Highly substituted cyclohexanes: Strong proximity effects influence synthetic access to 1,3,5-tris(bromomethyl)-1,3,5-trialkylcyclohexanes (alkyl = methyl, n-propyl)

Article abstract of DOI:10.1016/j.tetlet.2006.02.033

1,3,5-Tris(bromomethyl)-1,3,5-trialkylcyclohexanes (alkyl = methyl, n-propyl) were prepared. These are the first examples of 1,3,5-tris(halomethyl)- 1,3,5-trialkylcyclohexanes. One synthetic method directly converted the corresponding triols with PPh

Full text of DOI:10.1016/j.tetlet.2006.02.033

Tetrahedron Letters 47 (2006) 2607–2610
Highly substituted cyclohexanes: strong proximity effects
influence synthetic access to 1,3,5-tris(bromomethyl)-
1,3,5-trialkylcyclohexanes (alkyl = methyl, n-propyl)
Andreas Hofmann,a Rui Ren,a Alan Loughb and Ulrich Fekla,*
aUniversity of Toronto, UTM, Mississauga, Ontario, Canada L5L 1C6
bUniversity of Toronto, X-ray Facility, 80 St. George Street, Toronto, Ontario, Canada M5S 3H6
Received 23 December 2005; revised 2 February 2006; accepted 6 February 2006
Available online 24 February 2006
Abstract—1,3,5-Tris(bromomethyl)-1,3,5-trialkylcyclohexanes (alkyl = methyl, n-propyl) were prepared. These are the first exam-
ples of 1,3,5-tris(halomethyl)-1,3,5-trialkylcyclohexanes. One synthetic method directly converted the corresponding triols with
PPh3Br2, where an excess of the bromination reagent and high temperature (175 ꢀC) were required. Stoichiometric use of PPh3Br2
under mild conditions, successfully employed for the synthesis of the parent tris(bromomethyl)cyclohexane, did not lead to the
desired tribromides but rather to cyclic ethers. Proximity effects triggered by the 1,3,5-alkyl groups strongly influence the reactivity
of such highly substituted cyclohexanes. An alternative synthetic access to the tris(bromomethyl) compounds was also developed,
using 1,3,5-tris(triflatomethyl)-1,3,5-trialkylcyclohexanes (triflato = F3CSO3) as synthetic intermediates. An X-ray crystal structure
of 1,3,5-tris(bromomethyl)-1,3,5-trimethylcyclohexane was obtained.
ꢁ 2006 Elsevier Ltd. All rights reserved.
1,3,5-Trisubstituted cyclohexanes have proven tremen-
dously useful as scaffolds for the preparation of complex
molecular architectures, such as clefts for molecular
recognition,1 dendrimers,2 and polycyclic cages.1a,3 The
formation of cage-type structures typically involves that
a linker bridges substituents in axial position of the
cyclohexane ring. For this purpose, trifunctional cyclo-
hexanes bearing additional alkyl substituents at the
1,3,5-carbons are particularly useful: cis,cis-1,3,5-tri-
methylcyclohexane-1,3,5-tricarboxylic acid (‘Kemp’s
triacid’)4 is probably the best-known example for such
1,3,5-trifunctional 1,3,5-trialkylcyclohexanes. These
alkyl-substituted derivatives undergo reactions leading
to cage-type structures generally more easily than the
parent compounds,3a,c,e where the bonds leading to
closure of the cage might be covalent bonds,3c metal–
ligand interactions,3e or hydrogen bonds.3a This can
reasonably be attributed to the effect the alkyls exert
on conformational equilibria.3a,c,e Conformations
involving axial functional groups, generally disfavored
for an unsubstituted system (Scheme 1A), might become
more favored with the help of 1,3,5-alkyls (Scheme 1B),
such that the functional groups (X) are forced into
closer proximity. While this effect is often desirable, it
regularly makes synthetic procedures rather challenging.
