Effect of Remote Substitution on Face Selection in Addition Reactions of Nor- And Homoadamantan-9-ones and of Several Analogues

Article abstract of DOI:10.1021/jo9718197

The effects of 3-halo substitution on face selection in borohydride reductions of both nor- and homoadamantan-9-ones 1-X and 3-X, respectively, have been compared with those of 5-halo substitution in the parent 5-haloadamantan-2-ones 2-X. The differences between the product ratios in these three compounds are small, but the selectivity of 2-X is somewhat larger than that of either of the homologues. This is also true of the corresponding aza- and diazaadamantanones, and of an electrophilic addition. It is concluded that the approach angle of the reagent is not a sensitive variable in addition reactions of cyclohexanone and its derivatives, and that well-aligned antiperiplanar bonds are preferred to achieve maximal remote substituent induced selectivities.

Full text of DOI:10.1021/jo9718197

3218
J . Org. Chem. 1998, 63, 3218-3223
Effect of Rem ote Su bstitu tion on F a ce Selection in Ad d ition
Rea ction s of Nor - a n d Hom oa d a m a n ta n -9-on es a n d of Sever a l
An a logu es
Mira Kaselj,† Elena M. Gonikberg,‡ and William J . le Noble*
Department of Chemistry, State University of New York, Stony Brook, New York 11794-3400
Received October 2, 1997
The effects of 3-halo substitution on face selection in borohydride reductions of both nor- and
homoadamantan-9-ones 1-X and 3-X, respectively, have been compared with those of 5-halo
substitution in the parent 5-haloadamantan-2-ones 2-X. The differences between the product ratios
in these three compounds are small, but the selectivity of 2-X is somewhat larger than that of
either of the homologues. This is also true of the corresponding aza- and diazaadamantanones,
and of an electrophilic addition. It is concluded that the approach angle of the reagent is not a
sensitive variable in addition reactions of cyclohexanone and its derivatives, and that well-aligned
antiperiplanar bonds are preferred to achieve maximal remote substituent induced selectivities.
In tr od u ction
The addition of many reagents to 4-tert-butylcyclohex-
anone1 and to the corresponding olefin2 takes place in
contrasteric fashion: the principal product is often that
stereoisomer which results from axial approach of the
reagent to the trigonal carbon. This fact has stimulated
much research and discussion. It is generally agreed that
it has an electronic basis, but the exact nature of the
effect has remained controversial. To remove the concur-
rent steric factor as well as conformational questions as
much as possible, we have studied 5-substituted ada-
mantan-2-ones 2-X and their derivatives.3 The results
of many studies of addition and elimination processes of
every description4 have led us to the conclusion that the
newly forming (or breaking) bond is electron-deficient in
the transition states, and that the complex is stabilized
by electron donation from antiperiplanar vicinal bonds.5
The donation derives from the σ component of these
bonds; the recipient orbital is the σ* component of the
bond in transition (Figure 1).
F igu r e 1. (a) The effect of remote substitution on face
selection in additions to a trigonal site in adamantane deriva-
tives according to the Cieplak model.2,5 (b) General structure
of the probes used in this study.
proposed by Gung8 as a possible cause of the observed
selectivity. Ring flattening as part of the reaction
pathway of nucleophilic addition to simple cyclohex-
anones has been a long-standing conjecture9 in the
Felkin-Anh account, and this leaves our results open to
the objection that the rigid adamantanones cannot supply
any insight in the stereochemistry of addition to the more
flexible monocyclic species. In this paper, we address the
latter of these arguments.
This picture is not universally accepted, however. The
possibility of electrostatic interactions has been advanced
by Adcock6 and others.7 Distortions of the trigonal site
caused by substituents in the probe molecule have been
The net effect of ring flattening in simple cyclohex-
anones is that it reduces the hindrance offered by the
axial hydrogen atoms at C-3 and C-5 to the reagent and
thus allows a more favorable approach angle. This effect
would be mimicked in noradamantan-9-ones 1-X, not by
flattening the ring but by pinning back the offending
hydrogens. In contrast, the effect would be exacerbated
in homoadamantan-9-ones 3-X. To be sure, this is true
for both the en and zu faces; however, if the approach
angle is really the crucial factor envisioned for simple
cyclohexanones, one might expect a systematic shift in
E/Z product ratio toward unity or less for 1-X, and an
increase for 3-X.
† Present address: Geo-Centers, Inc., 200 Valley Road, Suite 102,
Mt. Arlington, NJ 07856.
‡ Albany Molecular Research, 21 Corporate Circle, Albany, NY
12203.
(1) Reetz, M. T.; Stanchev, S. J . Chem. Soc., Chem. Commun. 1993,
328 and references given there.
(2) J ohnson, C. R.; Tait, B. D.; Cieplak, A. S. J . Am. Chem. Soc.
1987, 109, 5875.
(3) (a) Cheung, C. K.; Tseng, L. T.; Lin, M.-H.; Srivastava, S.; le
Noble, W. J . J . Am. Chem. Soc. 1986, 108, 1598 and 1987, 109, 7239.
For recent examples and references, see (b) Kaselj M.; le Noble, W. J .
J . Org. Chem. 1996, 61, 4157 and (c) Gonikberg, E. M.; Picart, F.; le
Noble, W. J . J . Org. Chem. 1996, 61, 9588.
(4) See for example, Mukherjee, A.; Wu, Q.; le Noble, W. J . J . Org.
Chem. 1994, 59, 3270.
(5) Li, H.; le Noble, W. J . Recl. Trav. Chim. Pays-Bas 1992, 111,
199.
(6) Adcock, W. Cotton, J .; Trout, N. A. J . Org. Chem. 1994, 59, 1867.
(7) Mehta, G.; Khan, F. A.; Adcock, W. J . Chem. Soc., Perkin Trans.
2 1995, 2189.
(8) Gung, B. W.; Wolf, M. A. J . Org. Chem. 1996, 61, 232.
(9) See for example, Wu, Y.-D.; Houk, K. N.; Paddon-Row: M. N.
Angew. Chem., Int. Ed. Engl. 1992, 31, 1019, and references given
there.
S0022-3263(97)01819-7 CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/18/1998

Addition Reactions of Nor- and Homoadamantan-9-ones
J . Org. Chem., Vol. 63, No. 10, 1998 3219
With these thoughts in mind, we studied various
addition processes with 1-X, 3-X, 4, 5, and 6 for com-
parison with the known behavior of 2-X, 7, 8, and 9.
Exp er im en ta l Section
Gen er a l. IR spectra were measured in KBr pellets; absorp-
tions are given in cm-1. High and low resolution mass spectra
were recorded by the Stony Brook Mass Spectrometry Facility.
The signals are recorded below only where the identities of
the nuclei are obvious. NMR spectra were measured in CDCl3
except where noted, by means of 250 and 300 MHz spectrom-
eters. 19F Chemical shifts are given in ppm relative to CFCl3.
