A New Preparative Route to Organic Halides from Alcohols via the Reduction of Polyhalomethanes

Full text of DOI:10.1021/jo970767i

J . Org. Chem. 1997, 62, 7061-7064
7061
was converted into benzyl bromide nearly quantitatively
(1.5 h at room temperature). With BrCCl3 a 4:1 mixture
of benzyl chloride and bromide was obtained in 80%
overall yield (1.5 h at room temperature). Apart from
the efficiency of the process, two other features are worth
mentioning, i.e. the acidic medium obtained at the end
of the reaction and the evolution of carbon monoxide.
Other solvents were tested using benzyl alcohol and
CBr4 at room temperature. As summarized in eq 2, the
preparation of benzyl bromide can be carried out in N,N-
dimethylacetamide (DMAC) or in acetonitrile (CH3CN),
though with a slower rate and a slightly lower efficiency
than in DMF.
A New P r ep a r a tive Rou te to Or ga n ic
Ha lid es fr om Alcoh ols via th e Red u ction of
P olyh a lom eth a n es
E. Le´onel,* J . P. Paugam, and J . Y. Ne´de´lec
Laboratoire d’Electrochimie, Catalyse et Synthe`se
Organique, UMR CNRS no. 28, 2, Rue Henri-Dunant,
B.P. 28, F-94320 Thiais, France
Received April 30, 1997
Many methods of preparation of alkyl halides from
alcohols are known. The most commonly used prepara-
tive routes are the reactions of alcohols either with
hydracids or with phosphorus halides.1a-c We have found
a
convenient alternative which involves polyhalo-
methanes as the source of halogen in the presence of a
redox system. This process was first observed during the
electroreduction of CCl4 and BrCCl3 in N,N-dimethylfor-
mamide (DMF) in the presence of various functional
groups including aldehydes and alcohols. The electroly-
ses were conducted in an undivided cell and in the
presence of a stainless-steel cathode and a sacrificial
anode. The conversion of alcohols into the corresponding
halides occurred mainly with copper as the sacrificial
anode, and the reaction was found to be even more
efficient with some 1,10-phenanthroline (phen) added to
the reaction mixture. We also noticed an overconsump-
tion of the anode (based on the charge passed) along with
an unexpected consumption of the cathode. We have
found that the reaction does not require the supply of
electricity and can occur in a purely chemical mode.
Indeed the reaction can be conducted in a round-bottom
flask using either piece, or a powder of copper and iron
and small amounts of CuI(phen)2.
We report in this paper the details of a simple and
efficient method to prepare primary alkyl halides from
primary alcohols and â-halo-R,â-unsaturated ketones
from â-diketones. We also discuss two possible reaction
mechanisms, which can be discriminated according to the
nature of the solvent. The conversion of aldehydes into
gem-dihalides is currently under study, and the results
will be reported shortly.
We also used tetrahydrofuran (THF) as solvent with
1-decanol and obtained products derived from the open-
ing of THF (eq 3).
To test the generality of the procedure to prepare alkyl
halides, we carried out the reaction using several alco-
hols. The results are given in Table 1.
It can be seen that the procedure is applicable to
primary alcohols, whereas the main or only products
formed from secondary alcohols (Table 1, entries 7 and
8) or cyclohexanol (eq 4) were the corresponding alkyl
formates.
Resu lts
The best reaction conditions were determined using
benzyl alcohol or 1-decanol as model systems. Thus the
preparation of benzyl chloride was conducted in DMF
using CCl4 as halogenating agent (eq 1). Under the
Thus in a mixed primary and secondary diol, only the
primary group is converted into the halide while the
secondary group is protected in the form of the corre-
sponding formate (Table 1, entries 13 and 14) as already
observed by Boeckman et al.2 A higher yield of the halo
formate was obtained at higher reaction temperature (e.g.
100 °C, Table 1, entry 14).
The sole secondary alcohol which gave the correspond-
ing chloride (30%) with CCl4 or bromide (85%) with CBr4
is ethyl lactate (Table 1, entries 9 and 10). We expected
a stereospecific reaction with the commercially available
(S)-ethyl lactate, but we only obtained the racemic
bromide or chloride, in keeping with reaction mechanisms
discussed below.
standard reaction conditions (for 20 mmol of PhCH2OH:
CuBr (0.5 mmol), 1,10-phenanthroline (1 mmol), copper
powder (30 mmol), iron powder (50 mmol), 2 equiv of CCl4
(40 mmol), 3 h at room temperature), benzyl chloride was
produced in 77% isolated yield. With CBr4, benzyl alcohol
(1) (a) For a list of reagents with references, see: Larock, R. C.
