Sterically congested macrobicycles with heteroatomic bridgehead functionality

Article abstract of DOI:10.1016/j.tet.2013.10.018

A series of cyclophanes composed of two triarylelement caps linked by two-atom bridges has been synthesized. The bridgehead functional groups include phosphines in combination with amines, hydrosilanes, methylsilanes, and ethoxysilanes. Computational studies accurately predicted that when the bridgehead substituents are small (lone pairs or protons), an in_,in bridgehead stereochemistry is strongly favored, but larger bridgehead substituents favor the formations of in_,out stereoisomers. The X-ray structures, spectra, and reactivity of these compounds are discussed, as well as the resolution of one of the cyclophanes into pure enantiomers.

Full text of DOI:10.1016/j.tet.2013.10.018

Tetrahedron 69 (2013) 10316e10321
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Sterically congested macrobicycles with heteroatomic bridgehead
functionality
,
Jie Zong a, Joel T. Mague a, E. Celeste Welch b, Isaac M.K. Eckert b, Robert A. Pascal, Jr. a
*
a Department of Chemistry, Tulane University, New Orleans, LA 70118, USA
b Lotus Separations LLC, Princeton, NJ 08544, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 31 August 2013
Received in revised form 4 October 2013
Accepted 7 October 2013
Available online 14 October 2013
A series of cyclophanes composed of two triarylelement caps linked by two-atom bridges has been
synthesized. The bridgehead functional groups include phosphines in combination with amines, hy-
drosilanes, methylsilanes, and ethoxysilanes. Computational studies accurately predicted that when the
bridgehead substituents are small (lone pairs or protons), an in,in bridgehead stereochemistry is strongly
favored, but larger bridgehead substituents favor the formations of in,out stereoisomers. The X-ray
structures, spectra, and reactivity of these compounds are discussed, as well as the resolution of one of
the cyclophanes into pure enantiomers.
Keywords:
Cyclophanes
In/out isomerism
Non-bonded interactions
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
We recently reported the synthesis of the macrobicyclic in,in-
bisphosphine 1, a molecule with two interacting bridgehead
phosphines that show strong spinespin coupling and resist re-
action with reagents larger than protons.1 The in,in stereochemistry
of compound 1 was hardly unexpected, based on the conforma-
tional restrictions imposed by the six aryl rings and the previously
observed conformational preferences of other sterically congested
in-cyclophanes,2 but the incorporation of larger bridgehead func-
tionality must eventually lead to the formation of in,out or even
out,out stereoisomers.3 Indeed, while the preparation of 1 was
a logical extension of our studies of triarylphosphine-containing
cyclophanes,4e8 we have also made cyclophanes capped with tri-
arylamines,9 triarylhydrosilanes,5 and triarylmethylsilanes;7 thus
macrobicycles containing mixed bridgehead heteroatoms might be
prepared from precursors already in hand.
P
X
P
S
S
S
S
S
S
Y
P
1
2 X = N, Y = lp
3 X = Si, Y = H
4 X = Si, Y = Me
5 X = Si, Y = OEt
Scheme 1.
2. Results and discussion
We have long been interested in the effects of strain and con-
gestion on the spectra and reactivity of various functional groups in
cyclophane structures, and we now report the synthesis and
characterization of four new macrobicycles (Scheme 1, compounds
2e5), each containing mixed bridgehead functionality, which dis-
play both in,in and in,out geometries, and which possess unusual
structures, spectra, and reactivity.
2.1. Computational evaluation of possible macrobicyclic in/
out isomers
Diphosphine 1 contains two strongly pyramidalized bridgehead
atoms, and B3PW91/6-31G(d) calculations10,11 indicate that an in,in
geometry is favored by more than 18 kcal/mol1 over either of the
possible in,out-isomers, with the out,out-isomer even more un-
favorable (see Table 1). The first question we ask is: would the
in,in geometric preference be affected by substitution of a trigonal
planar nitrogen (as it is in triphenylamine) for a pyramidal
phosphorus? It seems not. Although the calculated degree of
inward pyramidalization of the bridgehead nitrogen of the
* Corresponding author. Tel.: þ1 504 862 3547; fax: þ1 504 865 5596; e-mail
address: rpascal@tulane.edu (R.A. Pascal Jr.).
