
system and in the former system (entries 9 and 10), indicating that
* Corresponding author.
** Corresponding author.
E-mail address: gb.hammond@louisville.edu (G.B. Hammond).
0022-1139/$ – see front matter ß 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jfluchem.2012.05.010

Z. Jin et al. / Journal of Fluorine Chemistry 143 (2012) 226–230
227
Scheme 2. Nicolaou’s proposal of amide oxidation by DMP.
give the final product, 2. Initially, we thought that our oxidation
proceeded through a similar reaction mechanism, since the
ultimate source of the newly formed oxygen comes from trace
water in the reaction mixture [6]. In order to test the above
hypothesis, we added an aromatic compound with moderate
electron density, N-arylacetamide, to the reaction mixture in an
attempt to capture the conjugated imine intermediate 3; two
attempts are outlined in Scheme 3.
Amide 1h, N-arylacetamide, CuBr, and F-TEDA-PF6 were mixed
under catalytic conditions (Condition B in Table 1, Eq. 1 in Scheme 3).
Surprisingly, amide 1h was not oxidized into 2h; instead, N-
arylacetamide was fluorinated by electrophilic aromatic substitu-
tion. No evidence for the trapped imine product was observed. In a
secondattempttotraptheputativeimineintermediate,theoxidation
of 1h was performed under normal catalytic conditions, at which
Scheme 1. Cu-catalyzed amide oxidation with F-TEDA-PF6 and F-TEDA-BF4.
a
-methylene radical stability or metal ion coordination might be
the factor that controls the initial steps of the oxidation reaction.
Unfortunately, both methodologies failed to oxidize a lactam into a
cyclic imide (entry 11), suggesting that only acyclic amides are
suitable substrates for these two oxidative systems.
Nicolaou has proposed a mechanism [4e], for amide oxidation
by DMP, in which an amide was dehydrogenated to form an imine
intermediate 3 (Scheme 2). Compound 3 reacts with water to
generate a hemiaminal 4, which can be further oxidized by DMP to
Table 1
Comparison of the effectiveness of F-TEDA-PF6 and F-TEDA-BF4 in amide oxidation.
Entry
% Yielda (SM)b
O
O
O
R1
N
H
R2
R1
N
H
R2
1
2
Amides
Imides
Ac
Bd
1
2
88 (7)
90 (2)
O
O
O
O
N
N
H
H
1a
2a
83
79
O
O
N
H
N
H
1b
2b
3
4
79 (13)
77 (10)
83 (5)
90 (3)
O
O
O
O
O
O
N
H
N
H
5
5
1c
2c
N
H
N
H
1d
2d
5
6
84 (11)
66 (30)
85 (6)
O
O
O
O
O
O
N
H
N
H
1e
2e
71 (12)
N
H
N
H
3
3
5
5
2f
1f

228
Z. Jin et al. / Journal of Fluorine Chemistry 143 (2012) 226–230
Table 1 (Continued )
Entry
% Yielda (SM)b
O
O
O
R1
N
H
R2
R1
N
H
R2
1
2
Amides
Imides
Ac
Bd
7
8
50 (10)
65 (15)
O
O
O
O
O
O
N
N
H
H
F
F
1g
2g
80 (5)
91
N
H
N
H
F
F
1h
2h
9
84 (6)
82 (6)
70 (10)
O
O
O
O
O
O
O
N
O
N
O
O
H
H
4
4
2i
1i
10
78 (8)
O
O
O
N
H
N
H
0
O
1j
2j
11
0 (96)
0 (95)
O
O
NH
NH
1k
2k
a
Yield was based on starting amide.
Starting material recovery.
b
c
Condition A: Amide 1 (0.25 mmol), Selectfluor (F-TEDA-BF4) (0.625 mmol, 2.5 equiv.) and CuBr (0.3 mmol, 1.2 equiv., added in six portions over 40 min) reacted in
acetonitrile (5 mL) at room temperature for 1 h.
d
Condition B: Amide 1 (0.25 mmol), F-TEDA-PF6 (0.55 mmol, 2.2 equiv.) and CuBr (0.025 mmol, 0.1 equiv.) reacted in acetonitrile (5 mL) at room temperature for 3–6 h.
time we added N-arylacetamide to the reaction mixture (Eq. 2).