The 13C NMR signals are assigned according to the outcome
of DEPT experiments, on the basis of chemical shift values
and on 19F coupling constants (in some cases) as discussed in
the text. The preparations of 1-bromo-9-noradamantan-3-
carboxylic acid (10),10 5-carbomethoxyadamantan-2-one eth-
ylene ketal (12-COOMe),11 3-azanoradamantan-9-one (15)12
and 5,8-diazahomoadamantan-2-one (16)14 have been de-
scribed.
Syn th esis
At the outset, compounds 1-X were not known but
derivative 10 had been reported.10 After the keto func-
tion had been protected as the ketal 11, hydrogenation
with palladium on carbon followed by hydrolysis gave the
keto acid 1-COOH. Hunsdiecker decarboxylation con-
verted this into 1-Br, and treatment of the bromide with
silver fluoride in hot cyclohexane yielded 1-F. The
synthesis of compounds 3-X was based on the known11
ester 12-COOMe, previously reported by this laboratory.
Lithium aluminum hydride reduction converted it into
12-CH2OH, and the corresponding alkoxide, generated
by means of tert-butyllithium, upon treatment with tosyl
chloride gave 12-CH2OTs. Hydrolysis in basic trifluoro-
ethanol led to an equimolar mixture of 12-CH2OH and
the ring-expanded homoadamantanols 13 and 14, which
are readily distinguished by means of their 13C NMR
spectra. Treatment of 14 with hydrofluoric acid gave 3-F.
Wittig reaction of this fluoride provided the methylene
analogue 4. Salt 5 was easily prepared from the known12
parent 15. Finally, since we had data13 on 5,7-diazaada-
mantan-2-one N-oxide 9 and since the parent diaza-
homoadamantanone 16 is known,14 we included 6 in our
repertoire (see Experimental Section).
1-Br om o-9-n or a d a m a n ta n -3-ca r boxylic Acid Eth ylen e
Keta l (11). A mixture of acid 10 (110 mg, 0.39 mmol),
ethylene glycol (0.5 mL), and zinc chloride (15 mg) was heated
in a commercial microwave oven (Samsung RE-552D) for 1
min. The mixture was allowed to cool to room temperature,
diluted with 6.25 N aqueous sodium hydroxide and extracted
with chloroform. The water layer was acidified with hydro-
chloric acid and extracted again; the combined extract was
dried over sodium sulfate. Evaporation of solvent gave white
solid 11 (100 mg, 85%), mp 145 °C. MS, m/z 304, 302 (1:1,
M+), 223 (M+ - Br); HRMS, fd 304.0135; calcd 304.0135; IR,
3200-2500 (m), 2958 (m), 2886 (m), 1691 (s), 1147 (s) cm-1
;
1H NMR, δ 10.15 (COOH), 4.26-4.12 (m, 2H), 4.10-3.96 (m,
2H), 2.72 (t, 1H, J ) 5.5 Hz), 2.54-2.15 (m, 5H), 1.99 (s, 2H),
1.84-1.73 (m, 2H); 13C NMR, δ 181.63 (COOH), 110.06 (OCO),
67.58 (C-COOH), 63.34 and 66.27 (CH2O), 52.09, 49.66, 42.21
and 38.70 (CH2), 46.56 and 43.89 (CH).
(10) (a) Vogt, B. R.; Hoover, J . R. E. Tetrahedron Lett. 1967, 2841.
(b) Stetter, H.; Thomas, H. G.; Meyer, K. Chem. Ber. 1970, 103, 863.
(c) Fleming, I.; Hanson, S. W. J . Chem. Soc., Perkin Trans 1 1973,
1669. (d) Hoover, J . R. E. Chem. Abst. 1970, 72, 110904. The Beilstein
Structure Editor incorrectly shows the structure of 1-Br in a reference
to the compound 1-bromonoradamantan-9-one reported by Hoover in
ref 10a.
(11) le Noble, W. J .; Srivastava, S.; Cheung, C.-K. J . Org. Chem.
1983, 48, 1099.
(12) (a) Reints Bok, T.; Speckamp, W. N. Heterocycles 1970, 12, 343.
(b) Speckamp, W. N.; Dijkink, J .; Dekkers, A. W. J . D.; Huisman, H.
O. Tetrahedron 1971, 27, 3143.
9-Oxo-n or a d a m a n ta n e-3-ca r boxylic Acid (1-COOH). A
solution of 11 (100 mg, 0.33 mmol) in absolute methanol (4
mL) containing potassium hydroxide (100 mg) was hydroge-
nated in the presence of 10% palladium on carbon (30 mg) at
room temperature and 1.5 atm for 2.5 h. The mixture was
filtered through Celite, and solvent was removed in vacuo. The
residue was dissolved in a mixture of concentrated hydrochlo-
ric acid (0.5 mL) and acetic acid (1 mL) and heated to reflux
for 1.5 h. The solution was concentrated to dryness and the
residue extracted with acetone. Evaporation of solvent gave
white solid 1-COOH (50 mg, 84%), mp 128-129 °C (lit.10b,10c
133-7 °C, 132-4 °C). MS, m/z 180 (M+), 162 (M+ - H2O); IR,
(13) Gonikberg, E. M.; le Noble, W. J . J . Org. Chem. 1995, 60, 7751.
(14) Kuznetsov, A. I.; Vladimirova, I. A.; Basargin, E. B.; Ba, M.
K.; Moskovkin, A. S.; Botnikov, M. Y. Chem. Heterocycl. Comput. (Engl.
Transl.) 1990, 26, 572.
3100-2500 (w), 2962 (m), 1721 (s), 1693 (s), 1295 (m) cm-1
;

3220 J . Org. Chem., Vol. 63, No. 10, 1998
Kaselj et al.
