Elementalfluorine. Part 14.1 Electrophilic fluorination and nitrogen functionalisation of hydrocarbons

Article abstract of DOI:10.1039/b204776b

Selective fluorination of a range of hydrocarbons was achieved by reaction with either elemental fluorine or Selectfluor, an electrophilic fluorinating reagent of the N-F class. An electrophilic mechanism is envisaged. On prolonged reaction, the strongly acidic reaction medium that is formed upon substitution of hydrogen by fluorine when Selectfluor is used as the fluorinating reagent, promotes loss of fluoride from the initial fluorinated product. Trapping of the subsequent carbocation by the acetonitrile solvent in a Ritter type process gives overall nitrogen functionalisation of hydrocarbons. Amidation of hydrocarbons could also be achieved in a one-stage process by reaction of the hydrocarbon with fluorine and a Lewis acid, such as boron trifluoride-diethyl ether, in acetonitrile.

Full text of DOI:10.1039/b204776b

View Article Online / Journal Homepage / Table of Contents for this issue
Elemental fluorine. Part 14.1 Electrophilic fluorination and
nitrogen functionalisation of hydrocarbons
Richard D. Chambers,*a Alan M. Kenwright,a Mandy Parsons,a Graham Sandford*a and
John S. Moillietb
1
a Department of Chemistry, University of Durham, South Road, Durham, UK DH1 3LE
b F2 Chemicals Ltd., Lea Lane, Lea Town, Preston, UK PR4 OXJ.
E-mail: r.d.chambers@durham.ac.uk; graham.sandford@durham.ac.uk
Received (in Cambridge, UK) 16th May 2002, Accepted 30th July 2002
First published as an Advance Article on the web 3rd September 2002
Selective fluorination of a range of hydrocarbons was achieved by reaction with either elemental fluorine or
Selectfluor, an electrophilic fluorinating reagent of the N–F class. An electrophilic mechanism is envisaged. On
prolonged reaction, the strongly acidic reaction medium that is formed upon substitution of hydrogen by fluorine
when Selectfluor is used as the fluorinating reagent, promotes loss of fluoride from the initial fluorinated product.
Trapping of the subsequent carbocation by the acetonitrile solvent in a Ritter type process gives overall nitrogen
functionalisation of hydrocarbons. Amidation of hydrocarbons could also be achieved in a one-stage process by
reaction of the hydrocarbon with fluorine and a Lewis acid, such as boron trifluoride–diethyl ether, in acetonitrile.
able for scale-up procedures. Nevertheless, these pioneering
studies suggest that direct fluorination could be a viable method
for the selective functionalisation of hydrocarbons, provided
Introduction
The search for efficient methodology for the selective conver-
sion of saturated hydrocarbons to functionalised derivatives
that more appropriate reaction media and temperature could be
has been the focus of substantial effort from many research
used. Furthermore, the fact that many selectively fluorinated
groups worldwide over the last 20 years.2–4
Effective processes for the selective transformation of hydro-
molecules can possess significantly different chemical, bio-
logical and physical properties compared to the corresponding
carbons to functionalised derivatives are urgently required.
non-fluorinated derivatives, features that find many useful
Many approaches have been investigated2–4 but low yields, long
applications in the life-science industries,9 provides a further
reaction times, lack of selectivity and high expense, limit the use
of many of these approaches. In general, the functionalised
alkane products are typically more reactive than the alkane
starting materials and reactions must be carried out to very low
conversion in order to minimize by-product formation.
Halogenation of alkanes offers direct methodology for the
stimulus for the development of efficient and selective direct
fluorination methodology.
The transformation of a carbon–hydrogen bond to a carbon–
fluorine bond using fluorine is a very exothermic process10 (∆H
= Ϫ430.5 kJ molϪ1) and there is much discussion in the liter-
ature11 concerning whether the mechanism of such substitution
functionalisation of hydrocarbons since derived haloalkanes
reactions follow either a free radical or electrophilic pathway.
can, of course, be used as substrates for the synthesis of a wide
range of functional systems. In particular, chlorination of
hydrocarbons and other saturated systems has been extensively
Since fluorine is less than 1% dissociated at room temperature,10
the concentration of fluorine atoms may not be sufficient to
initiate a radical chain process. An alternative initiation step
studied and the effects of solvent, temperature, stereochemistry,
(∆H = 16.3 kJ molϪ1, Fig. 1), originally suggested by Miller,12
rearrangement and relative reactivities of various C–H bonds in
saturated systems towards chlorine for a wide range of sub-
strates are well documented.5 Both thermal and photochemical
chlorination processes are established and are usually operated
at high temperature (>300 ЊC) in the vapour phase but, due to
the harsh reaction environment, free radical chlorination of
hydrocarbons exhibit low selectivity and, in general, give all
possible monochlorinated products upon reaction with alkanes.
In contrast, the replacement of C–H by a C–F bond in satur-
ated systems upon reaction with fluorine has received little
attention and a systematic survey of reactions between fluorine
and a variety of hydrocarbons has not been reported. Tedder6
studied gas phase reactions between, for example, fluorine and
n-butane and noted only minor differences in the reactivities of
each carbon site. More recently Barton7 and Rozen,8 demon-
Fig. 1 Thermodynamic data for the fluorination of methane.
strated that tertiary C–H bonds may be selectively transformed
to C–F bonds. However, in all of the reactions reported by these
workers only replacement of tertiary C–H bonds was reported
and reactions were carried out in a solvent mixture consisting
of chloroform and CFCl3 (a solvent now unavailable under the
terms of the Montreal Protocol) at low temperatures, unsuit-
probably occurs but conclusive evidence for this pathway has
not been established.