0040-4020/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2013.10.018

J. Zong et al. / Tetrahedron 69 (2013) 10316e10321
10317
(chloromethyl)phenyl]phosphine13 (6) was condensed with each of
the appropriate tris(mercaptophenyl)element caps (7e9) by treat-
ment with KOH at high dilution in refluxing benzene and ethanol
(Scheme 2). The yields were variable and always low.
Table 1
Calculated relative energies of in/out isomers of bis(triarylelement) cyclophanes
The most favorable reaction was the condensation of 6 and
tris(2-mercaptophenyl)amine9 (7) to give a 15% yield of amino-
phosphine 2, which was easily isolated. The silane-capped phanes
were more highly problematic. The reaction of tris(2-
mercaptophenyl)silane5 (8) gave only a 0.7% yield of the desired
in,in-hydrosilane 3, but other macrocyclic products were also
present. The best characterized of these is the out,in-ethoxysilane 5
(1.6%) presumably formed via the reaction of ethoxide with pre-
cursor 8 prior to cyclization. An out,in-hydroxysilane also appeared
to be present, but was not fully characterized. It should be noted
that even before the X-ray structures of these compounds were
obtained (see below), their bridgehead configurations were sug-
gested by their 1H NMR spectra, which are quite characteristic.
Attempts to improve the yield of 3 by carrying out the synthesis in
THF with hindered bases, in order to avoid the solvolysis of the
hydrosilane, were not successful. In contrast to the hydrosilane cap
8, tris(mercaptophenyl)methylsilane (9) is not vulnerable to sub-
stitution at silicon by alkoxides; nevertheless, condensation of 9
with phosphine 6 still gave only a low 2.2% yield of the expected
out,in-methylsilane 4. Despite the predicted high energy of in,in-
MeSi,P, the reaction mixture was carefully searched for such
a product, but without success.
P
N
S
S
S
S
S
S
P
P
1
2
Me
Si
Si
S
S
S
H
S
S
S
P
P
3
4
Compound
EþZPE (au)a
D
E (kcal/mol)
in,in-P,P (1)
in,out-P,P-A
in,out-P,P-B
out,out-P,P
ꢀ3380.091662b
ꢀ3380.062618b
ꢀ3380.061870b
ꢀ3380.043331b
0.0
þ18.2
þ18.7
þ30.3
The NMR spectra of all of these compounds show diastereotopic
methylene resonances, indicating that the enantiomerization of
these molecules is slow on the NMR time scale. In addition, all of
them show numerous doublets in their 13C NMR spectra, but the
31Pe13C coupling is perhaps more confusing than useful, because it
is not always easy to determine, which lines are paired in the ar-
omatic regions of the 13C NMR spectra! However, the observed
spinespin coupling among the central atoms (P, H, and Si) in
compound 3 is more informative. In the 1H NMR spectrum of 3, the
in,in-N,P (2)
in,out-N,P
out,in-N,P
out,out-N,P
ꢀ3093.505473
0.0
þ15.3
ꢀ3093.481031
Not a potential minimum
Not a potential minimum
in,in-HSi,P (3)
in,out-HSi,P
out,in-HSi,P
out,out-HSi,P
ꢀ3328.848381
ꢀ3328.817926
ꢀ3328.832399
ꢀ3328.813156
ꢀ3368.054971
ꢀ3368.056685
ꢀ3368.111798
ꢀ3368.092861
0.0
þ19.1
þ10.0
þ22.1
þ35.7
þ34.6
0.0
in,in-MeSi,P
in,out-MeSi,P
out,in-MeSi,P (4)
out,out-MeSi,P
central proton resonance is a doublet (
d
9.31, JPH¼25 Hz) due to
þ11.9
coupling with the phosphorus, but in addition there exist easily
visible 29Si side bands that yield JSiH¼248 Hz (see the
Supplementary data). Most interesting is the proton-decoupled 29Si
NMR spectrum of 3, consisting of a lone doublet with strong cou-
pling to phosphorus (JSiP¼76 Hz).
a
All calculations were performed at the B3PW91/6-31G(d) level; all compounds
possess C3 symmetry and display zero imaginary frequencies.
b
Ref. 1.
