10822 J. Am. Chem. Soc., Vol. 118, No. 44, 1996
Bordwell and Liu
Table 5. Equilibrium Acidities of Hydroxylic Acids Determined
are within (2 kcal of the literature values. We have no
explanation for the 6.4-kcal difference in BDE for Et2NOH,
but our value of 75.9 kcal is consistent with the BDEs of seven
other R2NOH hydroxylamines, estimated by eq 1, which fall in
the range of 74-77 kcal.19 It is clear from these comparisons
that estimates of BDEs of the O-H bonds in most hydroxylic
acids made by eq 1 agree remarkably well with literature values,
and that there is no reason to believe that this method leads to
“serious errors” as has been suggested.7 Our value for the O-H
bond in phenol of 90.4 kcal is, however, 2.1 kcal higher than
the value obtained by Mahoney et al. in chlorobenzene,5 and
3.4 kcal higher than the 87 kcal estimated to be the best gas
phase value.7 Furthermore, persual of the data in Table 4 shows
that BDE values estimated by eq 1 are usually slightly higher
than the literature values suggesting that the BDEs may be
subject to small solvent effects in DMSO.
by the Overlapping Indicator Methoda
c
d
acid
Et2NOH
t-Bu2NOH
TMPOH
NOH
indicator (pKIn)b
pKHA
pKhb
DDH (29.4)
TH (30.6)
TH (30.6)
TH (30.6)
29.60 ( 0.05 3.3 ( 0.1
31.15 ( 0.05 e,f
31.00 ( 0.02 e,f
32.4 ( 0.1
e,g
PhCON(i-Pr)OH
PhCON(t-Bu)OH 2-NPANH (20.66) 19.6 ( 0.1
t-Bu(i-Pr)C NOH TP2H (25.6) 25.5 ( 0.1
CNAH (18.9)
18.70 ( 0.02 3.7 ( 0.1
3.6 ( 0.1
3.25 ( 0.03
a See ref 14. b Indicators: TH ) triphenylmethane. DDH ) bis(p-
phenylphenyl)methane. MFH ) 9-methylfluorene. CNAH ) 4-chloro-
2-nitroaniline. 2-NPANH ) 2-naphthylacetonitrile. TP2H ) 1,3,3-
triphenylpropene. The pKHIn values are given in parentheses. c pKHA
corrected for homo-H-bonding.14 d Homo-H-bonding constant.14 e The
pKHA is too high to obtain pKhb values. f The homo-H-bonding effect
was observed to be small (pKhb < 1). g The homo-H-bonding effect
was observed to be large (pKhb ∼ 3-4).
The relationship between the BDE in DMSO solution, BDE-
(HA)s, and the corresponding gas-phase value, BDE(HA)g, can
be expressed by eq 3:
BDE(HA)s ) BDE(HA)g + ∆Hsolv(H•)s +
∆Hsolv(A•)s - ∆Hsolv(HA)s (3)
N-Hydroxy-2,2,6,6-tetramethylpiperidine was prepared by the
method of Ingold et al.9 N-Oxyl-2,2,6,6-tetramethylpiperidine (0.78
g, 5 mmol) and 1,2-diphenylhydrazine (1.0 g, 5.8 mmol) were mixed
in 5 mL of benzene under a nitrogen atomosphere. The mixture was
stirred at room temperature for 20 min, then fractionally distilled to
give 0.5 g (63%) of a pale yellow liquid (bp 60 °C/10τ) which solidified
upon standing, mp 37-39 °C (lit.22 mp 38-40 °C). 1H NMR
(CDCl3): δ 1.22 (12H, s, Me), 1.56 (6H, s, CH2), 4.03 (1H, br, OH).
N,N-Di-tert-butylhydroxylamine was similarly prepared from di-
tert-butyl nitroxide in 70% yield, bp 50 °C/10τ, mp 37-38 °C.23 1H
NMR (C6D6): δ 1.22 (18H, s, t-Bu), 3.97 (1H, br, OH).
tert-Butyl isopropyl ketoxime was similarly prepared by a method
reported earlier10 from tert-butyl isopropyl ketone imine (which was
obtained from a reaction of tert-butyllithium and isobutyronitrile) and
hydroxylamine hydrochloride in 80% yield, mp 140-141 °C (lit.24 mp
140-141 °C). 1H NMR: δ 1.10 (6H, d, J ) 7.1 Hz, i-Pr), 1.30 (9H,
s, t-Bu), 2.59 (1H, m, CH), 8.83 (1H, s, NH).