Barton and Rozen suggested7,8 that the replacement of
hydrogen by fluorine in saturated systems proceeds by an
electrophilic mechanism involving 3 centre, 2 electron bond
2190
J. Chem. Soc., Perkin Trans. 1, 2002, 2190–2197
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b204776b

View Article Online
Table 1 Survey of solvents for fluorination of hydrocarbons.
cis-9-Fluorodecalin, 2
Solvent
F2–substrate
Yield %a
Conv. (%)
Tar c (%)
CH3CN
CH3CN
CH3CN
1 : 1
5 : 1
16 : 1
<10
57
10
56
52
27
0
60
66
50
0
13b
64
0
0
13
0
0
0
99b
63
CH3CN–CFCl2CF2Cl (1 : 4)
CH3CN–CFCl2CF2Cl (1 : 9)
CH3CN–CFCl2CF2Cl (1 : 19)
CH3CN–CH2Cl2 (1 : 19)
CH3CN–p-dinitrobenzene (50 : 1)
CH3CN–p-dinitrobenzene (13 : 1)
CH3CH2CN
HCOOH
H2SO4
CF3COOH
CFCl2CF2Cl–CF3SO3H (1 : 1)
CFCl2CF2Cl–CF3SO3H(19 : 1)
BF3–CH2Cl2 (1 : 20)
BF3–CH2Cl2 (1 : 5)
CF3CH2OH
CFCl2CF2Cl
CH2Cl2
92
40
2
83
0
85
78
0
trace
25
33
28
28
9
8
0
28
1 : 1
2b
0
25b
69b
100b
80b
91b
92b
61b
85b
4b
<1
<1
<1
0
0
0
8
1 : 5
1 : 10
<1
<1
33
CH2Cl2
CH3NO2
11
97
24
a After work-up. b Conversion was estimated using GC-MS. c % tar by weight.
Scheme 1
carbocationic intermediates (Scheme 1), analogous to aliphatic
electrophilic nitration processes (SE1 process) described by
Olah.13 Furthermore, the direct fluorination of trans-decalin†
proceeded with retention of stereochemistry which, it was
suggested, could only be attributed to an electrophilic process.
However, it must be noted that various free radical reactions,
including chlorination of decalin at the tertiary position, can
also occur with retention of stereochemistry.14
In an attempt to address the problem of the mechanism of
direct fluorination (radical or electrophilic?), we reasoned that a
comparison of fluorination reactions between alkanes and both
fluorine and established electrophilic fluorinating reagents
would provide an indication as to the nature of the process. A
range of electrophilic fluorinating agents15 are now com-
mercially available, e.g. Selectfluor (Air Products) and a series
of radical clock experiments by Differding16 and Wong17 estab-
lished that quinuclidine fluorinating agents such as Select-
fluor are electrophilic in character. If similar product profiles
could be obtained in reactions between hydrocarbons and both
fluorine and Selectfluor, we could reasonably suggest that the
mechanism of both processes are electrophilic in nature. How-
ever, until our recent communication,18 fluorination of alkanes
direct fluorination processes. Furthermore, we aimed to estab-
lish viable reaction conditions (solvent and temperature, etc.)
and extend the range of hydrocarbons that could be fluorinated
to those that do not possess tertiary sites.
Results and discussion
In earlier parts of this series, we established that selective direct
fluorination of aromatics19,20 and a variety of dicarbonyl
derivatives21 is significantly affected by the nature of the reac-
tion medium and can be accomplished by using either an acidic
or a high relative permittivity reaction medium, such as formic
acid and acetonitrile respectively, which promote selective
electrophilic fluorination processes.
In order to establish a suitable solvent that could be used as
a reaction medium for selective direct fluorination of hydro-
carbons at convenient temperatures, we carried out a series
of reactions between fluorine and cis-decalin 1 in a range of
solvents at 0 ЊC (Table 1). In each case the conversion of the
starting material, the yield of cis-9-fluorodecalin 2 and tar
formation were recorded for comparison. cis-Decalin 1 was
chosen as the model substrate for comparison with Rozen’s
results (see above).
by N–F reagents had not been reported.
In this paper, we describe our studies directed towards the
fluorination of various hydrocarbons and compare the results
of direct fluorination reactions using elemental fluorine with
those carried out using an established electrophilic fluorinating
agent, Selectfluor, with a view to probing the mechanism of
Several points from the data collected in Table 1 are worth
noting; (1) acid reaction media, used effectively in electrophilic
fluorination reactions of aromatic19 and 1,3-dicarbonyl sub-
strates,21 gave low yields of 9-cis-fluorodecalin and substantial
amounts of tarry material; (2) low yields of fluorodecalin were
obtained in chlorinated solvents; (3) nitromethane, a solvent
with high relative permittivity allows the preparation of
† The IUPAC name for decalin is decahydronaphthalene.
J. Chem. Soc., Perkin Trans. 1, 2002, 2190–2197
2191

View Article Online
Table 2 Fluorination of hydrocarbons
Substrate
Conditions A or B Conversion (%)
Product
Others
A (70)
B (100)
65%
43%
23%
8%
14%
5%
others
A (53)
B (100)
63%
21%
4%
A (60)
41%, 10a exo, 10b endo 5.3 : 1
A (61)
B (84)
12, 63%, 2.4 :1.3 : 1 :1.1 : 0.4a
12, 58%, 2.4 :1.3 :1 :1.1a
a Not individually assigned (see text).
reasonable yields of 9-fluorodecalin but tar formation limits the
effective use of this solvent; (4) the most suitable solvents for
direct fluorinations, of those surveyed, are the nitriles and, for
subsequent direct fluorination reactions, our reaction medium
of choice was, therefore, acetonitrile; (5) addition of p-dinitro-
benzene, a radical inhibitor, to the acetonitrile reaction medium
made no significant difference to the outcome of the fluorin-
ation, indicating that an electrophilic, rather than a radical,
process is more likely.
The effectiveness of nitriles as reaction media does not, there-
fore, depend solely on relative permittivity because nitro-
methane is not a suitable medium. It is tempting to think that
interaction between acetonitrile and fluorine is occurring and
an electrophilic N–F reagent 3 is generated in situ (Scheme 2)
passing fluorine through acetonitrile, followed by addition of
decalin, then no product was observed.
Direct fluorinations of several hydrocarbons were carried out
(Conditions A, Table 2) and the products of these reactions
compared with those obtained by fluorination of the same
hydrocarbon substrates using Selectfluor (Conditions B,
Table 2). Direct fluorinations were carried out at 0 ЊC, but reflux
temperature of the acetonitrile was required in order that fluor-
inations with Selectfluor proceeded at a reasonable rate.