aminophosphine 2 is much less than that of phosphorus, the in,in-
isomer (in,in-N,P12) is still favored by 15 kcal/mol over the sole
in,out-isomer (in,out-N,P12). Interestingly, the triarylamine can
adopt only an in-conformation in this system; no computational
minimum was found for out,in-N,P or out,out-N,P.12
2.3. X-ray structures of macrobicycles with mixed bridgehead
functionality
Generally speaking, the bis(triarylelement)-capped phanes
crystallize easily, and satisfactory X-ray structures were obtained
for compounds 2, 3, 4, and 5. The structures of the two in,in-mac-
robicycles (2 and 3) are illustrated in Fig. 1, and the in,out-macro-
bicycles (4 and 5) are illustrated in Fig. 2. Of the cyclophanes
reported here, 2, 3, and 4 possess C3 symmetry in gas phase cal-
culations, and they have at least approximate C3 symmetry in the
solid state. Of these three, however, only compound 3 crystallized
on a special position (in space group P3) and thus its X-ray structure
exhibits exact, crystallographic C3 symmetry.
All of the in,in- and in,out-macrobicycles have both of the tri-
arylelement propellers in same configuration; that is, the racemic
mixture of each cyclophane contains molecules with (P,P) and
(M,M) propeller configurations. As it happens, compound 5 crys-
tallized in the chiral space group P1, but instead of providing easy
access to pure enantiomers by selection of individual crystals, the
asymmetric unit of this structure contains two independent mol-
ecules of 5 that by chance have opposite configurations; thus it too
is a racemate.
A second, more interesting question is: how large a functional
group may be contained within the macrobicycle? Compounds 1
and 2 possess only lone pair electrons. A proton can surely fit, be-
cause the hydrochloride salt of 1 is in-protonated in its X-ray
structure.1 Replacement of one phosphine with a hydrosilane again
yields (at least computationally) a preference for an in,in-isomer
(in,in-HSi,P), but it is notable that out,in-HSi,P, with the proton
outside, is only 10 kcal/mol higher in energy. However, when the
phosphine is replaced by a methylsilane, the in,in-isomer (in,in-
MeSi,P) is very strongly disfavored, by almost 36 kcal/mol, with
respect to out,in-MeSi,P. Furthermore, flipping the phosphine out, in
order to provide more room for an in-methyl group, is not effective;
in,out-MeSi,P is 35 kcal/mol less stable than the out,in-isomer.
These computational predictions were tested by synthesis.
2.2. Synthesis of macrobicycles with mixed bridgehead
functionality
The syntheses of all of the target cyclophanes employed the
The aminophosphine 2 adopts an in,in-conformation with ap-
proximate C3 symmetry, although, as expected, the amine is more
same general method used to make diphosphine 1.1 Thus, tris[2-

10318
J. Zong et al. / Tetrahedron 69 (2013) 10316e10321
Cl
Cl
Cl
N
P
S
S
P
+
N
S
HS
SH
KOH
HS
1:2 EtOH-
benzene
7
6
2
EtO
Si
Si
S
S
S
H
S
6
+
S
S
Si
H
HS
SH
KOH
P
HS
HS
1:2 EtOH-
benzene
P
3
8
5
Me
Si
S
6
Si
S
S
HS
Me
KOH
SH
1:2 EtOH-
benzene
P
9
4
Scheme 2.
Fig. 1. Molecular structures of compounds 2 (top) and 3 (bottom); thermal ellipsoids
have been drawn at the 50% probability level.
Fig. 2. Molecular structures of compounds 5 (top) and 4 (bottom); thermal ellipsoids
have been drawn at the 50% probability level.

J. Zong et al. / Tetrahedron 69 (2013) 10316e10321
10319
nearly planar than the highly pyramidalized phosphine (Fig. 1). The
We have previously reported the chromatographic resolution of
diphosphine 1, but it racemized quickly on the laboratory time
scale.1 This proved not to be the case for compound 2. Indeed no
racemization of enantiomerically pure (ꢀ)-2 was observed upon
heating for 24 h at 100 ꢁC in toluene, and complete decomposition
to unknown products was observed upon heating for 24 h at 180 ꢁC
ꢀ
ꢀ
NeP contact distance is 4.17 A, substantially longer than the 3.58 A
ꢀ
and 3.72 A PeP contacts observed in the structures of diphosphine
1.1 The molecular structure of 3 strongly resembles that of the hy-
drochloride salt of diphosphine 1 (1$HCl).1 The PeSi distance in 3 is,
ꢀ
ꢀ
at 4.08 A, slightly longer than the PeP distance in 1$HCl, 3.88 A, no
doubt because SieH bonds are slightly longer than PeH bonds, and
the PþeH/P interaction in 1$HCl is attractiveda hydrogen
bonddbut the SieH/P interaction in 3 is essentially repulsive.
