β-scission of tertiary alkyl hypochlorites promoted by phase-transfer catalysis

Full text of DOI:10.1021/jo9911847

2568
J . Org. Chem. 2000, 65, 2568-2571
Notes
condition observed to be propitious for efficient oxidative
decarboxylation of the trisubstituted acetic acids.
â-Scission of Ter tia r y Alk yl Hyp och lor ites
P r om oted by P h a se-Tr a n sfer Ca ta lysis
Discu ssion
J ennifer I. Dailey, Ryan S. Hays, Hua Lee,
Randall M. Mitchell, J ennifer J . Ries, and
Robert G. Landolt*
Reactions of a series of tertiary alcohols with aqueous
hypochlorite at pH ∼9 have been investigated to establish
a scope for TBASH catalyzed â-scissions and to clarify
the fate of alkyl groups cleaved. Results for reactions of
2.5 mmol of substrates 1-7 with 70 mmol of hypochlorite
and 0.59 mmol of TBASH are summarized in Table 1.
Under these conditions, reactions proceeded to comple-
tion in less than 1 h at ambient temperature and gave
high yields of â-scission ketone products. In cases where
they conveniently could be detected by gas chromatog-
raphy, alkyl halide products were characterized as well.
“Ω-chloro” ketone products from cyclic alcohols 5-7 were
characterized by NMR and conversion to 2,4-dinitrophen-
ylhydrazone derivatives. Under comparable conditions
except at pH ∼11, several starting alcohols, notably 3
and 7, essentially were inert in biphasic reactions with
aqueous hypochlorite and TBASH.
Department of Chemistry, Texas Wesleyan University,
Fort Worth, Texas 76105
Hollie H. Husmann, J oseph B. Lockridge, and
William H. Hendrickson
Department of Chemistry, University of Dallas,
Irving, Texas 75060
Received J uly 26, 1999
Tertiary alkyl hypochlorites, YZQCOCl (Y, Z ) R or
Ar, Q ) R), are known to fragment and form ketones
(YZCdO) and alkyl chlorides (QCl) when heated or
subjected to photolysis in a “â-scission” process whose
mechanism is projected to involve the corresponding
tertiary alkoxy radicals (YZQCO•) as intermediates.1
Comparable â-scissions have been observed in several
other processes in which alkoxy radicals have been
implicated.2 Beside cleavage, alkoxy radicals, as well as
alkyl radicals produced during â-scissions, may partici-
pate in chain-propagation steps yielding a variety of
products. In addition to heat or light, â-scission has been
induced by the action of tetraacetate-metal halides3 and
by both Ce(IV)4 and Fe(II) ions.5
In the process of identifying products of oxidative
decarboxylation reactions of trisubstituted acetic acids,
YZQCCO2H, with aqueous sodium hypochlorite and a
phase-transfer catalyst, ketones, YZCdO, were isolated,
along with traces of tertiary alcohols, YZQCOH.6 Ketones
detected were those expected to result from â-scissions
of t-ROCl derived from these alcohols (i.e., benzophenone
from triphenylmethanol), but products arising from
cleaved fragments, Q, were not isolated. Furthermore,
when subjected to two-phase reactions with aqueous
hypochlorite in the presence of the phase-transfer cata-
lyst, tetrabutylammonium hydrogen sulfate (TBAHS),
several tertiary alcohols, including triphenylmethanol,
1,1-diphenylethanol, and 2-phenyl-2-propanol, yielded
â-scission product ketones. These reactions were en-
hanced by adjusting the aqueous layer to pH ∼9, a
YZQCOH + NaOCl + TBAHS f YZCdO + QCl
At lower concentrations of TBASH, â-scission reaction
times were lengthened, and evidence was found for
reaction intermediates. Iodometric titration of aliquots
taken from organic phases of such reactions revealed the
presence of substantial quantities of alkyl hypochlorites.
Moreover, in contrast to earlier studies of alkyl hypochlo-
rite reactions in which chain decomposition was initiated
thermally or with light,1 the phase-transfer catalyst
brought about â-scission of the intermediate alkyl hy-
pochlorites in two-phase systems without additional
heating or photolysis.