For example, cis,cis-1,3,5-triaminocyclohexane has been
known since 1957, but the synthesis of cis,cis-1,3,5-tri-
methylcyclohexane-1,3,5-triamine was reported only in
2002, and complications due to an unusually stable
intramolecular acid anhydride were encountered during
its synthesis.3a
X
X
X
X
A)
X
X
Substituents (X) in axial position disfavored
X
X
R
X
R
X
R
R
X
B)
X
Keywords: Proximity effect; Cyclohexane; Axial; Equatorial; Alkyl
R
R
halide.
*
Corresponding author. Tel.: +1 905 828 3806; fax: +1 905 828
5425; e-mail addresses: alough@chem.utoronto.ca; ufekl@utm.
utoronto.ca
Substituents (X) in axial position more favored
Scheme 1.
0040-4039/$ - see front matter ꢁ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.02.033

2608
A. Hofmann et al. / Tetrahedron Letters 47 (2006) 2607–2610
Br
Mg
OH
Br
H
HBr or
H
H
Mg
PPh + Br2/PPh3Br2
3
Br
HO
H
Br
Mg
H
H
H
H
H
OH
Br
Mg
-H
1
2-H
3-H
Br
Scheme 2.
We are interested in cyclohexane-based alkyl trihalides.
1,3,5-tris(halomethyl)cyclohexanes are important inter-
mediates for the synthesis of cage-type compounds.
For example, 1,3,5-tris(bromomethyl)cyclohexane was
recently used to prepare a rare example of a tris-Grig-
nard reagent (compound 3-H in Scheme 2), useful for
the synthesis of silaadamantane cages.3f,g Given the
usefulness of alkyl halides as synthetic intermediates, it
is notable that this chemistry has not yet been extended
to the 1,3,5-trialkyl-substituted systems. In fact, we were
surprised to find that no 1,3,5-tris(halomethyl)-1,3,5-tri-
alkylcyclohexanes (Scheme 1B, X = –CH2-halogen, R =
alkyl) have been reported so far. In this contribution, we
report on the preparation of the first examples of such
compounds, using two different strategies.
R
decomposition
HBr
OH
O
R
3 eq
of PPh3Br2
R
R
Br
4-R
HO
R
4.5 eq of
PPh3Br2
R
Br
OH
175 o
C
1-R (R = Me, n-Prop)
R
TfOTf,
py
Br
R
OTf
R
R
(NEt4)Br
2-R
Br
Transformations on the unsubstituted 1,3,5-tris-
(hydroxymethyl)cyclohexane (1-H) are straightforward
and proceed as shown in Scheme 2. The conversion of
1-H into the 1,3,5-tris(bromomethyl)cyclohexane (2-H)
has been classically performed using PPh3 and Br2 (an
equivalent reagent is Ph3PBr2) but significant difficulties
in removing the triphenylphosphine oxide by-product
have been noted.3g We found that 2-H is easily made
by refluxing 1-H in 48% HBr overnight.5 A remarkable
transformation on the tribromide 2-H has been reported
recently, namely, the conversion into a tris-Grignard
compound 3-H, a very rare compound of a triply meta-
lated aliphatic framework.3f,g We found that the
reported formation of a tris-Grignard compound is
easily reproduced.
TfO
R
R
5-R
OTf
Scheme 3.
phosphorus, is being attacked, in a nucleophilic fashion,
by an adjacent CH2OH. This intramolecular attack
seems to be much faster than attack by external bro-
mide. This behavior contrasts with the reactivity of 1-
H, and we propose that the difference lies in the position
of the conformational equilibrium between structures
having CH2OH in either equatorial or axial position.
Such effects would not necessarily be visible in solid state
structures, and in the X-ray structure of compound 1-
Me, the CH2OH groups are still in equatorial posi-
tions.3e When we varied the reaction conditions, we
found that the desired tribromides 2-R are formed in
34–35% yield during 4 h if 50% excess of Ph3PBr2 (molar
ratio 4.5:1) and high temperature conditions (175 ꢀC)
are used.9 These reaction conditions are optimized,
and while low temperatures lead to incomplete bromin-
ation, higher temperatures lead to decomposition.