Under this revised protocol the imide (25%) and fluorinated N-
arylacetamidewasobserved,butagain, therewasnoevidence forthe
capturedimineintermediate.Furtherinvestigationsofthereactionof
N-arylacetamide with of F-TEDA-PF6 were conducted, since no
reaction had been observed previously between N-arylacetamides
and Selectfluor (F-TEDA-BF4) in acetonitrile at room temperature.
Control experiments revealed that N-arylacetamide could be
Scheme 3. Attempts to capture possible imine intermediate 3.

Z. Jin et al. / Journal of Fluorine Chemistry 143 (2012) 226–230
229
fluorinated by F-TEDA-PF6 under exceptionally mild conditions,
indicatingthatF-TEDA-PF6 isa significantly more active electrophilic
fluorinating agent than Selectfluor (F-TEDA-BF4). The above two
observationsindicatethatif theoxidation pathwayinvolvesanimine
or iminium intermediate, it cannot be trapped using such an electron
rich arene in the presence of F-TEDA-PF6. Lactam 1k could not be
oxidized using both systems (Table 1, entry 11), perhaps because its
imine intermediateis moredifficulttoformduetoenforcedeclipsing
interactions in the 7-membered ring imine.
explains the need for different CuBr loading when using the two
different counterions. Although both complexes C and D in Scheme
4 may be catalytically competent to initiate amide oxidation,
fluoride extraction by C to form complex E may result in a less
active complex. Therefore, large amounts of CuBr are needed for
oxidation in the case of F-TEDA-BF4, since the putatively active
species is being consumed during the course of the reaction.
However, for the high valence copper in D it is difficult to extract
the fluoride ion from PF6À to form complex E due to the stronger P–
F bond in the PF6À, therefore, D remains active in the reaction
system.
It is puzzling that the mere change of a counterion from
Selectfluor (F-TEDA-BF4) to F-TEDA-PF6 drastically reduces the
amount of CuBr needed to drive the amide oxidation to
completion. In previous studies of fluorination [7] and oxidation
[8] reactions, the difference between PF6À and BF4À is reported to
be insignificant. Nevertheless, this effect is real and significant in
the copper-catalyzed oxidation. It is not yet clear to us what the
nature of the catalytically active copper salt is in this oxidation.
The 1H NMR spectrum of the reaction mixture shows significant
line broadening as soon as Selectfluor (or F-TEDA-PF6) is added;
this phenomenon suggests the presence of paramagnetic Cu(II)
ions. However, we have demonstrated that Cu(II) salts were not
effective mediators of this process in our previous work [6]. It
might also be possible that a high valent Cu(III) ion generated
under the strongly oxidizing conditions used here is the principle
oxidant that initiates the reaction with amides. How, then, can one
explain the need for a stoichiometric amount of copper salts when
Selectfluor is employed, but the need for only a catalytic amount of
CuBr when of F-TEDA-PF6 is used? Cu(III) salts are typically not
very stable with simple halogen ligands; in the few cases where
Cu(III) salts have been isolated, a relatively small subset of ligands
have been reported to support the high oxidation state for copper
[9]. These considerations suggest that the initially formed Cu(III)
ion might undergo a relatively slow quenching reaction with
Selectfluor that is not available with F-TEDA-PF6. Since Cu(III) is
expected to be a relatively hard and fluorophilic Lewis acid, it is
reasonable to suspect that it is capable of extracting a fluoride ion
from the BF4 anion, whereas it might not be sufficiently Lewis
acidic to abstract a fluoride ion from the PF6 anion. PF5 is a much
more potent Lewis acid than BF3 on the F-scale; computational
studies reported by Christe and co-workers [10] show that PF5 has
stronger affinity (94.9 kcal/mol) towards fluoride ion than BF3
(83.1 kcal/mol). Such an explanation could account for the need
for a slow addition of CuBr when Selectfluor is used as the oxidant.
