Photochemical preparation of 1,3,5,7-tetracyanoadamantane and its conversion to 1,3,5,7-tetrakis(aminomethyl)adamantane

Article abstract of DOI:10.1021/ol036526g

Matrix presented. New adamantane derivatives 1 and 2 that bear functionalized one-carbon extensions at all four bridgehead positions have been prepared. Radical nucleophilic substitution (S_RN1) reaction of 1,3,5,7-tetrabromoadamantane with cyan

Full text of DOI:10.1021/ol036526g

ORGANIC
LETTERS
2004
Vol. 6, No. 11
1705-1707
Photochemical Preparation of
1,3,5,7-Tetracyanoadamantane and Its
Conversion to
1,3,5,7-Tetrakis(aminomethyl)adamantane
Ging S. Lee, Jude N. Bashara, Ghiwa Sabih, Asmik Oganesyan,
Gayane Godjoian, Hieu M. Duong, Eric R. Marinez, and Carlos G. Gutie´rrez*
Department of Chemistry and Biochemistry, California State UniVersity,
Los Angeles, California 90032-8202
cgutier@calstatela.edu
Received December 30, 2003
ABSTRACT
New adamantane derivatives 1 and 2 that bear functionalized one-carbon extensions at all four bridgehead positions have been prepared.
Radical nucleophilic substitution (SRN1) reaction of 1,3,5,7-tetrabromoadamantane with cyanide produces 1,3,5,7-tetracyanoadamantane (1),
which was reduced with borane reagents to 1,3,5,7-tetrakis(aminomethyl)adamantane (2). Improvements in the preparation of 1,3,5,7-
tetrahaloadamantanes (halogen ) Br, Cl, I) are also reported.
Tetra(bridgehead)-substituted adamantanes are valuable tetra-
hedral building blocks for the synthesis of engineered
materials.1 However, there are few examples of 1,3,5,7-
carbon-substituted adamantane cores, and they are limited
to tetraaryladamantanes2 and adamantane tetracarboxylic acid
[and its tetra(acid chloride),3,4 tetraester,3b,4 and tetraamide3b,5
derivatives]. Our work on the synthesis of organic and mixed
organic/inorganic dendrimers required the preparation of
1,3,5,7-tetracyanoadamantane (1) and 1,3,5,7-tetrakis(amino-
methyl)adamantane (2). We report here their efficient
synthesis from adamantane.
We initially prepared 2 (Scheme 1) by taking advantage
of Bashir-Hashemi’s photochemical conversion of either
adamantane carboxylic acid (3) or 1,3-adamantane dicar-
boxylic acid (4) to 1,3,5,7-tetrakis(chlorocarbonyl)adaman-
tane (5).4 This was converted to the tetrabenzylamide 6b,6
which was subsequently reduced to tetraamine 2, first by
deoxygenation with lithium aluminum hydride to the tetra-
benzylamine, followed by debenzylation by hydrogenolysis
over Pd-C. Tetramine 2 was isolated in 4% overall yield
(1) (a) Wang, S.; Oldhman, J. W.; Hudack, R. A.; Bazan, G. C. J. Am.
Chem. Soc. 2000, 122, 5695. (b) Martin, V. V.; Alferiev, I. S.; Weis, A. L.
Tetrahedron Lett. 1999, 40, 223. (c) Menger, F. M.; Migulin, V. A. J. Org.
Chem. 1999, 64, 8916. (d) Newkome, G. R.; Narayana, V. V.; Patri, A. K.;
Grob, J. Moorefield, C. N.; Baker, G. R. Polym. Mater. Sci. Eng. 1995,
737, 222. (e) Reddy, D. R.; Craig, D. C.; Desiraju, G. R. J. Chem. Soc.,
Chem. Commun. 1995, 3, 339. (f) Chen, S. H.; Mastrangelo, J. C.; Shi, H.;
Bashir-Hashemi, A.; Li, J.; Gelber, N. Macromolecules 1995, 28, 7775.
(2) (a) Li, Q.; Rukavishnikov, A. V.; Petukhov, P. A.; Zaikova, T. O.;
Keana, J. F. W. Org. Lett. 2002, 4, 3631. (b) Reichert, V. R.; Mathias, L.
J. Macromolecules 1994, 27, 7015.
(3) (a) Stetter, H.; Bander, O. E.; Neumann, W. Chem. Ber. 1956, 89,
1922. (b) Newkome, G. R.; Nayak, A.; Behera, R. K.; Moorefield, C. N.;
Baker, G. R. J. Org. Chem. 1992, 57, 358.