The molecular structures of the in,out-macrobicycles 4 and 5
(Fig. 2) are extremely similar. One must expect the internal cavity to
be larger than those in the in,in-phanes, but the effect is not par-
ticularly dramatic; the visual impression of the structures is that
the in,out-macrobicycles are ‘screwed together’ more tightly than
in DMSO. The 100 ꢁC experiment implies a Grzac >31 kcal/mol for
D
compound 2, much greater than the 20.7 kcal/mol barrier esti-
mated for compound 1.1 The reason for this difference is not at all
obvious, and we have thus far failed to define computationally the
racemization pathway for either molecule, despite numerous at-
tempts and the expenditure of an enormous amount of computer
time.
The resolution of compounds 3, 4, and 5 was not pursued be-
cause of the very limited amounts of material available.
ꢀ
the in,in-isomers. Thus, the PeSi separation in 4 is 4.90 A, and in the
ꢀ
ꢀ
two independent molecules of 5 it is 5.05 A and 5.06 A, but the
internal cavity is too narrow to accommodate functionality even as
large as an in-methyl group. As noted previously, in,in-MeSi,P and
in,out-MeSi,P are both (computationally) about 35 kcal/mol less
stable than the observed out,in-MeSi,P (4).
3. Conclusion
Three cyclization reactions, between triarylelement precursors
containing phosphorus, nitrogen or silicon as the central atoms,
gave four macrobicyclic, bis(triarylelement)-containing cyclo-
phanes, each bearing two different bridgehead heteroatoms. In
every case the observed product was the thermodynamically fa-
vored in/out isomer as judged by DFT calculations. Since cycliza-
tions based on SN2 displacements of halides by thiolates are
essentially irreversible and therefore under kinetic control, the
much greater stability of the observed products is clearly reflected
in the transition states for the cyclizations. Both computational
studies and the results of the syntheses indicate that only lone pairs
or protons as bridgehead substituents are small enough to form
in,in-isomers in this series of molecules, and substitution of even
a single methyl group forces the formation of an in,out-isomer. It
seems that, in order to place a methyl or some larger functional
group in so small a macrobicyclic cage, an entirely different syn-
thetic strategy must be developed.
2.4. Protonation of compounds 2, 3, and 4
When compound 1 was treated with HCl gas in chloroform
solution at room temperature, protonation was slow; the reaction
was first order with t1/2¼26 min, and the X-ray structure of the
product showed that the proton was on the more basic, sulfur-
substituted triarylphosphine.1 We naturally wondered how the
new macrobicycles would behave. Protonation of 2 under the same
conditions proceeded almost identically: the reaction was first or-
der with t1/2¼26 min, and it gave a single product that was char-
acterized by 1H NMR, 13C NMR, and HRMS data. Thus the slightly
larger cavity of 2 does not permit an increased rate of protonation
relative to that of 1. The new doublet at
d
9.31 in the product’s 1H
NMR spectrum exhibited a coupling constant of 532 Hz, clearly
indicating that protonation had occurred on phosphorus. Although
glistening crystals of 2$HCl were obtained upon evaporation of the
solvent, they only poorly diffracted in X-ray experiments.
4. Experimental
4.1. General
In contrast to the smooth, simple reaction observed with com-
pound 2, protonation experiments with silanes 3 and 4 were more
highly problematic. Both compounds did undergo protonation, but
somewhat more slowly than compounds 1 and 2. Unfortunately,
the rates of these reactions were inconsistent, and there was evi-
dence of partial decomposition in both cases. In the case of com-
pound 3, there appeared to be more than one product formed by
NMR analysis, but when the solvent was evaporated to leave some
large, single crystals, X-ray analysis showed them to be only
starting material. Thus, compound 3 had must have reacted, but
later lost the added proton upon slow evaporation of chloroform
and HCl. In the case of compound 4, whose single methyl is an
excellent reporter group for NMR spectroscopy, protonation yiel-
ded spectra with at least three new methyl resonances, with per-
haps traces of four more. No good-quality crystals were obtained
from this mixture, and the characterization of these products,
which must result in part from ring fragmentation at silicon, was
not pursued.