Fluorination of adamantane 4 by fluorine and Selectfluor
both gave mixtures consisting of 1- and 2-fluoroadamantane, 5
and 6, and small quantities of other products. In each case, the
major monofluorinated product 5 could be isolated by column
chromatography. Fluorination of cyclohexane 7 gave fluoro-
cyclohexane 8 and trace amounts of other products. The low
isolated yield of fluorocyclohexane 8 obtained upon fluorination
by Selectfluor is due to handling losses of the highly volatile
product during reaction (reflux at ∼80 ЊC) and/or work-up.
Direct fluorination of norbornane‡ 9 gave a mixture of endo
and exo-2-fluoronorbornane, 10a and 10b, in a ratio of 5.3 : 1.
Direct fluorination of n-decane 11 gave a mixture of all five
monofluorinated products 12 in the ratio 2.4 : 1.3 : 1 : 1.1 : 0.4
which gave 19F NMR signals at Ϫ173.4, Ϫ181.4, Ϫ181.7,
Scheme 2
but, while complexes involving interaction of acetonitrile and
fluorine have been observed at low temperatures by Legon,22 we
could not observe the presence of an N–F type complex by 19
F
NMR under the present reaction conditions. Moreover, when
we attempted the process in two separate steps, that is, first
‡ The IUPAC name for norbornane is bicyclo[2.2.1]heptane
2192
J. Chem. Soc., Perkin Trans. 1, 2002, 2190–2197

View Article Online
Table 3 Fluorination of cis- and trans-decalin
Substrate
Conditions A or B
A
Product
Yield (%)
54
Conversion (%)
68
B
A
B
15, 30
15a :15b :15c :15d
1 :1.14 :1.36 :2.24
75
64
81
57
16, 25
1.6 : 1.3 : 1 : 1.4a
a Not individually assigned (see text).
Ϫ182.7 and Ϫ217.6 ppm respectively and many unidentified
products in trace quantities. The triplet observed at Ϫ217.6
ppm is readily identified as product 12e arising from conversion
of CH3 sites to CH2F. The four other monofluorinated deriv-
atives 12a–d bearing fluorine atoms at secondary carbon sites
were isolated as an isomeric mixture by preparative scale
GC but it is very difficult to assign with certainty which 19F
resonance corresponds to each isomer. However, the high
frequency 19F NMR resonance (Ϫ173.4 ppm) can be assigned
to the 2-fluorodecane 12a by comparison with literature data of
a similar system (2-fluorooctane23 δF Ϫ170.8 ppm).
in which the fluorine atom is attached to the secondary sites and
no evidence of fluorination at the tertiary position was detected
1
by 19F NMR. In this case, the resonances observed in the H
NMR spectrum that are assigned to the CHF hydrogen atoms
are too broad to be of diagnostic use (see above) but the iso-
mers 16a–d could be tentatively identified by a comparison of
the chemical shift values observed in the 19F NMR spectrum at
Ϫ166.0 (16c), Ϫ173.4 (16b), Ϫ179 1 (16d) and Ϫ183.9 (16a)
in the ratio 1 : 1.3 : 1.4 : 1.6 with those obtained for the
trans-decalin isomers described above.
From the fluorination reactions described above, we can
see that for fluorination of cyclohexane 7, adamantane 4 and
decane 11 the product profiles obtained when using either
fluorine or Selectfluor are very similar and so an electrophilic
process for direct fluorination can be considered more likely.
However, fluorination of the decalin substrates 1 and 13 gave
totally different products depending on whether fluorine or
Selectfluor was used. The preferential fluorination of the
secondary sites of decalin by Selectfluor is probably due to
the increased steric demand of this fluorinating agent, which is
too bulky to access the electronically preferred tertiary site.
As shown above, reactions between Selectfluor and hydro-
carbons gave reasonable yields of the desired fluorinated
products. However, an attempt to increase the yield of fluoro-
adamantane by heating adamantane 4 with Selectfluor over a
prolonged period actually led to the isolation of N-acetylated
adamantane 17 as the major product and only a small quantity
of the expected fluoroadamantane 5 was observed (Scheme 3).
In order to establish the course of this reaction, the reaction
was repeated and samples were taken at regular intervals and
analysed by GC/MS. This analysis showed that the yield of
fluoroadamantane 5 reaches a maximum value after approxi-
mately five hours and then decreases as the reaction time
progresses. Concomitantly, the yield of acetamide 17 is
negligible until after three hours and then steadily increases as
reaction time increases. Therefore, we can reasonably suggest
Fluorination of n-decane 11 by Selectfluor gave a similar
product distribution; fluoro-decanes 12 were obtained in low
yield in a ratio of 2.4 : 1.3 : 1.0 : 1.1, also showing a slight
preference for fluorination at the 2-position to give 12a.
In contrast to the reactions described above, fluorination of
cis- and trans-decalin, 1 and 13, led to differing products which
depended upon the fluorinating agent used (Table 3). Reaction
of fluorine with cis- and trans-decalin gave cis- and trans-9-
fluorodecalin, 2 and 14, respectively arising from substitution
of the tertiary hydrogen atoms with retention of configuration,
consistent with Rozen’s observations.8 However, reaction of
trans-decalin 13 with Selectfluor gave products 15a–d arising
from fluorination at the CH2 sites only and no fluorination of
tertiary sites was observed, as determined by 19F NMR. The
four monofluorinated products 15a–d were obtained in the ratio
1 : 1.14 : 1.36 : 2.24 and each isomer gave a doublet (2JHF ∼49
Hz) in the 19F NMR spectrum at Ϫ196.6 (15a), Ϫ177.3 (15b),
Ϫ167.9 (15c) and Ϫ183.1 (15d) ppm respectively consistent
with the presence of a CHF group. The structures of isomers
15a–d were established by analysis of the 1H NMR resonances
centred at 4.54 (15a), 4.08 (15b), 4.47 (15c) and 4.87 (15d) ppm
after a consideration of the relevant coupling constants (typical
values 2JHF 50 Hz, 3JHaxial–Haxial 10 Hz, 3JHaxial–Hequatorial 5 Hz).
