An in-triphenylaminophane

Article abstract of DOI:10.1021/ol3014073

The synthesis and characterization of the triphenylamine-capped cyclophane 3 are described. It proved to be a conformationally rigid molecular propeller, with an inwardly pyramidalized, unreactive amine.

Full text of DOI:10.1021/ol3014073

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
An in-Triphenylaminophane
Shyam Sathyamoorthi, Joel T. Mague, and Robert A. Pascal Jr.*
Department of Chemistry, Tulane University, New Orleans, Louisiana 70118, United States
rpascal@tulane.edu
Received May 21, 2012
ABSTRACT
The synthesis and characterization of the triphenylamine-capped cyclophane 3 are described. It proved to be a conformationally rigid molecular
propeller, with an inwardly pyramidalized, unreactive amine.
For 25 years, we have pursued the synthesis of sterically
congested, mostly C3-symmetric in-cyclophanes (1, Figure 1)
bearing functional groups (XꢀY = CꢀH, CꢀMe, SiꢀH,
SiꢀF, and Pꢀlp) projected toward aromatic rings.1,2 Pro-
minent among these compounds are the in-phosphaphanes
2aꢀd, which possess unusual structure, spectra, and re-
activity: the shortest reported phosphorusꢀarene non-
bonded contact distances,1,3ꢀ5 strong spinꢀspin coupling
between the phosphorus atoms and the carbons of the
basal aromatic rings,3ꢀ5 strong circular dichroism of the
resolved enantiomers,5 and exceptional resistance to pro-
tonation or oxygenation of the in-phosphine.3,4 Only the
in-isomers of these phosphines have been observed, but the
out-isomer of a related cyclophane was trapped by sulfura-
tion of the phosphorus.6
pyramidalized and have high barriers to inversion,6,7 but
free triphenylamine has an essentially planar, sp2-hybridized
nitrogen atom,8 and even pyramidal amines have low
barriers to inversion. One may ask, will compound 3 exhibit
an inwardly pyramidalized nitrogen atom? Will 3 be con-
formationally flexible or relatively rigid? Will it be reactive?
Cyclophane 3 has long been elusive. The synthesis of the
phosphaphanes was facilitated by Block’s one-step synthe-
sis of tris(2-mercaptophenyl)phosphines,9 but this meth-
od is not applicable to amines. Earlier attempts to prepare
the key precursor of 3, tris(2-mercaptophenyl)amine (9,
Scheme 1), from the readily available10 tris(2-chlorophenyl)-
amine (5) invariably failed. In this paper we report at last a
four-step synthesis of the triphenylaminophane 3 and its
crystallographic and spectroscopic characterization.
The analogous aminophane 3 would provide interesting
points for comparison. For example, triarylphosphines are
(1) Review: Pascal, R. A., Jr. Eur. J. Org. Chem. 2004, 3763–3771.
(2) Selected examples: (a) Pascal, R. A., Jr.; Grossman, R. B.; Van
Engen, D. J. Am. Chem. Soc. 1987, 109, 6878–6880. (b) Pascal, R. A., Jr.;
Winans, C. G.; Van Engen, D. J. Am. Chem. Soc. 1989, 111, 3007–3010.
(c) Dell, S.; Ho, D. M.; Pascal, R. A., Jr. J. Org. Chem. 1999, 64, 5626–
5633. (d) Song, Q.; Ho, D. M.; Pascal, R. A., Jr. J. Am. Chem. Soc. 2005,
127, 11246–11247. (e) Qin, Q.; Mague, J. T.; Pascal, R. A., Jr. Org. Lett.
2010, 12, 928–930.
(3) Pascal, R. A., Jr.; West, A. P., Jr.; Van Engen, D. J. Am. Chem.
Soc. 1990, 112, 6406–6407.
(4) L’Esperance, R. P.; West, A. P., Jr.; Van Engen, D.; Pascal, R. A.,
Jr. J. Am. Chem. Soc. 1991, 113, 2672–2676.
(5) West, A. P., Jr.; Smyth, N.; Kraml, C. M.; Ho, D. M.; Pascal,
Figure 1. in-Cyclophanes.
R. A., Jr. J. Org. Chem. 1993, 58, 3502–3506.
(6) Chen, Y. T.; Baldridge, K. K.; Ho, D. M.; Pascal, R. A., Jr. J. Am.
Chem. Soc. 1999, 121, 12082–12087.
(7) Baechler, R. D.; Mislow, K. J. Am. Chem. Soc. 1970, 92, 3090–
3093.
Nucleophilic substitution of the chlorine atoms of 5 with
sodium isopropylthiolate to give thioether 8 appears to fail
r
10.1021/ol3014073
XXXX American Chemical Society