The BDEs of the C-H bond in carbon acids are generally
assumed to be solvent independent because the last two terms
in eq 3 cancel one another.20 It is also generally agreed that
the remaining ∆Hsolv(H•)s term is small. Its value is unknown,
and it is often neglected. This leaves BDE(HA)s ) BDE(HA)g
for carbon acids, as has been observed experimentally in
DMSO,2 and also in aqueous medium.20
The hydrogen bond donor properties of the hydroxyl group
in hydroxylic acids are likely to cause stronger bonding with
the hydrogen-bond-acceptor DMSO solvent than occurs with
the A• radical. The BDE of the hydroxyl group will then be
strengthened thereby, but the effects are smaller than estimated
for the hydrogen bond energies (3-4 kcal for alcohols)21
suggesting that solvation of the A• radicals may also be involved
to some extent. The hydrogen bond energy of phenol in DMSO
has been estimated to be 6-7 kcal,21 which may account for
the greater difference in our BDE value in DMSO versus
literature values than is observed for other hydroxylic acids.
We conclude that solvent corrections of about 2-3 kcal are
needed for the BDEs of the O-H bonds in phenol and its simple
derivatives as estimated by eq 1 in DMSO, and that smaller (1
to 2 kcal) solvent corrections may possibly be needed for some
other hydroxylic acids. There is no evidence to indicate that
“serious errors must be present in many other EC bond
energies”, as suggested by Wayner et al.7 Solvent corrections
for H-C, H-N, or H-S acids should ordinarily not be needed
since the hydrogen-bonding donor abilities of these functions
are weak.
N-Hydroxylnortropane. First, a reaction of tropane with ethyl
chloroformate in benzene solution25 gave N-ethoxycarbonylnortropane
in 95% yield, bp 109-111 °C/8τ (lit.25 bp 122-124 °C/13τ). Then,
hydrolysis of N-ethoxycarbonylnortropane in refluxing concentrated
hydrochloric acid gave nortropane hydrochloride in 90% yield. The
hydrochloride (1.5 g, 10 mmol) was dissolved in 10 mL of 1 N NaOH
aqueous solution and warmed to 50 °C, then 2.0 mL of 30% hydrogen
peroxide (18 mmol) and 2 mg of phosphotungstic acid were added.
After the gas evolution ceased (ca. 15 min), the resulting solution was
extracted with 20 mL of CH2Cl2 containing 1.2 g (6 mmol) of
PhNHNHPh. The organic layer was dried with anhydrous K2CO3. After
removal of the solvent, sublimation of the residue at 10 τ gave 0.6 g
(48%) of N-hydroxynortropane, mp 114-116 °C (lit.26 mp 118 °C).
1H NMR (CDCl3): δ 1.3-1.9 (10H, m), 3.44 (2H, s), 4.4 (1H, br,
OH).
N-Oxynortropane was prepared by the method of Ingold et al.26
The reduction potential of this radical was measured successfully using
the crude product.
N-tert-Butylbenzohydroxamic acid was prepared by a revised
method of Perkins et al.17 Into an acetonitrile solution (10 mL)
containing N-tert-butylhydroxylamine hydrochloride (0.55 g, 5 mmol)
and triethylamine (2.8 mL, 20 mmol) was added dropwise an acetonitrile
solution of benzoyl chloride (1.2 mL, 10 mmol). The resulting mixture
was stirred overnight and then mixed with 20 mL water and extracted
with CH2Cl2 twice (10 mL each). The organic layer was dried and the
Experimental Section
NMR spectra were recorded on a Gemini XL-300 (300 MHz) or
XLA 400 (400 MHz) spectrometer. Melting points were measured on
a Thomas Hoover capillary melting point apparatus and are uncorrected.
N-Isopropylhydroxylamine hydrochloride, N-tert-butylhydroxylamine
hydrochloride, N-hydroxyldiethylamine, N-oxyl-2,2,6,6-tetramethyl-
piperidine, di-tert-butyl nitroxide, and tropane (N-methyl-8-azabicyclo-
[3.2.1]octane) are commercially available (Aldrich).
(22) Paleos, C. M.; Dais, P. J. Chem. Soc., Chem. Commun. 1977, 345.
(23) Klages, F.; Sitz, H. Chem. Ber. 1959, 92, 2606.
(24) Jones, W. H.; Tristram, E. W.; Benning, W. E. J. Am. Chem. Soc.
1959, 81, 2151-2154.
(25) Sandoz Ltd. The Netherlands patent application 6,510,033; Chem.
Abstr. 1966, 65:3846.
(19) Bordwell, F. G.; Liu, Wei-Zhong. Unpublished results.
(20) Kanabus-Kaminska, J. M.; Gilbert, B. C.; Griller, D. J. Am. Chem.
Soc. 1989, 111, 3311-3314.
(21) Arnett, E. M.; Jaris, L.; Mitchell, E. J.; Murty, T. S. S. R.; Gorrie,
T. M.; Schleyer, P. v. R. J. Am. Chem. Soc. 1970, 92, 2365-2377.
(26) Mendenhall, G. D.; Ingold, K. U. J. Am. Chem. Soc. 1973, 95,
6395-6400.