Figure 1 profiles the fate of substrate and product
formed during a reaction of 2-phenyl-2-propanol (1) with
minimal TBASH present. The 2-phenyl-2-propyl hy-
pochlorite intermediate (8) was formed rapidly (<5 min).
Under an air atmosphere, 8 gradually declined during
an induction period of 10-15 min and then underwent
rapid decomposition via â-scission to give acetophenone
(9). The presence of 8 was confirmed by 1H NMR analysis
of a reaction of 1 in CDCl3. The results of complementary
GC, NMR, and titrimetric analysis are summarized in
Table 2. The corresponding methyl substituents for 1, 8,
and 9 are well resolved (δ ) 1.59 for 1, δ ) 1.70 for 8,7
and δ ) 2.6 for 9). Chloromethane (δ 3.02, lit.8 δ 3.05)
(1) (a) Walling, C.; McGuinness, J . A. J . Am. Chem. Soc. 1969, 91,
2053. (b) Kim, S. S.; Kim, C. J .; Youn, S. J .; Ra, H. S.; Lee, J . C.
Tetrahedron Lett. 1991, 32, 4725 and references therein.
(2) (a) Falvey, D. E.; Khambatta, B. S.; Schuster, G. B. J . Phys.
Chem. 1990, 94, 1056. (b) Beebe, T. R.; Boyd, L.; Fonkeng, S. B.; Horn,
J .; Mooney, T. M.; Saderholm, M. J .; Skidmore, M. V. J . Org. Chem.
1995, 60, 6602.
(3) Kapustina, N. I.; Spektor, S. S.; Kikishin, G. I. Izv. Akad. Nauk
SSSR, Ser. Khim. 1983, 7, 1541; Chem. Abstr. 1984, 100, 5578).
(4) Trahanovsky, W. S.; Macaulay, D. B. J . Org. Chem. 1973, 38,
1497.
1
was also detected by H NMR.
In reactions conducted under conditions comparable to
Figure 1 but with an argon atmosphere, the inhibition
period was dramatically reduced, with over 50% ketone
production within 5-8 min. By contrast, with an atmos-
(7) Reported chemical shifts for methyl substituents of 2-propanol
and 2-propyl hypochlorite are δ 1.17 and 1.30, respectively, in CDCl3.
McGillivray, G.; ten Krooden, E. S.-Afr. Tydskr. Chem. 1992, 45, 75.
(8) Tiers, G. V. D. J . Phys. Chem. 1958, 62, 1151.
(5) Cekovic, Z.; Djokic, G. Tetrahedron 1981, 37, 4263.
(6) Elmore, P. R.; Reed, R. T.; Terkle-Husllig, T.; Welch, J . S.; Young,
S. M.; Landolt, R. G. J . Org. Chem. 1989, 54, 970.
10.1021/jo9911847 CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/30/2000

Notes
J . Org. Chem., Vol. 65, No. 8, 2000 2569
Ta ble 1. Hyp och lor ite-TBAHS In d u ced â-Scission of Ter tia r y Alcoh ols YZQC-OH + Na OCl + TBAHS f YZCdO + Q-Cl
substrate
Y
Z
Q
productsa (%)
detected substrate
pH range
1
Ph-
Me-
Me-
acetophenone (71-80)
trace-<10%
8.6-9.4
chloromethaneb
2
3
4
Ph-
Me-
Ph-
Et-
Et-
propiophenone (82-93)
trace
trace
trace
9.0-9.3
8.9-9.5d
8.7-9.3
chloroethanec
Me-
Me-
PhCH2-
PhCH2-
acetonec
benzyl chloride (65-73)e
acetophenone (85-98)
benzyl chloride (88-98)
7-chloro-2-heptanone (65-73) [>90]f
5-chloro-1-phenyl-1-pentanone [70-90]f
6-chloro-1-phenyl-1-hexanone
(40-65; 80-90 in 60 min) [>90]f
5
6
7
Me-
Ph-
Ph-
-CH2- - -
-CH2- - -
-CH2- - -
-(CH2)4-
-(CH2)3-
-(CH2)4-
trace
trace
9.1-10.3
8.6-9.6
8.6-9.2
15-50% (trace-3%,
60 min)
a
d
Results after 30 min unless otherwise specified. b Detected by NMR. c Undetectable under conditions employed. The pH, which tended
to rise, was lowered with concentrated HCl to 9.0 during reaction. e Benzal chloride detected also, <1%, 2% after 1 h. f Crude product, see
the Experimental Section.