due to steric impediments. However, the Ullmann synthe-
sis of amines is relatively insensitive to crowding; thus we
decided to prepare 8 from precursors already containing
the bulky isopropyl thioethers. Treatment of bis(2-chloro-
phenyl)amine (4), available as an abundant byproduct of
the synthesis of 5,10 with sodium isopropylthiolate gave
thioether 6 in fair yield after purification (28%). A similar
monosubstitution of o-diiodobenzene gave the thioether 7.
Ullmann reaction of 6 and 7 in refluxing NMP produced
only N-arylphenothiazines (data not shown), but when the
reaction was carried out under the milder condition of
refluxing DMF, the crowded thioether 8 was obtained in a
low but tolerable 14% yield.11 The crowding in 8 is evident
in its 1H NMR spectrum, which shows broadened methyl
resonances due to the slow interconversion of diastereo-
meric conformations of the molecule. Finally, reductive
cleavage of the thioethers by sodium in ammonia gave the
key trithiol 9 in good yield (73%).
low yield stands in contrast to the much higher, but still
modest, yields (typically 15ꢀ20%) for the syntheses of
phosphaphanes 2.3ꢀ5
The 1H NMR spectrum of 3 is similar in most respects to
that of the phosphaphane 2a. However, 2a displays two
distinct resonances for its diastereotopic methylene
protons,3 due to the lack of enantiomerization of the
molecular propeller at room temperature, but the methy-
lene protons of 3 exhibit a single sharp resonance at δ 3.78
in CDCl3, suggesting that 3 is conformationally much
more mobile than 2a. This at first seemed reasonable, given
that easy inversion at nitrogen provides a conformational
degree of freedom not available to the phosphaphanes.
However, no low-energy pathway for enantiomerization
was found in our computational studies (see below), pointing
to a possible accidental isochrony of the diastereotopic
methylene resonances in 3. Indeed, when the spectrum was
recorded in benzene-d6, two sharp doublets were observed at
δ 3.55 and δ 3.67, clearly indicating that enantiomerization
of 3 is slow (at least on the NMR time scale). Other
spectroscopic data provide little to distinguish the properties
of compounds 3 and 2a. For example, given the difference in
the UV spectra of triphenylamine (λmax = 303 nm in CHCl3)
and triphenylphosphine (260 nm), the UV spectra of 3 and
2a are surprisingly similar: in CHCl3, aminophane 3 has
strong absorption bands at 292 and 336 nm, while phospha-
phane 2a displays absorptions at 292 and 341 nm,4 with
comparable extinction coefficients.
Scheme 1. Synthesis of Cyclophane 3
An X-ray structural analysis of compound 3 was thus of
prime importance. Single crystals of 3 were obtained by
addition of methanol to an NMR sample in CDCl3, and
the crystals yielded a high-quality determination without
complications. The structures of the two crystallographi-
cally independent molecules of 3 are illustrated in Figure 2;
both have approximate C3 symmetry, as expected. The
shallow inward pyramidalization of the apical nitrogen
atoms is immediately apparent, but to a degree that is
much less pronounced than in the phosphaphanes 2. Thus,
the average pyramidality14 of the triarylamines in the two
˚
molecules of 3 is 0.248 A, but the average pyramidality of
˚
the phosphaphanes is 0.741 A. As noted previously, tri-
phenylamine’s nitrogen is not pyramidalized; however, a
search of the Cambridge Structural Database15 found that
triarylamines with three ortho-substituted benzene rings16
are pyramidalized to nearly the same degree (average
˚
pyramidality, 0.235 A) as compound 3. With such a
flattened amine in 3, it is not surprising that the distance
Condensation of 9 with 1,3,5-tris(bromomethyl)benzene13
(10) gave cyclophane 3 in an abysmal 0.7% yield, but the
product was easily isolated by chromatography. This very
of the apical nitrogen from the center of the basal aromatic
˚
ring (dNꢀAr), 3.419 and 3.398 A in the two independent
˚
molecules, is nearly 0.5 A greater than the phosphorusꢀ
(8) Sobolev, A. N.; Belsky, V. K.; Romm, I. P.; Chernikova, N. Yu.;
Guryanova, E. N. Acta Crystallogr., Sect. C 1985, C41, 967–971.
(9) Block, E.; Ofori-Okai, G.; Zubieta, J. J. Am. Chem. Soc. 1989,
111, 2327–2329.
(14) Glaser, R.; Blount, J. F.; Mislow, K. J. Am. Chem. Soc. 1980,
102, 2777–2786.
(10) Frye, C. L.; Vincent, G. A.; Hauschildt, G. L. J. Am. Chem. Soc.
1966, 88, 2727–2730.
(15) (a) Allen, F. H.; Kennard, O.; Taylor, R. Acc. Chem. Res. 1983,
16, 146–153. (b) Allen, F. H. Acta Crystallogr., Sect. B 2002, B58, 380–
388.
(11) We are not unaware of modern methods for the synthesis of
arylamines, but, for example, when Pd(OAc)2/DPEphos/NaOtBu/-
toluene12 was used for the coupling, no triarylamine was formed.
(12) Sadighi, J. P.; Harris, M. C.; Buchwald, S. L. Tetrahedron Lett.
1998, 39, 5327–5330.
(16) (a) Field, J. E.; Combariza, M. Y.; Vachet, R. W.; Venkataraman,
€
€
D. Chem. Commun. 2002, 2260–2261. (b) Muller, E.; Burgi, H.-B. Acta
Crystallogr., Sect. C 1989, C45, 1400–1403. (c) Stoudt, S. J.; Gopalan, P.;
Bakulin, A.; Kahr, B.; Jackson, J. E. Inorg. Chem. 1996, 35, 6614–6621. (d)
Kelly, B. V.; Tanski, J. M.; Anzovino, M. B.; Parkin, G. J. Chem. Cryst.
2005, 35, 969–981.
(13) Vogtle, F.; Zuber, M.; Lichtenthaler, R. G. Chem. Ber. 1973,
106, 717–718.
B
Org. Lett., Vol. XX, No. XX, XXXX

