Assembly of π conjugated phosphole azahelicene derivatives into chiral coordination complexes: An experimental and theoretical study

Article abstract of DOI:10.1002/chem.200903234

Aza[n]helicene phosphole derivatives have been prepared from aza[n]helicene diynes by the FaganNugent route. Their photophysical properties (UV/Vis absorption and emission behavior) have been evaluated. Their behavior as P,N chelates towards coordination to Pd^II and Cu^I has been investigated: metal-bis-(aza[n]helicene phosphole) assemblies are formed by a highly stereoselective coordination process, as demonstrated by X-ray crystallography. An aza[6]helicene phosphole bearing an enantiopure helicene part has been obtained, which allows the preparation of enantiopure Pd ^II and Cu^I complexes with original topologies and high molar rotation (MR) and circular dichroism (CD). The structure-property relationship established from the experimental data has been studied in detail by theoretical studies (TDDFT calculations of UV/Vis, CD, and MR). Aza[n]helicene phosphole derivatives show π conjugation extended over the entire molecule, and its influence on the MR of aza[6]helicene phosphole 5 c has been demonstrated. Finally, it has been shown that the nature of the metal (coordination geometry and electronic interaction) can have a great impact on the amplitude of the chiroptical properties in metal-bis(aza[n]helicene phosphole) assemblies.

Full text of DOI:10.1002/chem.200903234

DOI: 10.1002/chem.200903234
Assembly of p-Conjugated Phosphole Azahelicene Derivatives into Chiral
Coordination Complexes: An Experimental and Theoretical Study
Sebastien Graule,[a] Mark Rudolph,[b] Wenting Shen,[a] J. A. Gareth Williams,[c]
Christophe Lescop,[a] Jochen Autschbach,*[b] Jeanne Crassous,*[a] and Rꢀgis Rꢀau*[a]
Abstract: Aza[n]helicene phosphole
derivatives have been prepared from
aza[n]helicene diynes by the Fagan–
Nugent route. Their photophysical
properties (UV/Vis absorption and
emission behavior) have been evaluat-
ed. Their behavior as P,N chelates to-
wards coordination to PdII and CuI has
pure helicene part has been obtained,
which allows the preparation of enan-
tiopure PdII and CuI complexes with
original topologies and high molar ro-
tation (MR) and circular dichroism
(CD). The structure–property relation-
ship established from the experimental
data has been studied in detail by theo-
retical studies (TDDFT calculations of
UV/Vis, CD, and MR). Aza[n]helicene
phosphole derivatives show p conjuga-
tion extended over the entire molecule,
and its influence on the MR of aza[6]-
helicene phosphole 5c has been dem-
onstrated. Finally, it has been shown
that the nature of the metal (coordina-
tion geometry and electronic interac-
tion) can have a great impact on the
amplitude of the chiroptical properties
been
investigated:
metal–bis-
ACHTUNGTRENNUNG(aza[n]helicene phosphole) assemblies
Keywords: chiroptical properties ·
are formed by a highly stereoselective
coordination process, as demonstrated
by X-ray crystallography. An aza[6]he-
licene phosphole bearing an enantio-
density functional calculations
·
·
in metal–bis
assemblies.
ACHTUNGTNER(NUNG aza[n]helicene phosphole)
helical structures
P,N ligands
· palladium
Introduction
most appealing attribute of helicene derivatives is their huge
chiroptical properties (optical rotation, circular dichroism)[2]
that make them promising building blocks for the synthesis
of functional molecular materials (e.g., nonlinear optical
materials,[3a,h] chiral wave guides,[3b] and so on). They also
display interesting phosphorescent properties due to large
spin–orbit coupling,[4] a rather uncommon behavior for or-
ganic molecules. [n]Helicenes have also been used as organ-
ic catalysts for enantioselective transformations,[3c,g] chiral
derivatizing agents,[3d] and building blocks for the design of
chiral ligands for homogeneous transition metal catalysis.[3e,f]
The potential use of helicene derivatives in devices depends
greatly on the facility of generating different frameworks to
modulate and optimize their electronic and optical proper-
ties.[5] Despite remarkable progress in helicene chemistry in
the last decade, the development of short and efficient strat-
egies to diversify helicene structures still remains a chal-
lenge to synthetic chemists.[1,6]
Helicenes are fascinating molecules endowed with helical
chirality and p-conjugated backbones.[1] In recent years, the
chemistry of these compounds has grown from the stage of
an academic curiosity to an important field of research
owing to their extraordinary optical and electronic proper-
ties provided by their unique helicoidal p structures. The
[a] S. Graule, W. Shen, Dr. C. Lescop, Dr. J. Crassous, Prof. R. Rꢀau
Sciences Chimiques de Rennes, UMR 6226 CNRS
Universitꢀ de Rennes 1, Campus de Beaulieu
35042 Rennes Cedex (France)
Fax : (+33)2-23-23-69-39
E-mail: jeanne.crassous@univ-rennes1.fr
regis.reau@univ-rennes1.fr
[b] M. Rudolph, Prof. J. Autschbach
Department of Chemistry, 312 Natural Sciences Complex
State University of New York at Buffalo
Buffalo, NY 14260-3000 (USA)
An appealing alternative to stepwise organic chemistry
for the construction of complex architectures is coordination
chemistry.[7] Indeed, metals are powerful templates for as-
sembling p-conjugated ligands into well-defined molecular
structures such as octupolar complexes,[8] metallocyclo-
phanes,[9] and molecular knots,[10] to name but a few. Consid-
E-mail: jochena@buffalo.edu
[c] Dr. J. A. G. Williams
Department of Chemistry, University of Durham
Durham, DH13LE (UK)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200903234.
5976
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

FULL PAPER
ering a given target architecture, this approach simply in-
volves careful design of the p-conjugated ligands guided by
the basic concepts of coordination chemistry (trans effect,
hemilability, etc.). Furthermore, coordination chemistry
offers a simple way to tune the optical and electronic prop-
erties of the p ligands, since both the coordination-sphere
geometry and the nature of the metal–ligand interaction can
be readily modified by varying the metal center. Surprising-
ly, this general approach has rarely been used for the tailor-
ing of helicoidal p-conjugated systems, and little is known
about the coordination and organometallic chemistry of hel-
icene derivatives.[1a,3e,6b,11,12] We have therefore investigated
the synthesis of helicene derivatives that could form stable
metal bisACHTUNGTRENNUNG(helicene) complexes. Compared to organic synthe-
sis, this approach offers two main advantages. First, starting
from one helicene-based ligand prepared according to a
stepwise synthesis, it would be possible to readily generate
structural variety simply by changing the metal template
(e.g., variation of coordination geometry, electronic proper-
ty, and charge number). Second, one can expect that the
nature of the metal will impact the chiroptical properties of
the helicene-based ligand. Note that this effect will be maxi-
mized if the donor atoms are included in the p-conjugated
systems.
Scheme 1. Synthesis of mixed phosphole–azahelicene p-conjugated com-
pounds 5a–e, and known 2-(2-pyridyl)phosphole 5 f.
and 5-positions. The P atom is electronically coupled with
this conjugated framework, as shown by the fact that chemi-
cal modifications, including coordination to metal centers, of
the P center results in fine-tuning of the optical and elec-
tronic property of the p system.[17]
In spite of its apparent simplicity, the above approach
poses a severe problem, that is, if a ligand bearing an enan-
tiomerically pure helicene is used, the final metal–bis-
Here we describe the synthesis, structural characteriza-
tion, and elucidation of electronic (UV/Vis, fluorescence
and phosphorescence spectroscopy) and chiroptical [molar
rotation (MR), circular dichroism (CD)] properties by both
experimental and theoretical methods of these novel mixed
phosphole–aza[n]helicene p-conjugated systems (Scheme 1)
and of the corresponding CuI and PdII bis(phosphole aza[n]-
helicene) complexes. The high stereoselectivity of the coor-
dination process leading to the metal–bis(phosphole aza[n]-
helicene) assemblies is discussed on the basis of experimen-
tal and theoretical results. The study of this series of com-
plexes bearing helicoidal ditopic ligands reveals that metallic
ions can impact the optical rotations and CD of metal–bi-
s(phosphole-aza[n]helicene) assemblies, which offers an un-
precedented way to tune the chiroptical properties of [n]hel-
icene derivatives.
ACHTUNGTRENNUNG(helicene) assembly must be obtained as a single enantio-
mer. In other words, the coordination process leading to the
target complexes must be highly stereoselective.[13] To meet
this requirement, a classical method in coordination chemis-
try is to tailor the electronic and steric properties of the
ligand donor sites. Amongst the numerous types of ligands
that have been developed, it is well established that hetero-
ditopic donors such as P,N chelates are well suited to stereo-
selective coordination.[14a] For example, the trans effect is
particularly effective in square-planar d8 metal centers and
enables high cis/trans stereoselectivity.[14b,c] Furthermore,
heteroditopic ligands exhibit hemilabile behavior, a dynamic
property which is very important for obtaining the thermo-
dynamically more stable metal–bisACTHNUTRGNEUNG(helicene) assembly.
Therefore, we envisaged appending aza[n]helicenes with a
phosphole ring to obtain rigid 1,4-P,N chelating ligands 5a–e
(Scheme 1). The phosphole moiety was selected because of
its unique electronic properties.[15] This heterole exhibits
weak aromatic character, due to the pyramidal geometry of
the P atom, which prevents the dienic moiety from efficient-
ly interacting with the heteroatom lone pair. According to
theoretical calculations, planar phosphole is not stable, al-
though it has aromatic stabilization comparable to that of
pyrrole (ca. 20–21 kcalmolꢀ1).[16] In fact this aromatic stabili-
zation is not sufficient to overcome the planarization barrier
at the tricoordinate P atom (ca. 35 kcalmolꢀ1), but is respon-
sible for the unusually low P-inversion barrier in phosphole
(ca. 16 kcalmolꢀ1). The weakly aromatic character of the
phosphole ring allows the dienic moiety of this heterole to
be highly conjugated with aromatic substituents at the 2-
Results
Synthesis and characterization of phosphole aza[n]helicene
derivatives: Target s3-phospholes 5a–e bearing an aza[n]he-
licene substituent were prepared according to the Fagan–
Nugent route[18] from diynes 4a–e (Scheme 1). Compounds
5a,b and 5d–f can be regarded as model molecules, since
their azahelicene moiety is not configurationally stable in
solution.
The diyne starting materials have two different structures
with the pyridine moiety in the terminal (4a–c; Scheme 2)
or at an inner position of the helix (4d,e; Scheme 2). They
were obtained by classical synthetic routes. Sonogashira cou-
pling of 1-phenyl-1,7-octadiyne[17] with commercially avail-
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5977

R. Rꢀau, J. Autschbach, J. Crassous et al.
port the proposed structure. The most important spectro-
scopic information for compounds 4a–e are 1) the presence
of four singlets in the 13C{1H} NMR spectra between d=
79.0 and 95.5 ppm, showing the presence of two unsymmet-
rical triple bonds, and 2) that protons, especially H1 (see
Scheme 2 for numbering), experience an anisotropic cone
effect, a feature proving the formation of helices. For exam-
ple, as observed for other azahelicene derivatives,[6] H1,12
which are at the inner positions of the benzo[f]quinoline
and naphtho[1,2-f]quinoline moieties of aza[4]helicenes 4b
,
AHCTUNGTRENNUNG
1
and 4d (Scheme 2) display strongly deshielded H NMR sig-
nals (Table 1). In contrast, aza[6]helicene diyne 4c has a re-
versed anisotropic cone effect, and protons H1,16 display
shielded chemical shifts (Table 1). Amongst these new deriv-
atives, only 4c bears a helicene moiety that is configuration-
ally stable. This aza[6]helicene diyne was successfully re-
solved into its pure enantiomers by HPLC on a Chiralcel
OD-H stationary phase.[12a] Single crystals of the pure enan-
tiomer (ꢀ)-4c suitable for X-ray diffraction study were
grown from a heptane/AcOEt solution at room temperature
and were studied by X-ray crystallography (Figure 1,
Table 2). As observed for other helical conjugated mole-
cules, (ꢀ)-4c aggregates in the solid state (Figure 1). This
aggregation is due to a combination of intermolecular CH–p
interactions between two helicene moieties (T-shaped organ-
ization; CH···Ph 3.5 ꢂ) and the diyne arms and helicene
fragments (CH···Ph 3.5 ꢂ).
Scheme 2. Synthesis of helicene-containing diynes 4a–e. i) nBuLi, THF,
Ar, RT, 3 h. ii) hn, cat I2, toluene, RT, 15 h.
The next step towards the target mixed phosphole–heli-
cene derivatives was oxidative coupling of diynes 4a–e with
zirconocene dichloride followed by [ZrCp2]/PPh exchange
(Scheme 1). This reaction sequence was conducted in THF
according to a known procedure,[17,18] giving rise to air-stable
phospholes 5a–e (Scheme 1), which were isolated in moder-
ate yields (Table 1) after purification by column chromatog-
raphy. The structures of phosphole derivatives 5a–e were
supported by high-resolution mass spectrometry and ele-
able 6-bromopyridine-2-carboxaldehyde and 2-chloro-3-
quinoline carboxaldehyde afforded diynes 1a (63% yield)
and 1b (61% yield; Scheme 2), respectively. These com-
pounds were engaged in Wittig reactions with the appropri-
ate phosphonium bromides 2a–d, prepared according to lit-
erature procedures, to give olefins 3a–e as cis/trans isomer
mixtures (Scheme 2).
1
mental analyses. They exhibited classical H and 13C NMR
Sequential photocyclization and in situ oxidation[1a] of stil-
benoid derivatives 3a–e was performed in toluene in the
presence of catalytic amounts of iodine. Helicene-containing
diynes 4a–e (Scheme 2) were isolated in moderate yields
(Table 1). Note that similar yields were obtained by using
propylene oxide with stoichiometric quantities of iodine.[19]
All of the novel compounds gave satisfactory elemental
analyses. They were fully characterized by high-resolution
mass spectrometry, and the consistence of their NMR data
with those of other diyne and azahelicene derivatives sup-
data, with signals associated with the helicene moieties that
compare well with those recorded for the corresponding
diynes. Their 31P{1H} NMR spectra showed sharp singlets in
the range expected for P-aryl phospholes (Table 1),[15,17] the
signals for 5d and 5e being slightly deshielded, probably be-
cause the phosphole moiety is located in the anisotropy
cone of the azahelicene.
The two stereogenic elements (P center, azahelicene) of
5a,b,d (Scheme 1) undergo rapid inversion in solution, and
only one set of NMR signals is observed at room tempera-
ture. For compound 5e, two
sets of NMR signals are ob-
served due to the presence of
Table 1. Selected NMR data and yields for diynes 4a–e (Scheme 2) and phospholes 5a–e (Scheme 1).
the
enantiomerically
pure
Cmpd
d(1H)[a] [ppm]
Yield [%]
Cmpd
d
(31P{1H})[c] [ppm])
Yield [%]
pinene substituent. Phosphole
5c (Scheme 1), which is the
only derivative featuring a con-
figurationally stable aza[6]heli-
cene moiety, was investigated in
more detail. Its 31P{1H} NMR
4a
4b
4c
4d
4e
8.81, 8.52 (H1, H10)
9.17, 8.76 (H1, H12)
7.81, 7.61 (H1, H16)[b]
9.09 and 8.96 (H1, H12)
8.97 (H1)
50
47
48
62
25
5a
5b
5c
5d
5e
+12.3
+12.2
+14.0, +14.5
+20.7
+20.4, +20.7
73
63
47
45
27
[a] At 300 MHz in CDCl3. [b] At 500 MHz in CDCl3. [c] At 81 MHz in CDCl3.
5978
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
spectrum displays two sharp
singlets of comparable intensity
(Table 1). In fact, when starting
with a diyne bearing an enan-
tiomerically pure helix (for ex-
ample
diastereomeric phospholes (i.e.,
(P,SP)-5c1 (P,RP)-5c2,
and
P configuration), two
Table 3) are obtained due to
the presence of the stereogenic
P atom. Their mirror images
(i.e., (M,RP)-5c1 and (M,SP)-
5c2, Table 3) were synthesized
from the
M
azahelicene.
Indeed, a variable-temperature
31P{1H} NMR study revealed an
inversion barrier between 5c1
and 5c2 of 16 kcalmolꢀ1 at
330 K. Phospholes 5a–c were
also characterized by an X-ray
diffraction study confirming the
proposed structures (Tables 2
and 4, Figure 1).
Figure 1. Packing of M-(ꢀ)-4c in the crystal along the x axis. Molecular structures of phospholes 5a–c deter-
mined by X-ray crystallography.[12]
Table 2. X-ray data of 4c, 5a–c.
4c
5a
5b
5c
formula
CCDC number
Mr
C39H27N1
736269
509.62
C33H26N1P1
736270
467.52
C37H28N1P1
651119
517.57
C45H32N1P1
707415
617.69
T [K]
100(2)
100(2)
100(2)
100(2)
crystal system
space group
a [ꢂ]
b [ꢂ]
c [ꢂ]
monoclinic
P21
9.233(1)
14.253(2)
11.032(1)
90
111.861(1)
90
1347.4(4)
2
triclinic
P1
9.831(1)
13.655(1)
19.098(2)
triclinic
P1
triclinic
P1
9.701(2)
11.374(2)
14.341(2)
78.216(10)
87.051(11)
83.926(12)
1539.6(4)
2
¯
¯
¯
9.5107(4)
10.9272(5)
14.1036(6)
72.146(3)
70.845(3)
84.199(3)8
1317.89(10)
2
a [8]
b [8]
74.935ACTHUGNETRNNU(G 4
76.380(3)
83.273(3)
2401.8(4)
4
g [8]
V [ꢂ3]
Z
Color
colorless
0.40ꢃ0.30ꢃ0.13
1.256
yellow
0.10ꢃ0.05ꢃ0.05
1.293
yellow
0.2ꢃ0.05ꢃ0.03
1.304
yellow
0.08ꢃ0.04ꢃ0.02
1.332
crystal size [mm]
1calcd [Mgmꢀ3
]
F
G
536
984
544
648
m
G
]
0.072
0.137
0.132
0.126
l [ꢂ]
0.71069
0.71069
0.71069
0.71069
diffractometer
index ranges hkl
APEX II Bruker-AXS
ꢀ11ꢁhꢁ11,
ꢀ13ꢁkꢁ17,
ꢀ13ꢁlꢁ13
1.99ꢁqꢁ26.38
9496
2873
2633
2873/1/362
1.134
R1=0.0364,
wR2=0.0776
R1=0.0413,
wR2=0.0841
0(4)
APEX II Bruker-AXS
ꢀ12ꢁhꢁ12,
ꢀ17ꢁkꢁ17,
ꢀ24ꢁlꢁ24
2.12ꢁqꢁ27.57
34837
11042
7168
11042/0/632
1.028
R1=0.0452,
wR2=0.1050
R1=0.0892,
wR2=0.1257
–
APEX II Bruker-AXS
ꢀ11ꢁhꢁ11,
ꢀ12ꢁkꢁ12,
ꢀ16ꢁlꢁ16
5.24ꢁqꢁ24.71
18731
4428
3255
4428/0/352
1.022
R1=0.0393,
wR2=0.0795
R1=0.0659,
wR2=0.0907
–
APEX II Bruker-AXS
ꢀ12ꢁhꢁ12,
ꢀ14ꢁkꢁ14,
ꢀ12ꢁlꢁ18
1.45–27.56
11764
6896
3479
6896/0/424
1.022
R1=0.0791,
wR2=0.1615
R1=0.1713,
wR2=0.1921
–
q range [8]
reflections collected
independent reflections
reflections [I>2s(I)]
data/restraints/parameters
goodness of fit on F2
final R indices [I>2s(I)]
R indices (all data)
absolute structure parameter
largest diff. peak and hole [eꢂꢀ3
]
0.187 and ꢀ0.165
0.368 and ꢀ0.480
0.394 and ꢀ0.387
0.451 and ꢀ0.388
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5979

