Steric control of the in/out sense of bridgehead substituents in macrobicyclic compounds: Isolation of new "crossed chain" variants of in/out isomers

Article abstract of DOI:10.1039/c5cc05620a

Isomers of the cage like dibridgehead diphosphine P((CH₂)₁₄)₃P (1) are treated with Ph₃PAu(2,6-C₆H₃(Trip)₂) (2 equiv.; Trip = 2,4,6-C₆H₂(iPr)₃). With out,out-1, workup gives out,out-1·(Au(2,6-C₆H₃(Trip)₂))₂ (46%), as confirmed by a crystal structure. With in,out-1, crystallization affords not in,out-1·(Au(2,6-C₆H₃(Trip)₂))₂, but rather an out,out isomer in which one of the (CH₂)₁₄ segments threads through the macrocycle formed by the other two. Implications for mechanisms of interconversion of in,out isomers are analyzed.

Full text of DOI:10.1039/c5cc05620a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Steric control of the in/out sense of bridgehead
substituents in macrobicyclic compounds:
isolation of new ‘‘crossed chain’’ variants of in/out
isomers†
Cite this: Chem. Commun., 2015,
1, 16053
5
Received 8th July 2015,
Accepted 28th August 2015
Michael Stollenz,‡ Deeb Taher,§ Nattamai Bhuvanesh, Joseph H. Reibenspies,
DOI: 10.1039/c5cc05620a
Zuzana Baranova and John A. Gladysz*
´
www.rsc.org/chemcomm
Isomers of the cage like dibridgehead diphosphine P((CH
treated with Ph PAu(2,6-C (Trip) ) (2 equiv.; Trip = 2,4,6-C
With out,out-1, workup gives out,out-1ꢀ(Au(2,6-C (Trip) ))
confirmed by a crystal structure. With in,out-1, crystallization affords
not in,out-1ꢀ(Au(2,6-C (Trip) )) , but rather an out,out isomer in
which one of the (CH 14 segments threads through the macrocycle
2
)
14
)
3
P (1) are
(iPr) ).
2 2
(46%), as
3
6
H
3
2
6
H
2
3
6
H
3
6
H
3
2 2
2
)
formed by the other two. Implications for mechanisms of inter-
conversion of in,out isomers are analyzed.
When two non-planar trivalent or tetravalent atoms are bridged
by three ‘‘chains’’ or ‘‘tethers’’ of sufficient lengths, in,out
1
stereoisomers become possible. Although this type of isomerism
has intrigued chemists for some time, a variety of nuances has
remained unexplored. Scheme 1 shows the familiar limiting stereo-
chemistries for structures with identical bridgeheads E–0, with the
blue ovals denoting substituents or lone pairs: in,in (I), out,out (III),
and in,out (V). The last has two degenerate forms (in,out vs. out,in),
but only one is depicted. Except in the case of dibridgehead
diamines (E–O = N:), pyramidal inversion barriers are high.
Scheme 1 In,out equilibria for macrobicyclic compounds.
We and others have established that at sufficient chain via one bridgehead ‘‘inversion’’ or ‘‘half turn’’ at a time, as repre-
lengths, such molecules are capable of turning themselves sented by I - II - III in Scheme 1. The hypothetical intermediate II
2
,3
‘‘inside out’’,
in what have been termed homeomorphic may be viewed as having a chain that is ‘‘crossed’’ with respect to the
4
isomerizations. These fascinating topological processes, in other two, and an in,out disposition. The corresponding reaction
which one chain is drawn through the macrocycle defined by coordinate for the in,out isomer V is not shown in its entirety, but
the other two, appear to interconvert the configurations of both would likely involve initial isomerization to the crossed chain out,out
bridgehead atoms without pyramidal inversions. Although no species VI, avoiding the bridgehead/bridgehead interactions possible
experimental or computational details are yet available regarding in the crossed chain in,in species IV. Additional conformational
these reaction coordinates, they may be visualized as proceeding manipulations involving the green chain in VI would ‘‘invert’’ the
upper bridgehead and consummate the degenerate isomerization
(
see also graphical abstract).