The decomposition of the catalytically active, Lewis acidic
Cu(III) species, outlined in Scheme 4, is a possible rationale that
In conclusion, we have developed
a mild and efficient
methodology for copper-mediated oxidation of amides to imides
by Selectfluor. Simply changing the counterion on Selectfluor (F-
TEDA-PF6 instead of F-TEDA-BF4) can significantly reduce the
amount of copper (10 mol%) that is needed. The oxidative ability of
copper bromide (10 mol%) and F-TEDA-PF6 is as efficient as that of
stoichiometric amount of CuBr and Selectfluor, but more
environmentally friendly. Both methodologies could be used to
synthesize unsymmetrical imides from amides, however, the
mechanistic considerations that explain the marked differences in
reactivity of the two salts have not been fully elucidated; thorough
mechanistic studies are needed to clarify the profound counterion
effect.
3. Experimental
3.1. General
All chemicals were commercially obtained from Alfa or Adrich,
and were used without further purification. 1H NMR (400 MHz),
13C NMR (100 MHz) and 19F NMR (376 MHz) spectra were
recorded on a Varian MR400 NMR spectrometer. Chemical shifts
(d) were reported as part per million (ppm). d 7.26, d 77.00 of CHCl3,
0.00 of CFCl3 were used as internal standards for 1H NMR, 13C NMR
and 19F NMR spectra, respectively. High-resolution mass spectra
(HRMS) were performed at mass spectrometry facility of Center for
Regulatory and Environmental Analytical Metabolomics, Univer-
sity of Louisville. Melting points of imides are measured by a
DigiMelt MPA160 melting point apparatus. FTIR spectra were
recorded in ATR (attenuated total reflection) solid mode using a
Perkin Elmer Spectrum 100.
General procedure for copper-mediated oxidation of amide 1
into imide 2 by Selectfluor (F-TEDA-BF4) (Condition A):
Amide 1 (0.25 mmol, 1 equiv.) and Selectfluor (0.625 mmol,
2.5 equiv.) were dissolved in acetonitrile (5 mL) at room
temperature, and CuBr (0.3 mmol, 1.2 equiv.) was added over
a 40 min period in 6 portions. After all the CuBr was added, the
resulting mixture was stirred for extra 20 min, and then
acetonitrile was evaporated under reduced pressure. Saturated
ammonium chloride solution (20 mL) was added into reaction
mixture and extracted with diethyl ether (25 mL Â 4); the ether
layers were combined and dried over Na2SO4, filtered, evapo-
rated under reduced pressure to give the crude product. Silica
gel flash chromatography of the crude product [hexanes–ethyl
acetate (10:1) to hexanes–ethyl acetate (4:1)] yielded pure
imide 2.
General procedure for copper-catalyzed oxidation of amide 1
into imide 2 by F-TEDA-PF6 (Condition B):
Amide
1 (0.25 mmol, 1 equiv.), F-TEDA-PF6 (0.625 mmol,
2.5 equiv.) and CuBr (0.025 mmol, 0.1 equiv.) were dissolved in
acetonitrile (5 mL) and stirred at room temperature for 3–6 h,
monitored by TLC until the reaction showed no further progress.
The work-up was the same as in Condition A.
À
À
Scheme 4. Proposed rationale for different effects of BF4 and PF6 counterions.
N-(3-Methyl-butyryl)-benzamide, (2a) [3b]

230
Z. Jin et al. / Journal of Fluorine Chemistry 143 (2012) 226–230
(c) A.P. Gledhill, C.J. McCall, M.D. Threadgill, Journal of Organic Chemistry 51
1H NMR (400 MHz, CDCl3):
d 8.92 (bs, 1H), 7.82 (d, J = 7.6 Hz,
(1986) 3196–3201;
2H), 7.51 (t, J = 6.4 Hz, 1H), 7.42 (t, J = 7.6 Hz, 2H), 2.82 (d, J = 6.8 Hz,
2H), 2.18 (m, 1H), 0.96 (d, J = 6.4 Hz, 6H).
13C NMR (100 MHz, CDCl3):
127.7, 46.2, 24.8, 22.52.
(d) A. Schnyder, A.F. Indolese, Journal of Organic Chemistry 67 (2002) 594–597;
(e) F. Wang, H. Liu, H. Fu, Y. Jiang, Y. Zhao, Advanced Synthesis and Catalysis 351
(2009) 246–252.
d 175.9, 165.6, 133.1, 132.9, 128.9,
[4] (a) C. Altomare, A. Carotti, G. Casini, S. Cellamare, M. Ferappi, E. Gavuzzo, F.