We tried to perform similar transformations on the
1,3,5-trialkyl derivatives 1-Me and 1-Prop (Scheme 3,
compounds 1-R).6 However, we found that the latter
compounds exhibit reactivity very different from that
of 1-H. Reacting compounds 1-R (R = Me, n-Prop)
with HBr did not form the desired tribromides 2-R,
but led to complete decomposition instead. The reaction
with Ph3PBr2, however, led to the following observa-
tions: room temperature reactions where 1-R were
reacted with a slight excess of PPh3Br2 (1:3.2 molar
ratio), yielded a mixture of compounds, which contained
no 2-R. We were able to isolate the bicyclic esters 4-R
(Scheme 3) as the main products, unambiguously char-
acterized by NMR spectroscopy.7 Clearly, the alkyl
groups in the 1,3,5-positions exert a profound effect on
the reactivity of the hydroxyls, by facilitating intra-
molecular ring-closure reactions.8 Apparently, the ester
is formed when one CH2–O functionality, activated by
An alternative access to 2-R was also developed. We
reasoned that care should be taken since once a good
leaving group is formed at one position, nucleophilic
attack by an adjacent group should be prevented. This
might be achievable by efficient functionalization of
the OH groups at low temperature, and the triflate
species 5-R are promising synthetic intermediates. We
prepared the known3e compound 5-Me, along with the
new compound 5-Prop.10 The tris-triflates 5-R were
reacted with three equivalents of (NEt4)Br in CHCl3

A. Hofmann et al. / Tetrahedron Letters 47 (2006) 2607–2610
2609
to obtain 2-R.11 These reactions are less straightforward
than envisioned, since the tris-triflates 5-R are tempera-
ture-sensitive, decomposing within days at room
temperature (even under inert atmosphere). Unfortu-
nately, elevated temperature is needed to facilitate the
nucleophilic attack by bromide, and the reaction is
essentially a competition between decomposition and
SN2 substitution. Thus, it seems unlikely that this
method can be optimized to give high yields of 2-R.
However, a small amount of very pure 2-Me was
obtained by means of slow recrystallization from crude
2-Me, prepared by this method. This crystalline material
was suitable for X-ray crystallography.
117.4 and 119.9, as well as endocyclic torsions of 48.5
and 47.3 were previously observed for compounds
1-Me and for cis,cis-1,3,5-tris[(diphenylphosphanyl)-
methyl]-1,3,5-trimethylcyclohexane, respectively.3e
In conclusion, we have synthesized and characterized
1,3,5-tris(bromomethyl)-1,3,5-trialkylcyclohexanes (alk-
yl = methyl, n-propyl), the first examples of the class
of 1,3,5-tris(halomethyl)-1,3,5-trialkylcyclohexanes. The
target compounds were obtained, despite the fact that
the alkyl groups in 1,3,5-positions render bromination
reactions of the triols unusually challenging. We antici-
pate that the new compounds described here are very
useful precursors for the synthesis of novel hydrocar-
bon-based cages.
In order to gain insight into the structural features of
the 1,3,5-tris(bromomethyl)-1,3,5-trialkylcyclohexanes,
a single-crystal X-ray structure determination of 2-Me
was undertaken (Fig. 1).12 The compound adopts a
chair conformation in the solid state, where the methyl
groups occupy the axial positions and the bromomethyl
groups occupy the equatorial positions. This can reason-
ably be expected from the relative size of these groups.