1H NMR, δ 9.72 (COOH), 3.04 (t, 1H, J ) 6.1 Hz), 2.79 (s,
2H), 2.36-1.80 (m, 8H); 13C NMR, δ 212.36 (C-9), 181.88
(COOH), 52.62 (C-3), 52.14 (C-1,5), 46.42 (C-2,4), 43.56 (C-
6,8), 43.13 (C-7).
chloroform and dried over magnesium sulfate. Evaporation
yielded 0.47 g (80%) of a 1:1:1 mixture of 12-CH2OH, 13, and
14; column chromatography (neutral alumina, petroleum
ether:ether ) 11:9) gave pure samples of all three compounds,
in that order. Compound 13, mp 91-92 °C. MS, m/z 224 (M+);
HRMS, fd 224.1411; calcd 224.1412; IR, 3550-3150 (m), 2900
(s), 1455 (m), 1105 (s), 1040 (m), 1012 (m). 1H NMR, δ 3.83
(m, 4H), 2.05-1.30 (m, 16H); 13C NMR, δ 112.37 (C-9), 72.24
(C-3), 64.31 and 63.91 (CH2O), 46.69 (CH2), 41.12 (CH2), 40.98
(CH2), 40.39 (CH), 36.75 (CH), 35.83 (CH2), 33.07 (CH2), 26.77
(CH), 23.86 (CH2). Compound 14: mp 91-9 °C. MS, m/z 224
(M+). HRMS, fd 224.1414, calcd 224.1412; IR, 3400-3100 (m),
2905 (s), 1445 (m), 1115 (s), 1035 (m), 1015 (m). 1H NMR, δ
3.92 (s, 4H), 2.30-2.15 (m, 4H), 2.06-1.63 (m, 10H), 1.44 (d,
J ) 13.5, 2H). 13C NMR (benzene), δ 110.62 (C-9), 71.67 (C-
3), 64.11 and 63.96 (CH2O), 45.05 (C-2,11), 42.05 (C-4), 35.93
(C-1,8), 35.75 (C-7,10), 30.28 (C-6), 30.08 (C-5).
3-F lu or oh om oa d a m a n ta n -9-on e (3-F ). Solid 14 (200 mg)
was added to a solution of 70% poly-hydrogen fluoride/pyridine
(Aldrich, 4 mL) and ammonium bifluoride (160 mg, 2.81 mmol)
in a polyethylene bottle at 0 °C. The mixture was stirred for
20 min, quenched with ice and water, and extracted with
pentane. The pentane solution was dried over magnesium
sulfate; removal of solvent gave white solid 3-F (100 mg, 61%),
mp (with dec) 240 °C. MS, m/z 182 (M+); HRMS, fd 182.1111,
calcd 182.1107; IR, 2910 (s), 2860 (m), 1739 (s), 1711 (s), 1450
(m), 1012 (m). 1H NMR, δ 2.42-2.15 (m, 11H), 1.90-1.80 (m,
4H); 13C NMR, δ 217.40 (C-9), 94.45 (J ) 170.32 Hz, C-3), 44.33
(J ) 23.6 Hz, C-2,11), 43.67 (J ) 12.60 Hz, C-1,8), 39.87 (C-
7,10), 38.20 (J ) 25.27 Hz, C-4), 29.83 (C-6), 27.29 (J ) 16.9
Hz, C-5); 19F NMR, δ -122.43 (bs, 1F).
3-Br om on or a d a m a n ta n -9-on e (1-Br ). A mixture of car-
boxylic acid 1-COOH (100 mg, 0.56 mmol) and red mercuric
oxide (254 mg, 1.12 mmol) in methylene bromide (50 mL) was
heated to reflux, and bromine (50 µL, 1.12 mmol) was added.
Heating was continued for 2 h; the solvent was removed by
distillation and the residue purified by passage through a
column of neutral alumina. Elution with petroleum ether
containing chloroform (0-30%) gave white solid 1-Br (80 mg,
67%), mp 119 °C. MS, m/z 216, 214 (1:1, M+), 135 (M+ - Br);
HRMS, fd 215.9975, calcd 215.9974; IR, 2962 (s), 2877 (m),
1
1726 (s), 1716 (s), 877 (m) cm-1; H NMR, δ 2.94 (t, J ) 6.22
Hz, 1H), 2.61 (s, 2H), 2.44-2.20 (m, 6H), 1.76 (d, J ) 11.1 Hz,
2H); 13C NMR, δ 210.20 (C-9), 60.36 (C-3), 54.27 (C-2,4), 53.10
(C-1,3), 47.62 (C-7), 43.19 (C-6,8).
3-F lu or on or a d a m a n ta n -9-on e (1-F ). A mixture of 3-bro-
monoradamantan-9-one 1-Br (75 mg, 0.35 mmol) and anhy-
drous silver fluoride (150 mg) in dry cyclohexane (4 mL) was
heated to reflux for 5 days. After cooling, the silver salt
precipitate was filtered and washed with warm cyclohexane.
The solvent was removed by distillation and the residue passed
through a column of neutral alumina with petroleum ether
containing chloroform (30-50%) to give white solid 1-F (35
mg, 65%), which sublimes in a sealed tube at 88 °C. MS, m/z
154 (M+), 135 (M+ - F); HRMS fd 154.0792, calcd 154.0794;
IR, 2962 (s), 1730 (s), 1260 (s), 1072-1020 (bs), 800 (s) cm-1
.
1H NMR, δ 2.75-2.63 (m, 3H), 2.33-2.16 (m, 6H), 1.77 (d, 2H,
J ) 11.2 Hz); 13C NMR, δ 211.20 (C-9), 104.07 (J ) 211.08
Hz, C-3), 51.72 (J ) 8.52 Hz, C-1,5), 46.99 (J ) 21.25 Hz,
C-2,4), 42.31 (J ) 3.48 Hz, C-6,8), 40.34 (J ) 3.15 Hz, C-7);
19F NMR, δ -168.34 (s, 1F).
5-F lu or o-9-m eth ylen eh om oa d a m a n ta n e (4). Triphenyl-
methylphosphonium bromide (490 mg, 1.37 mmol, vacuum-
dried over phosphorus pentoxide) in THF (5 mL) was treated
dropwise with n-butyllithium (0.88 mL, 1.6 M, 1.4 mmol) under
nitrogen. After 30 min, a solution of 3-F (250 mg, 1.37 mmol)
in THF (3 mL) was added dropwise; the mixture was heated
to reflux for 2 h, allowed to cool to room temperature, and
quenched with water. The precipitate (phosphonium salt) was
filtered and washed with ether. The aqueous part of the
filtrate was extracted with ether. The organic solution was
dried over magnesium sulfate. Removal of solvent gave white
solid 4 (175 mg, 71%), mp 94 °C. MS, m/z 180 (M+); HRMS,
fd 180.1311, calcd 180.1314; IR, 3070 (w), 2977 (w), 2908 (s),
2853 (m), 1658 (m), 1446 (m), 1010 (s), 887 (s). 1H NMR, δ
4.61 (s, 2H), 2.50 (s, 2H), 2.18-1.88 (m, 9H), 1.84-1.60 (m,
4H); 13C NMR, δ 154.71 (C-9), 102.65 (dCH2), 96.55 (J )
168.01 Hz, C-3), 45.52 (J ) 20.57 Hz, C-2,11), 39.68 (C-7,10),
38.57 (J ) 25.41 Hz, C-4), 37.08 (J ) 12.76 Hz, C-1,8), 30.59
(C-6), 27.64 (J ) 17.81 Hz, C-5); 19F NMR, δ -117.22 (s, 1F).