Comprehensive Organic Transformations, VCH: New York, 1989, p
353. (b) Wiley, G. A.; Hershkowitz, R. L.; Rein, B. M.; Chung, B. C. J .
Am. Chem. Soc. 1964, 86, 964. (c) Appel, R. Angew. Chem., Int. Ed.
Engl. 1975, 14, 801.
Attempts to convert allylic alcohols into allylic halides
were unsuccessful. The main product obtained from
(2) Boeckman J r., R. K.; Ganem, B. Tetrahedron Lett. 1974, 11, 913.
S0022-3263(97)00767-6 CCC: $14.00 © 1997 American Chemical Society

7062 J . Org. Chem., Vol. 62, No. 20, 1997
Notes
Ta ble 1. Con ver sion of Alk a n ols in to Alk yl Ha lid esa
Ta ble 2. P r ep a r a tion of â-Ha lo-r,â-u n sa tu r a ted Keton esa
Sch em e 1
cinnamyl alcohol was the dimer, likely formed via the
intermediate cinnamyl halide (eq 5).
have already been used for the halogenation of alcohols.
The reactions likely proceed according to a mechanism
proposed by Tabushi et al.4 In the presence of a base,
the deprotonation of chloro- or bromoform generates
dihalocarbenes which in the presence of an olefin lead
to dihalocyclopropanes. Tabushi4 reported in 1971 that
dichlorocarbene reacts readily with alcohols to give the
corresponding chlorides according to the mechanism
depicted in Scheme 1: a formal insertion of CCl2 into the
OH bond of ROH leads to the R,R-dichloro ether which,
when R is a primary alkyl group, decomposes either
concertedly or via the formation of a carbocation into the
corresponding alkyl chloride RCl with the evolution of
CO (path a). The formation of alkyl formate from
secondary alcohols, though not explained by Tabushi,4
can also be accounted for by the involvement of a carbene
intermediate (Scheme 1, path b). The formation of benzyl
chloride by electroreduction of CCl4 in the presence of
benzyl alcohol (Petrosyan’s work3a,b) or by chemical
reduction (our copper/iron system) can also be explained
by a carbene route.
This reaction scheme does not however take into
account the possible role of the solvent. We have
observed that the reaction is more rapid and more
efficient in DMF than in DMAC or acetonitrile,5 that the
evolution of CO is more facile in DMF than in the other
solvents, and finally that the halogenation of dimedone
was unsuccessful in all solvents used except DMF. We
postulated a possible reaction between dihalocarbenes
and DMF, and indeed found that the production of CO
On the contrary, an easily isolable dihalide was
obtained in good yield from 2-butyne-1,4-diol (Table 1,
entries 11 and 12).
The procedure can also be used to convert â-diketones
into â-halo-R,â-unsaturated ketones. The results are
summarized in Table 2.
2,4-Pentanedione gave a mixture of E and Z stereo-
isomeric â-halo enones (in 3:1 E:Z ratio for â-chloro
enone, Table 2, entry 1; and 4:1 E:Z ratio for â-bromo
enone, Table 2, entry 2). With dimedone the â-halo enone
was formed in quite good yields (Table 2, entries 3 and
4). For these two reagents the reaction gave higher yields
with CBr4 than with CCl4. Attempts to convert ethyl
acetoacetate under our standard reaction conditions
failed. The halogenation of dimedone conducted in
acetonitrile or in DMAC was also unsuccessful.
Discu ssion
Polyhalomethanes, either haloforms in the presence of
a strong base or CCl4 in an electroreductive process,3a,b
(3) (a) Petrosyan, V. A.; Niyazymbetov, M. E.; Baryshnikova, T. K.
Izv. Akad. Nauk SSSR 1988, 1, 91. (b) Petrosyan, V. A.; Niyazymbetov,
M. E. Russ. Chem. Rev. 1989, 58, 644.
(4) Tabushi, I.; Yoshida, Z. I.; Takahashi, N. J . Am. Chem. Soc. 1971,
93, 1820.