out-isomer, whether with C3 or C1 symmetry, were unsuc-
cessful. This represents a significant difference from the in-
phosphaphanes, which possess high-energy out-isomers.6
There exists insufficient compound 3 at present to
attempt a resolution into pure enantiomers, but DFT
calculations suggest that the barrier to racemization is very
high. A Cs-symmetric transition state structure was located
at the B3LYP/6-31G(d) level for the direct interconversion
of the two enantiomeric C3 ground state conformations;
however, the calculated free energy of this transition state
gives ΔG‡rac = 85 kcal/mol! A more reasonable pathway
for enantiomerization is a two-step sequence, C3 to C1 to
enantiomeric C3 conformations, for which the calculated
transition state free energies are “merely” 44 and 63 kcal/mol
(see the Supporting Information). We do not exclude the
possibility of lower energy pathways, but compound 3 is
almost certainly resolvable.
Finally, there is the question of protonation of com-
pound 3. Triphenylamine itself is far less basic than tri-
phenylphosphine (for the protonated species, the pKa’s are
ca. ꢀ5 and 2.7, respectively). Furthermore, at the B3LYP/
6-31G(d) level, the out-protonated amine is calculated to
be a full 26 kcal/mol higher in energy than the sterically
inaccessible in-protonated species, due chiefly to severe
angle strain in the out-isomer. In the event, dry HCl gas was
bubbled into a solution of 3 in CDCl3 for 30 s, and the
sample was left to stand for 2 h. 1H NMR analysis at this
time showed no change in the cyclophane resonances, and
after 24 h, the spectra showed evidence of extensive decom-
position and still no evidence of a protonated cyclophane.
In conclusion, the substitution of nitrogen for phos-
phorus in a triaryl-element-capped cyclophane results in
reduced inward pyramidalization of the apical nitrogen,
but the compound remains a rigid, unreactive, molecular
propeller.
Figure 2. Molecular structures of the two crystallographically
independent molecules of cyclophane 3 (50% thermal ellipsoids
have been employed).
˚
arene contacts of 2.90ꢀ2.98 A observed in the various
phosphaphanes 2.3ꢀ5
DFT-computed structures of 3 accurately mimic its
overall crystal conformation and the shallow pyramidali-
zation of the nitrogen. These calculations do overestimate,
to varying degrees, the nitrogenꢀarene contact distance,
Acknowledgment. This work was supported by Na-
tional Science Foundation Grant CHE-0936862, which is
gratefully acknowledged.
˚
with B3PW91/6-311G(d,p) calculations (dNꢀAr = 3.437 A)
giving the best agreement with experiment among many
methods.17 An important aspect of these calculations was
the finding of only two potential minima for compound
3, one with the observed C3 symmetry, and a higher
energy conformation [þ11.0 kcal/mol at the B3LYP/
6-31G(d) level] with C1 symmetry. All attemptstolocatean
Supporting Information Available. Experimental pro-
cedures and NMR spectra for compounds 6ꢀ9 and 3;
DFT-calculated pathways for the enantiomerization of 3;
a crystallographic information file (CIF) for the structure
determination of 3; and an ASCII text file containing the
atomic coordinates and energies of the calculated struc-
tures of 3. This material is available free of charge via the
Internet at http://pubs.acs.org.
(17) Other methods gave the following distances: B3LYP/6-31G(d),
˚
3.515 A; B3LYP/6-311G(d,p), 3.502 A; B3LYP/cc-pVDZ, 3.503 A;
˚
˚
˚
B3LYP/cc-pVTZ, 3.516 A; B3PW91/6-31G(d), 3.459 A; B3PW91/
˚
˚
cc-pVDZ, 3.438 A; B3PW91/cc-PVTZ, 3.460 A.
˚
The authors declare no competing financial interest.
Org. Lett., Vol. XX, No. XX, XXXX
C