3-Aza n or a d a m a n ta n -9-on e (15) was prepared in seven
5-(Hyd r oxym eth yl)a d a m a n ta n -2-on e Eth ylen e Keta l
(12-CH2OH). A solution of 12-COOMe (1.0 g, 4.0 mmol) in
ether (10 mL) was added dropwise to a suspension of lithium
aluminum hydride (0.130 g, 3.4 mmol) in dry ether (20 mL).
The mixture was heated to reflux for 4 h. After cooling, a
saturated aqeous ammonium chloride solution (30 mL) was
added slowly with stirring which was continued for 30 min
after the addition was complete. The mixture was filtered and
the filtrate extracted with ether. The organic layer was dried
over magnesium sulfate and evaporation of solvent gave
colorless oil 12-CH2OH (0.87 g, 98%), bp 79 °C at 0.5 mmHg.
MS, m/z 224 (M+), 193 (M+ - CH2O); HRMS, fd 224.1412, calcd
224.1412; IR, 3550-3140 (m), 2904 (s), 2851 (s), 1120 (s). 1H
NMR, δ 3.93 (s, 4H), 3.21 (s, 2H), 2.03-1.30 (m, 14H). 13C
NMR, δ 111.10 (C-2), 72.36 (CH2OH), 63.99 and 63.90 (CH2O),
38.27 (C-6), 36.09 (C-4,9), 35.94 (C-1,3), 34.15 (C-8,10), 33.51
(C-5), 26.58 (C-7).
1
steps from 4-hydroxypiperidine as described.12 Its H and 13C
5-(Tosyloxym eth yl)a d a m a n ta n -2-on e Eth ylen e Keta l
(12-CH2OTs). A 1.7 M solution of tert-butyllithium in hexane
(2.9 mL, 4.9 mmol) was added to a solution of 12-CH2OH (1.0
g, 4.5 mmol) in dry ether (25 mL) at 0 °C. A white precipitate
formed immediately. The mixture was stirred for 90 min after
which a solution of prepurified p-toluenesulfonyl chloride
(0.860 g, 4.5 mmol) in ether (12 mL) was added. The mixture
was stirred for 2.5 h at 0 °C, quenched with water and
extracted with ether. The ether solution was dried over
magnesium sulfate; removal of solvent gave white solid 12-
CH2OTs (1.77 g, 87%), mp 66-68 °C. HRMS, fd 378.1490,
calcd 378.1501; IR, 2908 (s), 2857 (m), 1597 (m), 1358 (s), 1175
NMR spectra were observed in methanol-d4; in this medium,
it is in equilibrium with the E- and Z-hemiketals.15 1H NMR
(CD3OD) δ 3.88 (t, 1H, J ) 6.0 Hz), 3.58 (t, 1H, J ) 6.0 Hz),
3.20-3.00 (m, 8H), 2.80-2.67 (m, 6H), 2.33-2.16 (m, 6H),
2.04-1.68 (m, 10H); 13C NMR (CD3OD), δ 98.63 (C), 67.15,
63.30 and 63.00 (CH2), 61.29, 61.21, 60.77, 53.14, 45.50 and
45.28 (CH), 44.74, 39.64 and 39.39 (CH2); CdO was not
observed.
N-Meth yl-3-a za n or a d a m a n ta n -9-on e Iod id e (5). Iodo-
methane (0.4 mL) was added to a solution of 15 (42 mg, 0.3
mmol) in methylene chloride (5 mL). A white precipitate
formed in the next 0.5 h. Evaporation of solvent gave white
solid 5 (76 mg, 90%). NMR analysis of a D2O solution of the
salt showed it to be present in the form of a diol. 1H NMR
(D2O), δ 4.26 (t, 1H, J ) 6.6 Hz), 3.76 (d, 2H, J ) 10.5 Hz),
3.26 (d, 2H, J ) 11.7 Hz), 3.13 (s, 3H), 2.51 (s, 2H), 2.26-2.02
(m, 4H); 13C NMR (D2O), δ 92.06 (C-9), 75.76 (C-7), 69.91 (C-
2,4), 48.02 and 46.67 (C-1,5 or CH3), 46.22 (C-6,8).
1
(s), 954 (s); H NMR, δ 7.70 (d, 2H, J ) 8.1 Hz), 7.26 (d, 2H,
J ) 8.1 Hz), 3.85 (m, 4H), 3.49 (s, 2H), 2.31 (s, 3H), 1.88-1.30
(m, 13H). 13C NMR, δ 144.48, 132.78, 129.66 (2CH), 127.72
(2CH), 110.43 (C-2), 78.75 (CH2OTs), 64.16 and 64.04 (CH2O),
37.98 (C-6), 35.83 (C-4,9), 35.77 (C-1,3), 33.74 (C-8,10), 32.48
(C-5), 26.30 (C-7), 21.48 (CH3).
3-Hyd r oxyh om oa d a m a n ta n -7-on e Eth ylen e Keta l (13)
a n d 3-H yd r oxyh om oa d a m a n t a n -9-on e E t h ylen e Ket a l
(14). A solution of 12-CH2OTs (1.0 g, 2.65 mmol) and sodium
carbonate (0.7 g, 6.5 mmol) in aqueous trifluoroethanol (20%
water by volume, 80 mL) was heated at 180 °C for 15 h. The
solvent was removed in vacuo and the residue extracted with
3,6-Dia za h om oa d a m a n ta n -9-on e (16). This compound
was prepared as described.14 The authors’ 1H NMR spectrum
(15) This was also the case with 5; see Hahn, J .; le Noble, W. J . J .
Am. Chem. Soc. 1992, 114, 1916.

Addition Reactions of Nor- and Homoadamantan-9-ones
J . Org. Chem., Vol. 63, No. 10, 1998 3221
was confirmed. 13C NMR, δ 214.5 (CdO), 59.9 (C-2,7,10,11),
2.50-1.55 (m, 30H), 1.27 (d, J ) 13.76 Hz, 2H). E-20: GCMS,
m/z 184 (w, M+), 166 (M+ - H2O); 13C NMR, δ 96.39 (J )
167.74 Hz, C-3), 72.20 (J ) 1.62 Hz, C-9), 43.47 (J ) 21.71
Hz, C-2,11), 38.40 (J ) 25.52 Hz, C-4), 32.89 (J ) 12.37 Hz,
C-1,8), 30.24 (C-7,10), 29.87 (C-6), 27.74 (J ) 18.82 Hz, C-5);
19F NMR, δ -119.45 (m, 1F). Z-20: GCMS, m/z 184 (w, M+),
166 (M+ - H2O); 13C NMR, δ 96.62 (J ) 166.89 Hz, C-3), 72.01
(J ) 1.74 Hz, C-9), 38.40 (J ) 25.52 Hz, C-4), 37.53 (J ) 21.76
Hz, C-2,11), 36.96 (C-7,10), 33.97 (J ) 1.51, C-1,8), 29.78 (C-
6), 27.82 (J ) 17.77 Hz, C-5); 19F NMR, δ -114.34 (bs, 1F).