(5) The solvent effect was tested on benzyl alcohol.

Notes
J . Org. Chem., Vol. 62, No. 20, 1997 7063
also occurs when CX4 is reduced chemically in DMF in
the absence of alcohol. CO was first characterized by
infrared spectroscopy. We also measured the amount of
CO evolved using an electroanalytical method as fol-
lows: CO was transferred from the reaction flask to an
electrochemical cell containing a given amount of an
electrochemically generated nickel(0)-bipyridine complex
which reacts rapidly with CO to give a well defined
Ni0(CO)2bpy complex, the formation of which is easily
monitored using standardized chronoamperometry.6 The
volume of CO was also measured using a measuring test
tube connected to the reaction flask.6 In acetonitrile, or
in DMAC, CO was only detected when the alcohol was
present.
authors have also shown that the intermediate 1 might
racemize at the R group by interaction with DMF.
Evidence consistent with the formation of the Vilsmeier
reagent was obtained by performing the formylation of
2-methoxynaphthalene in 50% isolated yield by replacing
ROH by the aromatic compound (eq 7).
Con clu sion
A similar process has already been reported by Burton
and Wiemers7 for the reduction of CF2Br2, CF2Cl2, and
CF2BrCl by Zn in DMF. The reaction of the difluorocar-
bene with DMF leads to CO and (difluoromethyl)dim-
ethylamine in equilibrium with a dimethyliminium fluo-
ride (eq 6).
It is now clear that there are at least two mechanisms
leading to the formation of alkyl halides from alcohols
or of alkenyl halides from â-dicarbonyl compounds via
the reduction of CX4 according to the nature of the
solvent. A typical carbene route (Scheme 1) would be
operative in DMAC or acetonitrile but not efficient with
â-dicarbonyl compounds, whereas in DMF the dihalocar-
bene would be rapidly trapped by the solvent leading to
the Vilsmeier reagent which can react with alcohols as
well as â-dicarbonyl compounds to give the corresponding
halides (Scheme 2). It is finally worth noting that the
reduction of CX4 in DMF is a source of CO. Thus caution
is required when one has to use a solution of CCl4, CBr4,
or BrCCl3 in DMF in the presence of reducing agents.
Exp er im en ta l Section
1H and 13C NMR spectra were recorded on a Brucker AC-200
(200 MHz) spectrometer. Mass spectra (electron impact) were
obtained on a Finnigan ITD 800 spectrometer coupled to a
Varian 3300 chromatograph with a CP Sil-5 capillary column.
Gas chromatography was performed on a Varian 3300 chro-
matograph equipped with a sil-5 CP capillary column. The
solvents were used as received.
This iminium salt is a typical Vilsmeier reagent which
has already been prepared for X ) Cl or Br from PCl5 or
(COX)2 and DMF and used to convert primary alcohols
into the corresponding halides,8a,b â-dicarbonyl com-
pounds into â-halo enones,9 whereas secondary alkanols
or cycloalkanols are converted into alkyl formates2 ac-
cording to Scheme 2.
P r ep a r a tion of Alk yl Ha lid es. Gen er a l P r oced u r e.
A
solution of CX4 (CCl4 or CBr4) diluted in 20 mL of DMF was
added to a well-stirred solution of DMF (50 mL) containing CuBr
(0.5 mmol), 1,10-phenanthroline monohydrate (1 mmol), copper
powder (30 mmol), iron powder (50 mmol), and the alcohol or
the dicarbonyl compound (20 mmol) under argon. CAUTION!!
The reaction is exothermic and leads to the formation of CO.
The temperature was controlled by an ice/water bath. The
reaction was monitored by GC and stopped after complete
consumption of the alcohol (0.5-3 h). The reaction mixture was
then hydrolyzed with hydrochloric acid (1 N). The aqueous layer
was extracted with diethyl ether (3 portions of 30 mL). The
combined organic layers were washed with water, dried over
MgSO4, and concentrated. Products were isolated by column
chromatography (silica gel, 70-230 mesh; eluent pentane). All
products are known and gave satisfactory NMR, IR, and mass
spectral data.