By a similar procedure, fluorination of cis-decalin 1 by
Selectfluor gave a mixture of four monofluoroisomers 16a–d
J. Chem. Soc., Perkin Trans. 1, 2002, 2190–2197
2193

View Article Online
Table 4
Substrate
Product
Yield (%)
51
Conversion (%)
53
Scheme 3
that N-acetamide 17 derives from the fluoroadamantane 5. An
indication as to the mechanism of this process was provided by
measuring the acidity of the reaction mixture which became
increasingly acidic, reaching pH < 1, over the course of the
reaction due to the formation of the highly acidic quinuclidine
derivative 18 (Scheme 4). Ionisation of fluoroadamantane 5 to
the corresponding tertiary carbocation 19 occurs when the
concentration of 18 is sufficiently high (after five hours) and
the cation 19 then reacts with the nucleophilic acetonitrile in a
Ritter type process. Aqueous work-up finally yields the
acetamide (Scheme 4). In contrast to these results, Lawrence
and co-workers suggested a radical mechanism to account
for the formation of fused bicyclic systems upon reaction
of menthol and Selectfluor in various nitriles, by a related
transformation.24
This experiment demonstrated that amidation could be
achieved in one-pot by the reaction of the alkane with a
fluorinating agent, a strong acid and acetonitrile. We then
sought to accomplish the same transformation but using fluor-
ine as the fluorinating agent, a Lewis acid and acetonitrile.
Indeed, we found that N-functionalisation could be achieved in
good isolated yield by passing fluorine through a mixture con-
sisting of the hydrocarbon substrate, boron trifluoride–diethyl
ether and acetonitrile (Table 4) and a representative mechanism
for this process is outlined in Scheme 5.
45
49
49
67
21
45
54
60
74
Amidation of cis- and trans-decalin led to the formation of
the trans-amide 21 in both cases, consistent with the inter-
mediacy of a carbocation species (Scheme 6).
Experimental
All starting materials were obtained commercially (Aldrich)
and all solvents were dried using literature procedures. NMR
spectra were recorded in deuteriochloroform, unless otherwise
stated, on a Varian VXR 400S NMR spectrometer with tetra-
methylsilane and trichlorofluoromethane as internal standards.
Spectral assignments were made with the aid of data collected
by 1H–1H COSY and 1H–13C HETCOR experiments and coup-
ling constants are given in Hz. In 19F NMR spectra, upfield
shifts are quoted as negative. Mass spectra were recorded on
either a VG 7070E spectrometer or a Fissons VG Trio 1000
spectrometer coupled with a Hewlett Packard 5890 series II gas
chromatograph. Accurate mass measurements were performed
by the EPSRC National Mass Spectrometry service, Swansea,
In summary, both tertiary and secondary sites of hydro-
carbons may be selectively fluorinated by both fluorine and
Selectfluor. Since the product profiles of both processes are
similar we can reasonably suggest that the mechanism of both
fluorination processes are electrophilic in nature, confirming
the suggestion of Barton and Rozen. Fluorination of alkanes
by Selectfluor is, however, complicated by the formation of
amide derivatives after prolonged reaction due to the highly
acidic nature of the reaction medium. The formation of amides
promoted by either Selectfluor or fluorine–Lewis acid does
provide effective methodology for the efficient N-functionalis-
ation of hydrocarbons.
Scheme 4
2194
J. Chem. Soc., Perkin Trans. 1, 2002, 2190–2197

View Article Online
Scheme 5
120 (Mϩ, 12%), 99 (10), 85 (55) and unidentified products (3%).
Purification of the crude product by preparative scale GC gave
an analytical sample of fluorocyclohexane 8 as a colourless
liquid; bp 101–103 ЊC (lit.,26 62 ЊC/180 mmHg) (Found: Mϩ,
102.0845. C6H11F requires: Mϩ, 102.0845); δF Ϫ171.47 (br s);
2
δH 1.30–1.75 (10 H, m, CH2), 4.57 (1 H, dm, JHF 48.8, CHF);
3
4
2
δC 22.8 (d, JCF 7.7, C-3), 25.2 (d, JCF 1.5, C-4), 32.3 (d, JCF
18.7, C-2), 91.5 (d, 1JCF 169.8, CHF); m/z (EIϩ) 102 (Mϩ, 8%),
82 (40, Mϩ Ϫ HF), 67 (100).
2-Fluoronorbornane 10a,b. Norbornane 9 (4.93 g, 51 mmol),
fluorine (195 mmol) and acetonitrile (140 ml) gave a reaction
mixture that was poured into iced water (150 ml), neutralised
(NaHCO3) and filtered to remove the solid brown crude prod-
uct (7.02 g) which contained exo- and endo-2-fluoronorbornane
10a,b (1.43 g, 41%, 60% conv.) in the ratio 10a : 10b, 5.3 : 1.0
and, unidentified products (16%). Purification of the crude
product by preparative scale GC gave a sample of exo- and
endo-2-fluoronorbornane 10a,b as an isomeric mixture and as a
white solid; mp 54–60 ЊC (sealed tube); exo-2-fluoronorbornane
10a: δH 0.91–2.49 (10 H, m, CH2 and CH), 4.58 (1 H, dm, 2JHF
56.4, CHF); δF Ϫ160.48 (m); δC 22.3 (d, 3JCF 10.3, C-6), 27.9 (s,
Scheme 6
UK. Elemental analyses were obtained on either a Perkin-
Elmer 240 or a Carlo Erba Elemental Analyser. Melting points
were recorded at atmospheric pressure and are uncorrected.
Column chromatography was carried out on silica gel (Merck
No. 1–09385, 230–400 mesh) and TLC analysis was performed
on silica gel TLC plates using dichloromethane as eluant.