The 13C signals were assigned on the basis of DEPT experi-
ments, 19F couplings, and the effect of the C-9 oxygen atom
on the chemical shifts of C-2,11 and C-7,10.
57.72 (C-4,5), 53.7 (C-1,8).
3,6-Dia za h om oa d a m a n ta n -9-ol (17). This compound was
prepared as described.16 1H NMR, δ 3.74, (s, 1H, H-9), 3.61
(d, J ) 14 Hz, 2H), 3.15 (d, J ) 13.5 Hz, 2H), 3.00 (s, 4H,
H-4,5), 2.79 (d, J ) 14 Hz, 2H), 2.40 (d, J ) 13.5 Hz, 2H), 1.69
(s, 2H, H-1,8); 13C NMR, δ 71.0 (C-9), 57.9, 57.7 (C-4,5), 57.5
(C-7,10), 50.7 (C-2,11), 37.1 (C-1,8).
3,6-Dia za h om oa d a m a n ta n -9 N-Oxid e (6). This com-
pound was prepared by means of the procedure used for
compound 9.13 1H NMR (D2O, pH adjusted to 13), δ 4.06 (d, J
) 13 Hz, 2H), 3.66 (d, J ) 13 Hz, 2H, H-2,11, overlaps with t,
J ) 8 Hz, 2H, H-5), 3.38 (d, J ) 14.5 Hz, 2H), 3.02 (t, J ) 8
Hz, 2H, H-4), 2.86 (d, J ) 14.5 Hz, 2H, H-7,10), 1.72 (s, 2H,
H-1,8); 13C NMR (D2O), δ 74.7 (C-4), 72.4 (C-2,11), 52.9 (C-
7,10), 48.9 (C-5), 40.4 (C-1,8); CdO was not detected.
Meth yla tion of 3-F . Methyllithium (0.17 mL,1.4 M, 0.24
mmol) was added to a solution of 3-F in ether (2 mL) at 0 °C.
After 12 h, the mixture was treated with saturated aqueous
ammonium chloride and extracted with ether; the organic layer
was dried over magnesium sulfate. Removal of solvent gave
a white solid mixture (37 mg, 85%) of E- and Z-5-fluoro-9-
methylhomoadamantan-9-ol (E- and Z-21) which was analyzed
by means of GC and of 19F NMR. IR, 3100-3600 (m), 2927
Selectivity Stu d ies. Red u ction of 1-Br . Sodium boro-
hydride (0.2 mmol) was added to a stirred solution of ketone
1-Br (0.2 mmol) in methanol (1 mL) at 0 °C. After 3 h, the
mixture was poured into an aqueous ammonium chloride
solution and extracted with chloroform. The extract was dried
over sodium sulfate; evaporation of solvent gave a mixture of
alcohols E- and Z-3-bromonoradamantan-9-ol (E- and Z-18).
The ratio of the isomers was obtained from 1H NMR spectra
measured in the presence of 0.15 eq of Eu(fod)3; the broad
singlet at δ3.84 is well resolved into two singlets at δ5.7 and
6.2 under these conditions. This ratio was found to be in
excellent agreement with the one deduced from GC analysis.
1
(s), 1455 (m), 1381 (m), 1017 (m); H NMR, δ 2.72-1.20 (m).
E-21: MS, m/z 183 (M+ - CH3); 13C NMR, δ 96.54 (J ) 167.27
Hz, C-3), 71.45 (C-9), 42.14 (J ) 21.99 Hz, C-2,11), 39.25 (J )
25.40 Hz, C-4), 37.41 (J ) 11.80 Hz, C-1,8), 32.89 (C-7,10),
29.02 (C-6), 28.03 (J ) 18.23 Hz, C-5), 27.42 or 26.21 (J )
1.92) (CH3). 19F NMR, δ -118.31 (m, 1F). Z-21: MS, m/z 183
(M+ - CH3). 13C NMR, δ 96.43 (J ) 166.52 Hz, C-3), 71.24 (J
) 1.62 Hz, C-9), 39.82 (J ) 21.91 Hz, C-2,11), 38.79 (J ) 26.29
Hz, C-4), 38.03 (J ) 11.95 Hz, C-1,8), 35.45 (C-7,10), 29.21
(C-6), 28.58 (J ) 18.11 Hz, C-5), 27.42 or 26.21 (J ) 1.92 Hz)
(CH3). 19F NMR, δ -116.40 (m, 1F). The 13C NMR assign-
ments were made in the same way as those described above
for compounds 20.
IR, 3450-3100 (m), 2960 (s), 1460 (m), 1073 (s), 785 (s) cm-1
.
1H NMR, δ 3.84 (bs, 2H-9), 2.70-2.52 (m, 4H), 2.25-1.83 (m,
16H), 1.57-1.43 (m, 4H). E-18: MS, m/z 137 (M+ - Br), 119
(M+ - Br-H2O); 13C NMR, δ 71.36 (C-9), 63.91 or 63.74 (C-3),
51.68 (C-2,4), 47.87 (C-7), 44.21 (C-1,5), 38.23 (C-6,8). Z-18:
MS, m/z 137 (M+ - Br), 119 (M+ - Br-H2O); 13C NMR, δ 70.47
(C-9), 63.91 or 63.74 (C-3), 50.22 (C-2,4), 46.61 (C-7), 44.68
(C-1,5), 39.61 (C-6,8). The chemical shift assignments are
based unambiguously on DEPT experiments, the effects of
Eu(fod)3 on the chemical shifts and on the well-known effect
of the oxygen atom at C-9 on the signals of C-2,4 and C-6,8
(see also text below). The major isomer is assigned on the
basis of 13C signal intensities.
n -Bu tyla tion of 3-F . n-Butyllithium (0.18 mL, 1.6 M, 0.29
mmol) was added to a solution of 3-F (50 mg, 0.27 mmol) in
THF (2 mL) at room temperature. After 2 h, aqueous workup
as described above provided a white solid mixture (56 mg, 85%)
of E- and Z-n-butyl-5-fluorohomoadamantan-9-ol (E- and Z-22).