R. Rꢀau, J. Autschbach, J. Crassous et al.
Table 3. Relative energies [kcalmolꢀ1] and corresponding mole fractions
Xi at 298.15 K of four optimized configurations (gas phase: BP/SV(P)
level; BP/SV(P) level with COSMO model for chloroform).
large 31P NMR downfield chemical shifts due to coordina-
tion (d=60 and 72 ppm) are consistent with the formation
of five-membered palladacycles, and these complexes dis-
1
play similar H NMR characteristics for the 1,5-diphenyl-2-
(2-pyridyl)phosphole fragments to those of the related [(2-
pyridylphosphole)PdCl2] complexes.[20] The proposed struc-
tures were confirmed by an X-ray diffraction study on [Pd-
AHCTUNGTRENNUNG
AHCTUNGTREN(GUNN CH3CN)4]-
Gas phase
Xi
COSMO
Xi
(P,SP)-s-trans-5c1
(P,SP)-s-cis-5c1
(P,RP)-s-trans-5c2
(P,RP)-s-cis-5c2
0.0
0.7
2.0
2.0
72.7
22.3
2.4
0.0
0.3
2.4
1.3
59.5
33.0
1.1
AHCTUNGTRENNUNG
A
E
N
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
2.6
6.4
analyses are consistent with the formation of Pd–bis(phosp-
hole azahelicene) assemblies. Their 31P{1H} NMR spectra ex-
hibit sharp singlets ([Pd
(5b)2]2+, d=+76.6 ppm; [Pd(5c)2]2+
ACHTUNGTRENNUNG ACHTUNGTRENNUNG ,
d=+81.6 ppm; [Pd
AHCTUNGTRENNUNG
Table 4. Selected bond lengths [ꢂ] and angles [8] for 5a–c and the corre-
sponding calculated values [in brackets] for 5c.[a]
set of 1H signals attributable to the phosphole azahelicene li-
gands is observed. High-frequency 31P NMR chemical shifts
are typical for dicationic PdII complexes bearing (2-pyridyl)-
phosphole ligands in a mutually cis configuration.[12,20] The
simplicity of these NMR data shows that these complexes
are formed as single diastereomers. Complex [PdACHTUNGTRENNUNG
(5d)2]2+
5a
5b
5c
was characterized by X-ray diffraction (Table 5 and 6,
Scheme 4). A view showing the supramolecular arrangement
ꢀ
P1 C1
ꢀ
C1 C2
1.816
1.3713
1.463
1.372
1.795
1.472
1.794
1.370
1.465
1.367
1.805
1.468
1.474
1.816 [1.826]
1.373 [1.398]
1.479 [1.468]
1.354 [1.390]
1.815 [1.827]
1.494 [1.469]
1.474 [1.470]
of [PdACHTNUGRTNEUNG
(5d)2]2+ is provided in Figure 2.
ꢀ
C2 C7
The reaction of 5b–d with a CuI source (2:1 molar ratio)
ꢀ
C7 C8
in CH2Cl2 solution at room temperature gave rise to the
ꢀ
C8 P1
new complexes [Cu
structures were supported by elemental analysis, and com-
plex [Cu
(5c)2]+ was characterized by HRMS. The 31P{1H}
(5b–d)2]+ (Scheme 5). The proposed
ACHTUNGTRENNUNG
ꢀ
C8 C9
ꢀ
C1 C15
1.464
C7-C8-C9-C10
N1-C15-C1-C2
Helical
38.40(5)–42.82(6)[c]
ꢀ37.70(8) ꢀ38.22(1) [ꢀ40.395]
AHCTUNGTRENNUNG
ꢀ0.44(3) to 1.71(6)[c] 39.92(5)
ꢀ26.27(4) [ꢀ4.721]
NMR spectra of these derivatives in CH2Cl2 display a broad
singlet at d=5–6 ppm (n1/2 =85 Hz), a typical shift for cat-
ionic CuI (2-pyridyl)phosphole complexes.[20] Upon exposure
to pentane vapors, air-stable single crystals of complexes
5.5
30.1
45.8 [48.6]
dihedral angle[b]
[a] Calculated structure (P,SP)-s-trans-5c1. [b] Angle between the planes
defined by the terminal pyridyl and phenyl groups. [c] Two values found
in the X-ray structure.
[CuACTHNUGRTNENUG ACHUTNGTRENNUGN
(5b)2]+ and [Cu(5d)2]+ (Scheme 5, Tables 5 and 6) suita-
ble for X-ray diffraction studies (Table 2) were obtained
from the homogeneous reaction mixtures (yields >77%).[12b]
Figure 2 shows the supramolecular organization of columns
Ligand behavior of mixed azahelicene phosphole derivatives
of [CuACTHGNUTERNNUG
(5d)2]+ complexes.
towards PdII and CuI centers: The neutral complexes [Pd-
A
E
Photophysical properties of azahelicene diynes, phospholes,
and metal assemblies
CTHUNGTRENNUNG
Azahelicene diynes: Initially, the UV/Vis and luminescence
properties of the structurally isomeric diynes 4b,d
(Scheme 2) were studied in detail to gain insight into the op-
tical properties of this novel helicene series. The profiles of
their UV/Vis spectra are characteristic of helicene deriva-
tives, with intense absorptions at high energy (lmax =270–
300 nm, e>104 mꢀ1 cmꢀ1), and
a series of well-defined
weaker bands beyond 350 nm (eꢂ103 mꢀ1 cmꢀ1; Figure 3). In
solution at room temperature, both compounds display
highly structured fluorescence spectra (Figure 3 and
Table 7), with vibrational progressions of 1300 and
1100 cmꢀ1 and emission quantum yields of 8 and 5% for 4b
and 4d, respectively. The fluorescence lifetimes on the order
of 5 ns are typical of extended aromatic p systems. In a rigid
Scheme 3. Synthesis of neutral complexes [PdACTHNUGTRENNUG(5b)Cl2] and [PdACHTUNGTREN(NUGN 5d)Cl2].
5980
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
Table 5. X-ray structure determination summary of complexes [Pd
[Pd(5d)2](SbF6)2
G
E
E
(PF6), [Cu
R
ACHTUNGTNER(NUNG 5b)Cl2].
N
G
E
G
N
PdACTHUNGTER(UNNG 5b)Cl2·CH2Cl2
formula
CCDC number
Mr
C75.5H59Cl3Pd1N2P2F12Sb2
651120
1740.44
C76.5H61Cu1N2P6F6Cl5
651121
1455.97
C153H122Cu4N4P4Cl14
651122
2890.89
C38H28Cl4N1P1Pd1
736271
777.78
T [K]
100(2)
100(2)
100(2)
100(2)
crystal system
space group
a [ꢀ]
b [ꢀ]
c [ꢀ]
tetragonal
I41/a
18.695(5)
18.695(5)
81.28(2)
90
90
90
triclinic
P1
monoclinic
C2/c
32.938(2)
19.965(1)
24.449(1)
90
124.683(2)
90
13221.0(12)
4
orthorhombic
Pcab
12.586(2)
22.480(3)
23.240(3)
90
90
90
6575(2)
8
¯
14.431(2)
21.614(3)
22.007(3)
82.315(4)
85.456(5)
75.858(4)
6588.7(16)
4
a [8]
b [8]
g [8]
V [ꢂ3]
28408(11)
16,
Z
color
red
orange
yellow
red
crystal size [mm]
0.2ꢃ0.1ꢃ0.05
1.628
0.1ꢃ0.1ꢃ0.03
1.468
0.2ꢃ0.2ꢃ0.1
1.452
0.20ꢃ0.08ꢃ0.04
1.571
1calcd [Mgmꢀ3
]
F
G
13808
2988
5928
3136
m
G
]
1.240
0.672
0.102
0.968
l [ꢂ]
0.71069
0.71069
0.71069
0.71069
diffractometer
index ranges hkl
APEX II Bruker-AXS
ꢀ20ꢁhꢁ20,
ꢀ17ꢁkꢁ20,
ꢀ90ꢁlꢁ90
5.10ꢁqꢁ23.258
86356
10112
5634
10112/397/1022
1.128
R1=0.1074,
wR2=0.2213
R1=0.1827,
wR2=0.2542
–
APEX II Bruker-AXS
ꢀ16ꢁhꢁ16,
ꢀ24ꢁkꢁ24,
ꢀ20ꢁlꢁ24
5.10ꢁqꢁ23.268
37977
18196
9642
18196/788/1944
1.028
R1=0.0793,
wR2=0.1887
R1=0.1628,
wR2=0.2282
–
APEX II Bruker-AXS
ꢀ32ꢁhꢁ41,
ꢀ24ꢁkꢁ24,
ꢀ30ꢁlꢁ26
5.11ꢁqꢁ26.378
68081
13411
10702
18196/325/969
1.028
R1=0.0485,
wR2=0.1187
R1=0.0647,
wR2=0.1302
–
APEX II Bruker-AXS
ꢀ15ꢁhꢁ15,
ꢀ28ꢁkꢁ17,
ꢀ27ꢁlꢁ27
2.43ꢁqꢁ26.59
34974
6656
3833
6656/0/425
1.083
R1=0.0834,
wR2=0.1709
R1=0.1551,
wR2=0.1949
–
q range [8]
reflections collected
independent reflections
reflections [I>2s(I)]
data/restraints/parameters
goodness of fit on F2
final R indices [I>2s(I)]
R indices (all data)
absolute structure parameter
largest diff. peak and hole [eꢂꢀ3
]
1.324 and ꢀ1.249
0.842 and ꢀ0.911
1.580 and ꢀ1.127
1.580 and ꢀ1.127
glass at 77 K, the fluorescence
is not shifted, and this reflects
the rigidity of the p skeleton
and the localized nature of the
excited state. At this tempera-
ture, the fluorescence is accom-
panied by well-defined struc-
tured phosphorescence in the
green region of the spectrum,
with a vibrational progression
of about 1500 cmꢀ1 (Figure 3
and Table 7). The temporal
decay of the phosphorescence is
easily visible to the eye, occur-
ring on a timescale of seconds.
The singlet-state fluorescence is
lower in energy for iso deriva-
tive 4d than for its linear
isomer 4b (DE=830 cmꢀ1),
whereas the triplet phosphores-
cence shows the opposite trend
(DE=350 cmꢀ1): the singlet–
triplet energy gaps are estimat-
Scheme 4. Synthesis of complexes [PdACHTNUTRGENNUG HCATUNGTRENNGUN
(5b–d)2]2+. View of complex [Pd(5d)2]2+ established by X-ray diffraction
study (H atoms, counteranions, and solvent molecules have been omitted). One of the two symmetrically inde-
pendent complexes is shown.
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5981

R. Rꢀau, J. Autschbach, J. Crassous et al.
(5b)2]2+, [Cu(5b)2]+, [Cu(5d)2]+, and [Pd
Table 6. Selected bond lengths [ꢂ] and angles [8] for [PdACHTUGNTNREUNNG ACHUTTGNRNENUG ACHTNUEGRTNNUNG ACTHUNGTREN(NGUN 5b)Cl2].
elongating the chromophore.
The influence on the triplet
state is smaller (ca. 650 cmꢀ1),
based on the shift in the phos-
phorescence spectrum.
Azahelicene phospholes: The
[Pd
G
[Pd
N
[Cu
A
[Cu
A
[PdACHTNUGTRENNNUG
ꢀ
P1 C1
ꢀ
C1 C2
1.753
1.322
1.504
1.360
1.801
1.495
1.509
58.765
54.299
28.46–32.90
1.835
1.380
1.485
1.385
1.835
1.478
1.453
55.227
53.03
1.783
1.361
1.498
1.349
1.804
1.468
1.463
52.806
39.085
21.92–22.3
1.801
1.359
1.468
1.359
1.809
1.475
1.466
51.629
52.447
28.52–28.84
1.789
1.347
1.490
1.343
1.810
1.484
1.464
12.73
41.123
33.14
play strong absorption in the
visible range of the electronic
spectrum (Table 7, Figure 4).
According to theoretical calcu-
lations, these bands can be as-
signed to p–p* transitions (vide
infra). A detailed analysis of
the structure/absorption proper-
ty relationship is provided in
the Discussion section. The
study on the emission proper-
ties of mixed phosphole–azahe-
licene derivatives started with
investigations on the known
1,2-diphenyl-5-(2-pyridyl)-
ꢀ
C2 C7
ꢀ
C7 C8
ꢀ
C8 P1
ꢀ
C8 C9
ꢀ
C1 C15
C7-C8-C9-C10
C2-C1-C15-C16
helical dihedral
angle[c]
49.91–53.22
ꢀ
P1 M1
2.248
2.136
2.248
2.109
81.031
96.248
168.177
81.572
167.342
17.96
2.302
2.200
2.319
2.188
79.934
98.612
162.285
79.725
165.276
24.75
2.280
2.084
2.295
2.083
85.020
117.964
125.250
87.514
126.297
83.58
2.295
2.105
2.298
2.090
86.277
120.280
122.511
86.330
125.904
86.35
2.239
2.096
–
–
ꢀ
N1 M1
ꢀ
P2 M1
ꢀ
N2 M1
N1-M1-P1
N1-M1-N2
N1-M1-P2
N2-M1-P2
N2-M1-P1
twist angle
81.520
–
–
–
–
–
phosphole (5 f)[17] (Scheme 1) as
a model compound with which
comparison can be made. In so-
lution at 298 K, 5 f displays a
structureless, broad fluores-
cence band centered at 475 nm
(F=3%, t<1 ns) with a large
Stokes shift of 7000 cmꢀ1. At
77 K, the fluorescence emission
is blueshifted (l=436 nm, t=
4.7 ns) and no phosphorescence
could be detected. At room
temperature, phosphole 5b
bearing the linear aza[4]heli-
cene moiety (Scheme 1) dis-
[a] Data obtained by X-ray crystallography. [b] Data obtained by TDDFT.[c] Angle between the planes defined
by the terminal pyridyl and phenyl groups.
plays
a
broad fluorescence
emission centered at 499 nm
(F=8%, tꢂ0.8 ns), irrespec-
tive of the excitation wave-
length, with a good match be-
tween the absorption and exci-
tation spectra (Figure 5). At
77 K, the behavior is more com-
plicated. Upon excitation at
short wavelengths (ca. 350 nm),
helicene-like fluorescence is ob-
served (l=378 nm, t=12 ns)
accompanied by a broad, rather
Figure 2. Simplified views of the supramolecular arrangement of complexes [Pd
(b).
(5d)2]2+ (a) and [Cu(5d)2]+
ACHUTGTNRENNUG CAHTUNGTRENNUGN
ed at 6300 and 5100 cmꢀ1 for 4b and 4d, respectively. The
more extended helicene diyne 4c (Scheme 2) displays simi-
lar spectra to the shorter parent 4b, but with a significant
redshift in the absorption and fluorescence spectra indicat-
ing stabilization of the S1 state by around 2000 cmꢀ1 upon
ill-defined emission between 450 and 600 nm (Figure 5).
The emission decay in this region displays short-lived
components on the order of 10 ns, but also a long-lived com-
ponent of 1.5 s. The latter is comparable to the phosphores-
cence displayed by the azahelicene diynes, and the time-re-
5982
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
solved emission spectrum re-
corded at time delays greater
than 100 ms is also similar to the
phosphorescence of the azaheli-
cene unit. Thus, it appears that,
under these conditions at low
temperature, emission can ema-
nate from helicene-like excited
states, as well as from lower-
energy singlet states delocalized
across the molecule.
Its isomer 5d bearing the iso-
helicene substituent (Scheme 1)
displays similar behavior at
77 K. In this case, however, the
dual emission is also evident at
298 K (Figure 6). Thus, excita-
tion at 350 nm leads to struc-
Scheme 5. Synthesis of cationic complexes [Cu
[Cu
(5d)2]+ (one stereoisomer; H atoms, counteranions, and solvent molecules have been omitted for clarity).
(5b–d)2]+. Solid-state structure of complexes [Cu(5b)2]+ and
ACHUTGTNRENNUG CAHTUNGTRENNUGN
ACHTUNGTRENNUNG
tured
helicene-like
bands
around 400 nm, in addition to a
Figure 3. Absorption (solid line, left), excitation (lem =410 nm, dotted
line) and emission (lex =360 nm, dashed line) spectra of diyne 4b in
CH2Cl2 at 298 K, and the emission spectrum (solid line, right) in a diethyl
ether/isopentane/ethanol (2:2:1 v/v) glass at 77 K.
Figure 5. Absorption (solid line, left), excitation (lem =500 nm, dotted
line) and emission (lex =360 nm, dashed line) spectra of 5b in CH2Cl2 at
298 K, and the emission spectrum in an diethyl ether/isopentane/ethanol
(2:2:1 v/v) glass at 77 K (solid line, right).
broader emission band at 460 nm. The emission bands have
different lifetimes (Table 7) and significantly different exci-
tation spectra, confirming that the units responsible for the
two emission bands are not fully coupled. Phosphole 5c
(Scheme 1) similarly displays a broad fluorescence emission
centered at 500 nm at 298 K, with biexponential decay kinet-
ics (Table 7). At 77 K, the emission spectrum of 5c consists
of a broad band centered at 490 nm (t=2.6 ns, with a long
phosphorescence component of 1.3 s also detectable in the
low-energy tail), together with some structured, helicene-
like emission to higher energy (l=425 nm, t=8.5 ns).
PdII and CuI complexes: Compared to the spectrum of free
ligands 5b,d, the p–p* transitions of the corresponding
Figure 4. Electronic spectra of mixed azahelicene phosphole derivatives
5a–e in CH2Cl2.
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5983