The C–Br vectors are parallel to C–C bonds in the cyclo-
hexane framework, and anti-geometry is observed for
those bonds. While this might be an indication of hyper-
conjugation, packing effects cannot be excluded, of
course. Compared to unsubstituted cyclohexane, the
cyclohexane ring in this heavily substituted compound
undergoes flattening. Steric repulsion enforces widening
of the endocyclic bond angles around CH2, leading to a
value of 117.3(6) (average of three values, see legend to
Fig. 1) for these angles, compared to the literature
value13 of 111.5ꢀ for cyclohexane. The flattening
becomes manifested in the six endocyclic torsion angles,
where the average of the absolute values is 49(1)ꢀ
compared to the cyclohexane value of 55ꢀ. The values
we observe seem typical for 1,1-3,3-5,5-hexasubstituted
cyclohexanes: endocyclic bond angles to methylene of
Acknowledgments
Funding by the Natural Science and Engineering
Research Council (NSERC) of Canada, the Canadian
Foundation of Innovation, and the University of Toron-
to (Connaught Foundation) is gratefully acknowledged.
We thank Dr. Gordon Hamer and Dr. Peter Mitrakos
for assistance with the NMR instruments. We thank
Mr. Asad A. Merchant and Mr. Jimmy Kung for assist-
ing in the synthesis of triols 1-H, 1-Me, and 1-Prop.
References and notes
1. Examples: (a) Jeong, K.-S.; Kim, J. H. Bull. Korean Chem.
Soc. 1996, 17, 564; (b) Conn, M. M.; Deslongchamps, G.;
de Mendoza, J.; Rebek, J., Jr. J. Am. Chem. Soc. 1993,
115, 3548; (c) Jeong, K. S.; Tjivikua, T.; Muehldorf, A.;
Deslongchamps, G.; Famulok, M.; Rebek, J., Jr. J. Am.
Chem. Soc. 1991, 113, 201; (d) Askew, B.; Ballester, P.;
Buhr, C.; Jeong, K. S.; Jones, S.; Parris, K.; Williams, K.;
Rebek, J., Jr. J. Am. Chem. Soc. 1989, 111, 1082; (e)
Rebek, J., Jr.; Askew, B.; Killoran, M.; Nemeth, D.; Lin,
F.-T. J. Am. Chem. Soc. 1987, 109, 2426.
2. Jang, W.-D.; Aida, T. Macromolecules 2004, 37, 7325.
3. (a) Menger, F. M.; Bian, J.; Azov, V. A. Angew. Chem.,
Int. Ed. 2002, 41, 2581; (b) Wrackmeyer, B.; Milius, W.;
Klimkina, E. V.; Bubnov, Y. N. Chem. Eur. J. 2001, 7,
775; (c) Izumi, H.; Futamura, S. J. Org. Chem. 1999, 64,
4502; (d) Mayer, H. A.; Otto, H.; Kuhbauch, H.; Fawzi,
¨
R.; Steimann, M. J. Organomet. Chem. 1994, 472, 347; (e)
Mayer, H. A.; Fawzi, R.; Steimann, M. Chem. Ber. 1993,
126, 1341; (f) Boudjouk, P.; Sooriyakumaran, R.; Kapfer,
C. A. J. Organomet. Chem. 1985, 281, C21; (g) Boudjouk,
P.; Kapfer, C. A.; Cunico, R. F. Organometallics 1983, 2,
336.
4. Kemp, D. S.; Petrakis, K. S. J. Org. Chem. 1981, 46, 5140.
5. Compound 1-H (500 mg) in 60 mL of 48% aqueous HBr.
The product mixture was extracted with 3 · 25 mL of
hexane, and the organic phase dried over MgSO4.
Removal of solvent under vacuum yeilded crystallize
crude product. After recrystallization from MeOH,
331 mg (32%, yield not optimized) of pure 2-H were
Figure 1. Thermal ellipsoid plot (30% probability) for 2-Me. Selected
˚
distances (A) and angles (ꢀ): C–Br: C10–Br1, 1.946(5); C12–Br2,
1.957(6); C8–Br3, 1.979(5); exocyclic angles: C7–C1–C8, 107.4(4);
C10–C3–C9, 108.2(4); C11–C5–C12, 108.8(4); endocyclic angles to
quaternary carbons: C2–C1–C6, 109.0(4); C2–C3–C4, 110.5(4);
C4–C5–C6, 108.9(4); endocyclic angles to methylenes: C1–C2–C3,
117.8(4); C5–C4–C3, 116.7(4); C5–C6–C1, 117.3(4); torsion angles:
Br3–C8–C1–C6, ꢀ175.8(6); Br1–C10–C3–C2, 179.6(6); Br2–C12–C5–
C6, 179.1(6); average of the absolute values of the six endocyclic
torsions: 49(1).