Department of Chemistry, Texas A&M University, P.O. Box 30012, College Station,
Texas 77842-3012, USA. E-mail: Gladysz@mail.chem.tamu.edu
We have sought to probe the fine details of Scheme 1
through the agency of compounds with bridgehead phosphorus
†
Electronic supplementary information (ESI) available: Experimental procedures and
crystallographic data. CCDC 1410646 (3), 1410647 (out,out-1ꢀ(Au(2,6-C (Trip) ))
). For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c5cc05620a
Current address: Department of Chemistry and Biochemistry, Kennesaw State
University, 370 Paulding Building NW, MD#1203, Kennesaw, Georgia 30144, USA.
2
,5,6
6
H
3
2
2
), atoms.
In previous papers, dibridgehead diphosphines with
0
and 1410648 (out,out-1 ꢀ(Au(2,6-C
6 3 2 2
H (Trip) ))
three (CH2)14 chains have been synthesized (1), and found to
undergo homeomorphic isomerizations with very low energy
barriers (11.5–8.5 kcal mol ). Interestingly, in,in-1 proves more
stable than out,out-1 (97 : 03, toluene, ꢁ80 1C; see Scheme 2,
bottom); in,out-1 is of comparable free energy at 150 1C, where
‡
ꢁ
1 2
§
Current address: Chemistry Department, The University of Jordan, Amman
1942, Jordan.
1
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun., 2015, 51, 16053--16056 | 16053

View Article Online
Communication
ChemComm
Fig. 1 Thermal ellipsoid plot of the molecular structure of 3 (50% prob-
ability level, dominant conformation where disordered).
compounds) were far greater than those associated with aurophilic
11
interactions (2.50–3.50 Å).
Scheme 2 Syntheses of out,out-1ꢀ(Au(2,6-C
6 3 2 2
H (Trip) )) .
Next, (in,in/out,out)-1 and 3 (2.0 equiv.) were combined at room
temperature as shown in Scheme 2 (A, bottom). As expected, the
7
more basic trialkylphosphine replaced the Ph P ligand in 3,
3
pyramidal inversion at phosphorus kicks in. Adducts of the gold(I)
5
affording the 1 : 2 adduct out,out-1ꢀ(Au(2,6-C H (Trip) )) in 46%
6
3
2 2
Lewis acids AuCl and AuMe have also been isolated (1ꢀ(AuX) ).
2
yield after preparative TLC. An alternative synthesis involving
Those synthesized from (in,in/out,out)-1 crystallize as out,out
isomers, as would be intuitively expected, and these appear to
be the dominant species in solution.
5
the bis(gold chloride) complex out,out-1ꢀ(AuCl)
reagent Li(2,6-C (Trip) (2.0 equiv., ꢁ78 1C) was investi-
)ꢀOEt
gated, As shown in Scheme 2 (B, top), workup gave out,out-1ꢀ
Au(2,6-C H (Trip) )) in 40% yield. Crystals were obtained as
2
and aryl lithium
6
H
3
2
2
For this study, we sought to test whether complexes with
‘crossed chains’’ might be isolated. Any documentation of this
(
6
3
2 2
‘
described in the ESI.† The structure was solved as with 3, and
the representations provided in Fig. 2 show a quite symmetrical
cage structure. A diethyl ether molecule occupies some of the void
heretofore unobserved type of in,out isomer would strengthen the
case for its intermediacy on the reaction coordinates proposed in
Scheme 1. Towards this end, we asked whether introducing pro-
gressively bulkier bridgehead substituents in the in,out isomer V
space within the cage. A solvate molecule, methylcyclopentane,
5
similarly crystallized within the cage of out,out-1ꢀ(AuMe)
2
.