Mazza, G. Pantaleoni, R. Giorgi, Journal of Medicinal Chemistry 31 (1988)
2153–2158;
(b) A. Gorka, L. Hazai, C. Szantay Jr., V. Hada, L. Szabo, C. Szantay, Heterocycles 65
(2005) 1359–1371;
Acknowledgments
(c) F. Minisci, C. Punta, F. Recupero, F. Fontana, G.F. Pedulli, Journal of Organic
Chemistry 67 (2002) 2671–2676;
(d) H. Nakayama, A. Itoh, SYNLETT (2008) 675–678;
(e) K.C. Nicolaou, C.J.N. Mathison, Angewandte Chemie International Edition 44
(2005) 5992–5997;
(f) J.P.N. Papillon, R.J.K. Taylor, Organic Letters 2 (2000) 1987–1990;
(g) T. Sueda, D. Kajishima, S. Goto, Journal of Organic Chemistry 68 (2003) 3307–
3310;
We are grateful to the National Science Foundation for financial
support (CHE-1111316), and acknowledge the support provided
by the CREAM Mass Spectrometry Facility (University of Louisville)
funded by NSF/EPSCoR (EPS-0447479).
(h) J.-R. Wang, L. Liu, Y.-F. Wang, Y. Zhang, W. Deng, Q.-X. Guo, Tetrahedron
Letters 46 (2005) 4647–4651;
References
(i) L. Xu, S. Zhang, M.L. Trudell, Chemical Communications (2004) 1668–1669;
(j) S. Yoshifuji, Y. Arakawa, Chemical and Pharmaceutical Bulletin 37 (1989)
3380–3381.
[1] (a) K. Krohn, C. Franke, P.G. Jones, H.J. Aust, S. Draeger, B. Schulz, Liebigs Annalen
der Chemie (1992) 789–798;
(b) H.B. Maruyama, Y. Suhara, J. Suzuki-Watanabe, Y. Maeshima, N. Shimizu, M.
Ogura-Hamada, H. Fujimoto, K. Takano, Journal of Antibiotics 28 (1975) 636–647;
(c) L. Ross, L.A. Guarino, Virology 232 (1997) 105–113;
[5] E. McNeill, J.D. Bois, Journal of the American Chemical Society 132 (2010) 10202–
10204.
[6] Z. Jin, B. Xu, G.B. Hammond, Tetrahedron Letters 52 (2011) 1956–1959.
[7] T. Furuya, A.E. Strom, T. Ritter, Journal of the American Chemical Society 131
(2009) 1662–1663.
[8] Q. Michaudel, D. Thevenet, P.S. Baran, Journal of the American Chemical Society
134 (2012) 2547–2550.
[9] (a) L.M. Huffman, S.S. Stahl, Journal of the American Chemical Society 130 (2008)
9196–9197;
(b) A.E. King, L.M. Huffman, A. Casitas, M. Costas, X. Ribas, S.S. Stahl, Journal of the
American Chemical Society 132 (2010) 12068–12073.
[10] K.O. Christe, D.A. Dixon, D. McLemore, W.W. Wilson, J.A. Sheehy, J.A. Boatz,
Journal of Fluorine Chemistry 101 (2000) 151–153.
(d) Y. Suhara, H.B. Maruyama, Y. Kotoh, Y. Miyasaka, K. Yokose, H. Shirai, K.
Takano, P. Quitt, P. Lanz, Journal of Antibiotics 28 (1975) 648–655;
(e) J. Thirkettle, E. Alvarez, H. Boyd, M. Brown, E. Diez, J. Hueso, S. Elson, M.
Fulston, C. Gershater, M.L. Morata, P. Perez, S. Ready, J.M. Sanchez-Puelles, R.
Sheridan, A. Stefanska, S. Warr, Journal of Antibiotics 53 (2000) 664–669.
[2] (a) E.L. Eliel, R.T. McBride, S. Kaufmann, Journal of the American Chemical Society
75 (1953) 4291–4296;
(b) G.E. Ficken, R.P. Linstead, Journal of the Chemical Society (1952) 4846–4854.
[3] (a) H. Alper, C. Mahatantila, F. Einstein, A. Willis, Journal of the American
Chemical Society 106 (1984) 2708–2710;
(b) D.A. Evans, P. Nagorny, R. Xu, Organic Letters 8 (2006) 5669–5671;