(4) Bashir-Hashemi, A.; Jianchang, L. Tetrahedron Lett. 1995, 36 (8),
1236.
(5) Stetter, H.; Krause, M. Tetrahedron Lett. 1967, 19, 1843.
(6) 6b: 1H NMR (300 MHz, CDCl3/MeOD-d3) δ 1.72 (s, 12H), 4.16 (s,
8H), 6.85-7.02 (m, 20H), 7.41 (t, 4H); 13C NMR (CDCl3) δ 38.66, 42.06,
42.95, 126.88, 126.99, 128.17, 137.99, 176.04; IR (cast) νmax 3405, 2892,
1664, 1452 cm-1
.
10.1021/ol036526g CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/23/2004

Scheme 1a
Scheme 2a
a Key: (a) (COCl)2, hν; (b) NH3; (c) PhCH2NH2; (d) LiAlH4,
THF; (e) H2, Pd-C; (f) HCl.
a Key: (a) Br2, 2AlBr3; (b) Br2, 4AlCl3; (c) NaCN, DMSO, hν;
(d) CCl4, AlCl3; (e) CH3I, AlBr3; (f) BH2Cl-SMe2; (g) HCl, MeOH
from 3 as the tetrahydrochloride 2‚4(HCl).7 This approach
was unsatisfying for two reasons. The modest yield of the
photocarbonylation step was aggravated by the considerable
expense of adamantane monocarboxylic acid starting material
(or by use of the even more costly 1,3-adamantanedicar-
boxylic acid, which gave higher, yet still modest, yields of
product). Furthermore, the insolubility of parent tetrabenz-
amide 6a necessitated the atom-inefficient introduction of
nitrogen as benzylamine (rather than ammonia), which
required an additional step to remove the four benzyl groups.
We sought an alternative procedure.
small amounts (3-12%) of 1-chloro-3,5,7-tribromoadaman-
tane (8).14 We sought to avoid halogen exchange in the
synthesis of 7 by using aluminum tribromide.15 We found
that equimolar amounts of adamantane and AlBr3 produced
1,3,5-tribromoadamantane in 86% yield. We also found that
two equiv of AlBr3 produced clean 7 in 85% yield, but
manipulations involving this Lewis acid are experimentally
more difficult.
Direct functionalization of adamantane through carbon-
carbon bond formation at the bridgehead has previously
resulted in limited success. Bridgehead monocyanation of
adamantane has been reported by Olah to occur in high yield
with trimethylsilylcyanide/SnCl4,8 yet even 1,3-dicyanation
occurs only sparingly, as introduction of the first electron-
withdrawing cyano group deactivates the molecule toward
formation of subsequent carbocation intermediates. Replace-
ment of halogen with other functional groups has been
observed in bridgehead monohaloadamantanes under condi-
tions that generate radical,9 carbanionic,10 or carbocationic11
intermediates. Reactions of polyhaloadamantanes have gen-
erally been limited to arylation.2 We report here that 1,3,5,7-
tetrabromoadamantane (7) reacts photochemically with so-
dium cyanide to produce 1,3,5,7-tetracyanoadamantane (1)
in good yield (Scheme 2).
Conversion of tetrabromoadamantane 7 to tetracyano-
adamantane 1 appears to occur by an SRN1 process.10b,16 No
reaction occurs in the dark, yet photolysis at 254 nm of a
0.03 M solution of 7 and 16 equiv of sodium cyanide in
DMSO (in quartz) in a Rayonet reactor produced a mixture
where tetracyanoadamantane 1 was the predominant product.
This was accompanied by small amounts of hypocyanated
haloadamantanes (mass spectrometric analysis was consistent
with the production of dibromodicyanoadamantane and
bromotricyanoadamantane) and others that appear to be the
result of decomposition of 1 under the reaction conditions.
Indeed, a sample of pure tetracyanoadamantane 1 decom-
posed slowly on photolysis to several unidentified products
that gave rise to 1H NMR signals comparable to those seen
in the reaction mixture of the photolysis of 7 and NaCN.
Our best yields in converting 7 to 1 resulted from photolysis
until only a small amount of 7 remained as indicated by TLC
(generally 5-6 h). The residue, after distillation of DMSO
at reduced pressure and removal of excess cyanide,17 was
chromatographed on silica gel. Removal of unreacted 7 by
elution with hexanes, followed by mixtures of dichloro-
methane and acetone, allowed isolation of 1 in 63% yield.