3
2
C-5), 34.7 (d, JCF 1.2, C-7), 34.8 (s, C-4), 39.9 (d, JCF 19.3,
C-3), 42.0 (d, 2JCF 19.4, C-1), 96.2 (d, 1JCF 181.6, C-2); endo-2-
fluoronorbornane 10b: δH 0.91–2.49 (10 H, m, CH2 and CH),
Reactions with elemental fluorine
2
2
5.02 (1 H, dm, JHF 57.9, CHF); δF Ϫ189.85 (ddd, JHF 57.9,
General procedure. Elemental fluorine, as a 10% (v/v) mixture
with nitrogen, was passed at a rate of ca. 50 ml minϪ1 through a
stirred, cooled (0 ЊC) mixture which consisted of the substrate
and acetonitrile. After addition of the fluorine, the reaction
mixture was poured into water (100 ml), neutralised (NaHCO3)
and extracted with dichloromethane (3 × 40 ml). The com-
bined, dried (MgSO4), organic extracts were evaporated to give
a crude product. The composition of a weighed crude reaction
mixture was determined by GCMS analysis and the conversion
of start material was calculated from GC peak intensities. The
amount of fluorinated product in the crude product was deter-
mined by adding a known amount of fluorobenzene to a
weighed amount of the crude product mixture. Comparison of
the relative intensities of the appropriate 19F NMR resonances
gave the yield of fluorinated derivative, based upon the conver-
sion obtained above. Analytical samples of fluorinated prod-
ucts were obtained by either preparative scale GC or column
chromatography. Yields of fluorinated products are quoted as
yields based on the conversion of starting material.
3JHF 29.3, JHF 16.2); δC 19.9 (d, JCF 10.3, C-6), 29.1 (s, C-5),
36.5 (d, 3JCF 2.3, C-4), 36.9 (d, 3JCF 4.9 C-7), 37.2 (d, 2JCF 22.1,
C-3), 41.2 (d, 2JCF 17.5, C-1), 94.7 (d, 1JCF 184.0, C-2); m/z (EIϩ)
114 (Mϩ, 11%), 99 (20).
3
3
Fluorodecane 12. n-Decane 11 (11.08 g, 78 mmol), fluorine
(78 mmol) and acetonitrile (140 ml) gave a brown crude product
mixture (12.07 g) which contained 2-, 3-, 4- and 5-fluorodecane
(4.80 g, 63%, 61% conv.) in the ratio of 1.0 : 1.1 : 1.3 : 2.4
(unassigned, see text), a trace amount of 1-fluorodecane;
δF Ϫ217.56 (m) and, unidentified products (26%). Purification
of the crude product by preparative scale GC gave an analytic-
ally pure sample of 2-, 3-, 4- and 5-fluorodecane as an isomeric
mixture and as a colourless liquid (Found: C, 74.7; H, 13.4.
C10H21F requires: C, 74.95; H, 13.2%); δF Ϫ172.46 (m, JHF
19.2), Ϫ180.38 (m, JHF 17.3), Ϫ180.71 (m, JHF 17.2), Ϫ181.62
(m, JHF 18.8); δH 0.85–0.98 (4.5 H, m, CH2 and/or CH3), 1.24–
1.70 (14.8 H, m, CH2 and/or CH3), 4.30–4.74 (1.0 H, m, CHF);
m/z (EIϩ) 140 (Mϩ Ϫ HF, 1%), 111 (9), 97 (22).
1-Fluoroadamantane 5. Adamantane 4 (4.00 g, 29 mmol),
fluorine (205 mmol) and acetonitrile (140 ml) gave a pale brown
crude product which contained 1-fluoroadamantane 5 (51%),
2-fluoroadamantane 6 (18%); m/z (EIϩ) 154 (Mϩ, 75%), 111 (19)
and, unidentified products (11%). Purification of the mixture
by column chromatography on silica gel using 6 : 1 cyclo-
hexane–DCM as the eluent gave 1-fluoroadamantane 5 (2.06 g,
65%, 70% conv.) as a white solid; mp 256–258 ЊC (lit.,25 257–259
ЊC); Rf 0.52 (Found: C, 77.8; H, 9.9. C10H15F requires: C, 77.9;
H, 9.8%); δF Ϫ128.8 (s); δH 1.62 (2 H, m, CH2), 1.88 (2 H, m,
trans-9-Fluorodecalin 14. trans-Decalin 13 (3.00 g, 22 mmol),
fluorine (110 mmol) and acetonitrile (140 ml) gave a yellow
crude product (5.00 g) which contained trans-9-fluorodecalin 14
(1.26 g, 54%, 68% conv.) and trace amounts of difluorodecalin.
Purification of the crude product by preparative scale GC gave
an analytically pure sample of trans-9-fluorodecalin 14 as a
colourless liquid; bp 197–198 ЊC (Found: C, 76.8; H, 11.0.
C10H17F requires: C, 76.9; H, 10.9%); δF Ϫ177.11 (m); δH 1.15–
3
1.80 (m); δC 21.6 (d, JCF 1.5, C-2), 25.8 (s, C-3), 28.7 (d,
3JCF 1.1, C-4), 37.1 (d, 2JCF 23.4, C-1), 43.1 (d, 2JCF 21.5, C-10),
3
CH2CF), 2.23 (1 H, s, CH); δC 31.5 (d, JCF 9.6, CH), 36.1 (d,
1
94.5 (d, JCF 172.7, C-9); m/z (EIϩ) 156 (Mϩ, 15%), 136 (100,
4JCF 1.9, CH2), 43.0 (d, 2JCF 17.1, CH2CF), 92.8 (d, 1JCF 183.2,
CF); m/z (EIϩ) 154 (Mϩ, 75%), 97 (100).
M Ϫ HF).