1
obtained; H NMR identical to literature data.3g
6. Compound 1-Prop was synthesized using LiAlH4 reduc-
tion of the commercially available (Aldrich) trimethyl-
cis,cis-1,3,5-tripropyl-1,3,5-cyclohexane-tricarboxylate,
analogously to the reported synthesis3g of 1-H. Compound
1-Me was similarly obtained from the trimethyl ester of

2610
A. Hofmann et al. / Tetrahedron Letters 47 (2006) 2607–2610
cis,cis-1,3,5-trimethyl-1,3,5-cyclohexanetricarboxylic acid,
CHCl3, and 0.563 mL (6.97 mmol) of pyridine were added.
The solution was cooled to ꢀ35 ꢀC, and 1.165 mL
(6.90 mmol) of trifluoromethanesulfonic anhydride were
slowly added to this cold solution. The solution was
stirred for 2 h at room temperature, and a precipitate
formed. The precipitate was removed by filtration, and the
brown solution passed through a column of silica gel. The
resulting yellow solution was used for the next step. In
order to establish purity, the solvent was removed from a
sample of this solution under vacuum, and CDCl3 was
added by vacuum transfer. The 1H NMR spectrum proved
identical to that previously reported for 5-Me.3e Com-
pound 5-Prop: this compound was synthesized in a
completely analogous fashion, and characterized by 1H
prepared by esterification of Kemp’s triacid, using the
two-step esterification procedure (SOCl2, followed by
ROH) described in: Weygand, F.; Klinke, P.; Eigen, I.
Chem. Ber. 1957, 90, 1896.
7. Compound 4-Me3: 1H NMR (200 MHz, CDCl3) d 3.83 (s,
2
2H, CH2Br), 3.35 (d, br, 2H, J = 11 Hz, CHH0–O), 2.89
2
(d, br, 2H, J = 11 Hz, CHH0–O), 1.71 (d, apparent
JHH = 14 Hz, 3H, CHaHe, CH2 of cyclohexane ring)
1.07 (d, apparent JHH = 16 Hz, 3H, CHaHe, CH2 of
cyclohexane ring), 0.95 (s, 3H, CH3), 0.613 (s, 6H, CH3).
Compound 4-Prop: 1H NMR (200 MHz, CDCl3) d 4.06
(s, 2H, CH2Br), 3.54 (dd, 2H, 2J = 11 Hz, apparent
2
4J = 1.6 Hz, CHH0–O), 3.10 (dd, 2H, J = 11 Hz, appar-
4
1
ent J = 1.9 Hz, CHH0–O), 1.0–1.8 (m, 18H, CH2 of
NMR spectroscopy: H NMR (200 MHz, THF-d8) d 4.35
cyclohexane ring+CH2 of isopropyls), 0.86 (t, 9H,
(s, 6H, CH2OTf), 1.1–1.9 (m, 18H, CH2 of cyclohexane
ring+CH2 of isopropyls), 0.93 (t, 9H, J = 7 Hz, CH3 of
2
2J = 7 Hz, CH3 of isopropyls).
8. A somewhat similar proximity effect was recently observed
when an undesired intramolecular ester was obtained in a
system having methyl groups in 1,3,5-positions.3c
isopropyls).
11. Compounds 2-Me and 2-Prop from 5-Me and 5-Prop: all
reactions were performed under inert conditions (Ref. 9).