Column chromatography (silica, 200-400 mesh, pentane:ether
) 9:1) gave initially pure E-22; later fractions were mixtures
of E- and Z-22. The ratio of isomers was determined from the
19F NMR spectrum of the crude product as well as by GC
Red u ction of 1-F . The reaction was carried out in pre-
cisely the same way as that of 1-Br, to give a mixture of E-
and Z-3-fluoronoradamantan-9-ol (E- and Z-19). The ratio of
the isomers was determined by means of integration of both
the 1H and the 19F NMR signals; the two were in excellent
agreement. IR, 3550-3100 (m), 2958 (s), 2874 (m), 1321 (s),
1050 (s), 730 (s) cm-1; 1H NMR, δ 3.86 (s, E-H-9), 3.80 (s, Z-H-
9), 2.36-2.27 (m, 8H), 2.05-1.09 (m, 14H), 1.42 (d, J ) 11.40
Hz, 2H). E-19: MS, m/z 156 (M+), 138 (M+ - H2O); 13C NMR,
δ106.48 (J ) 207.81 Hz, C-3), 72.63 (C-9), 44.61 (J ) 20.13
Hz, C-2,4), 43.23 (J ) 8.34 Hz, C-1,5), 40.48 (J ) 23.23 Hz,
C-7), 37.89 (J ) 2.91 Hz, C-6,8); 19F NMR, δ -168.63 (d, J )
1
analysis. IR, 3200-3600 (s), 2921 (s), 1468 (m), 1023 (s); H
NMR, δ 2.66-1.15 (m, 5H), 0.97-0.88 (m, 45H). E-22: MS,
m/z 183 (M+ - Bu); IR: 3190-3530 (s), 2949 (s), 2923 (s), 1461
(m), 1170 (m), 1028 (s). 1H NMR, δ 2.42-1.95 (m, 7H), 1.86-
1.60 (m, 8H), 1.40-1.18 (m, 7H), 0.97-0.88 (m, 3H); 13C NMR,
δ 96.50 (J ) 167.74 Hz, C-3), 72.18 (C-9), 41.71 (J ) 21.83
Hz, C-2,11), 39.36 (J ) 25.57 Hz, C-4), 37.71 (C-12), 35.26 (J
) 11.59 Hz, C-1,8), 32.93 (C-7,10), 29.64 (C-6), 28.06 (J ) 18.09
Hz, C-5), 24.28 (C-13), 23.22 (C-14), 14.15 (CH3); 19F NMR,
δ-117.59 (m, 1F). Z-22: MS, m/z 183 (M+ - Bu). 13C NMR, δ
96.93 (J ) 166.49 Hz, C-3), 71.96 (C-9), 39.88 (J ) 21.59 Hz,
C-2,11), 38.85 (J ) 26.32 Hz, C-4), 36.85 (C-12), 35.89 (J )
1.89 Hz, C-1,8), 34.90 (C-7,10), 29.04 (C-6), 28.79 (J ) 18.07
Hz, C-5), 24.46 (C-13), 23.28 (C-14), 14.15 (CH3); 19F NMR, δ
-115.56 (m, 1F). The 13C NMR signals were assigned on the
basis of DEPT experiments, the effect of the oxygen atom at
C-9 on the chemical shifts of the flanking carbons, and the
application of Eu(fod)3 as described in the text.
12.00 Hz, 1F). Z-19: MS, m/z 156 (M+), 138 (M+ - H2O); 13
C
NMR, δ 105.47 (J ) 208.75 Hz, C-3), 71.50 (J ) 1.70 Hz, C-9),
44.08 (J ) 7.53 Hz, C-1,6), 43.73 (J ) 19.20 Hz, C-2,4), 39.18
(J ) 23.31 Hz, C-7), 39.12 (J ) 3.16 Hz, C-6,8); 19F NMR, δ
-161.36 (q, J ) 12.71 Hz, 1F). The 13C assignments are
facilitated in this case by the 19F splittings which are steeply
dependent on the number of intervening bonds.
Red u ction of 3-F . Sodium borohydride (8.5 mg, 0.22
mmol) was added to a stirred solution of ketone 3-F (40 mg,
0.22 mmol) in methanol (1 mL) at 0 °C. After 3 h, the mixture
was poured into saturated aqueous ammonium chloride and
extracted with methylene chloride. The organic layer was
dried over magnesium sulfate, and removal of solvent gave a
white solid mixture (34 mg, 83%) of E- and Z-5-fluoro-
homoadamantan-9-ol (E- and Z-20). Analyses by means of GC,
1H NMR (H-9), and 19F NMR gave virtually identical ratios.
IR, 3050-3600 (m), 2911 (s), 1458 (m), 1068 (s), 1014 (s), 744
Ep oxid a tion of 4. Olefin 4 (80 mg, 0.44 mmol) was
dissolved in methylene chloride (4 mL), and m-chloroperben-
zoic acid (95 mg, 80-85%) was added. After 24 h, the excess
of peracid was destroyed by adding a 10% aqueous solution of
sodium sulfite. The mixture was extracted with methylene
chloride and the organic layer washed successively with 5%
sodium bicarbonate, water, and saturated aqueous sodium
chloride; drying over sodium sulfate and removal of solvent
gave 70 mg (80%) of a mixture of E- and Z-5-fluoro-9-
methylenehomoadamantane epoxide (E- and Z-23). The ratio
of isomers was determined from the 1H NMR signals of the
oxirane protons in the presence of 0.15 equiv of Eu(fod)3 as
well as from the 19F spectrum and by GC analysis. IR, 2924
1
(s); H NMR, δ 3.84 (s, H-9E), 3.63 (q, J ) 3.50 Hz, Z-H-9),
(16) Kuznetsov, A. I.; Vladimirova, I. A.; Serova, T. M.; Moskovkin,
A. S. Chem. Heterocycl. Comput. (Engl. Transl.) 1991, 27, 631.
1
(s), 2870 (m), 1456 (m), 1017 (s); H NMR, δ 2.58 (t, J ) 15.0

3222 J . Org. Chem., Vol. 63, No. 10, 1998
Kaselj et al.
Ta ble 1. P r od u ct Ra tios in Ad d ition P r ocesses of
Ad a m a n ta n e Der iva tives
Hz, 2H), 2.32-0.75 (m with s at 2.30 and 2.23, 32H). E-23:
GCMS, m/z 196 (M+), 181 (M+ - CH3). 13C NMR (C6D6), δ
95.22 (J ) 170.08 Hz, C-3), 62.23 (C-9), 53.68 (CH2O), 41.88
(J ) 21.98 Hz, C-2,11), 38.61 (J ) 25.89 Hz, C-4), 36.97 (C-
7,10), 35.02 (J ) 12.33 Hz, C-1,8), 30.12 (C-6), 27.68 (J ) 16.84
Hz, C-5); 19F NMR, δ -117.25 (bs, 1F). Z-23: MS, m/z 196
(M+), 181 (M+ - CH3). 13C NMR (C6D6), δ 95.50 (J ) 169.44
Hz, C-3), 62.31 (C-9), 54.07 (CH2O), 43.80 (J ) 20.00 Hz,
C-2,11), 38.69 (J ) 25.67 Hz, C-4), 34.76 (C-7,10), 34.34 (J )
12.62 Hz, C-1,8), 29.91 (C-6), 27.56 (J ) 17.05 Hz, C-5); 19F
NMR, δ-119.40 (m, 1F).