(Scheme 1) might eventually yield such severe bridgehead/
The results in Scheme 2 set the stage for analogous reactions
substituent interactions that the out,out crossed chain isomer VI
would become favored. The 1 : 2 adduct of in,out-1 and AuCl
of in,out-1. One practical consideration is that in,out-1 is initially
2b
accessed as a ca. 50 : 50 mixture with (in,in/out,out)-1. The
borane adducts 1ꢀ(BH ) are then chromatographically sepa-
crystallized as in,out-1ꢀ(AuCl)
2
, but with a distorted cage structure
3
2
that relieved bridgehead/substituent interactions and allowed inter-
5
rated. Consequently, it saves steps to simply derivatize in,out-1
molecular AuClꢀꢀꢀHꢀꢀꢀCHP hydrogen bonding. What might hap-
as a mixture with (in,in/out,out)-1, as demonstrated earlier for the
pen with a bulkier gold(I) Lewis acid that would be incapable of
such hydrogen bonding?
5
adducts 1ꢀ(AuCl)
2
. Thus, as shown in Scheme 3, the isomers of 1
Our attention was drawn to gold adducts of supersized aryl
groups. Complexes with 2,6-disubstituted phenyl ligands, such
3 6 3 6 2 3 2
as Ph PAu(2,6-C H (2,4,6-C H Me ) ) with two proximal mesityl
8
moieties, had been previously reported. In this spirit, we targeted
a similar precursor, but with the methyl mesityl substituents
transposed to isopropyl, abbreviated Ph PAu(2,6-C H (Trip) ) (3,
3
6
3
2
9
Scheme 2). Towards this end, equimolar amounts of the corre-
sponding aryl lithium reagent, Li(2,6-C
Ph PAuCl were combined in THF at room temperature. Workup
10
2
6
H
3
(Trip)
2
)ꢀOEt
and
3
gave 3 as a white crystalline solid in 75% yield. This complex, and
all other new compounds below, were characterized by NMR
1
13
1
31
1
(
H, C{ H}, P{ H}) and microanalyses, as summarized in the ESI.†
The crystal structure of 3 was determined as described in the
ESI.† The thermal ellipsoid diagram, shown in Fig. 1, illustrates
the considerable bulk of the aryl substituent, and the shielding of the
phosphine ligand by the isopropyl groups of the C H moieties. The
6 2
metrical parameters of 3 and all other structurally characterized
compounds below were routine, and are summarized in the ESI.†
The closest intermolecular gold–gold distances (49.94 Å for all thermal ellipsoid plot (50% probability level); right, space filling representation.
Fig. 2 Molecular structure of out,out-1ꢀ(Au(2,6-C
6
H
3 2 2 2
(Trip) )) ꢀOEt : left,
16054 | Chem. Commun., 2015, 51, 16053--16056
This journal is ©The Royal Society of Chemistry 2015

View Article Online
ChemComm
Communication
crossed chain out,out isomer. Some smaller peaks were evident, a
pair of which (32.4 and ꢁ37.0 ppm) were considered consistent
12
with an out mono(adduct) of in,out-1.
Preparative TLC afforded out,out-1ꢀ(Au(2,6-C H (Trip) )) in
6
3
2 2
3
3
5% yield, and the new structurally unassigned 1 : 2 adduct in
2% yield (64% and 71% of theory). Alternatively, in a protocol
analogous to B in Scheme 2, a comparable mixture of (in,in/out,out)-
and in,out-1ꢀ(AuCl) was treated with Li(2,6-C (Trip) ) (2.0 equiv.).
(Trip) )) in 18%
2
6
H
3
2
An analogous workup gave out,out-1ꢀ(Au(2,6-C
6
H
3
2 2
yield, and the new 1 : 2 adduct in 11% yield (33% and 24% of
theory). In most respects, the NMR spectra were quite similar, with
each giving a single set of signals. However, the former exhibited
13
only seven methylene carbon C signals, consistent with an
idealized (time averaged) C axis and a perpendicular mirror plane.
3
In contrast, the latter gave fourteen methylene carbon signals, half
of which were more intense than the other, indicating two types of
methylene chains (2 : 1 ratio). This cannot be reconciled with an
in,out formulation such as V (Scheme 1). Hence, the new product
was assigned as a crossed chain out,out isomer (VI) and denoted
0
with a prime (out,out-1 ꢀ(Au(2,6-C H (Trip) )) ) to distinguish it
6
3
2 2
from the conventional topology.