R. Rꢀau, J. Autschbach, J. Crassous et al.
Table 7. Photophysical data for azahelicene diynes 4b–d, azahelicene phospholes 5b–d (with pyridyl phosphole 5 f for comparison), and their PdII and
CuI complexes. Data at 298 K in CH2Cl2; at 77 K in diethyl ether/isopentane/ethanol (2:2:1 v/v).
abs
max
RT
flou
fluo
RT
fluo
RT
77 K
flou
77 K
flou
77 K
phos
77 K
phos
Derivatives
l
[nm]
l
[nm]
F
t
[ns]
l
[nm]
t
[ns]
l
[nm]
t
(e
N
[%]
8
[ns]
1.9
4b
4c
4d
289 (28600), 308 (12300), 326 (7290), 364 (3070), 382 (3300)
388, 408, 432
4.2
5.9
4.9
0.8
388, 409,
433
420, 445,
473
401, 423,
449
378, 398
425 and
490
8.6
9.3
7.8
12
514, 556,
602
532, 578,
626
505, 546,
592
ca. 500
265 (48700), 321 (18000), 331 (17500), 352 (8510), 393 (1560), 422, 446, 472
416 (1270)
285 (43800), 313 (12800), 334 (4680), 351 (2340), 370 (3550),
388 (3670)
285 (66600), 406 (28800)
265 (67500), 326 (21000), 395 (15900), 430 (12700)
6
5
1.5
1.0
401, 419, 444
5b
5c1,c2
499
502
8
–
1.5
1.3
6, 1.2
8.5, 2.6 ca. 550
5d
266 (36200), 357 (5780), 375 (4900)
389, 404
(sharp)
–
3
406
8
475
2.5
456 (broad)
475
480
10
0.5
[a]
[a]
5 f
5b-PdCl2
5d-PdCl2
[Pd
[Pd
[Cu
[Cu
364 (7650)
3
–
–
–
–
–
–
436
4.7
[b]
3.6, 7.2
–
5
15
–
–
–
–
–
–
[a]
[a]
478 sh (2400)
482 sh (2120)
2.5, 6.3 [b]
3.5, 14 460
8.3, 2.4
4.8, 3
10, 3.3 417
3.6, 3 410
[a]
[a]
470
(5b)2]2+ 399 (11600), 525 (1400)
382 and 445
380 and 454
400 and 450
410 and 490
–
450
540
550
510
ca. 530
1.5
2.3
1.9
1.8
E
N
R
6.5
[a] No phosphorescence detected. [b] Series of ill-defined peaks.
Figure 7. Electronic absorption spectra of compound 5b and complexes
[(5b)PdCl2], [Pd
(5b)2]2+ and [Cu(5b)2]+ (5ꢃ10ꢀ5 m in CH2Cl2).
A
ACHTUNGTRENNUNG
of the CuI- and PdII(phosphole azahelicene)2 assemblies are
essentially comparable, with the Cu complexes having
higher molar extinction coefficients (Table 7, Figure 7). The
most notable difference between the absorption spectra of
these two series of complexes is the absence of long-wave-
length absorptions for the CuI(phosphole azahelicene)2 com-
Figure 6. Absorption and emission (lex =350 nm) spectra of 5d in CH2Cl2
at 298 K (solid lines, left and right), and the excitation spectra registered
at 405 and 460 nm (dotted and dashed lines respectively).
PdCl2 complexes [Pd
A
(5d)Cl2] are much
plexes [CuACTHNGUTRENNUG ACHTGNUTRENNNUG
(5b)2]+ and [Cu(5d)2]+ (Table 7, Figure 7).
weaker, and new absorptions of low intensity at about
480 nm are observed (Table 7). These transitions can be as-
signed to charge transfers (CT) involving the metal and the
azahelicene phosphole ligand, as already observed in related
PdII 2-pyridylphosphole complexes (vide infra).[20] The ab-
sorption behavior of dicationic Pd(phosphole azahelicene)2
The emission displayed by both series of complexes at
298 K is very weak, whilst the behavior at 77 K is complicat-
ed by dual emissions. It is thus difficult to draw a clear pic-
ture of the emission properties. In the case of [PdACTHUNRGTNEUNG(5b)Cl2], a
broad structured emission band centered at 480 nm is ob-
served, independent of the excitation wavelength, the tem-
poral decay of which exhibits two components on the order
of 2.5 and 6.3 ns. Note that the data for the complex 5d are
assemblies [PdACHTUNGTRENNUNG ACHUTNGTRENNGUN
(5b)2]2+ and [Pd(5d)2]2+ is essentially compa-
rable to that of the corresponding PdCl2 complexes. Upon
coordination to the PdII center, the intensity of the p–p*
transitions of the free ligands decreases and new transitions
of low intensity are observed at longer wavelengths
(Table 7, Figure 7). The position of the absorption maxima
largely comparable to those of 5b. For [PdACTHNUTRGNEUNG
(5b)2]2+ at 298 K,
excitation at short wavelengths (<350 nm) gives rise pre-
dominantly to structured helicene-like fluorescence (t=
8.3 ns) bands (Figure 8). For longer excitation wavelengths,
5984
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
The CD spectra were recorded in CH2Cl2 solutions at con-
centrations around 5ꢃ10ꢀ5
m and display mirror-image
values for the + and ꢀ enantiomers. The CD spectrum of P-
4c (Figure 9) shows two bands with high De values around
Figure 8. Normalized emission spectra of [PdACTHNUTRGNEUNG
(5b)2]2+ in CH2Cl2 at 298 K
following excitation at the wavelengths indicated.
Figure 9. CD spectra of compounds 4c, 5c, [PdCAHTUGTNRENNUG(5c)2]ACHTUNTGRNE(NUGN SbF6)2, and [Cu-
A
ACHTUNGTRENNUNG
a broad band centered around 445 nm appears (t=2.4 ns,
Figure 8). On the other hand, excitation into the low-energy
CT absorption bands gives rise to no detectable emission.
At 77 K, long-lived phosphorescence centered at 540 nm is
observed (t=1.5 s, Table 7).
270 nm (ꢀ) and around 330 nm (+), which allow unambigu-
ous assignment of the absolute configuration by comparison
with previously known helicenes.[21a,b] For phosphole 5c, the
mixture of diastereomers (P,SP)-5c1 and (P,RP)-5c2 gives
lower CD values together with additional very small CD
bands at l>370 nm. Moreover, the CD spectra of (P,P)-[Pd-
The emission behaviors of CuI complexes [Cu
(5b)2]+ and
ACHTUNGTRENNUNG
[Cu
ACHTUNGTRENNUNG
sponding PdII assemblies (Table 7). At 298 K, these com-
plexes are weakly fluorescent (t=2.4–10.0 ns, Table 7),
while at 77 K the spectra again seem to result from a super-
imposition of fluorescent and phosphorescent bands
(Table 7).
ACHUTNGRENNUG CAHTUNGTRENNUGN
(5c)2]2+ and (P,P)-[Cu(5c)2]+ are quite different. They both
display two intense CD bands around 270 nm (ꢀ) and
around 330 nm (+). However, the magnitude of the CD
spectrum of PdII complex 3 is much larger than that of CuI
Chiroptical properties (molar rotation and circular dichro-
ism) of azahelicene diynes, phospholes, and metal com-
plexes: The specific molar rotations (MR in degcm2 dmolꢀ1)
of enantiopure compounds were determined in CH2Cl2 solu-
tions at concentrations around 0.01 g/100 mL at 238C with
the sodium D line. The + and ꢀ enantiomers display
mirror-image values within an error of ꢃ2%. Enantiopure
aza[6]helicene diyne P-4c (Scheme 2) displays a high MR
value of ½ꢀꢄ2D3 =+10240 (Table 8), comparable to that of P-
hexacarbohelicene (½ꢀꢄ2D3 =11950).[21a] The specific MR of
phosphole P-5c1,2 is lower (½ꢀꢄ2D3 =+8200) and corresponds
to a mixture of diastereomers ((P,SP)-5c1 and (P,RP)-5c2). In
complex 4 (Figure 9). The CD spectrum of (P,P)-[PdACHTUNGTRENNUNG
(5c)2]2+
displays additional intense bands at about 370 nm as well as
weak bands at longer wavelengths (410–450 nm) that are
not observed for (P,P)-[CuACHTNUGRTNEUNG
(5c)2]+ (Figure 9).
Theoretical study: Theoretical calculations were performed
on phosphole P-5c and its complexes (P,P)-[Pd
(5c)2]2+ and
(P,P)-[Cu
(5c)2]+ to gain more insight into the structure–chi-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
roptical property relationships. The most stable structures of
5c and its complexes were first determined by DFT calcula-
tions, and then the molecular orbitals were obtained and the
UV/Vis and CD spectra calculated by time-dependent (TD)
DFT. Finally, the MR at 589 nm was computed. In this sec-
tion, molar rotations (in units of degcm2 dmolꢀ1) are scaled
by a factor of 1:1000, that is, molar rotations are reported in
units of 103 degcm2 dmolꢀ1, which we abbreviate as mru.
During the course of this study we discovered that the li-
gands themselves are highly in-
comparison, complex (P,P)-[PdACHTUNGTRNEUNG
(5c)2]2+ displays a huge spe-
cific MR (½ꢀꢄ2D3 =+23100), which is much larger than that of
its CuI analogue (P,P)-[Cu
ACHTUNGTRENNUNG
of diyne 4e and phosphole 5e are very weak, since they ex-
hibit central chirality.
teresting, flexible systems with
Table 8. Experimental and calculated molar rotations of enantiopure compounds.
(5c)2]2+
N
(5c)2]+
ACHTUNGTRENNUNG
strong chiroptical properties.
We show below that the main
role of the metal atom is to
direct the geometry of the li-
gands. Therefore, to better un-
derstand the chiroptical proper-
P-4c
E
U
P-5c
R
23[a,b]
½ꢀꢄD
+10240
(P,P,RP,RP,DPd)-[Pd
+26660
+8200
N
A
ACHTUNGTRENNUNG
[c]
½ꢀꢄ
[a] Measured in CH2Cl2. [b] In degcm2 dmolꢀ1 and within an error of ꢃ2%. [c] TDDFT calculation on diaste-
reomer 5c1.
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5985

R. Rꢀau, J. Autschbach, J. Crassous et al.
ties of the metal complexes it is necessary to perform a de-
tailed analysis of the azahelicene phosphole ligands first.
consider separately a symmetric part PSai =(X+Y)ai and an
antisymmetric part PaAi =(XꢀY)ai, from which the transition
density and current density can be computed in the MO
basis.[21b,30] TDDFT programs typically determine the coeffi-
cients Xai =Pai and Yai =Pia, or alternatively their sum and
difference, for each transition.[31,32] The normalization is
such that[33] Equation (3) holds.
Computational details: All structures were optimized with
the Turbomole 5.7.1 program[22a] using the double-z SV(P)
basis set,[23] which includes polarization functions on all of
the atoms in the system except hydrogen, and the Becke–
Perdew (BP) functional,[24] by using the RI-DFT method
and accompanying auxiliary basis sets.[22b,c] Absorption (UV/
Vis) and circular dichroism (CD) excitation spectra, as well
as optical rotations (ORs) at the sodium D line wavelength
(589.3 nm) were calculated with the Becke half-and-half
functional (BHLYP)[25] and SV(P) basis set. Benchmark cal-
culations showed that this basis set/functional combination
describes the valence excitations in the UV/Vis range rea-
sonably well.[12a] All spectra were broadened with Gaussian
functions with a s value of 0.26 eV. Solvation effects for the
phosphole ligands were investigated by inclusion of the con-
ductor-like screening model (COSMO)[26] for chloroform
(dielectric constant of 4.8). Further investigations of the
electronic structure of the Pd and Cu phosphole complexes
were performed by using the NBO[27] analysis capabilities of
the Gaussian 03[28] code (at the same BHLYP/SV(P) level of
theory).
unocc occ
X X
ðX þ YÞaiðX ꢀ YÞai ¼ 1
ð3Þ
a
i
Thus, each excitation can be analyzed conveniently in
terms of contributions from occupied–unoccupied MO pairs
adding up to 1 (100%). A relatively pure transition may
have a contribution on the order of about 80% or higher
from a single pair fifa. Many of the coefficient products for
the occ–unocc pairs will be small; however, typically there
are dominant contributions in low-lying transitions. For ex-
ample, an excitation with 85% HOMO-to-LUMO character
has (X+Y)HOMOꢀLUMO (XꢀY)HOMOꢀLUMO =0.85. Most
TDDFT programs print contributions from this, or a related,
analysis of the transitions.
Consider next an orthonormal set of LMOs l along with
the transformation that describes the canonical MOs in the
localized orbital basis [Eq. (4)].
In some cases a sum-over-states (SOS) analysis [Eq. (1)]
was applied to the optical rotations
X
k
fj ¼
lkTkj
X
R0n
w20n ꢀ w2
ð4Þ
k
½ꢀꢄw ¼ 91:43028 w2
ð1Þ
k
n¼1
Instead of the NLMOs one may use other localized sets
such as obtained from the Boys,[34a] Edmiston–Rueden-
berg,[34b] or von Niessen[34c] criteria, or Pipek–Mezey[34d]
LMOs, to name a few alternatives. Substituting Kronecker d
terms into Equation (3) yields Equation (5), where
a,b,c 2 unocc and i,j,k 2 occ, where a,b,i,j are MO indices and
c,k indices for the localized orbitals.
where [f] is the molar rotation in units of degcm2 molꢀ1, w
the angular frequency of light, w0n the frequency of the nth
excitation, R0n is the rotatory strength of the nth excitation
in units of 10ꢀ40 esu2 cm2 (c.g.s. system), and k the number of
excitations included in the SOS. The optical rotations com-
puted directly (referred to as linear-response optical rota-
tions from now on) correspond to the SOS limit whereby k
covers all excitations that are possible with a given basis set.
We also analyzed some CD spectra in terms of MO-to-
MO and LMO-to-LMO (LMO=localized molecular orbi-
tal) contributions. In the present study, natural LMOs
(NLMOs) were used, as devised by Weinhold et al.[29] In the
context of this work, the analysis in terms of LMOs allows
for convenient identification of the metal orbitals in each
excitation without referring to the primitive or contracted
basis set. In linear-response TDDFT, the transition density
for an excitation can be calculated from a density matrix
and the Kohn–Sham MOs, assumed here to be real, as
Equation (2)
unocc
occ
X
X
X
1 ¼
ðX þ YÞaiðX ꢀ YÞbj
X X
dab
dij
ai
b
j
ð5Þ
¼
ðX þ YÞaiðX ꢀ YÞbjTcaTcbTkiTkj
ai bckj
We define LSck and LcAk in Equations (6), which are the
symmetric and antisymmetric components of the transition
density matrix in the LMO basis
X
LSck
LAck
¼
¼
ðX þ YÞaiTcaTki
ðX ꢀ YÞbjTcbTkj
ð6aÞ
ð6bÞ
ai
X
X
bj
10!j
¼
PaiðwjÞfifa
ð2Þ
ai
and the LMO analogue of Equation (3) is now Equation (7).
where wj is the excitation frequency. It is convenient to
write the density perturbation in terms of ordered pairs ai,
where a is an unoccupied and i an occupied orbital, and to
unocc occ
X X
LSckLAck ¼ 1
ð7Þ
c
k
5986
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
Equation (7) can be used to analyze occ–unocc LMO con-
tributions for a given excitation. In general, for a larger mol-
ecule this procedure may yield many contributions with rel-
atively small weights, but it is now easy to group contribu-
tions from a particular atom or functional group, or analyze
in terms of specific one-, two-, or delocalized multicenter or-
bitals. For a metal atom, or other subgroup, contributions
can also be extracted from the MO–MO terms by grouping
MO coefficients with respect to the AO basis at each center.
However with multi-z polarized metal basis sets it would
not be as simple, for example, to extract which metal orbital
contributes, in particular if the metal environment is not
highly symmetric, as is the case here. For the present work
we have written a program that analyzes the TDDFT results
from Turbomole computations in terms of LMOs according
to the occ–unocc LMO contributions of Equation (7).
Calculated UV and CD spectra of aza[6]helicene phosphole
5c: Starting from the crystal structure of ligand 5c1
(Figure 1 and Table 2), Figure 10 shows some changes in the
Figure 10. Overlay of the (darker) BP/SV(P) optimized structure and the
(lighter) crystal structure for 5c1 (hydrogen atoms added).
geometry occur upon optimization with the SV(P) basis.
Overall, however, the optimized and single-crystal X-ray
structures match well, and the differences were tentatively
attributed to crystal-packing effects. We proceeded to use
the optimized geometry, as it may better represent the struc-
ture in solution. Further optimization with the more flexible
TZVP[35] basis and additional calculations with TZVP and
the COSMO solvation model showed that the geometry re-
mained virtually unchanged from the SV(P) calculation.
Thus, the more economic SV(P) basis was used for subse-
quent computations.
Four configurations were studied: (P,SP)-s-trans-5c1, the
structure with an inverted phosphole (P,RP)-s-trans-5c2, and
ꢀ
both of these with the phosphole rotated 1808 about the C
C bond of N-C-C-P, that is, (P,SP)-s-cis-5c1 and (P,RP)-s-cis-
5c2, respectively. Their relative BP/SV(P) optimized ener-
gies in both gas phase and COSMO calculations are given in
Table 3.
Figure 11. Boltzmann averaged BHLYP/SV(P) absorption spectrum of
the phosphole ligands (top) and the spectra of the individual components
(bottom). All calculated spectra are redshifted by 0.25 eV and Gaussian-
broadened (s=0.26). View of the HOMO and LUMO of (P,SP)-s-trans-
5c1.
Figure 11 shows the simulated BHLYP/SV(P) UV/Vis ab-
sorption spectrum averaged over the various conformers,
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5987

R. Rꢀau, J. Autschbach, J. Crassous et al.
and of the individual conformers. The computed “stick”
spectra were redshifted by 0.25 eV and Gaussian-broadened
(s=0.26). Absolute intensities are known to be difficult to
reproduce computationally, and a deviation by a factor of
two for absorption spectra is quite typical at the TDDFT
level of theory. Likewise, spectral shifts on the energy scale
of a few tenths of an electron volt are also not uncommon.
Therefore, overall the calculated averaged spectrum agrees
well with experiment within the predictive limits of the com-
putational model. The analysis of the MOs involved in the
main UV/Vis bands of 5c1[36] is provided in Table 9. The
Table 9. UV/Vis analysis of optimized 5c1[36] structure.[a]
Excitation
Energy [eV]
Osc. str.
From–to
%
1
4
2.90
3.76
74.7
16.2
162–163
160–164
162–165
162–166
162–164
161–164
161–163
161–166
157–164
159–164
85.4
41.2
12.0
10.3
22.8
17.0
13.9
29.5
20.2
17.3
5
3.84
22.3
19
21
4.82
4.95
20.2
62.8
[a] 157, 164 (PH); 159, 160, 162, 163, 165 (BPH); 161, 166 (H). BPH des-
ignates the categorization of the orbital: (enzene)/P(hosphole)/H-
(elicene). Excitation energies have been redshifted by 0.25 eV, and oscil-
lator strengths scaled by a factor of 100.
BN
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
HOMO–LUMO (162–163) are also depicted in Figure 11
and discussed below. The frontier orbitals of each ligand
conformation appear quite similar; all are conjugated from
the helicene through the phosphole and into the C-phenyl
group.
Figure 12. BHLYP/SV(P) CD spectrum of the Boltzmann-averaged
phosphole ligands (top) and the individual components (bottom). All
spectra are redshifted by 0.25 eV and Gaussian-broadened (s=0.26).
As shown in Figure 12, within acceptable error bars for
the intensities and overall energy shift, the calculated CD
spectrum also agrees well with experiment. This computa-
tionally large system requires calculation of a significant
number of excitations to cover the experimentally accessible
frequency range. For BHLYP/SV(P), 120 calculated excita-
tions extend only to the upper end of the experimentally re-
corded spectrum. Additional excitations were calculated to
ensure that there is no influence from the higher-lying exci-
tations near the high-energy cutoff of the spectrum.
The data from Figure 12 show a noticeable difference in
intensity of the CD from 5c1 in the low energy regime.
Compounds s-cis-5c1 and s-trans- and s-cis-5c2 have similar
CD band positions and intensities with the applied broaden-
ing, but 5c1 has a much more intense and redshifted peak.
There are also sizeable differences in the calculated MRs at
589.3 nm. The MR for trans-5c1 is roughly 27 mru, whereas
that of cis-5c1 is about 17 mru; trans-5c2 and cis-5c2 have
calculated MR values of approximately 11.7 and 16.4 mru,
respectively. Below we investigate the reasons for these pro-
nounced differences.
mixed phosphole–azahelicene ligand 5c. Four key features
that significantly affect the molar rotation were theoretically
studied, which we discuss individually below.
1) Conjugation of the ligand p system with the C-phenyl
substituent of the phosphole ring: To study how the
length of conjugation affects the MR, the terminal
phenyl group of optimized 5c1 structure was rotated in
108 intervals (Figure 13). No further optimizations were
performed at any of the rotation steps, and at each point
the MR was calculated. There are two pronounced
troughs when the C-phenyl group is not conjugated with
the phosphole group (70–808 and 240–2608) and two
peaks when the two moieties are conjugated (140–1608,
310–3208) in the rotation cycle. It is apparent in
Figure 13 that the MR maxima have somewhat different
values because the phenyl group is not perfectly symmet-
ric with respect to the rotation axis (the atoms of the
phenyl group at a and a+1808 have similar but not ex-
actly the same locations). Taking the difference between
the maximum and minimum peak values, the orientation
of the C-phenyl group can contribute up to approximate-
ly 5 mru to the variations in the calculated MR.
Influence of conjugation and substituents: The structure/chi-
roptical property relationship was investigated in detail for
5988
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
rotation cycle. The geometries, which correspond to the
phosphole groups being perpendicular to the helicene p
system (cutting off the conjugation between the helicene
and phosphole moieties), exhibit the lowest MRs. The
effect is very pronounced, causing overall variations of
about 25 mru.
3) Substituents attached to the phosphole moiety: To inves-
tigate how the P-phenyl group and the cyclohexane
group fused to the phosphole influence the MR, we per-
formed the same type of calculations as under point 2
while considering three different cases: a) the P-phenyl
group is removed, b) the cyclohexane group is removed,
c) both groups are removed. In each case the dangling
bonds were hydrogen-capped with no further optimiza-
tions performed on any of these systems. The MRs for
these structures as a function of the rotation angle of the
phosphole moiety relative to the helicene, in comparison
to 5c1 rotation, are shown in Figure 15. Clearly, the
model without the cyclohexane group (i.e., with the P-
phenyl group still attached) follows the response of the
unaltered ligand the closest (see Discussion).
Figure 13. Molar rotation (589.3 nm) as a function of the rotation of the
C-phenyl group for the optimized 5c1 ligand.
2) Conjugation of the helicene p system with the phosp-
hole: To study the effect of the conjugation of the heli-
cene p system with the phosphole, the optimized 5c1[36]
ligand was taken as a starting point. Subsequently, the
ꢀ
phosphole group was rotated about the C C bond of the
N-C-C-P linking segment in 108 intervals. As before, no
further optimizations were performed at any of the rota-
tion steps, and at each interval the MR was calculated
(Figure 14). We also considered scenarios in which the
C-phenyl group has been rotated to be roughly parallel
with the phosphole group (denoted 5c1a), and approxi-
mately perpendicular (denoted 5c1b). These two models
represent the extreme cases of allowing the conjugation
of the p system to continue through the 5-phenyl group.
All three systems have two peaks and two troughs in the
Figure 15. Molar rotation at 589.3 nm as a function of the rotation of the
1
ꢀ
phosphole group about the C C bond of N-C-C-P for the optimized 5c
ligand, 5c1 with the cyclohexane group removed, 5c1 with the phenyl
group (attached to P) removed, and 5c1 with both groups removed.
4) Location of the P-phenyl substituent relative to the heli-
cene chromophore: Another key piece of information is
how the same series of calculations turns out for the 5c2
derivative (phosphole inversion). The N-C-C-P dihedral
angles of the 5c1 and 5c2 structures optimized to the
same value within a few degrees. Therefore, the main dif-
ference between these isomers is the location of the P-
phenyl group. Figure 16 shows the influence on the MR
of the rotation of the phosphole moiety relative to the
helicene moiety for 5c1 and 5c2. Diastereomers 5c1 and
5c2 have similar troughs and a similarly strong peak in
the molar rotation for angles between 150 and 2008.
Figure 14. Molar rotation as a function of the rotation of the phosphole
1
1a
ꢀ
group about the C C bond of N-C-C-P for the optimized 5c ligand. 5c
indicates that terminal phenyl group is roughly parallel with the phosp-
hole group, and 5c1b indicates a perpendicular orientation.
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5989