Sch em e 2
An a lytica l Exp er im en ts. Electroanalytical experiments
were carried out using a three-electrode electrochemical cell
under argon. The DMF solution contained 0.1 mol dm-3 of
Bu4NBF4 and 0.1 mol dm-3 of a nickel divalent complex made
of either NiCl2bpy with 2 bipyridine or Ni(BF4)2bpy3. This
solution was electrolyzed at constant current of 400 mA to
convert the NiII complex into a Ni0 complex. In a separate flask
connected to the cell, a solution of CX4 (5 mmol) in DMF (2 mL)
was added to the DMF solution (10 mL) containing CuBr (0.07
mmol), 1,10-phenanthroline monohydrate (0.14 mmol), copper
powder (4 mmol), and iron powder (7 mmol). The flask was
flushed with argon in order to transfer CO into the electro-
chemical cell. The reaction with CO is very fast and quantitative
(eq 8).10
Thus a common intermediate (1) would lead to either
the alkyl halide or the alkyl formate. We have indeed
observed that even with primary alcohols the formate is
the major product if the reaction mixture is hydrolyzed
after less than 0.5 h, whereas only the alkyl halide is
isolated after a longer reaction time. Hudson8a reported
that the alkoxydimethyliminium intermediate (1) is
rapidly formed and can be isolated when R ) phenyl. The
In the presence of excess bipyridine the mixture of Ni0(CO)2bpy
and Ni0bpy2 is stable. The oxidation of Ni0(CO)2bpy occurs at
(6) For more details, see the Experimental Section.
(7) Burton, D. J .; Wiemers, D. M. J . Am. Chem. Soc. 1985, 107, 5014.
(8) (a) Hepburn, D. R.; Hudson, H. R. J . Chem. Soc., Perkin Trans
1, 1976, 754. (b) Fujisawa, T.; Iida, S.; Sato, T. Chem. Lett. 1984, 1173.
(9) Mewshaw, R. E. Tetrahedron Lett. 1989, 30, 3753.
(10) Oc¸afrain, M.; Devaud, M.; Troupel, M.; Pe´richon, J .; Ne´de´lec,
J . Y., unpublished work.

7064 J . Org. Chem., Vol. 62, No. 20, 1997
Nibpy2 + 2CO f Ni(CO)2bpy + bpy
Notes
formate [5451-52-5]; 10, sec-octyl formate [112360-52-8]; 11,
2-bromooctane, [557-35-7], c; 13, ethyl 2-chloropropanoate
[535-13-7], c; 14, ethyl 2-bromopropanoate [535-11-5], c; 16,
1,4-dichlorobut-2-yne [821-10-3]; 17, 1,4-dibromobut-2-yne
[2219-66-1]; 19, 1-chloro-2-octyl formate, n; 20, 1,2-octyl di-
formate, (Chem. Abstr. 1952, 46, 8146c P); 21, 1-bromo-2-octyl
formate [53067-12-2]; 23, cyclohexyl formate [4351-54-6]; 27,
(Z)-4-chloropent-3-en-2-one [49784-64-7], (E) [49784-51-2];
28, (Z)-4-bromopent-3-en-2-one [80356-21-4], (E) [80356-20-
3]; 30, 3-chloro-5,5-dimethylcyclohex-2-en-1-one [17530-69-7];
31, 3-bromo-5,5-dimethylcyclohex-2-en-1-one [13271-49-3]; 34,
2-methoxy-1-naphthalenecarbaldehyde [5392-12-1].
(8)
-0.2 V/SCE. So, we can measure the disappearance of Ni0bpy2
by chronoamperometry at constant potential of -1.2 V/SCE.
We also determined the volume of CO evolved using a
measuring test tube. Starting from 5.07 × 10-3 mol (0.5 mL) of
BrCCl3 in DMF (12 mL) containing CuBr (0.07 mmol), 1,10-
phenanthroline monohydrate (0.14 mmol), copper powder (4
mmol), and iron powder (7 mmol), we obtained 116 mL of CO at
20 °C, which corresponds to 95% yield based on BrCCl3.
Nu m ber , n a m e, a n d r egistr y n u m ber p r ovid ed by th e
a u th or s (c, commercial; n, new compound): 2, benzyl chloride
[100-44-7], c; 3, benzyl bromide [100-39-0], c; 6, 1-chlorode-
cane [1002-69-3], c; 7, 1-bromodecane [112-29-8], c; 8, decyl
J O970767I