Tris[2-(chloromethyl)phenyl]phosphine,13
tris(2-mercapto-
phenyl)amine,9 and tris(2-mercaptophenyl)silane5 were pre-
pared as described previously. All other solvents and reagents
were commercial, reagent grade materials, and they were used
without further purification. 1H NMR spectra were recorded at
400 MHz on a Varian Unity INOVA spectrometer; samples were
dissolved in CDCl3. 13C NMR spectra were recorded on the same
instrument at 101 MHz. 31P and 29Si NMR spectra were recorded
on a Bruker AVANCE 300 spectrometer at 121 MHz and 60 MHz,
respectively. GCeMS analyses were performed on coupled Varian
450-GC and Varian 300-MS instruments. High-resolution ESI-TOF
mass spectra were recorded on an Agilent 6220 spectrometer.
4.2. Data for compounds
4.2.1. in,in-N,P
(2). Tris[2-(chloromethyl)phenyl]phosphine13
(561 mg, 1.38 mmol) and tris(2-mercaptophenyl)amine9 (470 mg,
1.38 mmol) were mixed in 2:1 benzeneeethanol (1.2 L), and the
solution was heated to reflux. An argon-saturated solution of KOH
(372 mg, 6.62 mmol) in ethanol (55 mL) was added slowly over
5.5 h. After 17 h, the mixture was cooled, and the solvent was
evaporated under reduced pressure to leave a white precipitate.
The precipitate was extracted twice with chloroform. The com-
bined extracts were concentrated and the resulting light yellow
liquid was chromatographed on silica gel (solvent, 2:1 hex-
anesebenzene) and the fractions containing compound 2 were
2.5. Resolution of compound 2
Cyclophane 2 proved to be easily resolved by supercritical fluid
chromatography (SFC) on Chiralpak 1A (mobile phase: 40% EtOH/
CO2) (see Supplementary data). Ultimately, 25 mg of 2 were re-
solved into pure enantiomers, and these samples yielded specific
rotations ([
components, respectively.
a
]
25) of þ132 and ꢀ122 for the faster and slower eluting
D

10320
J. Zong et al. / Tetrahedron 69 (2013) 10316e10321
combined and further purified by preparative TLC (silica gel, 1:1
hexanesebenzene) to give cyclophane 2 as a white solid (136.8 mg,
(d, JPC¼6 Hz), 133.2, 135.9, 136.9 (d, JPC¼26 Hz), 139.7, 140.2 (d,
JPC¼27 Hz), 145.5 (15 of 15 expected resonances); HRMS (ESI) m/z
699.1430 (MþH), calcd for C41H36OPS3Si 699.1435.
0.214 mmol, 15.5%). 1H NMR (CDCl3)
d
4.14 (dd, J¼10 Hz, 2 Hz, 3H),
4.23 (dd, J¼10 Hz, 1 Hz, 3H), 6.73 (dd, J¼8 Hz, 1 Hz, 3H), 6.76 (dd,
J¼8 Hz, 2 Hz, 3H), 7.00 (td, J¼8 Hz, 1 Hz, 3H), 7.12 (m, 6H), 7.24 (td,
J¼8 Hz, 1 Hz, 3H), 7.30 (m, 3H), 7.64 (dd, J¼8 Hz, 2 Hz, 3H); 13C NMR
4.2.4. Tris(2-mercaptophenyl)methylsilane (9). Thiophenol (5.0 mL,
49 mmol) in cyclohexane (20 mL) was added to flask containing
cyclohexane (110 mL), TMEDA (15.7 mL) and n-BuLi (2.5 M in
hexane, 40.7 mL,101.8 mmol) at 0 ꢁC. The reaction was permitted to
warm to room temperature and left for two days; a white pre-
cipitate formed. The cyclohexane was siphoned away, and the
precipitate was washed with cyclohexane. After the precipitate
settled again, the cyclohexane was removed, and THF (55 mL) was
added. The solution was cooled to ꢀ78 ꢁC, and MeSiCl3 (1.35 mL) in
THF (15 mL) were added dropwise. The mixture was permitted to
warm to room temperature, and after 20 h it was acidified with 10%
sulfuric acid (100 mL). The mixture was concentrated in vacuo to
remove THF and the resulting aqueous mixture was extracted three
times with CHCl3. The combined organics were dried over Na2SO4
and concentrated in vacuo to give a yellow oil. This material was
purified by column chromatography on silica gel (3:1, then 1:1,
hexanesebenzene) to give compound 9 as a thick yellow oil (0.39 g,
(CDCl3)
d
42.3 (d, JPC¼25 Hz), 124.0, 124.4, 128.4, 128.7, 129.3, 130.7
(d, JPC¼5 Hz), 131.8, 134.4, 136.9, 137.0 (d, JPC¼28 Hz), 140.8 (d,
JPC¼29 Hz), 151.9 (13 of 13 expected resonances); HRMS (ESI) m/z
640.1350 (MþH), calcd for C39H31NPS3 640.1351. Single crystals,
suitable for X-ray analysis, were obtained from CHCl3eMeOH.