Red u ction of 15. The reductions were carried out under
conditions similar to those described above, in methanol-d4 and
in THF-d8. The resulting mixtures of E- and Z-3-azanorada-
mantan-2-ol (E- and Z-24) were analyzed by integrating the
H-9 signals, and the assignments of configuration based on
the chemical shifts of the flanking carbons in the 13C NMR
spectrum. 1H NMR (CD3OD) δ 4.00 (bs, E-H-9), 3.92 (bs, Z-H-
9), 3.76 (t, Z-H-7, J ) 6.1 Hz), 3.65 (t, E-H-7, J ) 6.7 Hz), 3.46
(d, 2H, J ) 10.9 Hz), 3.15-2.70 (m, 6H), 2.29 (bs, 4H), 2.17 (d,
2H, J ) 11.3 Hz), 2.05-1.71 (m, 6H). E-24: 13C NMR
(CD3OD), δ70.43 (C-9), 63.56 (C-2,4), 62.26 (C-7), 44.03 (C-
1,5), 38.79 (C-6,8); Z-24: 13C NMR (CD3OD), δ 70.38 (C-9),
62.46 (C-2,4), 61.43 (C-7), 43.76 (C-1,5), 39.90 (C-6,8). The
composition depended on the solvent used in the reduction;
see text.
Red u ction of 5. This reaction was done with sodium
borohydride at room temperature in D2O to give a mixture of
E- and Z-N-methyl-3-azanoradamantan-9-ol iodide (E- and
Z-25), which was analyzed by means of integratable 13C NMR
(INVGATE). E-25: 1H NMR (D2O) δ 4.36 (bs, 1H), 4.27 (bs,
1H), 3.83 (d, 2H, J ) 10.3 Hz), 3.49 (d, 2H, J ) 10.8 Hz), 3.30
(s, 3H), 2.74 (bs, 2H), 2.35 (m, 2H); 13C NMR (D2O) δ 80.16
(C-7), 73.69 (C-2,4), 69.88 (C-9), 51.43 (CH3), 45.89 (C-1,5),
38.99 (C-6,8). Z-25: 1H NMR (D2O) δ4.22 (bs, 1H), 4.15 (d,
2H, J ) 10.4 Hz), 3.10 (d, 2H, J ) 10.8 Hz), 3.26 (s, 3H), other
signals overlap those of the E-isomer; 13C NMR (D2O) δ 78.49
(C-7), 72.69 (C-2,4), 68.90 (C-9), 51.48 (CH3), 45.00 (C-1,5),
40.25 (C-6,8).
substrate reaction
conditions
analysis
E:Z
ref
1-Br
NaBH4
MeOH, 0 °C
GC
59:41
60:40
59:41
60:40
1H NMR
1H NMR
1H NMR
2-Br
1-F
NaBH4
NaBH4
MeOH, 0 °C
MeOH, 0 °C
3a
3a
19F NMR 60:40
2-F
3-F
NaBH4
NaBH4
MeOH, 0 °C
MeOH, 0 °C
1H NMR
62:38
19F NMR 67:33
1H NMR
GC
66:34
66:34
70:30
66:34
2-F
3-F
MeLi
MeLi
Ether, 0 °C
Ether, 0 °C
GC
GC
3a
19F NMR 67:33
3-F
n-BuLi
mCPBA
THF, 0 °C
CH2Cl2, rt
13C NMR 67:33
7
GC
66:34 19
66:34 19
58:42
1H NMR
1H NMR
4
m-CPBA CH2Cl2, rt
19F NMR 58:42
GC
58:42
66:34
46:54
62:38 20
45:55 20
15
NaBH4
MeOH-d4, 0 °C 1H NMR
THF-d8, 0 °C
MeOH, 0 °C
THF, 0 °C
D2O, rt
D2O, rt
MeOH, 0° C
MeOH, 0 °C
1H NMR
1H NMR
1H NMR
a
NaBH4
MeLi
NaBH4
NaBH4
NaBH4
NaBH4
5
8
9
6
13C NMR 87:13
1H NMR
1H NMR
1H NMR
1H NMR
1H NMR
96:4
88:12 13
83:17
28:72
31:69 13
20
17
b
m-CPBA CH2Cl2, rt
m-CPBA CH2Cl2, rt
a
b
5-Azaadamantan-2-one. 5,7-Diazaadamantan-2-ol.
singly bound oxygen atom at C-2 in adamantanes (or C-9
in nor- or homoadamantanes) tends to polarize the axial
C-H bonds “underneath” it, so that the protons appear
at lower field than normal17a and the 13C atoms show up
at higher field.17b An electron-withdrawing group at C-5
in adamantanes (C-3 in nor- and homoadamantanes)
shifts the signals of the neighboring carbons to lower
field, and as a result of the combined influence of these
two substituents, the Z- isomers have the signals of the
flanking carbons between them (C-4,9 in adamantanes)
fairly close together, whereas those of the E- isomers are
relatively far apart. In Z-5-bromoadamantan-2-ol, for
example, the C-4,9 and the C-8,10 signals appear at δ
42.54 and δ 34.28, respectively, whereas the same signals
in the E-isomer appear at δ 47.63 and δ 29.13, respec-
tively.18 These patterns reappear in virtually every
reaction we study, and hence the result is usually clear
at a glance. Comparison of the inner and outer signals
then reveals which isomer is the major one. This
approach was used throughout this study; it was backed
up in several cases by the use of Eu(fod)3, which shifts
the pair of carbons proximal to the oxygen much more
strongly to lower field than the distal pair. In those cases
where the remote substituent is fluoro, the 19F splittings
of the signals of all carbon atoms within three bonds were
also very helpful. The results are given in Table 1.
When the results for the haloadamantanone deriva-
tives 1-X-3-X are compared, it is at once obvious that
syn approach by the nucleophile to the carbonyl group is
preferred in all cases. The length of the bridge between
the remote bridgehead atoms makes at most a minor
difference. The same thing can be said about the single
Red u ction of 6 a n d Oxid a tion of 17. The reduction was
carried out in methanol as described above to give a mixture
of E- and Z-3,6-diazahomoadamantan-9-ol N-oxide (E- and Z-
26). These compounds were also formed in the oxidation of
17 with m-chloroperbenzoic acid. Since the isomer ratios in
1
the two reactions were very different, the H NMR signals of
the mixtures were readily sorted. E-26: 1H NMR (D2O, pH
adjusted to 13): δ 4.15 (s, 1H, H-9), 3.88 (d, J ) 13 Hz, 2H),
3.78 (m, 4H), 3.44 (d, J ) 14.5 Hz, 2H), 3.07 (t, J ) 7 Hz, 2H),
2.77 (d, J ) 14.5 Hz, 2H), 1.89 (s, 2H, H-1,8); 13C NMR (D2O),
δ 74.3 (C-4), 73.7 (C-2,11), 67.2 (C-9), 48.5 (C-5), 48.3 (C-7,10),
33.8 (C-1,8). Z-26: 1H NMR (D2O), δ 4.0-3.6 (m, 11H), 3.08
(m, 2H), 2.01 (s, 2H); 13C NMR (D2O), δ 74.2 (C-4), 67.7 (C-
2,11), 66.4 (C-9), 54.4 (C-7,10), 48.3 (C-5), 34.8 (C-1,8).