F igu r e 1. Reaction of 25 mL of 0.10 M1 in CH2Cl2 with 100
mL of bleach at pH 9 catalyzed by 0.125 mmol of TBAHS: 4
F igu r e 2. Addition of 0.125 mmol TBASH to a reaction
) 1, b ) 8, [ ) 9, g ) 1 + 8 + 9.
mixture of 25 mL of 0.10 M1 in CH2Cl2 with 100 mL of bleach
at pH 9 after 345 min: 4 ) 1, b ) 8, [ ) 9, g ) 1 + 8 + 9.
Ta ble 2. TBASH-Ca ta lyzed Rea ction of 0.099 M
2-P h en yl-2-P r op a n ol w ith Blea ch in DCCl3
When treated with aqueous bleach in the presence of
TBAHS, 8 was converted to acetophenone in 100% yield.
time
[1]
[1]
[8]
[8]
[9]
NMR
[9]
GC
total
NMR
(min) NMR GCa titrat.b NMR
The phase-transfer catalyst was shown to bring about
decomposition of the intermediate alkyl hypochlorite
efficiently under biphasic conditions with aqueous hy-
pochlorite. In the absence of catalyst, the alkyl hypochlo-
rite was formed more slowly and was stable under the
reaction conditions. After being stirred in dichloromethane
with bleach at pH 9 for 24 h with no catalyst present,
84% of 1 was converted to 8 (as shown by titrimetry),
but less than 0.3% of â-scission to acetophenone occurred.
Furthermore, as shown in Figure 2, the addition of
catalyst to a reaction mixture (after most of the alcohol
had been converted to alkyl hypochlorite) resulted in
rapid decomposition of 8 and formation of acetophenone.
0
0.099 0.101
0
0
0
0
0
0
0.099
60 0.034 0.031 0.065 0.064
180 0.029 0.035 0.064 0.064
600 0.018 0.015 0.057 0.059
1260 0.010 0.004 0.033 0.034
0.001 0.098
0.004 0.093
0.026 0.028 0.103
0.055c 0.051 0.099
a
After KI treatment, less titrated value of 8 (see the Experi-
b
mental Section). No blank correction; maximum concentration
of ROCl. c 0.032 M chloromethane also detected.
phere of O2, nearly 100% ROCl was unreacted after 73-
88 min and <5% ketone formed after 2.5 h.
The alkyl hypochlorite 8 and comparable intermediates
from other alcohol precursors were found to undergo
â-scission in a chromatograph injection port. To distin-
guish heat-promoted â-scission from that caused by
TBASH, samples were treated with acidic potassium
iodide, to reduce the hypochlorites to alcohols9 before GC
analysis. The results presented in Table 2 show excellent
agreement among the three methods of analysis of
reaction mixtures. To provide additional evidence for the
alkyl hypochlorite as an intermediate in these reactions,
authentic 8 was independently prepared directly from 1.
Both aqueous bleach and TBAHS appeared necessary
for the â-scission reaction of 8. When the latter was
dissolved in dichloromethane and mixed with TBAHS
with no bleach present, no decomposition occurred after
3 h. Consequently, it is possible that the phase-transfer
agent promoted formation of intermediates either that
facilitate formation of the alkyl hypochlorite or that
stimulate â-scission of the latter or both. Chlorine
monoxide (Cl2O) is illustrative of such an intermediate.