Fluorocyclohexane 8. Cyclohexane 7 (10.92 g, 130 mmol),
fluorine (65 mmol) and acetonitrile (140 ml) gave a yellow crude
product mixture (11.08 g) which contained fluorocyclohexane 8
(4.43 g, 63%, 53% conv.), difluorocyclohexane (1%); m/z (EIϩ)
cis-9-Fluorodecalin 2. cis-Decalin 1 (3.00 g, 22 mmol), fluor-
ine (110 mmol) and acetonitrile (140 ml) gave a yellow crude
product (7.44 g) which contained cis-9-fluorodecalin 2 (1.25 g,
57%, 64% conv.), spectral data below; difluorodecalin (5%),
J. Chem. Soc., Perkin Trans. 1, 2002, 2190–2197
2195

View Article Online
m/z (EIϩ) 174 (Mϩ, 66%), 154 (44, Mϩ Ϫ HF); an unsaturated
component (2%), m/z (EIϩ) 136 (Mϩ, 100%); and, unidentified
products (8%). Purification of the crude product by preparative
scale GC gave an analytically pure sample of cis-9-fluoro-
decalin 2 as a colourless liquid; bp 202–203 ЊC (Found: C,
76.65; H, 11.1. C10H17F requires: C, 76.9; H, 10.9%); δF Ϫ140.57
(m); δH 1.20–1.85 (m); δC 20.0–38.1 (broad overlapping signals,
δF Ϫ167.97 (dd, 2JHF 49.4, 3JHF 4.5); δH 0.60–2.10 (16 H, m, CH2
2 3 3
and CH), 4.47 (1 H, dtt, JHF 49.2, JHF 10.8, JHF 4.8, CHF);
2
3
15d: δF Ϫ183.07 (tq, JHF 47.8, JHF 10.2); δH 0.60–2.10 (16 H,
m, CH2 and CH), 4.87 (1 H, d sept, 2JHF 48.4, 3JHF 2.0, CHF);
m/z (EIϩ) 156 (Mϩ, 82%), 136 (76, Mϩ Ϫ HF).
cis-Fluorodecalin. Selectfluor (13.85 g, 39 mmol), cis-
decalin 1 (3.00 g, 22 mmol) and acetonitrile (120 ml), after
1.5 h, gave a yellow liquid (3.20 g) which contained cis-
fluorodecalin 16 (0.77 g, 30%, 75% conv.) and trace amounts of
other products. Purification of the crude product by prepar-
ative scale GC gave an analytically pure sample of cis-
fluorodecalin as an isomeric mixture of 1-fluoro(ax)-cis-decalin
16a, 1-fluoro(eq)-cis-decalin 16b, 2-fluoro(eq)-cis-decalin 16c and
2-fluoro(ax)-cis-decalin 16d in the ratio of 1.6 : 1.4 : 1.3 : 1.0
(unassigned, see text) and, as a colourless oil; (Found: C, 77.0;
H, 11.1. C10H17F requires C, 76.9; H, 11.0%); δF (Ϫ57 ЊC)
Ϫ166.55 (d, 2JHF 47.8), Ϫ173.43 (d, 2JHF 49.3), Ϫ179.08 (d, 2JHF
49.6), Ϫ183.92 (q, 2JHF 37.9 ); δH (Ϫ57 ЊC) 1.07–2.06 (16 H, m,
CH and CH2), 4.45–4.84 (1 H, m, CHF); m/z (EIϩ) 156 (Mϩ,
22%), 136 (86, MϩϪ HF).
2
1
CH2), 40.8 (d, JCF 19.5, C-10), 96.9 (d, JCF 169.8, C-9); m/z
(EIϩ) 156 (Mϩ, 15%), 136 (94, M Ϫ HF), 113 (100).
Reactions with Selectfluor
General procedure. A solution consisting of Selectfluor,
substrate and acetonitrile (130 ml) was stirred and heated (65
ЊC). After 24 h the reaction mixture was poured into water,
neutralised (NaHCO3) and extracted with dichloromethane
(3 × 50 ml). The combined, dried (MgSO4) organic extracts
were evaporated to give a crude product which was analysed by
GCMS and 19F NMR as described above. Purification of the
crude product by preparative scale GC gave an analytically pure
sample of the monofluorinated products.
1-Fluoroadamantane 5. Adamantane 4 (1.0 g, 7 mmol),
Selectfluor (4.7 g, 13 mmol) and acetonitrile (100 ml) gave a
pale brown crude product mixture (1.5 g) which contained
1-fluoroadamantane 5 and 2-fluoroadamantane 6 in the ratio
of 5.5 : 1 respectively, adamantanol (4 area%); m/z (EIϩ) 152
(Mϩ, 20%), 95 (100), N-(adamantyl)acetamide (1 area%); m/z
(EIϩ) 193 (Mϩ, 43%), 136 (53) and unidentified products. Puri-
fication of the mixture by column chromatography on silica gel
using 6 : 1 cyclohexane–DCM as the eluent gave 1-fluoro-
adamantane 5 (0.5 g, 68%, 73% conv.) as a white solid; physical
and spectral data as above.
Nitrogen functionalisation reactions
Method 1.
N-(1-Adamantyl)acetamide 17. Fluorine (208 mmol) as a 10%
(v/v) mixture with nitrogen was passed through a stirred, cooled
(0 ЊC) mixture consisting of adamantane 4 (7.07 g, 52 mmol)
and acetonitrile (140 ml). After addition of the fluorine, boron
trifluoride–diethyl ether (29.54 g, 208 mmol) was added and,
after 5 min at rt, the reaction mixture was poured into water,
neutralised (NaHCO3) and extracted using dichloromethane.
The combined dried (MgSO4) organic extracts were evaporated
to give a brown crude product mixture which, upon recrystal-
lisation from acetonitrile, gave N-(1-adamantyl)acetamide 17
(5.25 g, 74%, 70% conv.) as a white solid; mp 150–151 ЊC
(Found: C, 74.55; H, 9.9; N, 7.3. C12H19NO requires C, 74.55;
H, 9.9; N, 7.3%); δH 1.66 (6 H, br s, CH2), 1.89 (3 H, s, CH3),
1.97 (6 H, s, CH2CN), 2.05 (3 H, m, CH), 5.20 (1 H, br s, NH);
δC 24.7 (s, CH3), 29.4 (s, CH), 36.3 (s, CH2), 41.6 (s, CH2CN),
51.8 (s, CN), 169.3 (s, CO); m/z (EIϩ) 193 (Mϩ, 100%), 178 (24,
Mϩ Ϫ CH3), 150 (18, Mϩ Ϫ COCH3).