Alternatively, recrystallization of the crude residue from
Adamantane has been reported to react with 1 equiv of
aluminum trichloride and excess bromine to provide 1,3,5,7-
tetrabromoadamantane (7)12 in 47-58% yields. We found a
marked improvement in the yield (>90%) of tetrahalo-
adamantanes13 with the use of greater amounts AlCl3 (2, 3,
or 4 equiv), but with the generation of not only 7 but also
1
(7) 2‚4HCl: mp > 400 °C.; H NMR (300 MHz, D2O) δ 1.27 (s, 8H),
2.77 (s, 12H); 13C NMR (D2O) δ 39.00, 33.39, 48.98; IR (KBr) νmax 2923,
1596, 1520, 1459, 1376 cm-1
.
(13) 7: mp ) 246-248 °C (lit.12 245-247 °C); H NMR (CDCl3) δ
1
(8) Olah, G. A.; Wang, Q. Synthesis 1992, 1090.
(9) (a) Santiago, A. N.; Stahl, A. E.; Rodriguez, G. L.; Rossi, R. A. J.
Org. Chem. 1997, 62, 4406. (b) Kraus, G. A.; Siclovan, T. M. J Org. Chem.
1994, 59, 22.
2.71; 13C NMR (300 MHz, CDCl3) δ 54.62, 54.78; IR (KBr) νmax 487 cm-1
UV λmax) 262 nm.
;
(14) 8: 1H NMR (300 MHz, CDCl3) δ 2.55, 2.68; 13C NMR (DMSO)
δ 56.69, 55.69, 52.00, 51.00.
(10) (a) Wu, T.; Xiong, H.; Rieke, R. D. J. Org. Chem. 1990, 55, 5045.
(b) Lukach, A. E.; Rossi, R. A. J. Org. Chem. 1999, 64, 5836.
(11) Olah, G. A.; Farooq, O.; Prakash, S. Synthesis 1985, 1140.
(12) (a) Murray, R. W.; Rajadhyaksha, S. N.; Mohan, L. J. Org. Chem.
1989, 54, 5783. (b) Sollot, G. P.; Gilbert, E. E. J. Org. Chem.1980, 45,
5405.
(15) It has been reported that catalytic amounts of AlBr3 and Br2 react
with adamantane to produce 1,3,5-tribromoadamantane in low yield: (a)
Tolstikov, H. A.; Lerman, B. M.; Aref’eva, Z. Tetrahedron Lett. 1972, 31,
3191. (b) Baughman, G. L. J. Org. Chem. 1963, 29, 238.
(16) Lukach, A. E.; Santiago, A. N.; Rossi, R. A. J. Org. Chem. 1997,
62, 426.
1706
Org. Lett., Vol. 6, No. 11, 2004

acetone-water gave 1 as an off-white powder in 73% yield
and of similar purity as the chromatographed material.
Although DMSO absorbs in the photolysis region, and the
toxicity of cyanide in this solvent requires extreme care, we
have been unable to identify other solvents useful for this
transformation, largely because of the insolubility of the
cyanide and tetracyanoadamantane 1.
Useful photocyanation under these conditions appears to
be limited to tetrabromoadamantane (7). We prepared
tetrachloroadamantane 9 in 98% yield by a modified
literature procedure,18 and tetraiodoadamantane 10 in 91%
yield from 7 and iodomethane/AlBr3.19 Photolysis of 9 and
NaCN produced essentially no conversion to 1 after more
than 10 h. On the other hand, tetraiodoadamantane 10 reacted
within 1 h but led to a complex mixture of products, among
which was 1.
enolizable intermediates could be formed, thus avoiding
elimination pathways.