An overall reaction mechanism for phase-transfer catalyst-
(9) Fort, R.; Denivelle, L. Bull. Soc. Chim. Fr. 1954, 1109.

2570 J . Org. Chem., Vol. 65, No. 8, 2000
Notes
Sch em e 1
For either GC or NMR analyses, the difference in ROH
detected in reaction aliqouts before and after aqueous KI
treatment corresponds nicely with the levels of ROCl
observed directly by titrimetry and NMR.
Exp er im en ta l Section
Ma ter ia ls a n d Gen er a l P r oced u r es. All compounds em-
ployed were secured from commercial suppliers. Aqueous hy-
pochlorite (“5.25%”) was obtained in the form of commercial
Clorox bleach. In all cases, the phase-transfer catalyst (PTC)
used was tetrabutylammonium hydrogen sulfate (TBAHS). All
1H NMR spectra were recorded at 60 MHz. Gas chromatographic
(GC) instruments were equipped with capillary columns and
flame ionization detectors. The pH’s of aqueous layers of biphasic
systems were set and maintained at desired levels by addition
of aqueous NaOH or HCl, and the pH was monitored with pH
meters equipped with gel-filled plastic combination electrodes.
Progress of reactions was followed by GC using chlorobenzene
as an internal standard, which was shown to be stable to the
reaction conditions and to have retention times different from
both reactants and products. All reactions were conducted at
ambient temperatures.
Gen er a l Rea ction s of Ter tia r y Alcoh ols (Ta ble 1). Sub-
strate alcohols and chlorobenzene (2.5 mmol) were dissolved in
50 mL of CH2Cl2 and stirred magnetically with 100 mL (70
mmol) of hypochlorite adjusted to pH 8.8-9.3 and containing
0.20 g (0.59 mmol) of TBAHS. Reactions were conducted under
conventional fluorescent lighting and analyzed by GC to deter-
mine the types and percentages of products formed. Potassium
iodide (KI) workups were accomplished by treating a 0.5 mL
reaction aliquots with two drops of 1 M KI in 0.5 M HCl. Under
the comparable conditions, except with the aqueous phase pH
∼11 (maintained by addition of 5% aqueous NaOH), compounds
3 and 7 were found to be unreactive to bleach and TBAHS for 3
h and up to 3 days, respectively.
Ketone products from pH 9 reactions of substrates 5-7 were
isolated from two-phase systems that did not contain an internal
standard. After reactions had run for a minimum of 1 h, the
CH2Cl2 phase was extracted six times with water, poured
through fluted filter paper into a round-bottom flask, subjected
to rotary evaporation, and weighed. For 7-chloro-2-heptanone:
1H NMR (CCl4) δ 1.5 (br m 6H), δ 2.1 (s, 3H), δ 2.4 (t, 2H), δ 3.5
(t, 2H); 2,4-dinitrophenylhydrazone (DNP), mp 94-95 °C (lit.15
mp 95-96 °C). For 5-chloro-1-phenyl-1-pentanone: 1H NMR
(DCCl3) δ 1.6 (m, ∼4H), δ 2.7 (m, ∼2H), δ 3.4 (m, ∼2), δ 7.7 (m,
∼5H); DNP mp 176-178 °C, (lit.15 mp 76-178 °C). For 6-chloro-
1-phenyl-1-hexanone: 1H NMR (DCCl3) δ 1.6 (br m, 6H), δ 2.9
(t, 2H, J ≈ 6 Hz), δ 3.5 (t, 2H, J ≈ 6 Hz), δ 7.7 (m, 5H); DNP mp
139.5-141 °C (lit.15 mp 141.5-143 °C).