R. Rꢀau, J. Autschbach, J. Crassous et al.
Figure 16. Molar rotation as a function of the rotation of the phosphole
Figure 17. Molar rotation as a function of the rotation of benzene relative
ꢀ
and helicene moieties about the C C bond of N-C-C-P for the optimized
ꢀ
to azahelicene about an axis that would be the C C bond of N-C-C-P for
5c1 and 5c2 ligands.
the optimized 5c1 and 5c2 ligands.
However, there is a distinctively large difference in the
0ꢃ608 range. We therefore suggest that the location of
the phenyl group is one of the main factors responsible
for the lower MR of the optimized 5c2 structure com-
pared to 5c1.
To further probe the influence of the location of the P-
phenyl group on the MR, the optimized 5c1 and 5c2 struc-
tures were each separated into two molecules, one azaheli-
cene and one benzene (P-phenyl), kept where their carbon
frameworks would be positioned in the optimized ligand
structures and with the dangling bonds capped with hydro-
gen. The benzene moieties were then rotated with respect to
the azahelicene in the same manner as the other rotation
Figure 18. Benzene rotating around azahelicene; an overlay of the
models derived from 5c1 at 2708 (lighter) and 5c2 at 08 (darker).
ꢀ
models (i.e., about what would be the C C bond of the N-
C-C-P dihedral angle). Figure 17 shows the calculated MRs
as a function of the rotation angle. For comparison, the MR
of an azahelicene molecule with a geometry as adopted in
the 5c1 ligand is approximately 11.9 mru, which is close to
the experimental[21a] value of 12.0 mru for P-hexahelicene.
The BHLYP/SV(P)//BP/SV(P) calculated MR value for P-
hexahelicene is also 12.0 mru.
Figure 18 depicts this model and demonstrates that the
azahelicene substructures in 5c1 and 5c2 are similar enough
to compare directly. Figure 19 depicts the starting positions
of the benzene rings in the model with respect to the azahe-
licene in the R-(5c2) and S-(5c1) configurations of the li-
gands, viewed along the axis of rotation. Figure 17 shows the
troughs for the 5c2-like configuration to be in the 350ꢃ508
range, and for the 5c1-like configuration at the 250ꢃ508
range. Comparing this information to Figure 19, it is appar-
ent that when 5c2 is in the 08 range, it is in the same quad-
rant as the azahelicene. The same is also true when the 5c1-
like configuration is in the 2708 range. The main inference is
that when the phenyl group is spatially close to the terminal
ring of the azahelicene then the MR will drop significantly.
Figure 19. Depiction of the starting positions of benzene with respect the
helicenes for the rotation in the model of Figure 18 corresponding to 5c1
(P–S, lighter) and 5c2 (P–R, darker). Locations at other rotation angles
are also indicated by text. The view direction is along the axis of rotation.
The molar rotations calculated as a function of the rotational angle are
shown in Figure 17.
5990
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
Structural analysis of [Cu
The CuI bis
(aza[6]helicene phosphole) complex was exam-
ined in detail. Because no crystal structure of [Cu
(5c)2]+
has been obtained, the geometry was built from two of the
optimized 5c1[36] ligands, which were placed around the Cu
center and further optimized. The copper center adopts a
distorted tetrahedral geometry, so two metal configurations
were considered: One structure has a D configuration with
G
respect to the metal center [Cu
configuration [CuLA
(5c1)2]+ (Figure 20). At the BP/SV(P)
level of theory (gas phase) [CuLA
(5c1)2]+ is more stable than
[CuDA
(5c1)2]+ by 1.5 kcalmolꢀ1. The energy difference is
(5c1)2]+ and the other a L-
DACHTUNGTRENNUNG
G
CHTUNGTRENNUNG
CHTUNGTRENNUNG
CHTUNGTRENNUNG
small. Within the error bars of DFT we cannot conclude
that one of these conformers forms preferentially over the
other. Next, CD calculations were performed. Comparing
the calculated and experimental CD spectra, and consider-
ing the performance of the computations for the isolated
ligand, the computations for both configurations can be con-
sidered to be in good agreement with experiment (Fig-
ure 21a). The result is that the calculations both for the L
and the D Cu configurations are in similar agreement with
the experimental CD spectrum. Based on the relative stabil-
ity and similarity of the spectra, both species might be pres-
ent in solution unless the synthesis yields one of the isomers
preferentially.
We shall now discuss the contribution of the metal to the
main transitions. Plots of selected MOs and the Cu-, P-, and
N-centered natural localized molecular orbitals (NLMOs)
Figure 20. Optimized structures and relative energies for [Cu
and [CuLA
(5c1)2]+ by DFT calculations.
(5c1)2]+
DACHTUNGTRENNUNG
CHTUNGTRENNUNG
Figure 21. a) BHLYP/SV(P) CD spectrum of [Cu
analysis. All calculated spectra are redshifted by 0.25 eV and Gaussian-broadened (s=0.26). b) Partial SOS molar rotation for the [Cu
and [CuDA
(5c1)2]+ (bottom) configurations [Eq. (1)] plotted as a function of k, the number of excitations included in the SOS. The converged linear re-
sponse BHLYP/SV(P) result is indicated by a dotted line.
LACHTUTNGRENNUG DACTHUNGTRENNNUG
(5c1)2]+ (top) and [Cu (5c1)2]+ (bottom). Numbered excitations are those selected for MO and SOS
CTHUNGTRENNUNG
CHTUNGTRENNUNG
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5991

R. Rꢀau, J. Autschbach, J. Crassous et al.
are provided in the Supporting Information. Plots of the
canonical MOs involved in some of the more intense transi-
tions indicate that the L complex has more metal involve-
ment than the D configuration. The NLMOs of the ground-
state electronic structures of the two complexes show that
the qualitative aspects of the metal–nitrogen and metal–
phosphorus bonds are similar for both complexes. Although
isomer is 30.3. The experimental MR is only 13, which sug-
gests the other isomer is likely dominant in solution.
Considering the lowest-energy positive CD band in the D
complex, we find that there are no intense transitions, simi-
lar to above, that involve any of the partially metal centered
orbitals. There are four weaker transitions in the calculated
CD spectrum involving metal MO contributions greater
than 10%. Including these weak metal-centered transitions
along with the intense transitions which do not involve a
metal-centered orbital (see Figure 21a, bottom), the SOS
method yields a partial MR value of approximately 38.0 mru
of which the metal-centered transitions contribute only
ꢀ
there are no formal Cu N/P bonds assigned by the NBO
program, the ligand “lone pair” orbitals pointing directly at
the metal center clearly represent ligand–metal bonds, and
therefore these NLMOs are identified with the metal–nitro-
gen bonds. The NBO procedure indicates little s–d mixing
for Cu in the N/P Cu bonds; the metal hybrid AOs involved
ꢀ1.1. For [Cu
(5c1)2]+ the dominant SOS contributions to
DACHTUNGTRENNUNG
ꢀ
in the bonds have more than 95% s character in both [CuL-
the MR are transitions from the HPB backbone. Therefore,
there must be negative partial MRs from numerous low-in-
tensity transitions and/or higher lying excitations that bring
the MR down to near the calculated value. Including all cal-
culated excitations [k=120 in Eq. (1)], the SOS MR for the
D complex from the experimentally accessible frequency
range of the CD spectrum is on the order of 3.3 (see Fig-
ure 21b). The converged calculated linear-response molar
rotation is 13.2, which is close to the experimental value
(13.1 mru).
ACHTUNGTRENNUNG DACHTUNGTRENNUNG
(5c1)2]+ and [Cu (5c1)2]+. The NBO algorithms assign the
configuration of the metal center as 3d104s0.3 in both stereo-
isomers, that is, charge donation occurs from the ligands to
the CuI ion.
The metal contribution to each important transition and
to the final MR was calculated for both the [Cu
(5c1)2]+
DACHTUNGTRENNUNG
and the [CuLA
CHTUNGTRENNUNG
tions involving the metal center and using an SOS method
[Eq. (1)], the contributions to the optical rotation due to
metal-centered excitations can be identified. After visually
inspecting the MOs involved in these transitions they can be
categorized as “partially metal centered”, delocalized “heli-
cene–phosphole–benzene” (HPB), and any combination of
delocalized “helicene/phosphole/benzene” (H/P/B) orbitals.
“Helicene–phosphole–benzene” means that there are frag-
ment orbitals involved from each of these groups to form
extended conjugated MOs in the complex. Here, “partially
metal centered” means considerable participation from one
or more metal orbitals in addition to “helicene/phosphole/
benzene”, as there are no purely metal frontier orbitals.
Considering the second major positive CD band after the
The SOS results above do not completely quantify the
metal involvement in the optical activity, because many of
the partially metal centered (pMc) MOs extend over a large
part of the conjugated ligand backbone, making it difficult
to assess the metal contribution from the MO analysis. In
addition to looking at the percentage contribution of the
transitions from MO to MO [Eq. (3)], we also analyzed the
excitations in terms of NLMO-to-NLMO contributions
[Eq. (11)]. The individual percentages are small and the
analysis yields many terms, but it helps to quantify the metal
involvement in these transitions. For example, for excitation
7 of [CuLACHTUNGTRNEUNG
(5c1)2]+ the sum contribution of all transitions
shoulder in the [Cu
N
originating from occupied metal one-center NLMOs (metal
lone pairs, one-center refers to the NLMO centered around
one atom) is 14%, that from metal two-center NLMOs
(metal–ligand bonds) is 21%, and that to the metal lone-
pair one-center NLMOS is less than 1%. The metal orbitals
contribute a total of 35% to this excitation. A truncated
table of NLMO percentage contributions for the three com-
plexes can be found in Tables 10–12. Note that the metal
one-center unoccupied NLMOs (Table 10) include the for-
mally unoccupied metal orbitals as identified by the NBO
analysis, as well as some more diffuse Cu/Pd NLMOs that
the program specifically identified as metal-centered. The
tables indicate that, overall, the metal centers themselves do
not contribute significantly to the intense CD excitations.
Experimentally, the low-energy CD band is very weak for
the Cu complex, and compared to the Pd complex the Cu
system has a much smaller MR. The SOS of the five lowest
eight excitations strongly contribute to the spectrum (rota-
tory strengths of at least 150ꢃ10ꢀ40 esu2 cm2). In the occ–
unocc MO analysis, four of the excitations afford a contribu-
tion from a metal-centered MO that is greater than 10%
[per Eq. (3)]. Two of the excitations have a greater than
10% contribution from p orbitals of the conjugated HPB
backbone, and the other two contain no single MO-to-MO
transition contributing 10% or more. The strong mixing of
MOs in the excitations was already seen in the ligand spec-
trum. The calculated partial MR of these eight excitations
from the SOS method at the sodium D line is 33.7 mru, with
a breakdown of 8.1 from partially metal centered MO tran-
sitions, 1.2 from HPB MO transitions, and 24.4 from numer-
ous small contributions. From this data, approximately 24%
of the MR magnitude can be attributed to low-lying excita-
tions with some metal involvement, a contribution of 3.5%
comes from the HPB-centered MOs, and the bulk of 73.5%
comes from numerous excitations that are not easily ana-
lyzed. Including all calculated excitations [k=120 in Eq. (1)]
yields a reduction of the SOS rotation to 24.6 mru, as shown
in Figure 21b. The converged linear response value for this
calculated excitations for [CuLACTHNUTRGNEUNG
(5c1)2]+ is 23.25, and for
[CuDACHTGNUTRENNUNG
(5c1)2]+ it is ꢀ10.13 mru. These excitations, which have
less than 2% total NLMO contribution from the metal, as
mentioned above, distinguish the L and D configurations
with an MR difference of over 30 mru. The experimentally
5992
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
Table 10. NLMO-to-NLMO percentage contributions for selected transi-
FULL PAPER
measured MR is about 13.1. The calculated linear-response
tions of [Cu
(5c1)2]+.[a]
MR for [CuDACTHNUGTRENNUG LACHTUGNTRENNGUN
(5c1)2]+ and [Cu (5c1)2]+ are 13.23 and
Excitation
Metal
1-center [%]
Metal–(N,P)
2-center [%]
Total
[%]
30.34 mru, respectively, which are equivalent to the com-
plete SOS within the excitation space allowed by the basis
set. Figure 21a indicates that the excitations in this range
dominate the MR at 589.3 nm. Overall, the data suggest
that the involvement of the metal is relatively weak and
does not directly contribute to the MR in a significant
manner (see Discussion). However, the metal configuration
dictates the structure of the ligand in the complex.
Using our helicene–benzene model from the ligand analy-
sis (point 4), we can hypothesize that the MR is much
higher for the L configuration, since it does not have any P-
phenyl groups crowding near the helicene moiety, and that
the MR is much lower for the D configuration, as it has a P-
phenyl and a C-phenyl group (from the other ligand) spa-
tially close to it.
1
2
1
1
1
1
2
2
7
8
14
16
1
2
9
6
3
2
21
22
1
2
9
7
4
2
35
38
2
10
15
17
18
20
21
4
18
13
7
4
[a] 1-center refers to the NLMO centered around 1 atom (i.e., a Cu lone
ꢀ
pair). 2-center refers to a centering of 2 atoms (i.e., a Cu N bond). All
metal 1-center unocc. contributions for the above excitations are <1%,
and have been omitted.
Table 11. NLMO-to-NLMO percentage contributions for selected transi-
tions for [Cu
(5c1)2]+.[a]
Structural analysis of [Pd
For the structural stability calculations, NO crystallographic
(5c)2]2+ and CD/MR calculations:
ACHTUNGTRENNUNG
Excitation
Metal
1-center [%]
Metal
2-center [%]
Total
[%]
data of [PdACTHNUTRGNEUNG
(5c)2]2+ was available; therefore, the coordinates
were built from two of the optimized 5c1 ligands,[36] which
1
2
1
1
1
1
2
2
were placed around the Pd center and further optimized at
7
8
16
1
27
2
43
3
the BP/SV(P) level of theory giving [Pd
(5c1)2]2+. A [PdD-
DACHTUNGTRENNUNG
9
0
1
1
ACHTUNGTRENNUNG
12
18
25
28
11
4
6
16
5
8
27
9
14
9
manner. The geometry around Pd is distorted square-planar,
so there is less room for the bound ligands than in the dis-
torted tetrahedral copper complexes, which leads to signifi-
cantly more steric repulsion. As a consequence, [PdD-
4
5
Note: All metal 1-center unocc. contributions for the above excitations
are <1%, and have been omitted.
AHCTUNGTRENNUNG
(5c1)2]2+ is calculated to be approximately 19.4 kcalmolꢀ1
more stable than [PdDACTHNUTRGNEUNG
(5c2)2]2+, which shows a predisposition
for two 5c1 ligands binding to the PdD center over two 5c2 li-
gands (Figure 22). Note that we did not converge a distorted
Table 12. NLMO-to-NLMO percentage contributions for selected transi-
tions for [Pd
DA
.
Excitation
Metal
1-center [%]
Metal
2-center [%]
Total
[%]
1
2
9
0
0
5
0
1
1
0
0
0
0
1
0
0
1
0
1
0
1
1
1
0
1
0
0
18
1
2
1
0
0
0
0
3
1
0
0
0
0
0
0
1
2
0
3
0
0
23
1
3
2
0
0
0
0
4
1
0
1
0
1
0
1
2
3
0
4
17
18
19
20
23
32
35
36
38
50
58
59
60
62
70
72
83
87
90
Figure 22. Optimized structures and relative energies for [Pd
[Pd
(5c2)2]2+ from DFT calculations.
(5c1)2]2+ and
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
square-planar Pd center having a L configuration from two
P-helices, which suggests that the helix configuration dic-
tates the stereochemistry at the PdII center.
From the analysis of the calculated excitations,[12a] the re-
sults confirm that the dominant contributions to the CD
spectrum (Figure 23) are from p–p* transitions involving
phosphole/helicene/benzene MOs for the excitations having
[a] The majority of the metal 1-center unocc. contributions for the above
excitations are <1% and have been omitted.
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5993