Fluorocyclohexane 8. Selectfluor (13.91 g, 39 mmol), cyclo-
hexane 7 (3.00 g, 36 mmol) and acetonitrile (130 ml) gave a
colourless product (8.00 g) which contained fluorocyclohexane
8 (0.77 g, 21%, 100% conv.). Purification by preparative scale
GC gave an analytically pure sample of fluorocyclohexane 8 as
a colourless oil; physical and spectral data as above.
n-Fluorodecane 12. Selectfluor (21.42 g, 61 mmol),
n-decane 11 (7.81 g, 55 mmol) and acetonitrile (210 ml), after 18
h, gave an orange liquid (10.77 g) which contained fluorodecane
12 (4.28 g, 58%, 84% conv.) and small amounts of other
unidentified products. Purification by preparative scale GC gave
an isomeric mixture of 2-, 3-, 4- and 5-fluorodecane in the ratio
of 2.39 : 1.27 : 1.09 : 1.00 (unassigned, see text) and, as a colour-
less liquid (Found: Mϩ Ϫ HF, 140.1565. C10H21F requires Mϩ Ϫ
HF, 140.1565); δF Ϫ172.46 (m, JHF 19.2), Ϫ180.38 (m, JHF
17.3), Ϫ180.71 (m, JHF 17.2), Ϫ181.62 (m, JHF 18.8); δH 0.85–
0.98 (5 H, m, CH2 and/or CH3), 1.24–1.70 (15 H, m, CH2 and/
or CH3), 4.30–4.74 (1.00 H, m, CHF); m/z (EIϩ) 140 (Mϩ Ϫ HF,
1%), 111 (9), 97 (22).
Method 2 – General procedure. Fluorine as a 10% (v/v) mix-
ture with nitrogen was passed through a stirred, cooled (0 ЊC)
mixture consisting of the hydrocarbon, boron trifluoride–
diethyl ether and acetonitrile. After addition of the fluorine, the
reaction mixture was poured into water, neutralised (NaHCO3)
and extracted using dichloromethane. The combined dried
(MgSO4) organic extracts were evaporated to give a crude
product which was purified as stated below.
N-(Cyclohexyl)acetamide 20. Cyclohexane 7 (9.8 g, 117
mmol), fluorine (59 mmol), boron trifluoride–diethyl ether
(16.6 g, 117 mmol), and acetonitrile (120 ml) gave a crude
product mixture which was distilled to give cyclohexane (4.6 g,
55 mmol); bp 67 ЊC. The brown solid which remained gave, after
recrystallisation, N-(cyclohexyl)acetamide 20 (4.5 g, 51%, 53%
conv.) as a white solid; mp 104–105 ЊC (from acetonitrile) (lit.27
mp 104 ЊC); (Found: Mϩ, 141.1153. C8H15NO requires: Mϩ,
141.1153); δH1.06–2.0 (10 H, m, CH2), 1.94 (3 H, s, CH3), 3.74
(1 H, m, H-1), 5.43 (1 H, br s, NH); δC 23.6 (s, CH3), 24.8 (s,
trans-Fluorodecalin 15. Selectfluor (24.85 g, 70 mmol),
trans-decalin 13 (5.38 g, 70 mmol) and acetonitrile (250 ml)
after 4.5 h gave a brown liquid (4.37 g) which contained trans-
fluorodecalin 15 (2.21 g, 25%, 81% conv.) and trace amounts of
other products. Purification of the crude product by preparative
scale GC gave an analytically pure sample of trans-fluoro-
decalin 15 as an isomeric mixture of 2-fluoro(eq)-trans-decalin
15c, 1-fluoro(eq)-trans-decalin 15b, 2-fluoro(ax)-trans-decalin 15d
and 1-fluoro(ax)-trans-decalin 15a in the ratio of 1.36 : 1.14 :
2.24 : 1.00 respectively and, as a colourless oil (Found: C, 77.0;
H, 11.1. C10H17F requires C, 76.9; H, 11.0%); 15a: δF Ϫ196.61
C-4), 25.5 (s, C-3), 33.2 (s, C-2), 48.1 (s, C-1), 169.0 (s, C᎐O);
᎐
m/z (EIϩ) 141 (Mϩ, 8%), 98 (4, Mϩ Ϫ COCH3), 70 (6), 60 (58),
43 (100).
2
(mq, JHF 49.6); δH 0.60–2.10 (16 H, m, CH2 and CH), 4.54
(1 H, ddt, 2JHF 49.2, 3JHF 3.2, 3JHF 2.0, CHF); 15b: δF Ϫ177.34
N-(trans-9-Decalyl)acetamide 21. cis-Decalin 1 (2.2 g, 16
mmol), fluorine (65 mmol), boron trifluoride–diethyl ether
(2.3 g, 16 mmol), and acetonitrile (120 ml) gave a crude product
(d, 2JHF 49.3); δH 0.60–2.10 (16 H, m, CH2 and CH), 4.08 (1 H,
2
3
3
3
dddd, JHF 49.8, JHF 10.8, JHF 9.9, JHF 4.8, CHF); 15c:
2196 J. Chem. Soc., Perkin Trans. 1, 2002, 2190–2197

View Article Online
mixture which was distilled to give decalin (0.7 g, 5 mmol); bp
55 ЊC/10 mmHg. The brown solid which remained gave, after
recrystallisation, N-(trans-9-decalyl)acetamide 21 (0.8 g, 49%,
67% conv.) as a white solid; mp 183–184 ЊC (from acetonitrile)
(Found: C, 73.6; H, 10.9; N, 7.3. C12H21NO requires: C, 73.8; H,
10.9; N, 7.2%); δH 0.85–1.68 (18 H, m, CH and CH2), 1.94 (3 H,
s, CH3), 2.57 (2 H, m, CH2), 4.89 (1 H, br s, NH); δC 21.6 (s,
CH2), 24.6 (s, CH3), 26.1 (s, CH2), 28.6 (s, CH2), 34.4 (s, CH2),
45.2 (s, CH), 55.9 (s, CNH), 169.2 (s, CO); m/z (EIϩ) 195 (Mϩ,
1%), 152 (18), 136 (100), 121 (141).