Rea ction s of 2-P h en yl-2-P r op a n ol(1). (a ) Rea ction in
CDCl3. An a lysis by 1H NMR (Ta ble 2). Compound 1 (0.134
g, 0.984 mmol), TBAHS (0.0191 g, 0.056 mmol), chlorobenzene
(0.0552, 0.590 mmol), and tert-butylbenzene (0.0645 g, 0.481
mmol) were dissolved in 10 mL of CDCl3. The solution was
stirred with 40 mL (28 mmol) of bleach at pH 9. At intervals,
aliquots of the organic layer were removed and analyzed by GC
and NMR. A 0.2 mL aliquot was added to 15 mL of reagent-
grade 2-propanol containing 2 mL of acetic acid and 0.25 g of
NaI. The iodine was titrated to a colorless endpoint with 0.01
M sodium thiosulfate. The concentrations of the following
compounds were determined from NMR by integration of the
appropriate methyl signal relative to that of tert-butyl benzene
(δ ) 1.33): 2-phenyl-2-propanol (δ ) 1.60), 2-phenyl-2-propyl
hypochlorite (δ ) 1.70), acetophenone (δ ) 2.60), and chlo-
romethane (δ ) 3.02). When a reaction aliquot was treated with
aqueous KI, the signal at δ ) 1.70 disappeared, and the signal
at δ ) 1.60 increased by a corresponding amount.
induced reactions of hypochlorite with tertiary alcohols
is suggested in Scheme 1, with Cl2O potentially playing
the role indicated.10
The stability of YZQCOCl prior to addition of the
catalyst is not unexpected. Although it is generally
accepted that alkyl hypochlorites undergo a rapid light-
catalyzed chain decomposition, Walling and J acknow
have shown that the chain reaction is inhibited by
oxygen.11 Evidently, in a system that has a relatively low
alkyl hypochlorite concentration and is not deoxygenated,
chain reactions may not occur unless the initiation rate
is substantially increased through the addition of the
phase-transfer catalyst.
The quaternary salt may function in the traditional
phase-transfer catalytic role, increasing concentration of
ionic species, particularly including hypochlorite ion
(OCl-), in the organic phase to serve as a participant for
several steps. Rationales for increased efficiency at pH
8-10 vs pH > 11 include the following: enhanced
formation of Cl2O, suggested as the active agent in other
phase-transfer reactions of aqueous hypochlorite;12,13
enhanced efficiency of coextraction of hypochlorous acid
along with hypochlorite by catalyst at pH near the pKa
(7.5-7.6) of HOCl;14 or by influencing efficiency of
competing reactions,12 including the chain length of
radical propagating processes.
It is very important that instability of alkyl hypochlo-
rites in nonaqueous solvents during GC analysis be
recognized. Heat-induced â-scission of alkyl hypochlorites
takes place readily in the inlet ports of gas chromato-
graphs (typically > 100 °C). This is made apparent if
duplicate aliquots of reaction mixtures are analyzed by
GC after one is treated with KI. Alkyl hypochlorites
present have been shown to react with KI and revert (be
reduced) to the corresponding alcohol. This has permitted
another way to estimates levels of alkyl hypochlorites.
(10) Chlorine monoxide is known to react with alcohols to form
ROCl; see: Anbar, M.; Ginsburg, D. Chem. Rev. 1954, 54, 925 and
references therein. Other radical species, such as chlorine atoms, also
may participate in chain propagation steps analogous to those shown
in Scheme 1.
(11) Walling, C.; J acknow, B. B. J . Am. Chem. Soc. 1960, 82, 6108.
(12) Fonouni, H. E.; Krishnan, S.; Kuhn, D. G.; Hamilton, G. A. J .
Am. Chem. Soc. 1983, 105, 7672.
(13) Dneprovskii, A. S.; Eliseenkov, E. V. Russ. J . Org. Chem. 1994,
30, 235.
(14) Abramovici, A.; Neuman, R.; Sasson, Y. J . Mol. Catal. 1985,
29, 291. The pKa of HOCl reported therein is 7.53; elsewhere it is cited
as 7.58.
After KI treatment, levels of 1 directly measured by GC
included, in addition to unreacted starting material, some alcohol
due to reduction of 8 by KI. Direct GC analyses of aliquots (no
KI treatment) resulted in significant thermal decomposition of
8. For Table 2 and Figures 1 and 2, concentrations of unreacted
starting material were determined by subtracting the titrimetri-
(15) Wilt, J . W.; Hill, J . W. J . Org. Chem. 1961, 26, 3523.