R. Rꢀau, J. Autschbach, J. Crassous et al.
tion of the metal center as 4d95s0.4, that is, there is significant
accumulation of electronic charge at the metal center com-
pared to the formal PdII d8 configuration, but the d shell is
not completely filled. The MO analysis of the excitations
correspondingly identified excitations with unoccupied orbi-
tals having some metal 4d character.
Using the NLMO-to-NLMO percentage contributions to
analyze the transitions previously investigated in terms of
MO contributions[12a] (Table 12) suggests that in the palladi-
um complex metal orbitals contribute less to the intense ex-
citations than was found for the Cu complex. Only one in-
tense transition contains 23% total contribution from metal-
centered NLMOs; the rest have less than 5%. The lowest
eight excitations of [PdDACTHNUTRGNEUNG
(5c1)2 ]2+ have less than 2% metal
contribution according to the NLMO percentage contribu-
tion analysis. Using the SOS analysis, these first eight excita-
tions yield a partial MR of ꢀ10.36 mru; adding in the ninth
gives a partial MR of 1.97. The lowest 11 excitations, which
make up the first CD band, give a partial MR of 4.52, and
with all calculated excitations [k=150 in Eq. (1)] the partial
MR is 21.95 mru. The bottom panel of Figure 23 shows that
the first pair of calculated excitations contributes strongly to
the MR, on the order of ꢀ20 mru. Both of these excitations
have sizeable rotatory strengths, but they strongly cancel
due to an excitation-energy difference of only 0.01 eV. The
MO analysis shows that each of these excitations is dominat-
ed by a pair of HOMO- and HOMOꢀ1-to-LUMO transi-
tions (helicene-to-phosphole). The NLMO analysis
(Table 12), indicates that there is !1% metal contribution
in either of these excitations even though the LUMO con-
tains a metal-orbital contribution. Inclusion of additional
calculated excitations slowly converges the SOS MR closer
to the linear-response value. Considering the partial MRs,
the calculated linear response MR of 26.66, and the NLMO
analysis, the data suggest that the metal has little involve-
ment in the MR at 589.3 nm.
Figure 23.
with experiment. Numbered excitations are those selected for MO and
SOS analysis. (Bottom) Partial SOS molar rotation for [PdDA .
(5c1)2]2+
Equation (1), plotted as a function of k, the number of excitations includ-
ed in the SOS. The converged linear response BHLYP/SV(P) result is in-
dicated by a dotted line. All calculated spectra are redshifted by 0.25 eV
and Gaussian-broadened (s=0.26). Compared to our previous communi-
cation[12a] a larger number of excitations has been computed for this
work, which affects the high-energy cut-off of the spectrum.
(5c1)2]2+ compared
ACHTUNGTRENNUNG(Top) BHLYP/SV(P) CD spectrum of [PdDACHTUNGTRENNUGN
C
The location of the phenyl groups with respect to the heli-
cenes is the main correlation we can make between the opti-
mized structures of [Pd
(5c1)2]2+, [Cu (5c1)2]+, and [CuL-
DACHTUTNGRENNUG DACHUTNGTRENNUGN
AHCTUNGTRENNUNG
tions of these three structures reveals no striking similarities
that would allow us to make another analogy between the
Pd and Cu complexes. Considering the previous discussion
on how the phenyl groups of the phosphole ligands, and
therefore of the copper complexes, affect the computed
the strongest rotatory strength. There are numerous metal
orbitals involved in various excitations, but with low per-
centage contributions. It is also noticeable that the CD
strengths of the transitions that have the largest percentage
of pMc MOs in the transition vectors tend to be relatively
weak. Most of these excitations can be assigned as metal
bonding to metal antibonding (LUMO) excitations.
As done for the copper complexes, we performed an
NBO calculation and analyzed the NLMOs for the Pd com-
plex. Several metal-centered NLMOs are illustrated in the
Supporting Information. Again, the orbitals assigned by the
NBO algorithms as ligand “lone pairs” pointing directly at
the metal center represent the ligand–metal bonds. The
NBO analysis indicates s–d mixing at the palladium center;
the hybrid metal AOs involved in the bonds have 23% s
and 77% d character. The NBO code assigns the configura-
MR, we notice that in the [Pd
groups attached to the P center are close to the azahelicene.
They are not as close as in the [CuDA
(5c1)2]+ complex but not
quite as far away as in [CuLA
(5c1)2]+. Since the phenyl–heli-
cene orientation of [PdDA
(5c1)2]2+ is closer to that of [CuL-
(5c1)2]+, the calculated value of 26.60 falls in line with the
(5c1)2]2+ structure phenyl
DACHTUNGTRENNUNG
CTHUNGTRENNUNG
CHTUNGTRENNUNG
CTHUNGTRENNUNG
AHCTUNGTRENNUNG
results expected from this model. Within this model, it is the
flexibility of Cu to adopt a distorted tetrahedral geometry
which allows the phenyl groups to pack in next to the heli-
cenes (in the Cu D configuration), and cause a lower MR
than would be possible from a distorted square-planar ge-
ometry. It is also this flexibility that allows the phenyl
5994
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
groups to position themselves further away from the heli-
cenes (as in the Cu L configuration) in a manner that is sim-
ilarly possible for a distorted square-planar configuration.
is predicted to be the most stable by the theoretical calcula-
tions (Table 3). In fact, there is roughly a 2.0–2.5 kcalmolꢀ1
difference in energy between 5c1 and 5c2 (Table 3). These
energies are sufficiently close (considering the error bars of
relative DFT energies) to support the experimental observa-
tion of both phosphole configurations in solution by NMR
spectroscopy. Note that the low inversion barrier at P deter-
mined by NMR spectroscopy for 5c (16 kcalmolꢀ1 at 330 K)
is typical for those recorded in 1-phenylphospholes.[15] The
fact that P*-phospholes exist as two isomers of comparable
energy that interconvert at room temperature is a key prop-
erty of this heterole that is of great importance for the dia-
stereoselective formation of metal–bis(azahelicene phosp-
hole) assemblies (vide infra).
The UV/Vis data of compounds 5a–e (Figure 4) were in-
vestigated in order to probe the electronic interaction be-
tween the phosphole ring and the aza[n]helicene moieties.
The presence of extended p systems in 5a–e is suggested by
the fact that their lmax are bathochromically shifted com-
pared to those of their 1,7-diyne precursors (Table 7). The
absorption spectra of phospholes 5a,b (bearing terminal
azahelicenes) are considerably redshifted compared to those
of 5d,e, which have an inner pyridine moiety (Table 7,
Figure 4). This behavior can be attributed to the fact that
the former have longer, more extended p-conjugated sys-
tems than branched compounds 5d,e (Scheme 1). It is note-
worthy that a regular bathochromic shift occurs upon in-
creasing the size of the pyridine-based p system (see the
series 2-(2-pyridyl)phosphole 5 f/aza[4]helicene derivative
5b/aza[6]helicene derivative 5c; Scheme 1, Table 7). These
differences clearly show the impact of the helicene composi-
tion on the optical properties of mixed phosphole azaheli-
cene derivatives and indicate that these two p subunits are
electronically coupled.
The above conclusion based on experimental data is con-
firmed by TDDFT calculations with the BHLYP functional
and the SV(P) basis set that were conducted on 5c, which
bears the largest helicene moiety (Scheme 1). The calculated
UV/Vis spectrum of the more stable (P,SP)-s-trans isomer of
5c (Table 3, Figure 1) well reproduces the experimental one,
given a shift of ꢀ0.25 eV (Figure 11). Note that the Max-
well–Boltzmann averaged calculated spectrum (taking into
account the relative contribution of all isomers of 5c,
Table 3) is similar to the experimental spectrum presented
in Figure 11, within the error range of such calculations (see
the Supporting Information). Strong calculated UV/Vis exci-
tations of 5c1,2 were analyzed (in and slightly above the ex-
perimental frequency range). For all stereoisomers, the
lowest excitation (which is also responsible for the lowest
positive CD band; vide supra) is predominantly a HOMO-
to-LUMO transition of the p system (80% or greater contri-
bution). Interestingly these molecular frontier orbitals in-
volve both the azahelicene and the phosphole fragments
(Figure 11). Likewise, all the other UV/Vis excitations are
p–p* transitions involving combinations of molecular orbi-
tals of the p–system (HOMOꢀ7 to LUMO+7 MOs)
spreading over the helicene and phosphole moiety
Discussion
Metal-mediated oxidative coupling of functional 1,n-diynes
(n=6,7) followed by metal–dihalogenophosphine exchange
is a versatile synthetic route towards phosphole deriva-
tives.[17] Considering that our target compounds are mixed
phosphole azahelicenes that can act as 1,4-chelating ligands
(Scheme 1), this general synthetic approach firstly requires
the preparation of diynes bearing azahelicene substituents.
This novel family of helicenes, including 4-azahelicenes sub-
stituted by a diyne at the 3-position and 5-azahelicenes sub-
stituted at the 6-position, was obtained by using a Sonoga-
shira coupling–photocyclization–Wittig reaction sequence
(Scheme 2). Note that functionalized azahelicenes are still
rare,[6] and that these results show that it is possible to con-
duct the photocyclization–oxidation sequence in the pres-
ence of potentially reactive triple bonds. The solid-state
analysis of (ꢀ)-4c (Figure 1) reveals that the helical angle
(angle between the terminal pyridyl and phenyl rings, 56.78)
ꢅ
as well as the C C distances (1.191–1.194 ꢂ) are classi-
cal.[6,17] These compounds display photophysical properties
typical of regular helicenes (fluorescence at room tempera-
ture; fluorescence and phosphorescence at low tempera-
ture).[4] Chiral diynes 4c,e display different optical rotations.
23
The specific rotation of compound 4e is quite low (½aꢄD
=
42.9), as expected from its small helicoidal moiety, whereas
that of 4c is much higher (½aꢄ2D3 =1660) and comparable to
those recorded for other aza[6]helicene derivatives.[6]
Among the prepared mixed phosphole azahelicene deriv-
atives (Scheme 1), 5a,b,d can be regarded as model mole-
cules, since their helicene moiety inverts at room tempera-
ture. The impact of the helicoidal system length on the heli-
coidal curvature (angle between the two terminal aromatic
rings, hc), and therefore on the helicene inversion barrier, is
nicely illustrated by comparison of the solid-state structures
within the 5a–c series (Tables 2 and 4, Figure 1). The ben-
zo[f]quinoline part of 5a is completely planar, the naphtho-
ACHTUNGTRENNUNG[1,2-f]quinoline part of 5b displays a helical geometry with
an hc of only 30.18, while the hc of the aza[6]helicene frag-
ment in 5c is much more pronounced (45.88; Figure 1). For
all compounds, the P atom of the phosphole ring is pyrami-
dalized (ꢀ angles around P, 296–3028) and, as is usually ob-
served for 2-pyridylphosphole derivatives,[17] the P and N
atoms have an s-trans conformation in the solid state
(Figure 1). The twist angles between the central phosphole
ring and its aza[n]helicene and C-phenyl substituents are
small enough (<408, Table 4) to allow electronic interaction
between the three phenyl–phosphole–azahelicene p systems.
Interestingly, only one pair of enantiomers is observed in
the solid state for phospholes 5b,c (Scheme 1), which exist
as interconverting diastereomers in solution. Furthermore,
the (P*,SP*)-5c1 diastereomer that crystallizes is that which
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5995

R. Rꢀau, J. Autschbach, J. Crassous et al.
(Figure 11). Surprisingly, even the high-energy transition at
4.7 eV, which is normally the p–p* transition fingerprint of
helicenes (see, for example, compounds 4b–d, Table 7), is
not pure but involves a complex mixture of molecular p or-
bitals that include the phosphole moiety (HOMO,
HOMOꢀ1, …; see Supporting Information). Therefore, the
different absorption bands cannot be attributed to localized
excitations in the helicene or the phosphole moiety sepa-
rately. This theoretical study therefore confirms the conclu-
sion drawn from the UV/Vis experimental data that the p
subunits of mixed azahelicene phosphole derivatives are
strongly electronically coupled. The theoretical calculations
reveal that the lowest absorption bands of cis- and trans-5c1
are redshifted with respect to the corresponding bands of
cis- and trans-5c2. These data suggest that the configuration
at the P atom impacts on the HOMO–LUMO gap of phosp-
hole-based p-conjugated systems. Indeed, the calculated
HOMO–LUMO gaps for cis- and trans-5c1 (4.95 and
4.92 eV) are smaller than that of cis- and trans-5c2
(5.05 eV). To the best of our knowledge, this is the first time
that the influence of the configuration at P on the optical
properties of phosphole-based p-conjugated systems has
been demonstrated.
The emission behavior of mixed azahelicene phosphole
derivatives 5b–d is essentially comparable to that of the cor-
responding helicenes (fluorescence at RT; fluorescence plus
phosphorescence at low temperature from helicene part)
with an additional contribution of the fluorescent phosphole
moiety. Note that the quantum yields of these phospholes
bearing azahelicene substituents are much higher than that
of the simple phosphole-(2-pyridyl) derivative 5 f (Table 7).
The coordination chemistry of phospholes 5b–d was then
examined. Before investigating the synthesis of rather so-
phisticated metal–bis(phosphole-aza[n]helicene) assemblies,
it was necessary to check that phosphole-modified azaheli-
cene bearing either a terminal (5b, Scheme 1) or an inner
(5d, Scheme 1) P,N moiety can act as 1,4-chelating ligands
towards transition metal centers such as PdCl2 (Scheme 3).
lating ligands towards metal centers and illustrate that, due
to the ease of inversion at P, they can adapt their configura-
tion in metal coordination spheres to minimize steric inter-
actions.
The reaction of Pd2+ with two equivalents of phosphole
pyridine derivatives 5b–d affords the corresponding metal–
bis(azahelicene phosphole) assemblies as single diastereo-
mers according to multinuclear NMR data (Scheme 4). The
stereoselective organization of the P,N ligands in the PdII-co-
ordination sphere was investigated in detail by an X-ray dif-
fraction study of complex [Pd
theoretical calculations performed on complex [Pd
(Table 6). The PdII center of complex [Pd(5d)2]2+ shows a
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
highly distorted square-planar geometry (vide infra) and, in
accordance with the trans effect,[12] the P and N atoms have
a mutual trans arrangement (Scheme 4). The angle between
the two N-Pd-P planes is 17.38. This distortion, which is due
to overlap of the two coordinated azahelicene moieties, re-
sults in a chiral coordination sphere around the PdII center
with either or L or D configuration.[12b] It is likely that the
configuration at the metal is maintained in solution due to
the overlap of the helices (Scheme 4), which prevents “flop-
ping” of the ligands. Remarkably, amongst the numerous
possible stereoisomers (26), only one pair of enantiomers is
obtained in the solid state (Mhelix 1,Mhelix 2,SP1,SP2,LPd) and
(Phelix 1,Phelix 2,RP1,RP2,DPd). For obvious steric reasons, the
phenyl substituents of the phosphole ligands are in trans po-
sition with respect to the PdN2P2 coordination plane, and
the two azahelicene moieties are homochiral (Scheme 4).
Indeed, as observed for [PdACTHNUTRGNE(UNG 5b)Cl2], the phosphole ring can
adapt its configuration at P to minimize steric repulsion.
This solid-state analysis shows that both electronic (trans
effect) and steric (repulsion of phosphole substituents) fea-
tures participate in the high stereoselectivity of the coordi-
nation process leading to PdII(phosphole aza[n]helicene)2 as-
semblies (Scheme 4). It is furthermore likely that the hemi-
labile nature of P,N chelates 5b-d is a clue to obtaining the
more thermodynamically stable complex by establishing co-
ordination–decoordination equilibria.
The NMR data of complexes [PdACHTNUGRTENUNG(5b/d)Cl2] are consistent
with coordination of the mixed phosphole azahelicene li-
gands through their P and N atoms. This was confirmed by
an X-ray diffraction study on [PdACHTNUTRGNEUNG(5b)Cl2] (Scheme 3). The
The structure of the aza[4]helicene moieties is not per-
turbed by their assembly into the dicationic Pd bis(phosp-
hole azahelicene) complex. For example, the angles between
PdII center has an almost square-planar geometry with
angles and bond lengths in the metallacycle comparable
with those reported for other PdII complexes bearing 2-(2-
pyridylphosphole) chelates.[20] For example, the small twist
angle between the phosphole ring and the azahelicene
(18.18) allows electronic interaction between these two p
units. The helical shape of the aza[4]helicene part is not per-
turbed upon complexation, since the twist angle between
the two terminal aromatic rings of the aza[4]helicene (33.18)
is comparable to that of the free ligand (30.18). The phenyl
substituent of the P atom and the aza[4]helix have a mutual
anti arrangement with respect to the PdII coordination plane
in order to minimize steric repulsion between these two
phosphole substituents (Scheme 3). These results confirm
that mixed phosphole azahelicene derivatives act as 1,4-che-
the two terminal aromatic rings in [Pd
34.98) are comparable to those of [Pd(5b)Cl2] (30.18). Final-
(5d)2]2+ (30.7 and
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
ly, the rather large distance between the two helicenes
(shortest inter-ring distance, 4.1 ꢂ) indicates that no intra-
molecular p–p interaction takes place. An additional inter-
esting point is the organization of complex [PdACTHNUTRGNEUNG
(5d)2]2+ into
heterochiral columns in which each molecule stacks with its
enantiomeric neighbor (Figure 2). This type of supramolec-
ular organization, which is due to p stacking of the azaheli-
cene moieties, is well known in helical molecules.[1] This
result shows that the presence of the metal center does not
prevent metal phosphole aza[n]helicene complexes from be-
having as regular helicene derivatives.
The formation of complex [Pd
serves discussion, since in this case the phosphole ligand
(5c)2]2+ (Scheme 4) de-
ACHTUNGTRENNUNG
5996
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
bears optically pure helix moieties. The fact that complex
FULL PAPER
either L or D configuration.[38] Among the numerous theo-
[Pd
A
retically possible diastereomers, only two for complex [Cu-
(P,RP*)-5c or (M,RP*)-5c shows that the configuration of
the helix controls the configuration of the other stereogenic
centers, that is, the phosphole P and Pd atoms. This is nicely
confirmed by the fact that calculations at the BP/SV(P)
level of theory performed for the ligand having a P-helix re-
(5b)2]+ and one for [Cu(5d)2]+ were found in the solid
ACHUTGTNRENNUG CAHTUNGTRENNUGN
state. Furthermore, the two diastereomers of [CuACHTUNGTRENNUNG
(5b)2]+
ꢀ
N
ꢀ
differ only by the conformation of the flexible
moiety grafted onto the phosphole rings. Therefore, if one
focuses on the metal coordination sphere, only the
(Mhelix1,Phelix2,RP1,RP2,DCu) and (Phelix1,Mhelix2,SP1,SP2,LCu) dia-
stereomers are observed. These solid-state studies speak for
the fact that the coordination process leading to the CuI
complexes is stereoselective. However, it is difficult to draw
helpful conclusion from these experimental data, since these
complexes contain two heterochiral helicene moieties,
whereas our approach is to use optically pure helicene li-
gands that will afford complexes with homochiral helicene
moieties. In fact, DFT calculations at the BP/SV(P) level re-
vealed that the energy difference between the diastereo-
vealed that the [PdACHTUNGTRENNUNG
(5c1)2]2+ assembly, in which the P atom
has R configuration and the configuration around the Pd
center is D, is 19.4 kcalmolꢀ1 more stable than its SP,DPd dia-
stereomer [PdACHTUNGTRENNUNG
(5c2)2]2+ (Figure 22). Furthermore, the com-
plex having a LPd configuration (P-helix, RP) is not a mini-
mum on the potential-energy surface. This theoretical study
confirms the high stereoselectivity of the coordination pro-
cess and allows one to establish unambiguously that only
(P,P,RP,RP,DPd)-[PdACHTUNGTRENNUNG AHCTUNGTRENNGUN
(5c1)2]2+ and (M,M,SP,SP,LPd)-[Pd(5c1)2]2+
complexes were formed (Scheme 4). The large energy differ-
ence between the different diastereomers (Figure 22) can be
attributed to steric factors, as suggested by the fact that the
meric L and D [CuACTHNUTRGNEUNG
(5c1)2]+ complexes bearing the P,N li-
gands with an enantiomerically pure P-helix is quite small
square-planar PdII coordination sphere of [Pd
ACHTUNGTRENNUNG
(ca. 1.5 kcalmolꢀ1).
much more distorted than that of [Pd
between the N-Pd-P planes: [Pd
G
These results demonstrate that modification of azaheli-
cenes by a phosphole moiety to generate chelating P,N frag-
ments that can coordinate to metal centers is a versatile ap-
proach to readily generate structural diversity. In fact, start-
ing from one programmed heteroditopic P,N helicene, air-
stable chiral metal-bis(phosphole-aza[n]helicene) assemblies
can be readily obtained on coordination to metal ions
having different coordination geometries. Theoretical and
experimental results show that the coordination is highly
stereoselective in the case of square-planar metal centers.
E
E
,
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
(Table 6).[12b] These data suggest that there is no steric stress
associated with the assembly of phosphole-modified [6]hel-
icene ligand 5c in the Pd coordination sphere. The curvature
of the helices is retained in the coordination sphere of the
metal (5c, 45.88; theoretical calculations, 49.58), and the two
twist angles between the coordinated phosphole and the pyr-
idine rings of 19.4 and 23.48 allow electronic interaction be-
tween the phosphole and the azahelicene p systems to take
place. In conclusion, the coordination of the mixed phosp-
hole azahelicene derivatives to a d8 PdII center is highly ste-
reoselective due to a unique combination of specific elec-
tronic (trans effect), steric (presence of the bulky helicene
and P-substituent), and dynamic (fast inversion at P) prop-
erties. This result clearly shows that, in spite of their pecu-
liar structures and the presence of sterically demanding aza-
helicene substituents, the heteroditopic P,N chelating moiet-
ies including a phosphole moiety dictate the coordination
behavior of 5b–d towards the square-planar Pd2+ ion and
allow a highly stereoselective coordination process to take
place.
Complexes [PdACTHNUTRGENNUG ACHTUGNTRENNUGN
(5c)2]2+ and [Cu(5c)2]+ exhibit very differ-
ent chiroptical properties. The former has a much higher
specific molar rotation than the latter (Table 8). Likewise,
the CD spectrum of the palladium complex is globally more
intense than that of the copper analogue (Figure 9). Further-
ACTHNUTRGNEUNG
more, the CD spectrum of PdII assembly [Pd(5c)2]2+ dis-
plays intense bands at about 370 nm as well as weak bands
at low-energy wavelengths (410–450 nm) that are not ob-
served for [CuACTHNUTRGNEUNG
(5c)2]+ (Figure 1). These results clearly show
that 1) it is possible to perform coordination-driven tuning
of chiroptical properties of phosphole-modified azaheli-
cenes, and that 2) it is more efficient to organize these heter-
oditopic ligands around a d8 square-planar metal center than
around a tetrahedral one.
Regarding the optical activity of these metal-bis(phosp-
hole-aza[6]helicene) assemblies, the main questions are:
1) do metal orbitals directly participate in the excitations, or
is the only role of the metal to keep the ligands in their rela-
tive positions? 2) If so, does the metal boost the molar rota-
tion? 3) Does the nature of the metal dictate specific rela-
tive arrangements of the ligands that may boost or quench
the optical activity in the complex? The confrontation be-
tween the experiments and the theoretical calculations ena-
bled us to resolve all these questions.
The next step was to investigate whether, using the same
azahelicene phosphole ligands, metal–bis(phosphole-aza[n]-
helicene) assemblies with different topologies could be ob-
tained by varying the nature of the metal center. For this
purpose d10 tetrahedral CuI complexes were prepared
(Scheme 5). The chiral CuI centers display a distorted tetra-
hedral geometry with the two azahelicenes having an almost
perpendicular arrangement (angle between the two N-Cu-P
planes: [CuACHTUNGTRENNUNG ACHUTNGTRENNGUN
(5b)2]+, 83.78; [Cu(5d)2]+, 86.48; Scheme 5). The
Cu P and Cu N distances, as well as the P,N bite angles,
are unremarkable (Tables 5 and 6). Note that the CuI atoms
are stereogenic centers with spiro-type chirality, leading to
The computed[12a] CD spectrum of [Pd
(5c1)2]2+ calculated
ACHTUNGTRENNUNG
ꢀ
ꢀ
for the most stable diastereomer (P,P,RP,RP,DPd) agrees very
well with the experimental CD of (+)-[PdACTHNUTRGNEUNG
(5c)2]2+ after a
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5997