References
1 For Part 13, see R. D. Chambers, D. Holling, R. C. H. Spink and
G. Sandford, Lab Chip, 2001, 1, 132.
2 Activation and Functionalisation of Alkanes, ed.C. L. Hill, John
Wiley and Sons, New York, 1989.
3 J. A. Davies, Selective Hydrocarbon Activation: Principles and
Progress, VCH, New York, 1994.
4 G. A. Olah and A. Molnar, Hydrocarbon Chemistry, John Wiley and
Sons, New York, 1995.
5 M. L. Poutsma, in Free radicals, ed. J. K. Kochi, John Wiley and
Sons, New York, 1973, Vol. 2, p. 159.
6 J. M. Tedder, Adv. Fluorine Chem., 1961, 2, 104.
7 D. H. R. Barton, R. H. Hesse, R. F. Markwell, M. M. Pechet and
S. Rozen, J. Am. Chem. Soc., 1976, 98, 3034.
8 S. Rozen, Acc. Chem. Res., 1988, 21, 307.
9 Organofluorine Chemistry. Principles and Commercial Applications,
R. E. Banks, B. E. Smart, and J. C. Tatlow, Plenum, New York,
1994.
N-(trans-9-Decalyl)acetamide 21. trans-Decalin 13 (3.0 g, 22
mmol), fluorine (44 mmol), boron trifluoride–diethyl ether (3.1
g, 22 mmol) and acetonitrile (120 ml) gave, after heating
(82 ЊC), a crude product mixture which was distilled to give
decalin (1.5 g, 11 mmol); bp 50 ЊC/8 mmHg. The brown solid
which remained gave, after recrystallisation, N-(trans-9-decalyl)-
acetamide 21 (1.0 g, 45%, 49% conv.) as a white solid; spectral
and physical data as above.
10 R. J. Lagow and J. L. Margrave, Prog. Inorg. Chem., 1979, 26,
161.
11 J. Hutchinson and G. Sandford, Top. Curr. Chem., 1997, 193,
1.
12 W. T. Miller and S. D. Koch, J. Am. Chem. Soc., 1957, 79,
N-(exo-2-Norbornyl)acetamide 22. Norbornane 9 (2.4 g, 25
mmol), fluorine (59 mmol), boron trifluoride–diethyl ether
(3.6 g, 25 mmol) and acetonitrile (120 ml) gave a reaction mix-
ture which was poured into water, neutralised (NaHCO3) and
filtered to remove norbornane (1.0 g, 10 mmol). The liquid
which remained was worked up as above and gave, after
recrystallisation, N-(exo-2-norbornyl)acetamide 22 (1.0 g, 45%,
60% conv.) as a white solid; 141–142 ЊC (from acetonitrile)
(Found: C, 70.2; H, 10.2; N, 9.3. C9H15NO requires: C, 70.6; H,
9.8; N, 9.1%); δH 1.08–1.55 (8 H, m, CH2), 1.80 (1 H, m, H-3),
1.94 (3 H, s, CH3), 2.19 (1 H, m, H-2), 2.27 (1 H, m, H-4), 3.71
(1 H, m), 5.35 (1 H, br s, NH); δC 23.8 (s, CH3), 26.4 (s, C-6),
27.1 (s, C-5), 35.6 (s, C-7), 35.7 (s, C-4), 40.5 (s, C-3), 42.3 (s,
C-1), 52.8 (s, C-2), 169.4 (s, CO); m/z (EIϩ) 153 (Mϩ, 75%), 138
(21, Mϩ Ϫ CH3), 124 (23), 94 (100).
3084.
13 G. A. Olah, R. Malhotra, and S. C. Narang, Nitration: Methods and
Mechanisms, VCH, New York, 1989.
14 D. P. Curran, N. A. Porter, and B. Giese, Stereochemistry of Radical
Reactions, VCH, New York, 1996.
15 G. S. Lal, G. P. Pez and R. G. Syvret, Chem. Rev., 1996, 96,
1737.
16 E. Differding and G. M. Ruegg, Tetrahedron Lett., 1991, 32,
3815.
17 S. P. Vincent, M. D. Burkhart, C. Tsai, Z. Zhang and C. Wong,
J. Org. Chem., 1999, 64, 5264.
18 R. D. Chambers, M. Parsons, G. Sandford and R. Bowden,
Chem. Commun., 2000, 959.
19 R. D. Chambers, C. J. Skinner, J. Hutchinson and J. Thomson,
J. Chem. Soc., Perkin Trans. 1, 1996, 605.
20 R. D. Chambers, J. Hutchinson, M. E. Sparrowhawk,
G. Sandford, J. Moilliet and J. Thomson, J. Fluorine Chem., 2000,
102, 169.
21 R. D. Chambers, M. P. Greenhall and J. Hutchinson, Tetrahedron,
1996, 52, 1.
22 G. Cotti, S. A. Cooke, C. M. Evans, J. H. Holloway and A. C. Legon,
Chem. Phys. Lett., 1996, 260, 388.
23 D. Albanese, D. Landini and M. Penso, J. Org. Chem., 1998, 63,
9587.
24 R. E. Banks, N. J. Lawrence, M. K. Besheesh, A. L. Popplewell and
R. G. Pritchard, Chem. Commun., 1996, 1629.
25 G. A. Olah, X. Y. Li, Q. Wang and G. K. S. Prakash, Synthesis, 1993,
693.
N-(1-Adamantyl)acetamide 17. Fluorine (28 mmol), adam-
antane 4 (1.50 g, 11 mmol), boron trifluoride–diethyl ether
(24.57 g, 173 mmol) and acetonitrile (140 ml) gave a dark yellow
solid (1.52 g) which contained N-(1-adamantyl)acetamide 17
(26%); spectral and physical data as above, and a small number
of unidentified products.
Acknowledgements
26 D. S. Ashton and J. M. Tedder, J. Chem. Soc., B, 1971, 1723.
27 Dictionary of Organic Compounds, Chapman and Hall, New York,
1982.
We thank F2 Chemicals Ltd. (studentship to MP) and the
Royal Society (University Research Fellowship to GS).
J. Chem. Soc., Perkin Trans. 1, 2002, 2190–2197
2197