Notes
J . Org. Chem., Vol. 65, No. 8, 2000 2571
cally determined concentrations of 8 from alcohol levels in KI-
treated aliquots.
pH 7.5 was added with stirring. After the solution was stirred
for 25 min in the ice bath in the dark, the layers were separated,
and the organic layer was washed two times with 10% aqueous
sodium carbonate and dried over anhydrous sodium sulfate.
Iodometric titration indicated an alkyl hypochlorite concentra-
tion of 0.20 M. An aliquot of the alkyl hypochlorite solution and
chlorobenzene was diluted to 25 mL. This solution was treated
with 100 mL of bleach at pH 9 containing 0.140 mmol of TBASH.
The reaction progress was followed by titrimetry and GC, and
9 was formed in proportion to the loss of 8.
(b) Rea ction s in CH2Cl2 (F igu r es 1 a n d 2). Except as noted
below, illumination was minimized by operating either in a
darkened room or in a hood with covered window. The general
procedure for tertiary alcohols was followed using 2.5 mmol of
1 in 25 mL of CH2Cl2 and 0.125 mmol of TBASH. The catalyst
was added either at the beginning of the reaction (Figure 1) or
after 345 min (Figure 2). Aliquots were taken periodically, and
reactions were subjected to fluorescent room lighting for short
sampling periods. The concentration of 8 was determined by
iodometric titration, and a blank correction equivalent to 0.005
M hypochlorite, determined by titration of an aliquot taken from
a reaction without alcohol, was used. The concentrations of 1
and 9 after KI treatment were measured by GC. The concentra-
tion of 1 present prior to KI treatment were calculated as
described in (a) above.
For 1H NMR analysis, 8 was prepared as above from 15 mL
of a 0.40 M solution of 1 in CCl4 treated with 30 mL of bleach at
pH 7.5. Iodometric titration of this solution gave a hypochlorite
concentration of 0.36 M. Analysis by 1H NMR (CCl4) gave δ 1.53
(s, 0.5 H), δ 1.63 (s, 5.3 H), δ 7.37 (s, 5 H). Calculations using
the relative areas of the two methyl signals indicated that the
sample was 91% alkyl hypochlorite and 9% alcohol.
Conditions equivalent to Figure 1, except for exposure to
intense incandescent lighting, gave essentially the same results
as displayed in Figure 1. Experiments conducted comparable to
Figure 1, except for use of an argon atmosphere resulted in rapid
formation of ketone product; use of an oxygen atmosphere almost
eliminated ketone formation for a period of hours (see Discus-
sion).
(c) Rea ction in CH2Cl2 w ith ou t TBAHS. A 25 mL solution
of 0.103 M 1 was stirred with 100 mL of bleach at pH 9.
Iodometric titration showed 81% of 1 converted to 8 after 6 h
and 90% after 24 h. GC analysis after 24 h indicated only a 0.9%
conversion of 1 to 9.
Ack n ow led gm en t. All phases of this work were
supported by Chemistry Departmental Research Grants
from the Robert A. Welch Foundation (AZ0003, Texas
Wesleyan, and BA0015, University of Dallas) and by the
O’Hara Chemical Sciences Institute of the University
of Dallas. The effort was made possible in part by
instrumentation supplied by the National Science Foun-
dation under Instrumentation and Laboratory Improve-
ment Grants to Texas Wesleyan (USE-9250391 and
DUE 9650488). We also thank Marian Gooding and
Basel Zaitoun for demonstrating the reproducibility of
certain of the experimental results.
1-Meth yl-1-p h en eth yl Hyp och lor ite (8). The method of
Walling and McGuinness1a was modified. A solution of 2-phenyl-
2-propanol (409 mg, 3.00 mmol) in CH2Cl2 (15 mL) was cooled
in an ice bath, and aqueous hypochlorite (15 mL, 10 mmol) at
J O9911847