A portion of this material was resolved by preparative super-
critical fluid chromatography (SFC) on a chiral support (Chiralpak
IA; 40% EtOH/CO2 at 100 bar). The faster eluting component had
25
25
[
a]
þ132ꢂ5 (c 0.00031, EtOH), the slower [
a
]
ꢀ122ꢂ5 (c
D
D
0.00020, EtOH). The resolved enantiomers had chemical and opti-
cal purity in excess of 99%, as judged by analytical chiral SFC.
4.2.2. in,in-N,P hydrochloride (2$HCl). A stream of HCl gas was
bubbled into an NMR tube containing a CDCl3 solution of cyclo-
phane 2 for 5 s, and then 1H NMR spectra were recorded at in-
tervals. Protonation was a slow process (t1/2¼26 min), but it was
1.1 mmol, 6.7%). 1H NMR (CDCl3)
d 1.28 (s, 3H), 3.48 (s, 3H), 7.15 (t,
essentially complete in 6 h. 1H NMR (CDCl3)
d
3.91 (d, J¼10 Hz, 3H),
J¼8 Hz, 3H), 7.29 (m, 6H), 7.40 (d, J¼8 Hz, 3H); 13C NMR (CDCl3)
4.52 (br d, J¼10 Hz, 3H), 6.73 (br d, J¼7 Hz, 3H), 7.18 (m, 9H), 7.65 (br
d, J¼7 Hz, 6H), 7.73 (br s, 3H), 7.84 (br s, 3H),11.99 (d, J¼532 Hz,1H);
d 0.68, 125.9, 130.7, 132.3, 135.5, 138.1, 138.5 (7 of 7 expected reso-
nances); MS (EI) m/z 336 (MꢀH2S, 24), 321 (MꢀH2SꢀCH3, 12), 227
13C NMR (CDCl3)
d
40.4 (d, JPC¼8 Hz), 114.9 (d, JPC¼84 Hz), 125.0,
(MꢀH2SꢀC6H4SH, 100).
125.6,129.7,129.8,131.2 (d, J¼13 Hz),133.5 (d, J¼10 Hz),135.4,136.6
(d, J¼10 Hz), 137.1, 140.3 (d, J¼9 Hz), 150.7 (13 of 13 expected res-
onances); HRMS (ESI) m/z 640.1352 (MꢀCl), calcd for C39H31NPS3
640.1351.