Reduction of tetranitrile 1 was best accomplished in our
hands by monochloroborane-methyl sulfide in refluxing
THF (with removal of methyl sulfide). Isolation of product
was accomplished by decomposition of remaining hydride
with multiple additions of excess methanol and evaporation
to remove the volatile borate esters and solvent. Reaction
with dry methanolic HCl produced the tetrakis(amino-
methyl)adamantane tetrahydrochloride [2‚4(HCl)] in 98%
yield.7 The parent tetraamine 2 was prepared by deprotona-
tion of an aqueous solution of 2‚4(HCl) with NaOH.21
The present work also represents a (formal) new synthesis
of the very useful 1,3,5,7-adamantanetetracarboxylic acid
(12), which has been previously prepared by multistep
processes that required assembly of the polycyclic-substituted
adamantane cage from monocyclic precursors.3b A more
efficient synthesis of derivatives of 12 has been reported to
occur from the photocarbonylation of adamantane mono- or
dicarboxylic acids, but the expense of starting materials and
modest product yields are limitations.4 We have converted
Rossi has previously described the SRN1 substitution of
adamantane bridgehead halogen (Scheme 3).20 Photochemical
reaction of iodoadamantane (13) with acetophenone enolate
resulted in good yields of 14 (X ) H). However, attempted
tetracyanoadamantane
1 to 1,3,5,7-tetracarbomethoxy-
adamantane (11) in 72% yield by solvolysis with methanolic
anhydrous HCl.22
Scheme 3
In conclusion, we have synthesized new and useful 1,3,5,7-
tetrasubstituted adamantanes 1 and 2 in good yields from
(inexpensive) adamantane. Since 1 can be solvolyzed to
1,3,5,7-tetra(carbomethoxy)adamantane (11), this also rep-
resents a formal synthesis of 1,3,5,7-adamantane tetra-
carboxylic acid (12), which has previously found wide
application. Finally, we have also improved procedures for
making 1,3,5,7-tetrahaloadamantanes 7, 9, and 10.
We will describe in other reports the incorporation of the
tetrasubstituted adamantanes 1 and 2 into several designed
materials.
disubstitution of 1,3-diiodoadamantane (15) resulted in cage
fragmentation products (presumably via the intermediacy of
the enolate 16 (X ) I) formed subsequently from the
monosubstitution product 14 (X ) I) and its elimination to
form 17. Our substitution of all four bridgehead positions in
tetrabromoadamantane 7 with cyanide was possible as no
Acknowledgment. We would like to acknowledge the
National Institutes of Health (NIGMS MBRS S06 GM08101)
for generous support of this work. C.G.G. and E.R.M. also
thank the Camille and Henry Dreyfus Foundation for a
Scholar-Fellow Award.
(17) Excess NaCN is washed away with water and decomposed with
bleach. 1: mp > 400 °C.; 1H NMR (300 MHz, DMSO-d6) δ 2.38 (s, 12H);
13C NMR (DMSO-d6) δ 29.37, 36.19, 120.96; IR (KBr) νmax 2240 cm-1
;
HRMS-FAB (m/z) [M+] calcd 236.1062, found 236.1630; UV λmax) 257
nm.
(18) It has been reported that adamantane and catalytic AlCl3 refluxed
in CCl4 for 3 days produce 9 in 55% yield: Bach, R. D.; Badger, R. C.
Synthesis 1979, 7, 529. We refluxed 1:2 adamantane/AlCl3 in CCl4 for 1 h
to produce 9: mp ) 191-193 °C (lit.18 194 °C); 1H NMR (300 MHz,
CDCl3) δ 2.35 (s, 12H); 13C NMR (CDCl3) δ 46.56, 53.58; IR (KBr) νmax
Supporting Information Available: Complete experi-
mental details and data characterizing all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
502 cm-1
.
(19) It has been reported that reaction of tetrabromoadamantane and Al
foil with bromine in methylene iodide produced 10 in 65% yield: McKinley,
J. W.; Pincock, R. E.; Scott, W. B. J. Am. Chem. Soc. 1973, 95, 2030. We
stirred 1: 1 adamantane: AlBr3 in iodomethane at room temperature for
15h to yield 10: mp >360 °C (lit 370 °C19); 1H NMR (300 MHz, CDCl3)
δ 3.22 (s, 12H); 13C NMR (300 MHz, CDCl3) δ 57.21, 59.22; IR (KBr)
OL036526G
(21) 2: 1H NMR (300 MHz, D2O) δ 0.89 (s, 8H), 2.11 (s, 12H); 13C
NMR (D2O) δ 32.65, 38.99, 50.26; 1H NMR (300 MHz, CDCl3) δ 1.09 (s,
8H), 2.45 (s, 12H); IR (cast) νmax 3646, 3444, 2914, 1556, 1320 cm-1
;
+
479 cm-1
.
HRMS-FAB (m/z) [M + H] calcd 253.2391, found 253.2388.
(20) Borosky, G.; Pierini, A. B.; Rossi, R. A. J. Org. Chem. 1990, 12,
3705.
(22) 11: 1H NMR (300 MHz, CDCl3) δ 3.65 (s, 12H), 2.66 (s, 12H); IR
(cast) νmax 2985, 2878, 1736, 1179 cm-1
.
Org. Lett., Vol. 6, No. 11, 2004
1707