0
Crystals of out,out-1 ꢀ(Au(2,6-C
6 3 2 2
H (Trip) )) were obtained as
described in the ESI,† and X-ray analysis gave the representa-
tions in Fig. 4. These confirmed the crossed chain and out,out
features. An inversion center at the midpoint of the crossed
chain was also apparent. The C46–C52 segment was disordered,
but this could be refined to a 58 : 42 occupancy ratio. When the
bulk sample was ground and analysed by powder X-ray diffrac-
tion, identical unit cell dimensions were obtained, thereby
establishing morphological homogeneity.
0
Scheme 3 Synthesis of the ‘‘crossed chain’’ species out,out-1 ꢀ(Au(2,6-
(Trip) ))
Several structural features involving Fig. 3 and 4 and the
C
6
H
3
2 2
.
previously synthesized compounds 1ꢀ(AuX)
2
merit analysis.
(Trip) )) fills
0
First, the crossed chain in out,out-1 ꢀ(Au(2,6-C
6
H
3
2 2
3
1
were treated with 3 (2.0 equiv.) in THF. A P NMR spectrum the interior space of the ‘‘cage’’ in a similar manner as the ether
Fig. 3) of a separately conducted reaction showed the signal for solvate molecule in out,out-1ꢀ(Au(2,6-C H (Trip) )) ꢀOEt . The
(
6
3
2
2
2
out,out-1ꢀ(Au(2,6-C H (Trip) )) (31.6 ppm) and one that was con- volume of the unit cell of the latter is (after normalizing to Z)
6
3
2 2
sidered promising for in,out-1ꢀ(Au(2,6-C
6
H
3
(Trip)
2
))
2
(32.3 ppm) or a 2–3% greater, consistent with a less compact arrangement of
methylene chains. The intramolecular phosphorus–phosphorus
distances are 11.310 Å and 10.769 Å, respectively, with the ca. 5%
Fig. 4 Molecular structure of the dominant conformation of the ‘‘crossed
3
1
1
0
Fig. 3
P{ H} NMR spectrum of the reaction mixture in Scheme 3 prior to
chain’’ species out,out-1 ꢀ(Au(2,6-C
6 3 2 2
H (Trip) )) : left, thermal ellipsoid plot
workup (THF).
(50% probability level); right, space filling representation.
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun., 2015, 51, 16053--16056 | 16055

View Article Online
Communication
ChemComm
expansion in the former presumably aiding the accommodation of
the crossed chain. The gold–gold distances, 15.488 Å and 15.027 Å,
A. J. Nawara-Hultzsch, T. Fiedler, R. Laddusaw, N. Bhuvanesh and
J. A. Gladysz, Angew. Chem., Int. Ed., 2011, 50, 6647–6651 (Angew.
Chem., 2011, 123, 6777–6781).
0
trend in the same direction. In out,out-1 ꢀ(Au(2,6-C H (Trip) )) , the
6
3
2
2
3 (a) A. H. Haines and P. Karntiang, J. Chem. Soc., Perkin Trans. 1, 1979,
2577–2587; (b) R. S. Wareham, J. D. Kilburn, D. L. Turner, N. H. Rees and
D. S. Holmes, Angew. Chem., Int. Ed. Engl., 1995, 34, 2660–2662 (Angew.
Chem., 1995, 107, 2902–2904); (c) M. Saunders and N. Krause, J. Am.
Chem. Soc., 1990, 112, 1791–1795; (d) R. W. Alder, E. Heilbronner,
E. Honegger, A. B. McEwan, R. E. Moss, E. Olefirowicz, P. A. Petillo,
R. B. Sessions, G. R. Weisman, J. M. White and Z.-Z. Yang, J. Am. Chem.
Soc., 1993, 115, 6580–6591.
C. H. Park and H. E. Simmons, J. Am. Chem. Soc., 1968, 90,
2429–2431.
M. Stollenz, N. Bhuvanesh, J. H. Reibenspies and J. A. Gladysz,
Organometallics, 2011, 30, 6510–6513.
bridgeheads appear less aligned, as reflected by the offset of the two
phosphorus–gold vectors. Accordingly, the average value of the two
Au–P–P angles in out,out-1 (Au(2,6-C
6
H
6
H
3
(Trip)
(Trip)
2
))
))
2
(160.01) is closer to
(152.51).