Compound 2-Me: 1.457 g (6.93 mmol) of tetraethyl-
ammonium bromide were added to a CHCl3 solution of
tris-triflate 5-Me (2.31 mmol). The solution was heated to
75 ꢀC overnight, under stirring, in a closed pyrex reaction
vessel. The CHCl3 was removed under vacuum, and
25 mL of dry hexane were added via vacuum transfer. The
organic salts were removed by filtration, and the solution
was concentrated under vacuum. Yellow oil was obtained,
which partly crystallized. After 10 days at ꢀ35 ꢀC, a small
amount (62 mg, 6.6%) of very pure crystalline material,
suitable for X-ray crystallography, was obtained. This
procedure was analogously applied to synthesize
2-Prop. Both products were identified by NMR spectro-
scopy (full characterization given above (Ref. 9)).
Compound 2-Me was additionally characterized by
single-crystal X-ray crystallography. Compound 2-Prop
did not crystallize.
9. Compound 2-R from 1-R and Ph3PBr2: The reactions
were performed using dry solvents and air-free conditions,
using glove-box and high-vacuum line techniques. Syn-
thesis of 2-Me: 83 mg (0.384 mmol) of 1-Me and 728.8 mg
(1.727 mmol) of PPh3Br2 were suspended in 3 mL of dry
CH3CN and sonicated for 30 min at room temperature.
The mixture was stirred at 175 ꢀC for 4 h under nitrogen
(closed pyrex reaction vessel), after which the solvent was
removed under vacuum to yield brown oil. The mixture
was dissolved in hexane (30 mL) and stirred for 30 min.
Removal of triphenylphosphine oxide by filtration, fol-
lowed by evaporation of hexane, yielded 53.1 mg (34.2%)
of product, pure by NMR, in the form of an oil.
1H NMR (500 MHz, CDCl3) d 3.26 (s, 6H, CH2Br),
1.41 (‘AB’ multiplet, 6H, CH2), 1.24 (s, 9H, CH3). 13C
NMR (125.9 MHz, CDCl3): d = 27.0, 35.3, 43.4,
50.1 ppm. MS (EI), m/e 387, 323, 308, 244, 229. Synthesis
of 2-Prop: 140.7 mg (0.468 mmol) of 1-Prop and 982.9 mg
(2.33 mmol) of PPh3Br2 were subjected to an analogous
procedure as described for 2-Me above, to yield 80.7 mg
(35.2%) of product. 1H NMR (500 MHz, CDCl3) d 3.36 (s,
6H, CH2Br), 1.1–1.7 (m, 18H, CH2 of cyclohexane
ring+CH2 of isopropyls), 0.91 (t, 9H, 2J = 7.2 Hz, CH3
of isopropyls) 13C NMR (125.9 MHz, CDCl3): d = 14.53,
16.21, 37.2, 39.9, 43.3, 45.4 ppm. MS: m/e 486, 443, 407,
393, 363, 328. HRMS calcd for C18H33Br3 486.013, found
486.013.
12. Crystals were obtained as described above (Ref. 11).
C12H21Br3, MW = 405.0, clear cube, monoclinic, space
˚
group = P21, T = 150(1) K, a = 7.8052(4) A, b =
˚
˚
11.8722(6) A, c = 8.0925(4) A, b = 106.853(3)ꢀ, Z = 2, R1
(I > 2r(I)) = 0.0323, wR2 (all data) = 0.0784, GOF(F2) =
1.04. Crystallographic data (excluding structure factors)
have been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication number CCDC
293253. Copies of the data can be obtained, free of charge,
on application to CCDC, 12 Union Road, Cambridge
CB2 1 EZ, UK (fax: +44(0) 1223 336033 or e-mail:
deposit@ccdc.cam.ac.uk).
10. Compounds 5-Me and 5-Prop: 5-Me: all reactions were
performed under inert conditions (Ref. 9). Compound 1-
Me (500 mg, 2.31 mmol) was suspended in 25 mL of
13. Mann, G. Z. Chem. 1990, 30, 1.