4.2.5. out,in-MeSi,P (4). Tris[2-(chloromethyl)phenyl]phosphine
(413 mg, 1.01 mmol) and tris(2-mercaptophenyl)methylsilane (9,
376 mg, 1.01 mmol) were mixed in 2:1 benzeneeethanol (900 mL),
and the solution was heated to reflux. An argon-saturated solution
of KOH (280 mg, 4.99 mmol) in ethanol (100 mL) was added slowly
over 9 h. After another 15 h, the mixture was cooled, and the sol-
vent was evaporated under reduced pressure to leave a white
precipitate. This material was extracted twice with chloroform. The
combined extracts were concentrated, and the resulting light yel-
low liquid was chromatographed on silica gel (solvent, 1:1 hex-
4.2.3. in,in-HSi,P (3) and out,in-EtOSi,P (5). Tris[2-(chloromethyl)
phenyl]phosphine (563 mg, 1.38 mmol) and tris(2-
mercaptophenyl)silane5 (493 mg, 1.38 mmol) were mixed in 2:1
benzeneeethanol (1.2 L), and the solution was heated to reflux. An
argon-saturated solution of KOH (376 mg, 6.70 mmol) in ethanol
(55 mL) was added slowly over 6 h. After 18 h, the mixture was
cooled, and the solvent was evaporated under reduced pressure to
leave a white precipitate. The precipitate was extracted twice with
chloroform. The combined extracts were concentrated and the
resulting light yellow liquid was chromatographed on silica gel
(solvent, 1:1 hexanesebenzene) and the fractions containing
compound 3 were combined and further purified by preparative
TLC (silica gel, 1:1 hexanesebenzene) to give cyclophane 3 as
a white solid. This material exhibited a single component by TLC,
but the 1H NMR spectrum clearly indicated the presence of sig-
nificant impurities. Further chromatographic purification was un-
successful, but careful crystallization from CHCl3eCH2Cl2eMeOH
anesebenzene). The fractions containing compound
4 were
combined and further purified by preparative TLC (silica gel, 1:1
hexanesebenzene) to give cyclophane 4 as a white solid (15 mg,
0.022 mmol, 2.2%). 1H NMR (CDCl3)
d
0.77 (s, 3H), 3.66 (d, J¼10 Hz,
3H), 4.22 (d, J¼10 Hz, 3H), 6.70 (dd, J¼8 Hz, 2 Hz, 3H), 7.12 (m, 6H),
7.24 (m, 12H), 7.47 (d, J¼8 Hz, 3H); 13C NMR (CDCl3)
d 5.3, 41.5 (d,
JPC¼18 Hz), 125.1, 128.6,129.0,130.0,130.1, 130.9 (d, JPC¼5 Hz), 133.1,
135.7, 136.9 (d, JPC¼26 Hz), 140.2 (d, JPC¼27 Hz), 140.5, 146.4 (14 of
14 expected resonances). HRMS (ESI) m/z 669.1324 (MþH), calcd
for C40H34PS3Si 669.1324. Single crystals, suitable for X-ray analysis,
were obtained from benzeneeCHCl3eMeOH.
gave cyclophane 3 (6 mg, 9
mmol, 0.7%) as crystals suitable for X-ray
4.04 (dd, J¼10 Hz, 3 Hz, 3H), 5.09 (dd,
analysis. 1H NMR (CDCl3)
d
Acknowledgements
J¼10 Hz, 5 Hz, 3H), 7.06 (dd, J¼8 Hz, 3 Hz, 3H), 7.19 (m, 6H), 7.29 (m,
6H), 7.39 (m, 3H), 7.51 (m, 3H), 7.82 (d, J¼8 Hz, 3H), 9.31 (d, J¼25 Hz,
This work was supported by National Science Foundation Grants
CHE-0936862 and CHE-1265507, which are gratefully acknowl-
edged. The X-ray diffractometer used in this work was purchased
with funds from National Science Foundation Grant MRI-1228232.
1H); 13C NMR (CDCl3)
d
43.7 (d, JPC¼29 Hz), 128.0, 128.6, 129.6,
130.6, 131.7 (d, JPC¼6 Hz), 135.36, 135.40 (d, JPC¼17 Hz), 137.4, 137.6,
141.7 (d, JPC¼34 Hz), 141.8, 143.6 (d, JPC¼2 Hz) (13 of 13 expected
resonances); 31P NMR (CDCl3)
d
ꢀ48.9; 29Si NMR (CDCl3)
ꢀ35.8 (d,
d
JPSi¼76 Hz); HRMS (ESI) m/z 655.1173 (MþH), calcd for C39H32PS3Si
Supplementary data
655.1168.