R. Rꢀau, J. Autschbach, J. Crassous et al.
redshift of 0.25 eV (Figure 23). Therefore the absolute con-
more important in the sense that it also dictates the relative
positions of the ligands. Indeed, the fact that the MR de-
pends significantly on the stereochemistry and the shape of
the complex is an important issue that is evidenced in our
TDDFT calculations, and this conclusion encouraged us to
go back to the phosphole structure, which is easier to inves-
tigate by theoretical calculations.
The key to better understanding the optical properties of
the copper and palladium complexes lies predominantly in
the characteristics of the p system and the conformational
flexibility of the phosphole ligands. By studying the aspects
that affect the MR of the ligand, we can notice trends and
offer valuable insight into the metal complexes. Among a
variety of possibilities, we have identified four key features
that significantly affect the MR of the ligands and have im-
figuration (P,P,RP,RP,DPd)-(+) can be ascertained for PdII–bis-
ACHTUNGTRENNUNG(aza[6]helicene phosphole) assemblies. The over-/underesti-
mation of the CD intensity of the bands at 330 and 250 nm,
respectively, is similar to the case of free hexahelicene.[21]
On the other hand, the computed CD of diastereomers
(P,P,RP,RP,DCu)-[Cu
both reproduce reasonably well the experimental CD spec-
trum of (+)-[Cu
(5c)2]+ (after a redshift of 0.25 eV, Fig-
ure 21a), so that the absolute configuration of CuI–bis-
(aza[6]helicene phosphole) assemblies cannot be determined
ACHTUNGTRENNUNG ACHTUNGTRENNUGN
(5c1)2]+ and (P,P,RP,RP,LCu)-[Cu(5c1)2]+
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
by means of CD calculations. Interestingly, the CD spectra
of the two diastereomeric CuI complexes are similar. An im-
portant aspect is that, both in the PdII– and the CuI–bis-
ACHTUNGTRENNUNG(aza[6]helicene phosphole) assemblies, the most intense CD
bands involve essentially p–p* transitions of the extended
phosphole–azahelix p systems. In the MO–MO transitions
having the highest CD values, the metal centers contribute
to many transitions (Tables 10–12 and Supporting Informa-
tion) but to a smaller extent than the phosphole azahelix p
plications for the optical activity of metal–bis
phosphole) assemblies.
ACHTUNGTREN(NGNU aza[6]helicene
First, the MR calculations on aza[6]helicene phosphole 5c
show that the optimized diastereomeric structures have very
different MR values (trans-5c1, 11.7 mru; cis-5c2, 16.4 mru).
Furthermore, conjugation through the 5-phenyl group can
quite significantly impact the MR value of 5c, although not
in the same magnitude as the conjugation between the
phosphole and the aza[6]helicene moieties, for which the
MR values span a range of approximately 25 mru (as com-
pared to only about 5 from 5-phenyl rotation in Figures 13
and 14). The maximum MRs are obtained when the phosp-
hole group is in conjugation with the helicene p system. If
we consider the 5c1 structure, there are two other groups
which can perturb the system: the P-phenyl group and the
cyclohexyl group fused to the phosphole. From Figure 15 it
is clear that the extent of conjugation within the molecule
with the cyclohexane group removed (i.e., the P-phenyl
group still present) follows the MR of 5c1 more closely over
the range of rotations. From investigating the ligands, re-
moval of the P-phenyl group (PPh!PH) seems to have a
more profound effect in relation to its proximity to the aza-
helicene group. The HOMOs of the optimized ligands have
a node in the p system located at the P atom (see Support-
ing Information). The P-phenyl carbon atom, however,
makes some contribution to the LUMO of the optimized li-
gands. On the other hand, the cyclohexane group does not
significantly contribute either to the HOMO or to the
LUMO. Unlike the P-phenyl group, its removal does not
impact the frontier orbitals of the p system which we consid-
er to dominate the MR. To summarize, conjugation from
the helicene through the phosphole group can noticeably
affect the molar rotations, and even more so if the conjuga-
tion continues through the terminal phenyl group. The sub-
stituents on the phosphole group also affect the MR, both
via direct electronic effects on the frontier orbitals of the
conjugated p system, and more indirectly via through-space
interactions with the helicene moiety. Indeed, we have ob-
served that the P-phenyl group plays a particularly impor-
tant role, causing the MR to drop significantly in magnitude
when it is in spatial proximity to the azahelicene. This sug-
gests that the location of the phenyl group is one of the
system. In the [PdACHTUNGTRENNUNG
(5c)2]2+ complex, the PdII center contrib-
utes in particular to the low-energy tail of the first CD band
through partial metal-to-ligand charge transfer (excitation 9,
Table 12 and Figure 23). However, there is no such contribu-
tion in the case of [CuACTHNUTRGNEUNG
(5c)2]+, for which low-energy CD
bands involving the metal center are not observed (excita-
tions 1 and 2, Tables 10 and 11 and Figure 21a). These re-
sults show that overall the metal contribution to the CD is
much more important for the Pd than for the Cu assemblies.
The molar rotations of PdII– and the CuI–bis-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
(P,P,RP,RP,DPd)- [PdACHTUNGTRENNUNG
(5c2)2]2+ built from two phospholes
(P,SP)-5c1 (26.7 mru). If we use the structure built/optimized
from phosphole (P,RP)-5c2, then a molar rotation of ꢀ9245
is obtained. These results not only confirm the absolute con-
figuration assignment (P,P,RP,RP,DPd)-(+) for [PdACTHNUTRGNEUNG
(5c)2]2+, but
more importantly bring to light the high sensitivity of the
molar rotation to the diastereomeric structure of PdII–bis-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
on the basis of molar rotation calculations. The results indi-
cate that the isomer with the smaller metal contribution in
the orbitals and in the intense excitations has the lower
molar rotation and agrees well with experiment. However,
since the overall participation of the metal orbitals in the
optical activity of the Cu complex does not appear domi-
nant, the question arises whether the role of the metal is
5998
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
main factors responsible for the lower MR of the optimized
5c2 structure compared to 5c1.
Conclusion
At this stage, we can correlate the factors influencing the
A series of aza[n]helicene phospholes has been prepared,
and the photophysical properties of these p-conjugated com-
pounds investigated. The chiroptical properties of aza[n]he-
licene phosphole 5c were examined in detail. The influence
of the p substituents on the chiroptical properties of aza[6]-
helicene phosphole 5c is examined theoretically. Indeed,
1) by modifying the dihedral angles between the phosphole
ring and the substituents in the 2- and 5-positions of 5c
(aza[6]helicene and phenyl), the conjugation path is varied
and influences the molecular rotation (MR); 2) the position
of the P-phenyl group relative to the aza[6]helicene, de-
pending on the SP or RP configuration, shows a significant
influence on the MR. We found that the azahelicene phosp-
holes are interesting systems not only in their own right but
also as ligands in metal complexes. The coordination
chemistry of aza[n]helicene phospholes as new 1,4-P,N che-
lating ligands enabled the preparation of PdII– and CuI–bis-
MR of the [CuDACHTUNGTRENNUNG LAHCUTNGTRENNGUN
(5c)2]+ and [Cu (5c)2]+ complexes with the
position of their phenyl groups relative to the helicene moi-
eties. If we consider the four points influencing the CD of
phosphole 5c, some interesting results might be expected.
Considering the structure of [CuLACTHNUTRGNEUNG
(5c1)2]+ (Figure 20) the
phenyl groups are not stacked next to the helicene moiety.
The models from the previous sections would predict that
1) removing the phenyl groups attached to the P center
should not dramatically alter the MR, since there is no
phenyl group near the azahelicene lowering the MR, and
2) further removing the terminal phenyl groups should
make the MR lower due to decreasing the extent of p conju-
gation. This is exactly the trend seen from the calculations.
The MR values for [CuLACTHNUTRGNEUNG
(5c1)2]+ with the P-phenyl groups
removed (PPh!PH), and then with further removal of the
terminal phenyl groups are 30.3, 33.5, and 23.2 mru, respec-
tively.
AHCTUNGTREG(NNUN aza[n]helicene phosphole) assemblies. In the case of the Pd
In the optimized [Cu
N
complexes, experimental and theoretical studies revealed
high stereoselectivity thanks to a combination of electronic
(trans effect) and steric (repulsion of the P-Ph substituents
and the azahelicenes) factors in the metal coordination
hand, two phenyl groups are in close proximity to the azahe-
licene (Figure 20). This could be the cause of the low MR
for this isomer. It is actually the terminal phenyl group of
one ligand that is closest to the azahelicene of the other
ligand. Considering these facts, the ligand model would pre-
dict that 1) removing the phenyl groups attached to the P
center should increase the MR, since there is a phenyl
group near the azahelicene lowering the MR, and 2) further
removing the 5-phenyl groups should make the MR even
higher because it is these groups that are closest to the aza-
helicene (despite having shortened the length of conjuga-
tion). This is, again, exactly the trend seen from the calcula-
sphere. The chiroptical properties of [Pd
AHCTUNGTRENNUNG
(5c)2]2+ and [Cu-
ACHTUNGTRENNUNG
(5c)2]+ have been compared by means of TDDFT calcula-
tions, which showed that the metal makes small contribu-
tions to the CD absorption values and that the role of the
metal atom in such complexes is to organize the aza[6]heli-
cenes in space relative to each other. Indeed, it has been
demonstrated that the proximity in space of the phenyl
groups to the helicene moieties significantly influences the
MR values of the metal–bis
semblies.
ACHTUNGTRNE(NUNG aza[6]helicene phosphole) as-
tions. The MR values for [CuDACTHNUTRGNEUNG
(5c1)2]+ with P-phenyl groups
removed, and then with further removal of the terminal
phenyl groups are 13.2, 26.9, and 37.7 mru, respectively.
Considering these phenyl-group configurations, a ligand
without phenyl groups would cause MRs of the D and L iso-
mers that are more similar than those calculated for the full
systems. However, these isomers would still be quite
unique: with all phenyl groups removed the MRs of the two
isomers still differ by more than 10 mru.
The Pd system was investigated by us in a previous short
communication,[12a] mainly to assign the CD spectrum. It
was noted that the low-energy tail of the CD spectrum in-
volved MO contributions with some metal character, but so
far it has been unclear what the exact role of the metal orbi-
tals in the molar rotation of the complex is. The experimen-
Experimental Section
General information: All experiments were performed under an atmos-
phere of dry argon by using standard Schlenk techniques. Commercially
available reagents were used as received without further purification. Sol-
vents were freshly distilled under argon from sodium/benzophenone (tet-
rahydrofuran, diethyl ether) or from phosphorus pentoxide (pentane, di-
chloromethane). PPhBr2 was prepared as described in the literature.[40a]
Irradiation reactions were conducted with a Heraeus TQ 150 mercury
vapor lamp. Preparative separations were performed by gravity column
chromatography on basic alumina (Aldrich, Type 5016 A, 150 mesh,
58 ꢂ) or silica gel (Merck Geduran 60, 0.063–0.200 mm) in 3.5–20 cm col-
13
31
umns. 1H, C, and P NMR spectra were recorded on Bruker AM300
and DPX200 spectrometers. 1H and 13C NMR chemical shifts are report-
ed in parts per million (ppm) relative to Me4Si as external standard.
31P NMR downfield chemical shifts are expressed with a positive sign in
ppm, relative to external 85% H3PO4 and are proton-decoupled. High-
resolution mass spectra were obtained on a Varian MAT 311 or ZabSpec
TOF Micromass instrument at CRMPO, University of Rennes 1. Ele-
mental analyses were performed by the CRMPO, University of Rennes
1. CD Specific rotations (in degcm2 gꢀ1) were measured in a 1 dm ther-
mostated quartz cell on a Jasco-P1010 polarimeter. Circular dichroism (in
mꢀ1 cmꢀ1) was measured on a Jasco J-815 Circular Dichroism Spectrome-
ter. Derivatives 2c to 5c and their PdII and CuI complexes were prepared
as described in ref. [12a].
tally measured MR for [Pd
+23.1 mru. The computed MR value for [Pd
26.6, and that for [PdDA
ACHTUNGTRENNUNG
CTHUNGTRENNUNG
CTHUNGTRENNUNG
clearly indicate that the structure of the Pd complex in solu-
tion can be unambiguously assigned as containing two 5c1
centers and a D configuration at the Pd center and shed
light on the high stereoselectivity of formation of the PdII–
bisACHTUNGTRENNUNG(aza[6]helicene phosphole) assembly.
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5999