In the presence of a 2-fold excess of m-chloroperbenzoic acid,
N,N′-dioxide 27 is formed from alcohols 26. 1H NMR (D2O,
pH adjusted to 13), δ 4.01 (m, 3H), 3.89 (d, J ) 13 Hz, 2H),
3.77 (s, 4H), 3.70 (d, J ) 14 Hz, 2H), 3.62 (d, J ) 14 Hz, 2H),
2.19 (s, 2H); 13C NMR (D2O), δ 71.8 (C-7,10), 66.0 (C-9), 65.2
(C-2,11), 65.1 and 64.9 (C-4,5), 37.3 (C-1,8). Efforts to reduce
27 back to the mono-N-oxides 26 were unsuccessful.
Resu lts a n d Discu ssion
As noted in the Experimental Section, several analyti-
cal approaches were available to assay the mixtures: GC
1
and H, 13C, and 19F NMR. In some cases, Eu(fod)3 was
used to separate signals. Most of the crude mixtures
were analyzed by more than one technique, and the
results differed by no more than one percentage point in
all instances.
The assignment of the major components of these
mixtures was made virtually exclusively on the basis of
the 13C NMR spectra. As we have noted elsewhere, a
(17) Srivastava, S.; Cheung, C.-K.; le Noble, W. J . Magn. Reson.
Chem. 1985, 23, 232.
(18) (a) Mukherjee, A. K.; le Noble, W. J . J . Org. Chem. 1993, 58,
7955. (b) Li, H.; Silver, J . E.; Watson, W. H.; Kashyap, R. P.; le Noble,
W. J . J . Org. Chem. 1991, 56, 5932.

Addition Reactions of Nor- and Homoadamantan-9-ones
J . Org. Chem., Vol. 63, No. 10, 1998 3223
Sch em e 1
example of electrophilic addition (4 vs 7). No trend is
visible in any of these data that might reflect a gradual
response to the placement of the hominal hydrogen
atoms. One might wonder whether the homoadaman-
tane skeleton is indeed as rigid as pictured above, and
whether a deformation is introduced by the extra meth-
ylene. In that case, the substrate would be subject to a
conformational equilibration of the type shown in eq 1,
which would presumably raise one axial hydrogen and
lower the other. In these enantiomeric forms, the nu-
cleophile will seek out a path away from the plane
perpendicular to that of the carbonyl group, and favoring
the less hindering hydrogen. If that were so, the ho-
moadamantane skeleton would be more similar to the
noradamantane framework than to the adamantane
parent itself. Indeed, one observes that the adamantane
derivatives generally show somewhat larger ratios than
either of the analogues.
Schleyer21 has examined the conformation of homoada-
mantane by means of NMR as well as on a computational
basis. As he pointed out, the c2V symmetrical structure
suffers from a pair of eclipsed methylene groups (C-4 and
C-5); while this interaction is relieved in the equilibrating
structures, the relief is achieved at the cost of increased
strain throughout the rest of the molecule. Both the
spectral evidence and the calculations favored the sym-
metrical structure. It should be pointed out that what-
ever the positions of the axial hydrogens at C-2 and C-11
may be, they are repeated with the axial hydrogens at
C-7 and C-10: the two faces are identical in that respect.
As a result of these considerations, we favor the view-
point that the higher ratios with the adamantane skel-
eton are the result of the more perfect antiperiplanar
alignment of the bonds, which allows the remote sub-
stituent to make its influence felt more strongly. This
also accounts for the fact that when the substituent is
placed at the equatorial position of C-4 rather than at
C-5 in the adamantanone,3b its effect is not significantly
stronger, and even two bromine atoms at C-4 and C-9
cause only a modestly larger preference for syn approach
than does a single bromine atom at C-5.
unshared pair to serve as an electron donor. A compari-
son of compounds 5 and 8 shows a larger ratio for the
latter. We do not have information on 3-azahomoada-
mantan-9-one, but comparing the diaza analogues 6 and
9 again shows the adamantane analogue to exhibit the
larger ratio.
Finally, it may be noted that compounds 6 and 17,
when treated with sodium borohydride and m-chloroper-
benzoic acid, respectively, give the same products E- and
Z-26, but in very different ratios. Thus, the major
stereoisomer of 26 differs depending on which of the two
sequences is followed in Scheme 1. In both cases,
hyperconjugation involving a remote substituent is the
determining factor, as we have reported earlier13 in
another case. We do not know of any basis other than
transition state hyperconjugation on which these facts
could have been confidently predicted.
We conclude as follows. Our experiments do not
support the notion that ring flattening and the related
repositioning of axial hominal hydrogens are important
determinants in the stereochemistry of addition to cy-
clohexanone and its derivatives. Rigidly enforced anti-
periplanarity of the intervening bonds is preferred in the
transmission of remote substituent effects on the stereo-
chemistry of addition and elimination.
Since the differences between conpounds 1-3 were
small, we decided to expand the investigation by incor-
porating a quaternary nitrogen atom in the remote
bridgehead position(s); we have shown elsewhere16,20,22
that this greatly magnifies the ratios. The keto amines
15 and 5-azaadamantan-2-one themselves are not very
informative because the ratios are solvent dependent: in
hydrogen-bonding solvents, the zu face is favored, and
Ack n ow led gm en t. The National Science Founda-
tion supported this work.
Su p p or tin g In for m a tion Ava ila ble: NMR spectra of
1-COOH, 1-Br, 1-F, 3-F, 4-6, 11, 12-CH2OH, 12-CH2OTs, 13-
17, 27, and E,Z-mixtures of 18-26 (119 pages). This material
is contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
in others, the en face.23 This is easily explained:
a
hydrogen-bonded nitrogen atom carries an increased
positive charge, whereas a nonbonded one can use its
(19) Srivastava, S.; le Noble, W. J . J . Am. Chem. Soc. 1987, 109,
5874.
(20) Hahn, J . M.; le Noble, W. J . J . Am. Chem. Soc. 1992, 114, 7751.
(21) Schleyer, P. v. R.; Funke, E.; Liggero, S. H. J . Am. Chem. Soc.
1969, 91, 3965.
(22) Lau, J .; Gonikberg, E. M.; Hung, J .-t.; le Noble, W. J . J . Am.
Chem. Soc. 1995, 117, 11421.
J O9718197
(23) Senda, Y.; Morita, M.; Itoh, H. J . Chem. Soc., Perkin Trans. 2
1996, 221.