A second component proved to be the out,in compound 5
(1) 1H and 13C NMR spectra of compounds 2, 2$HCl, 3, 4, 5, and 9;
kinetics of protonation of 2, and chiral SFC chromatograms for 2. (2)
An ASCII text file containing the atomic coordinates and energies of
the calculated structures in Table 1. (3) Crystallographic data (ex-
cluding structure factors) for the structures in this paper have been
deposited with the Cambridge Crystallographic Data Centre as
supplementary publication numbers CCDC 955764e955767. Copies
of the data can be obtained, free of charge, on application to CCDC,
(15 mg, 21
m
mol, 1.6%), which was obtained as crystals suitable for
X-ray analysis from CHCl3eCH2Cl2eMeOH. 1H NMR (CDCl3)
d
1.21
(t, J¼7 Hz, 3H), 3.65 (d, J¼11 Hz, 3H), 3.84 (dq, J¼10 Hz, 7 Hz, 1H),
3.93 (dq, J¼10 Hz, 7 Hz, 1H), 4.20 (d, J¼11 Hz, 3H), 6.68 (dd, J¼7 Hz,
2.5 Hz, 3H), 7.11 (td, J¼7 Hz, 2 Hz, 3H), 7.16 (td, J¼7 Hz, 2 Hz, 3H),
7.25 (m, 12H), 7.65 (dd, J¼7.5 Hz, 1 Hz, 3H); 13C NMR (CDCl3)
d 18.8
(s), 41.6 (d, JPC¼18 Hz), 60.3, 125.2, 128.6, 129.1, 130.4, 130.6, 130.9

J. Zong et al. / Tetrahedron 69 (2013) 10316e10321
10321
and all potential minima were characterized by analytical frequency
calculations.
12 Union Road, Cambridge, CB2 1EZ, UK [fax: þ44(0)-1223-336033
or e-mail: deposit@ccdc.cam.ac.uk]. Supplementary data related
to this article can be found at http://dx.doi.org/10.1016/
j.tet.2013.10.018.
11. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Na-
katsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng,
G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery,
J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K.
N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.;
Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.;
Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dan-
References and notes
1. Zong, J.; Mague, J. T.; Kraml, C. M.; Pascal, R. A., Jr. Org. Lett. 2013, 15, 2179e2181.
2. For a review of in-cyclophanes, see: Pascal, R. A., Jr. Eur. J. Org. Chem. 2004,
3763e3771.
3. For a review of in/out stereoisomerism, see: Alder, R. W.; East, S. P. Chem. Rev.
1996, 96, 2097e2111.
€
nenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.;
4. Pascal, R. A., Jr.; West, A. P., Jr.; Van Engen, D. J. Am. Chem. Soc. 1990, 112,
6406e6407.
Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A.02; Gaussian: Wallingford CT,
2009.
5. L’Esperance, R. P.; West, A. P., Jr.; Van Engen, D.; Pascal, R. A., Jr. J. Am. Chem. Soc.
1991, 113, 2672e2676.
6. West, A. P., Jr.; Smyth, N.; Kraml, C. M.; Ho, D. M.; Pascal, R. A., Jr. J. Org. Chem.
12. The systematic names of these hexabenzo macrobicycles are extremely cum-
bersome, but since all of the compounds discussed herein possess the same
structure with the exception of the bridgehead functional groups, a simple
system of abbreviations is employed in Table 1, the Experimental section, and
elsewhere. For both the stereochemical designation and the nature of bridge-
head, the sulfur-substituted triphenylelement cap is listed first and the
methylene-substituted cap second. Thus, ‘in,out-N,P’ is a macrobicycle com-
posed of an inwardly-pyramidalized, sulfur-substituted triphenylamine and an
outwardly pyramidalized, methylene-substituted triphenylphosphine
13. Letsinger, R. L.; Nazy, J. R.; Hussey, A. S. J. Org. Chem. 1958, 23, 1806e1807.
1993, 58, 3502e3506.
7. Dell, S.; Ho, D. M.; Pascal, R. A., Jr. J. Org. Chem. 1999, 64, 5626e5633.
8. Chen, Y. T.; Baldridge, K. K.; Ho, D. M.; Pascal, R. A., Jr. J. Am. Chem. Soc. 1999, 121,
12082e12087.
9. Sathyamoorthi, S.; Mague, J. T.; Pascal, R. A., Jr. Org. Lett. 2012, 14, 3427e3429.
10. All hybrid density functional calculations were performed with GAUSSIAN 09;11
the built-in default thresholds for wave function and gradient convergence
were employed. All calculations were performed at the B3PW91/6-31G(d) level,