R. Rꢀau, J. Autschbach, J. Crassous et al.
Determination of optical data: UV/Vis spectra were recorded at room
temperature on a UVIKON 942 or a Biotek Instruments XS spectropho-
tometer in quartz cuvettes of 1 cm path length.
was warmed to room temperature and stirred for 30 min. Then the reac-
tion mixture was cooled to ꢀ788C and
a solution of aldehyde 1a
(500 mg, 1.74 mmol) in anhydrous THF (5 mL) was added dropwise.
After stirring under argon at room temperature for three hours, the reac-
tion mixture was filtered over Celite. Evaporation of the solvent and
column chromatography over silica gel (heptane/EtOAc 8:2 as eluent) af-
forded 3a (600 mg, 95%) as a mixture of Z/E isomers. trans isomer (first-
Steady-state luminescence spectra were measured on a Jobin Yvon Fluo-
roMax-2 spectrofluorimeter, fitted with a red-sensitive Hamamatsu R928
photomultiplier tube; the spectra shown are corrected for the wavelength
dependence of the detector, and the quoted emission maxima refer to
the values after correction. Samples for emission measurements were
contained within quartz cuvettes of 1 cm path length modified with ap-
propriate glassware to allow connection to a high-vacuum line. Degassing
was achieved by a minimum of three freeze–pump–thaw cycles whilst
connected to the vacuum manifold; final vapor pressure at 77 K was
<5ꢃ10ꢀ2 mbar, as monitored with a Pirani gauge. Luminescence quan-
tum yields were determined by the method of continuous dilution, with
1
eluted): H NMR (300 MHz, CDCl3): d=7.25–7.68 (m, 14H), 7.18 (d, J=
ꢅ
ꢅ
16.2 Hz, 1H, =CH), 2.53 (m, 4H, CCH2CH2), 1.85 ppm (m, 4H,
CCH2CH2); 13C NMR (75 Hz, CDCl3): 155.9, 143.6, 136.6, 133.4, 131.6,
ꢅ
ꢅ
128.8, 128.4, 128.25, 127.6, 127.2, 125.3, 124.0, 120.5, 90.3 ( C), 89.9 ( C),
ꢅ
ꢅ
ꢅ
ꢅ
ꢅ
81.1 ( C), 81.0 ( C), 29.8 ( CCH2), 28.0 ( CCH2), 19.2 ( CCH2CH2),
19.1 ppm ( CCH2CH2). cis isomer (second-eluted): 1H NMR (300 MHz,
CDCl3): d=7.25–7.43 (m, 11H), 7.19 (d, J=7.6 Hz, 1H), 7.04 (d, J=
7.9 Hz, 1H), 6.85 (d, J=12.4 Hz, 1H, =CH), 6.69 (d, J=12.4 Hz, 1H, =
ꢅ
quinine sulfate in 1m H2SO4 or [RuACHTNUTRGENUNG(bpy)3]Cl2 (bpy=2,2’-bipyridine) in
ꢅ
ꢅ
CH), 2.51 (m, 4H, CCH2CH2), 1.82 ppm (m, 4H, CCH2CH2);
air-equilibrated aqueous solution as standard (f=0.548 and 0.028 respec-
tively); estimated uncertainty in f is ꢃ20% or better.
Luminescence lifetimes shorter than 10 ms were measured by time-corre-
lated single-photon counting (TCSPC) following excitation at 374.0 nm
with an EPL-375 pulsed-diode laser. The emitted light was detected at
13C NMR (75 Hz, CDCl3): 156.6, 143.6, 136.6, 135.7, 133.5, 131.6, 130.5,
ꢅ
ꢅ
ꢅ
128.9, 128.4, 128.2, 127.6, 124.9, 123.9, 90.3 ( C), 89.7 ( C), 81.0 ( C),
ꢅ
ꢅ
ꢅ
ꢅ
ꢅ
80.7 ( C), 27.9 ( CCH2), 27.5 ( CCH2), 19.0 ( CCH2CH2), 19.0 ppm (
CCH2CH2); elemental analysis calcd (%) for C27H23N (361.478): C 89.71
H 6.41 N 3.87; found: C 89.77 H 6.47 N 3.62; HRMS (EI): m/z: calcd for
C27H23N: 361.18305; found: 361.1798.
908 with
through a monochromator. Phosphorescence lifetimes longer than 10 ms
were measured by multichannel scaling following excitation with
a Peltier-cooled R928 photomultiplier tube after passage
a
2-(8-Phenylocta-1,7-diyn-1-yl)benzo[f]quinoline (4a): A toluene solution
(340 mL) of substituted pyridine 3a (200 mg, 0.55 mmol) containing cata-
lytic amounts of iodine was irradiated overnight with a Heraeus TQ 150
mercury vapor lamp. Evaporation of the solvent followed by purification
by column chromatography over silica gel (heptane/EtOAc 8:2 as eluent)
pulsed xenon lamp. Measurements at 77 K were carried out in a diethyl
ether/isopentane/ethanol (2:2:1 v/v) matrix in 4 mm o.d. tubes in a liquid-
nitrogen dewar.
6-(8-Phenylocta-1,7-diyn-1-yl)pyridine 2-carboxaldehyde (1a): [PdCl2-
ACHTUNGTRENNUNG(PPh3)2] (94 mg, 134 mmol, 5%), CuI (25 mg, 134 mmol, 5%), and iPr2NH
1
afforded 4a (100 mg, 50%) as a beige solid. H NMR (500 MHz, CDCl3):
d=8.81 (d, J=8.5, 1H, H1), 8.52 (d, J=8.0 Hz, 1H, H10), 7.99 (d, J=
9.1 Hz, 1H, H5 or H6), 7.95 (d, J=9.1 Hz, 1H, H5 or H6), 7.9 (d, J=
7.5 Hz, 1H), 7.66 (m, 2H), 7.62 (d, J=8.5 Hz, 1H, H2), 7.45 (m, 2H),
(11 mL, 97 mmol) were added to a stirred solution of 2-bromopyridine-
carboxaldehyde (500 mg, 2.68 mmol) in anhydrous THF (10 mL) at room
temperature under argon. After 5 min, 1-phenyl-octa-1,7-diyne (489 mg,
2.68 mmol) was added. The resulting mixture was stirred overnight.
Evaporation of the solvents under reduced pressure and purification by
column chromatography over silica gel (heptane/CH2Cl2 1:1 as eluent) af-
forded 1a as a brown product (470 mg, 63%). Rf (heptane/CH2Cl2 1:1)=
0.1; 1H NMR (300 MHz, CDCl3): d=7.82 (m, 2H), 7.59 (dd, J=7.4 Hz,
ꢅ
7.30 (m, 3H), 2.62 (t, J=6.7 Hz, 2H, CCH2CH2), 2.53 (t, J=6.67 Hz,
2H, CCH2CH2), 1.83–1.93 ppm (m, 4H, CCH2CH2); 13C NMR (75 Hz,
ꢅ
ꢅ
CDCl3): 148.1, 143.3, 131.7, 131.6, 131.4, 130.85, 129.3, 128.7, 128.2, 127.7,
127.6, 127.5, 127.2, 124.5, 124.2, 123.9, 122.7, 91.7 ( C), 89.75 ( C), 81.2
( C), 81.1 ( C), 28.0 ( CCH2), 27.5 ( CCH2), 19.2 ( CCH2CH2),
19.1 ppm ( CCH2CH2); HRMS (EI): m/z: calcd for C27H21N: 359.16740;
found: 359.1674; UV/Vis (CH2Cl2): l (e)=280 (35000), 290 (26000), 304
(13400), 326 (6310), 342 (6780), 360 nm (8800).
ꢅ
ꢅ
ꢅ
ꢅ
ꢅ
ꢅ
ꢅ
ꢅ
ꢅ
J=1.1 Hz, 1H, Hpy), 7.40 (m, 2H), 7.27 (m, 3H), 2.53 (m, 4H, C CCH2),
1.84 ppm (m, 4H, C CCH2); 13C NMR (75 Hz, CDCl3): 193.1, 152.7,
ꢅ
144.4, 137.2 (CH), 131.5 (CH), 131.5 (CH), 131.0 (CH), 128.2 (CH), 128.2
ꢅ
ꢅ
ꢅ
ꢅ
(CH), 123.8 (CH), 120.0 (CH), 92.3 (C ), 89.6 (C ), 81.1 (C ), 79.8 (C ),
1-Phenyl-2-(benzo[f]quinol-2-yl)-5-phenyl-3,4-butanophosphole
(5a):
ꢅ
ꢅ
ꢅ
ꢅ
(
27.9 ( CCH2), 27.4 ( CCH2), 19.03 ( CCH2CH2), 19.0 ppm
nBuLi (1.6m in hexanes, 0.9 mL, 1.47 mmol) was added dropwise to a
THF solution (20 mL) of diyne 4a (220 mg, 0.61 mmol) and
[Cp2ZrCl2](179 mg, 0.61 mmol) under argon at ꢀ788C. The reaction mix-
ture was warmed to room temperature and stirred overnight. Freshly dis-
tilled PhPBr2 (139 mL, 0.67 mmol) was added to this solution at ꢀ508C.
The reaction mixture was allowed to warm to room temperature and was
stirred for 30 h. The mixture was then filtered over basic alumina (THF
as eluent) under inert atmosphere and the volatile substances were re-
moved in vacuo. Final purification by column chromatography over silica
gel (heptane/EtOAc 8:2 as eluent) afforded phosphole 5a (210 mg, 73%)
as a yellow solid. Single crystals were grown by slow diffusion of pentane
into a chloroform solution. M.p. 2058C; Rf (heptane/AcOEt 8:2)=0.75;
31P NMR (81 MHz, CDCl3): d=12.3 ppm; 1H NMR (300 MHz, CDCl3):
d=8.81 (d, J=8.7 Hz, 1H), 8.55 (d, J=8.0 Hz, 1H), 7.9–8.0 (m, 4H),
7.79 (d, J=8.7 Hz, 1H), 7.58–7.69 (m, 2H), 7.47 (d, J=8 Hz, 2H), 7.34 (t,
J=7.7 Hz, 4H), 7.21 (t, J=7.3 Hz, 1H), 7.1 (m, 2H), 3.6–3.79 (m, 1H),
2.88–3.1 (m, 2H), 2.72–2.79 (m, 1H), 1.65–1.92 ppm (m, 4H); 13C NMR
(75 MHz, CDCl3): 155.9 (C), 155.7 (C), 149.4 (C), 149.3 (C), 148.1 (C),
148.1 (C), 146.0 (C), 145.9 (C), 144.7 (C), 144.6 (C), 142.4 (C), 142.4 (C),
137.3 (C), 137.0 (C), 133.65 (CH), 133.4 (CH), 132.2 (C), 132.0 (C), 131.4
(C), 130.6 (CH), 130.5 (CH), 129.7 (C), 129.3 (CH), 129.2 (CH), 128.9
(CH), 128.6 (CH), 128.5 (CH), 128.4 (CH), 128.3 (CH), 128.2 (C), 126.9
(CH), 126.7 (CH), 126.5 (CH), 122.8 (C), 122.4 (CH), 122.1 (CH), 122.0
(CH), 29.7 (CH2), 29.0 (CH2), 28.2 (CH2), 23.2 ppm (CH2); HRMS (EI):
m/z: calcd for C33H26NP: 467.18029; found: 467.1785; UV/Vis (CH2Cl2): l
(e)=271 (25600), 291 (18200), 348 (10900), 393 nm (21200); elemental
analysis calcd (%) for C33H26NP: C 84.77, H 5.61, N 3.00; found: C 85.73,
H 5.60, N 3.02.
CCH2CH2); elemental analysis calcd (%) for C20H17NO (287.36): C 83.60,
H 5.96, N 4.87; found: C 83.27, H 5.96, N 4.87; HRMS (EI): m/z: calcd
for C20H17NO: 287.13101; found: 287.1315.
2-(8-Phenylocta-1,7-diyn-1-yl)quinoline 3-carboxaldehyde (1b): [PdCl2-
ACHTUNGTRENNUNG(PPh3)2] (91 mg, 130 mmol, 5%), CuI (25 mg, 130 mmol, 5%), and iPr2NH
(11 mL, 97 mmol) were added to a stirred solution of 2-chloro-3-quino-
line carboxaldehyde (500 mg, 2.6 mmol) in anhydrous THF (10 mL) at
room temperature and under argon. After 5 min 1-phenyl-octa-1,7-diyne
(475 mg, 2.6 mmol) was added. The resulting mixture was stirred over-
night. Evaporation of the solvents under reduced pressure and purifica-
tion by column chromatography over silica gel (heptane/EtOAc 9:1 as
eluent) afforded 2b as an orange oil (533 mg, 61%). Rf (heptane/AcOEt
9:1)=0.19; 1H NMR (300 MHz, CDCl3): d=10.71 (s, 1H, CHO), 8.71 (s,
1H, H4), 8.14 (d, J=8.7 Hz, 1H), 7.95 (d, J=7.9 Hz, 1H), 7.86 (ddd, J=
8.7 Hz, 7.1 Hz, 1.5 Hz, 1H), 7.63 (ddd, J=8.3 Hz, 6.8 Hz, 1.1 Hz, 1H),
7.40–7.65 (m, 2H, H Ph), 7.27–7.31 (m, 3H, H Ph), 2.69 (t, J=6.8 Hz,
ꢅ
ꢅ
ꢅ
2H, CCH2), 2.53 (t, J=6.8 Hz, 2H, CCH2), 1.67–1.97 ppm (m, 4H,
CCH2CH2); 13C NMR (75 MHz, CDCl3): 190.9 (CHO), 149.9 (C), 144.2
(C), 136.8 (CH), 132.8 (CH), 131.5 (CHꢃ2), 129.6 (CH), 129.1 (CH),
128.7 (C), 128.2 (CHꢃ2), 127.9 (CH), 127.6 (CH), 126.2 (C), 123.8 (C),
ꢅ
ꢅ
ꢅ
97.5 and 89.5 (C CCH2), 81.2 and 77.8 (C CCH2), 28.0 and 27.3 (C
ꢅ
CCH2), 19.3 and 19.0 ppm (C CCH2CH2); HRMS (EI): m/z: calcd for
C23H19N [MꢀCO] +: 309.15175; found: 309.1513.
C
6-(8-Phenylocta-1,7-diyn-1-yl)-2-(2-phenylethenyl)pyridine (3a): nBuLi
(1.6m in hexanes, 1.3 mL, 2.1 mmol) was added dropwise to a stirred sus-
pension of benzylphosphonium bromide 2a (904 mg, 2.1 mmol) in anhy-
drous THF (15 mL) cooled to ꢀ788C under argon. The reaction mixture
6000
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
A
m/z: calcd for C37H28NP: 517.19594; found: 517.1942; UV/Vis (CH2Cl2): l
(e)=284 (49000), 320 (17000), 332 (14300), 382 (22800), 406 nm
(28800); elemental analysis calcd (%) for C37H28NP: C 85.86, H 5.45, N
2.71; found: C 85.85, H 5.44, N 2.73.
(3b): nBuLi (1.6m in hexanes, 0.85 mL, 1.32 mmol) was added dropwise
to a stirred suspension of 2-naphthylmethylphosphonium bromide 2b
(600 mg, 1.2 mmol) in anhydrous THF (15 mL) cooled to ꢀ788C under
argon. The reaction mixture was warmed to room temperature and
stirred for 30 min. Then the reaction mixture was cooled to ꢀ788C and a
solution of aldehyde 1b (300 mg, 1 mmol) in anhydrous THF (10 mL)
was added dropwise. After stirring under argon at room temperature for
three hours, the reaction mixture was filtered over Celite. Evaporation of
the solvent and column chromatography over silica gel (heptane/EtOAc
8:2 as eluent) afforded 3b (377 mg, 92%) as a Z/E isomer mixture.
1H NMR (300 MHz, CDCl3): d=7.94 (s, 0.46H), 7.79–7.83 (m, 4.38H),
7.58 (t, J=7.9 Hz, 0.5H), 7.45–7.49 (m, 3.8H), 7.22–7.35 (m, 5.8H), 7.07
(d, J=7.5 Hz, 0.5H), 7.02 (d, J=12 Hz, 0.5H, =CH Z isomer), 6.82 (d,
2-(8-Phenylocta-1,7-diyn-1-yl)-3-[(Z/E)-2-phenylethenyl]quinoline (3d):
nBuLi (1.35m in hexanes, 1.05 mL, 1.42 mmol) was added dropwise to a
stirred suspension of benzylphosphonium bromide 2a (617 mg,
1.42 mmol) in anhydrous THF (10 mL) cooled to ꢀ788C under argon.
The reaction mixture was warmed to room temperature and stirred for
30 min. Then the reaction mixture was cooled to ꢀ788C and a solution of
aldehyde 1b (480 mg, 1.42 mmol) in anhydrous THF (5 mL) was added
dropwise. After stirring under argon at room temperature for three
hours, the reaction mixture was filtered over Celite. Evaporation of the
solvent and column chromatography over silica gel (heptane/EtOAc 8:2
J=12 Hz, 0.5H, =CH Z isomer), 2.49–2.61 (m, 4H, CCH2), 1.80–
1
as eluent) afforded 3d (570 mg, 97%) as a Z/E isomer mixture. H NMR
1.91 ppm (m, 4H, CCH2CH2); 13C NMR (75 MHz, CDCl3): 156.6 (C),
(300 MHz, CDCl3): d=8.37 (s, 1H, H4), 8.46 (d, J=8.7 Hz, 1H), 7.60–
7.80 (m, 5H), 7.53 (m, 1H), 7.39–7.42 (m, 4H), 7.26–7.32 (m, 5H), 6.93
(d, J=12 Hz, 0.07H, =CH Z isomer), 6.83 (d, J=12 Hz, 0.07H, =CH Z
155.9 (C), 143.7 (C), 143.6 (C), 136.6 (CH), 135.7 (CH), 134.2 (C), 134.1
(C), 133.6 (CH), 133.5 (C), 133.5 (CH), 133.4 (Cꢃ2), 132.8, 131.6, 130.7
(CH), 128.4 (CH), 128.3 (CH), 128.1 (CH), 128.0 (CHꢃ2), 127.8 (CH),
127.7 (CHꢃ2), 127.6 (CH), 126.9 (CH), 126.4 (CHꢃ2), 126.2 (CHꢃ2),
125.3 (CH), 125.0 (CH), 124.0 (C), 123.7 (CH), 122.8 (CH), 120.6 (CH),
90.4 (Cꢃ2), 89.8 (C), 81.1 (Cꢃ2), 81.8 (C), 28.0 (C CCH2 ꢃ2), 27.6 (C
CCH2 ꢃ2), 19.2 (C CCH2CH2), 19.1 ppm (C CCH2CH2 ꢃ3); HRMS
(EI): m/z: calcd for C31H25N: 411.19870; found: 411.1983.
ꢅ
ꢅ
isomer), 2.7 (t, J=6.8 Hz, 2H, CCH2), 2.52 (t, J=6.8 Hz, 2H, CCH2),
1.85–1.98 ppm (m, 4H, CCH2CH2); 13C NMR (75 MHz, CDCl3): 147.3
ꢅ
(C), 143.5 (C), 137.0 (C), 132.0 (CH), 131.6 (CHꢃ2), 130.8 (CH), 129.6
(CH), 129.0 (CH), 128.9 (CHꢃ2), 128.3 (CH), 128.2 (CHꢃ2), 127.6
(CH), 127.5 (CHꢃ2), 127.3 (CH), 127.2 (CH), 126.8 (CHꢃ2), 124.8
ꢅ
ꢅ
ꢅ
(CH), 123.9 (C), 95.5 and 89.6 (C CCH2), 81.1 and 79.9 (C CCH2), 28.0
2-(8-Phenylocta-1,7-diyn-1-yl)naphthoACTHNUGTRNEUNG[1,2-f]quinoline (4b): A toluene so-
ꢅ
ꢅ
and 27.6 (C CCH2), 19.4 and 19.1 ppm (C CCH2CH2); HRMS (EI): m/
z: calcd for C31H25N: 411.19870; found: 411.1983.
lution (570 mL) of substituted pyridine 3b (649 mg, 1.58 mmol) contain-
ing catalytic amounts of iodine was irradiated overnight with a Heraeus
TQ 150 mercury vapor lamp. Evaporation of the solvent followed by pu-
rification by column chromatography over silica gel (heptane/EtOAc 8:2
as eluent) afforded 4b (304 mg, 47%) as a beige solid. M.p. 1048C. Rf
(heptane/EtOAc 8:2)=0.4; 1H NMR (300 MHz, CDCl3): d=9.17 (d, J=
8.8 Hz, 1H), 8.76 (d, J=7.4 Hz, 1H), 8.12 (d, J=8.8 Hz, 1H), 7.98 (d, J=
8.9 Hz, 2H), 7.86 (d, J=8.5 Hz, 1H), 7.75 (d, J=8.6 Hz, 1H), 7.56–7.64
6-(8-Phenylocta-1,7-diyn-1-yl)benzo[k]phenanthridine (4d):
A toluene
solution (250 mL) of quinoline 3d (150 mg, 0.36 mmol) containing cata-
lytic amounts of iodine was irradiated overnight with a Heraeus TQ 150
mercury vapor lamp. Evaporation of the solvent followed by purification
by column chromatography over silica gel (heptane/EtOAc 8:2 as eluent)
afforded 4d (92 mg, 62%) as an orange oil. Rf (heptane/EtOAc 8:2)=
0.21; 1H NMR (300 MHz, CDCl3): d=9.09 (m, 1H), 8.96 (d, J=8.7 Hz,
1H), 8.48 (d, J=8.7 Hz, 1H), 8.32 (dd, J=8.3 Hz, 1.5 Hz, 1H), 8.0 (m,
1H), 7.94 (d, J=8.7 Hz, 1H), 7.68–7.78 (m, 4H), 7.44–7.47 (m, 2H),
ꢅ
(m, 3H), 7.45–7.48 (m, 2H), 7.28–7.31 (m, 3H), 2.64 (t, J=6.6 Hz, 2H,
ꢅ
CCH2), 2.54 (t, J=6.6 Hz, 2H, CCH2), 1.82–1.98 ppm (m, 4H,
CCH2CH2); 13C NMR (75 MHz, CDCl3): 148.7 (C), 142.6 (C), 135.4
(CH), 133.5 (C), 131.6 (CHꢃ2), 131.1 (CH), 131.0 (C), 129.9 (C), 128.8
(CH), 128.4 (CH), 128.3 (CHꢃ3), 127.5 (CH), 127.2 (CH), 126.7 (CH),
126.6 (CH), 126.5 (C), 126.3 (CH), 124.1 (C), 124.0 (C), 123.5 (CH), 91.8
ꢅ
7.28–7.30 (m, 3H), 2.76 (t, J=6.8 Hz, 2H, CCH2), 2.57 (t, J=6.8 Hz,
2H, CCH2), 1.91–2.04 ppm (m, 4H, CCH2CH2); 13C NMR (75 MHz,
CDCl3): 146.0 (C), 144.4 (C), 135.0 (C), 131.6 (CHꢃ2), 131.5 (C), 129.9
(CH), 128.8 (CH), 128.7 (CH), 128.5 (CH), 128.3 (CH), 128.2 (CHꢃ2),
127.9 (CH), 127.6 (CH), 127.0 (CH), 126.9 (CHꢃ2), 125.7 (C), 124.4
ꢅ
ꢅ
and 89.8 (C CCH2), 81.3 and 81.1 (C CCH2), 28.1 and 27.6 (C CCH2),
19.2 and 19.1 ppm (C CCH2CH2); HRMS (EI): m/z: calcd for C31H23N:
409.18305; found: 409.1837; UV/Vis (CH2Cl2): l (e)=289 (33700), 307
(14100), 324 (7700), 340 (6000), 363 (3000), 382 nm (3300).
ꢅ
ꢅ
(CH), 124.0 (C), 123.9 (C), 95.5 and 89.7 (C CCH2), 81.2 and 79.6 (C
ꢅ
ꢅ
CCH2), 28.1 and 27.6 (C CCH2), 19.5 and 19.1 ppm (C CCH2CH2);
HRMS (EI): m/z: calcd for C31H23N: 409.18305; found: 409.1837; UV/Vis
(CH2Cl2): l (e)=284 (50000), 315 (14500), 336 (5000), 349 (2500), 369
(3600), 388 nm (3670).
1-Phenyl-2-(naphthoACHTUNGTRENNUNG[1,2-f]quinol-2-yl)-5-phenyl-3,4-butanophosphole
(5b): nBuLi (1.4m in hexanes, 0.4 mL, 0.56 mmol) was added to a THF
solution (10 mL) of diyne 4b (96 mg, 0.23 mmol) and [Cp2ZrCl2] (69 mg,
0.23 mmol) under argon dropwise at ꢀ788C. The reaction mixture was
warmed to room temperature and stirred overnight. Freshly distilled
PhPBr2 (53 mL, 0.26 mmol) was added to this solution at ꢀ508C. The re-
action mixture was allowed to warm to room temperature and stirred for
30 h. The mixture was then filtered over basic alumina (THF as eluent)
under inert atmosphere and the volatile substances were removed in
vacuo. Final purification by column chromatography over silica gel (hep-
tane/EtOAc 8:2 as eluent) afforded phosphole 5b (76 mg, 63%) as a
yellow solid. Single crystals were grown by slow diffusion of pentane into
1-Phenyl-2-(benzo[k]phenanthrid-6-yl)-5-phenyl-3,4-butanophosphole
(5d): nBuLi (1.6m in hexanes, 0.4 mL, 0.66 mmol) was added dropwise to
a THF solution (10 mL) of diyne 4d (113 mg, 0.28 mmol) and [Cp2ZrCl2]
(81 mg, 0.28 mmol) under argon at ꢀ788C. The reaction mixture was
warmed to room temperature and stirred overnight. Freshly distilled
PhPBr2 (63 mL, 0.3 mmol) was added to this solution at ꢀ508C. The reac-
tion mixture was allowed to warm to room temperature and stirred for
30 h. The mixture was then filtered over basic alumina (THF as eluent)
under inert atmosphere and the volatile substances were removed in
vacuo. Final purification by column chromatography over silica gel (hep-
tane/EtOAc 8:2 as eluent) afforded phosphole 5d (65 mg, 45%) as a
yellow solid. M.p. 1798C; Rf (heptane/EtOAc 8:2)=0.2; 31P NMR
(81 MHz, CDCl3): d=20.7 ppm; 1H NMR (300 MHz, CDCl3): d=9.2 (m,
1H), 9.03 (d, J=8 Hz, 1H), 8.38 (d, J=7.7 Hz, 1H), 8.02 (m, 2H), 7.66–
7.87 8 (m, 5H), 7.58 (brd, J=8 Hz, 2H), 7.39 (brt, J=7.3 Hz, 2H), 7.26
(m, 3H), 7.02 (m, 3H), 2.95 (m, 2H), 2.65 (m, 1H), 2.25 (m, 1H), 1.85
a
31P NMR (81 MHz, CDCl3): d=12.2 ppm; 1H NMR (300 MHz, CDCl3):
d=9.23 (d, J=8.9 Hz, 1H), 8.92 (d, J=8 Hz, 1H), 8.16 (d, J=8.7 Hz,
1H), 8.03 (d, J=8.7 Hz, 2H), 7.9 (d, J=8.7 Hz, 1H), 7.86 (d, J=8.7 Hz,
1H), 7.78 (d, J=8.7 Hz, 1H), 7.66 (m, 2H), 7.49 (d, J=7.8 Hz, 2H),
7.32–7.40 (m, 4H), 7.08–7.22 (m, 1H), 6.96 (m, 3H), 3.66–3.74 (m, 1H),
2.95–3.12 (m, 2H), 2.79 (m, 1H), 1.6–1.95 ppm (m, 4H); 13C NMR
(m, 2H), 1.65 ppm (m, 2H); 13C NMR (75 MHz, CDCl3) (J
ACHTNUTRGNE(NUG P,C) coupling
(75 MHz, CDCl3, JACHTUNGTRENNUNG(P,C) coupling constants not assigned): d=155.4 (C),
constants not assigned): d=157.4 (C), 157.2 (C), 156.9, 148.6 (C), 148.4
(C), 146.2 (C), 145.0 (C), 143.6 (C), 143.5 (C), 141.9 (C), 137.3 (C), 137.1
(C), 134.7 (C), 133.2 (CH), 132.9 (CH), 132.0 (C), 131.6 (C), 131.4 (C),
130.3 (CH), 129.3 (CH), 129.2 (CH), 129.1 (C), 128.7 (CHꢃ2), 128.6
(CH), 128.4 (CHꢃ2), 128.3 (CH), 128.2 (CH), 127.9 (CH), 127.7 (CH),
126.9 (CH), 126.6 (CH), 126.4 (CH), 126.1 (CH), 125.0 (C), 124.9 (C),
155.1 (C), 154.8 (C), 149.8 (C), 149.7 (C), 146.3 (Cꢃ2), 144.7 (C), 144.6
(C), 137.0 (C), 135.2 (CH), 133.7 (CH), 133.5 (CH), 132.3 (C), 132.2 (C),
130.5 (CH), 130.4 (C), 130.1 (C), 129.3 (CH), 129.2 (CH), 128.9 (CH),
128.8 (CH), 128.4 (CH), 128.3 (CHꢃ2), 127.6 (CH), 127.5 (CH), 127.0,
126.8 (CH), 126.4 (CHꢃ2), 126.1 (CH), 123.0 (Cꢃ2), 121.1 (CH), 121.0
(CH), 29.7 (CH2), 29.2 (CH2), 28.2 (CH2), 23.3 ppm (CH2); HRMS (EI):
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6001