0
1801 than that in out,out-1 ꢀ(Au(2,6-C
3
2
2
In summary, this study has expanded the range of molecular
topologies that can be realized within the realm of in,out isomers.
Specifically, when the in bridgehead substituent associated with an
in,out isomer becomes sufficiently large, one of the tethers con-
necting the bridgeheads can thread through the other two, leading
to an out,out isomer with crossed chains. To our knowledge, this
type of conformational behavior has not previously been documen-
ted. These observations provide support for the intermediacy of
crossed chain species in homeomorphic isomerizations (II, IV, VI,
Scheme 1). Additional investigations of such equilibria – for
example, extensions to other methylene chain lengths and bridge-
head substituents – are in progress. These phenomena may be
much more common than currently appreciated. For example, the
capture of metal cations by cryptands that feature bridgehead
4
5
6
For macrobicyclic dibridgehead diphosphorus compounds based
upon triarylphosphines and phosphites, see (a) F. D ¨a britz, A. J ¨a ger
and I. Bauer, Eur. J. Org. Chem., 2008, 5571–5576; (b) F. D ¨a britz,
G. Theumer, M. Gruner and I. Bauer, Tetrahedron, 2009, 65, 2995–3002;
(c) Review of earlier papers in this series: I. Bauer and W. D. Habicher,
Collect. Czech. Chem. Commun., 2004, 69, 1195–1230; (d) J. Zong,
J. T. Mague, C. M. Krami and R. A. Pascal, Jr., Org. Lett., 2013, 15,
2179–2181.
7
R. D. Baechler and K. Mislow, J. Am. Chem. Soc., 1970, 92, 3090–3093.
8 G. W. Rabe and N. W. Mitzel, Inorg. Chim. Acta, 2001, 316, 132–134.
For some earlier applications of this bulky aryl ligand in transition
metal complexes, see, inter alia: (a) C.-S. Hwang and P. P. Power,
Organometallics, 1999, 18, 697–700; (b) C. Ni, J. C. Fettinger,
G. J. Long and P. P. Power, Dalton Trans., 2010, 39, 10664–10670;
9
13
nitrogen atoms can easily be envisioned as proceeding via an
out nitrogen lone pair (e.g., of III), followed by a homeomorphic
isomerization that occurs via a crossed chain species.
(
c) M. Carrasco, I. Mendoza, M. Faust, J. L o´ pez-Serrano, R. Peloso,
A. Rodr ´ı guez, E. Alvarez, C. Maya, P. P. Power and E. Carmona, J. Am.
Chem. Soc., 2014, 136, 9173–9180.
´
We thank the US National Science Foundation (CHE- 10 B. Schiemenz and P. P. Power, Organometallics, 1996, 15, 958–964.
1
1 H. Schmidbauer and A. Schier, Chem. Soc. Rev., 2012, 41,
70–412.
1
153085), the Fulbright Foundation (D.T.), and the University
3
of Jordan (D.T) for support.
1
2 The binding of a second gold(I) Lewis acid to in,out-1 would logically
be slower than to out,out-1. The chemical shift of the upfield signal
of the proposed mono(adduct) is close to those of in,out-1 (ꢁ32.4
Notes and references
1
2
and ꢁ37.8 ppm, CH
2
Cl
2
, ꢁ90 1C; ꢁ34.2 ppm, CH
2
Cl
2
, ꢁ50 1C (above
R. W. Alder and S. P. East, Chem. Rev., 1996, 96, 2097–2112.
T
coal); ꢁ31.3 ppm, CDCl
3
, ambient probe temperature).
(a) A. J. Nawara, T. Shima, F. Hampel and J. A. Gladysz, J. Am. Chem. 13 B. Dietrich, Cryptands, in Comprehensive Supramolecular Chemistry,
Soc., 2006, 128, 4962–4963; (b) M. Stollenz, M. Barbasiewicz,
ed. G. W. Gokel, Elsevier, Oxford, 1996, vol. 1, pp. 153–211.
16056 | Chem. Commun., 2015, 51, 16053--16056
This journal is ©The Royal Society of Chemistry 2015