R. Rꢀau, J. Autschbach, J. Crassous et al.
124.4 (Cꢃ2), 123.6 (C), 28.1 (CH2 ꢃ2), 23.3
(CH2), 22.3 ppm (CH2);
gel (heptane/EtOAc 8:2 as eluent) afforded phosphole 5e (42 mg, 27%)
as a yellow solid. Rf (heptane/EtOAc 8:2)=0.23; ½aꢄ2D3 =ꢀ11 (CH2Cl2, c=
0.83); 31P NMR (81 MHz, CDCl3): d=20.4 and 20.7 ppm (two diastereo-
mers); 1H NMR (300 MHz, CDCl3): d=8.96 (d, J=8 Hz, 1H), 8.27 (d,
J=7 Hz, 1H), 7.73 (m, 1H), 7.62–7.80 (m, 1H), 7.53 (d, J=7.7 Hz, 2H),
7.37 (t, J=7.7 Hz, 2H), 7.28–7.22 (m, 5H), 6.98 (m, 3H), 3.5 (m, 1H),
2.99 (t, J=5.4 Hz, 1H), 2.88 (s, 1H), 2.68–2.62 (m, 4H), 1.88 (m, 1H),
1.68 (m, 4H), 1.61 (s, 3H, H7), 1.46 (s, 1H), 0.88 ppm (d, J=8 Hz);
13C NMR (75 MHz, CDCl3): 171.1 (C), 159.2 (C), 159.1 (C), 150.8 (C),
145.6 (C), 137.4 (C), 137.2 (C), 133.1 (CH), 132.9 (CH), 131.9 (C), 130.8
(CH), 129.3 (CH), 129.2 (CH), 128.6 (CH), 128.5 (CH), 128.4 (C), 128.2
(CH), 127.6 (CH), 127.2 (CH), 126.8 (CH), 126.7 (CH), 126.3 (CH),
125.6 (CH), 125.2 (C), 77.2 (CH), 68.0 (CH2), 65.9 (C), 60.4 (CH2), 49.7
(CH), 77.3 (C), 41.2 (CH), 38.0 (CH2), 31.1 (CH2), 28.2 (CH2), 25.8 (CH),
26.5 (CH2), 23.3 (CH2), 22.8 (CH), 21.3 (CH3), 21.2 (CH3), 21.1 ppm
(CH3); HRMS (EI): m/z: calcd for C40H36NOP: 577.25345; found:
577.2493; UV/Vis (CH2Cl2): l (e)=350 (3500); elemental analysis (%)
calcd for C40H36NP: C 85.53, H 6.46, N 2.49; found: C 85.51, H 6.45, N
2.51.
HRMS (EI): m/z: calcd for C37H28NP: 517.19594; found: 517.1950; UV/
Vis (CH2Cl2): l (e)=278 (34200), 317 (12700), 336 (5800), 361 (5400),
379 nm (4900); elemental analysis (%) calcd for C37H28NP: C 85.86, H
5.45, N 2.71; found: C 85.81, H 5.41, N 2.78.
2-(8-Phenylocta-1,7-diyn-1-yl)-3-(Z,E)-[2-{(1R,5S)-6,6-dimethylbicyclo-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
bromide 2d (425 mg, 0.9 mmol) in anhydrous THF (10 mL) cooled to
ꢀ788C under argon. The reaction mixture was warmed to room tempera-
ture and stirred for 30 min. Then the reaction mixture was cooled to
ꢀ788C and a solution of aldehyde 1b (300 mg, 0.89 mmol) in anhydrous
THF (5 mL) was added dropwise. After stirring under argon at room
temperature for three hours, the reaction mixture was filtered over
Celite. Evaporation of the solvent and column chromatography over
silica gel (heptane/EtOAc 8:2 as eluent) afforded 3e (385 mg, 95%) as a
Z/E isomer mixture. 1H NMR (300 MHz, CDCl3): d=8.24 (s, 0.9H), 8.2
(s, 0.1H), 8.08 (d, J=8.4 Hz, 0.1H), 8.03 (d, J=8.4 Hz, 0.9H), 7.74 (d,
[PdACHTUNGRTENNUG(5b)Cl2]: [PdACHTUGNTERNN(UGN CH3CN)2Cl2] (25 mg, 96.6 mmol) was added to a solu-
J=8.1 Hz, 1H), 7.61 (m, 1H), 7.48 (d, J=7.5 Hz, 1H), 7.42ACTHNUTRGENUG(N m, 3H), 7.27
tion of phosphole (50 mg, 96.6 mmol) in dry CH2Cl2 (5 mL) under argon.
After stirring at room temperature under argon for 1 h, the solvent was
stripped off and the precipitate washed with dry Et2O, yielding [Pd
(5b)Cl2] as a dark red solid (62 mg, 92%). 31P NMR (81 MHz, CD2Cl2):
d=+60.0 ppm; 1H NMR (300 MHz, CD2Cl2): d=9.51 (d, 1H, J=
8.3 Hz), 9.21 (d, 1H, J=7.9 Hz), 8.8 (d, 1H, J=8.3 Hz), 8.03–8.21 (m,
3H), 7.65–7.92 (m, 7H), 7.19–7.4 (m, 7H), 3.9 (m, 1H), 3.4 (m, 1H), 3.2
(m, 2H), 2.98 ppm (m, 4H). UV/Vis (CH2Cl2): l (e)=295 (30000), 352
(10600), 393 (8310), 407 (8170, 478 nm (2400); elemental analysis calcd
(%) for C37H28Cl2NPPd: C 63.95, H 4.06, N 2.02; found: C 63.91, H 4.10,
N 1.99.
(m, 3H), 7.04 (d, J=16.2 Hz, =CH E isomer), 6.97 (d, J=16.2 Hz, =CH
E isomer), 6.61 (d, J=12 Hz, =CH Z isomer), 6.40 (d, J=12 Hz, =CH Z
isomer), 5.84 (s, 0.9H), 5.77 (s, 0.1H), 2.80 (t, J=5.2 Hz, 1H), 2.64 (t, J=
ꢅ
6.4 Hz, 2H), 2.53 (m, 4H, CCH2), 2.46 (m, 1H), 2.19 (m, 1H), 1.89 (m,
ꢅ
4H, CCH2CH2), 1.86 (s, 3H), 1.23–1.19 (m, 1H), 0.98 (d, J=14.7 Hz,
1H), 0.88 ppm (s, 3H); 13C NMR (75 MHz, CDCl3): 146.8 (C), 146.75
(C), 143.4 (C), 134.1 (C), 133.3 (CH), 132.5 (C), 131.6 (CH), 130.1 (CH),
129.2 (CH), 128.8 (CH), 128.6 (C), 128.4 (C), 128.2 (CH), 127.6 (CH),
127.5 (CH), 127.4 (CH), 127.2 (C), 127.0 (CH), 123.9 (C), 120.6 (CH),
ꢅ
ꢅ
ꢅ
ꢅ
95.4 (C CCH2), 89.5 (C CCH2), 81.2 (C CCH2), 79.9 (C CCH2), 77.3
ꢅ
(C), 41.4 (CH), 40.9 (CH), 37.9 (C), 32.3 (CH2), 31.2 (CH2), 27.9 (C
ꢅ
ꢅ
CCH2), 27.5 (C CCH2), 26.4 (CH3), 20.9 (CH3), 19.3 (C CCH2CH2),
19.0 ppm (C CCH2CH2); HRMS (EI): m/z: calcd for C34H34N:
456.26913; found: 456.2696.
[PdACHTUNGRTENNUG(5d)Cl2]: [PdACHTUGNTERNN(UGN CH3CN)2Cl2] (25 mg, 96.6 mmol) was added to a solu-
ꢅ
tion of phosphole (50 mg, 96.6 mmol) in dry CH2Cl2 (5 mL) under argon.
After stirring at room temperature under argon for 1 h, the solvent was
stripped off and the precipitate was washed with dry Et2O, yielding [Pd-
6-(8-Phenylocta-1,7-diyn-1-yl)-[{(4S,6R)-3,4,5,6-tetrahydro-5,5-dimethyl-
4,6-methano}benzoACHTUNGTRENNUNG[1,2-k]-phenanthridine] (4e): A toluene (700 mL) so-
A
a
dark red solid (57 mg, 85%). 31P NMR (121.5 MHz,
l
(e)=307 (27000), 325
lution of quinoline 3e (260 mg, 0.57 mmol) containing catalytic amounts
of iodine was irradiated overnight with a Heraeus TQ 150 mercury vapor
lamp. Evaporation of the solvent followed by purification by column
chromatography over silica gel (heptane/EtOAc 9:1 as eluent) afforded
4e (64 mg, 25%) as a yellow oil. ½aꢄ2D3 =ꢀ42.9 (CH2Cl2, c=0.95). Rf (hep-
tane/EtOAc 8:2)=0.25; 1H NMR (500 MHz, CDCl3): d=8.97 (d, J=
8.3 Hz, 1H), 8.53 (d, J=8.1 Hz, 1H), 8.24 (dd, J1 =8.1 Hz, J2 =1.5 Hz,
1H), 7.74 (m, 1H), 7.65 (m, 1H), 7.44 (m, 2H), 7.36 (d, J=8.1 Hz), 7.29
(m, 3H), 3.90 (AB system, dd, J1 =16.1 Hz, J2 =3.1 Hz, 1H), 3.70 (AB
system, dd, J1 =16.1 Hz, J2 =2.6 Hz, 1H), 3.0 (t, J=5.8 Hz, 1H), 2.7 (t,
(18300), 436 (5760), 482 nm (2120): elemental analysis calcd (%) for
C37H28Cl2NPPd: C 63.95, H 4.06, N 2.02; found: C 63.93, H 4.08, N 1.98.
[PdACHTUNGNRET(NUNG 5b)2]ACHTUTNGERNNUG ACHUTNGRETNNG(UN CH3CN)4]ACTHNUGTREN(UNGN SbF6)2 (39 mg, 0.87 mmol) was added
(SbF6)2: [PdII
to a solution of phosphole 5b (90 mg, 1.74 mmol) in dry CH2Cl2 (5 mL)
under argon. After stirring at room temperature under argon for 1 h, the
solvent was stripped off and the precipitate was washed with dry Et2O,
yielding [PdACHTUNGTRNNEUG(5b)2]ACHTUNGTRNEN(UGN SbF6)2 (70 mg, 61%) as a dark red solid. Single crys-
tals were grown by slow diffusion of pentane vapor into a CH2Cl2 solu-
tion, but they were too small for X-ray analysis. 31P NMR (81 MHz,
CDCl3): d=76.7 ppm; 1H NMR (200 MHz, CDCl3): d=9.11 (d, J=
8.9 Hz, 1H), 8.3 (d, J=8.7 Hz, 1H), 8.03–8.14 (m, 2H), 7.9 (d, J=8.6 Hz,
1H), 7.79 (m, 2H), 7.28–7.65 (m, 7H), 6.83–6.97 (m, 3H), 6.66 (m, 3H),
3.95 (m, 1H), 3.30 (m, 1H), 2.63 (m, 2H), 2.41 (m, 2H), 2.04–12.12 ppm
(m, 4H); HRMS (ES): m/z: calcd for C74H56N235ClP2106Pd ([C++,Clꢀ]+):
1175.26420; found: 1175.2674; UV/Vis (CH2Cl2): l (e)=280 (39400), 303
(32400), 320 (25800), 340 (17300), 399 (11600), 525 nm (1400).
[PdACHTUNGNRET(NUNG 5d)2]ACHTUTNGERNNUG ACHUTNGRETNNG(UN CH3CN)4CATHNUGTREN(NNGU SbF6)2] (12 mg, 24 mmol) was added to
(SbF6)2: [PdII
a solution of phosphole 5d (25 mg, 48 mmol) in dry CH2Cl2 (2 mL) under
argon. After stirring at room temperature under argon for 1 h, the sol-
vent was stripped off and the precipitate washed with dry Et2O, yielding
ꢅ
ꢅ
J=6.8 Hz, 2H, CCH2), 2.69–2.75 (m, 1H), 2.56 (t, J=6.8 Hz, 2H,
ꢅ
ꢅ
CCH2), 1.99 (m, 2H, CCH2CH2), 1.92 (m, 2H, CCH2CH2), 1.48 (s,
3H), 1.4 (d, J=9.2 Hz, 1H), 0.69 ppm (s, 3H); 13C NMR (125.8 MHz,
CDCl3): 151.7 (C), 146.3 (C), 145.5 (C), 131.7 (C), 131.6 (CH), 130.5
(CH), 128.2 (C), 127.9 (CH), 127.6 (CH), 127.4 (CH), 127.2 (C), 126.9
ꢅ
(C), 126.7 (CH), 126.5 (CH), 125.7 (C), 123.9 (C), 95.6 (C CCH2), 89.7
ꢅ
ꢅ
ꢅ
(C CCH2), 81.1 (C CCH2), 79.8 (C CCH2), 49.7 (CH), 41.1 (CH), 38.0
ꢅ
ꢅ
(C), 37.9 (CH2), 31.1 (CH2), 28.1 (C CCH2), 27.6 (C CCH2), 25.7 (CH3),
ꢅ
ꢅ
21.2 (CH3), 19.4 (C CCH2CH2), 19.1 ppm (C CCH2CH2); HRMS (EI):
m/z: calcd for C34H31N: 453.24565; found: 453.2455; UV/Vis (CH2Cl2): l
(e)=292 (3200), 350 (190), 364 (180).
[PdACTHUNTGRENNG(U 5d)2]ACHTUNTGERN(NUGN SbF6)2 (28 mg, 85%) as a dark red solid. Single crystals suitable
1-Phenyl-2-[{(4S,6R)-3,4,5,6-tetrahydro-5,5-dimethyl-4,6-methano}benzo-
for X-ray analysis were grown by slow diffusion of pentane vapor into a
CH2Cl2 solution. 31P NMR (81 MHz, CD2Cl2): d=81.6 ppm; HRMS (ES):
m/z: calcd for C74H56N235ClP2106Pd ([C++,Clꢀ]+): 1175.26420; found:
ACHTUNGTRENNUNG[1,2-k]phenanthrid-6-yl]-5-phenyl-3,4-butanophosphole (5e) (two diaste-
reomers): nBuLi 1.5m in hexanes (0.44 mL, 0.66 mmol) was added drop-
wise to a THF solution (15 mL) of diyne 4e (126 mg, 0.28 mmol) and
[Cp2ZrCl2] (81 mg, 0.28 mmol) under argon, at ꢀ788C. The reaction mix-
ture was warmed to room temperature and stirred overnight. Freshly dis-
tilled PhPBr2 (63 mL, 0.3 mmol) was added to this solution at ꢀ508C. The
reaction mixture was allowed to warm to room temperature and was
stirred for 30 h. The mixture was then filtered over basic alumina (THF
as eluent) under inert atmosphere and the volatile substances were re-
moved in vacuo. Final purification by column chromatography over silica
1175.2710; UV/Vis (CH2Cl2):
(15400), 500 nm (3400).
l (e)=284 (98900), 313 (57500), 402
(CH3CN)4+,PF6ꢀ) (10 mg, 0.28 mmol) was added to a
[CuACHTNURTGENN(UG 5b)2]PF6: [CuCAHUTNGTRENNGUN
solution of phosphole 5b (30 mg, 0.56 mmol) in dry CH2Cl2 (5 mL) under
argon. After stirring at room temperature under argon for 1 h, the sol-
vent was stripped off and the precipitate was washed with dry Et2O,
yielding [CuACHTNUGTRNEUNG(5b)2]PF6 (49 mg, 77%) as a yellow solid. Single crystals
6002
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
were grown by slow diffusion of pentane vapor into a CH2Cl2 solution.
31P NMR (81 MHz, CDCl3): d=5–6 ppm; UV/Vis (CH2Cl2): l (e)=292
(44400), 342 (20400), 384 (16600), 403 nm (17400).
1491–1509; c) J. Autschbach, T. Ziegler, S. J. A. van Gisbergen, E. J.
Baerends, J. Chem. Phys. 2002, 116, 6930–6940.
[3] a) E. Botek, J.-M. Andrꢀ, B. Champagne, T. Verbiest, A. Persoons,
J. Chem. Phys. 2005, 122, 234713; b) W. N. Herman, Y. Kim, W. L.
Cao, J. Goldhar, C. H. Lee, M. M. Green, V. Jain, M. J. Lee, J. Mac-
romol. Sci. Part A 2003, 40, 1369–1982; c) N. Takenaka, R. S. Sar-
angthem, B. Captain, Angew. Chem. 2008, 120, 9854–9856; Angew.
Chem. Int. Ed. 2008, 47, 9708–9710; d) D. J. Weix, S. D. Dreher, T. J.
Katz, J. Am. Chem. Soc. 2000, 122, 10027–10032; e) M. T. Reetz,
E. W. Beuttenmꢅller, R. Goddard, Tetrahedron Lett. 1997, 38, 3211–
3214; f) M. T. Reetz, S. Sostmann, J. Organomet. Chem. 2000, 603,
105–109; g) M. Samal, J. Misek, I. G. Stara, I. Stary, Collect. Czech.
Chem. Commun. 2009, 74, 1151–1159; h) A. Bossi, E. Licandro, S.
Maiorana, C. Rigamonti, S. Righetto, G. R. Stephenson, M. Spasso-
va, E. Botek, B. Champagne, J. Phys. Chem. C 2008, 112, 7900–
7907.
[4] a) K. Schmidt, S. Brovelli, V. Coropceanu, D. Beljonne, J. Cornil, C.
Bazzini, T. Caronna, R. Tubino, F. Meinardi, Z. Shuai, J. L. Brꢀdas,
J. Phys. Chem. A 2007, 111, 10490–10499; b) K. Schmidt, S. Brovelli,
V. Coropceanu, J. L. Brꢀdas, C. Bazzini, T. Caronna, R. Tubino, F.
Meinardi, J. Phys. Chem. A 2006, 110, 11018–11024; c) M. Sapir, E.
Vander Donckt, Chem. Phys. Lett. 1975, 36, 108–110; d) J. B. Birks,
D. J. S. Birch, E. Cordemans, E. Vander Donckt, Chem. Phys. Lett.
1976, 43, 33–36; e) O. E. Weigang, Jr. , J. A. Turner, P. A. Trouard, J.
Chem. Phys. 1966, 45, 1126–1134; f) Y. H. Tian, G. Park, M. Ker-
tesz, Chem. Mater. 2008, 20, 3266–3277; g) M. T. Reetz, S. Sost-
mann, Tetrahedron 2001, 57, 2515–2520.
[CuACHTUNGTRENNUNG(5d)2]Cl: CuCl (2 mg, 20 mmol) was added to a solution of phosphole
5d (20 mg, 39 mmol) in dry CH2Cl2 (2 mL) under argon. After stirring at
room temperature under argon for 1 h, the solvent was stripped off and
the precipitate was washed with dry Et2O, yielding [CuACTHNUTRGNEUNG
(5d)2]Clꢀ (18 mg,
82%) as a bright yellow solid. Single crystals suitable for X-ray analysis
were grown by slow diffusion of pentane vapor into a CH2Cl2 solution.
31P NMR (81 MHz, CDCl3): d=5–6 ppm; UV/Vis (CH2Cl2): l (e)=278
(71400), 322 (45200), 368 (29900), 403 nm (24800).
X-ray crystallography: Single crystals of (ꢀ)-4c, 5a, and [Pd
ACHTUNGTERNUN(NG 5c)Cl2] suit-
able for X-ray crystal analysis were obtained at room temperature by
slow diffusion of pentane vapor into dichloromethane solutions. Single-
crystal data collection was performed at 100 K with an APEX II Bruker-
AXS (Centre de Diffractomꢀtrie, Universitꢀ de Rennes 1, France) with
MoKa radiation (l=0.71069 ꢂ). Reflections were indexed, corrected for
Lorentzian and polarization effects, and integrated by the DENZO pro-
gram of the KappaCCD software package. The data merging process was
performed by using the SCALEPACK program.[17] Structure determina-
tions were performed by direct methods with the solving program
SIR97,[18] which revealed all the non-hydrogen atoms. SHELXL pro-
gram[19] was used to refine the structures by full-matrix least-squares
methods based on F2. All non-hydrogen atoms were refined with aniso-
tropic displacement parameters. Hydrogen atoms were included in ideal-
ized positions and refined with isotropic displacement parameters.
In the crystal lattice of [PdACHTUNTRGNEUNG(5c)Cl2], a dichloromethane molecule was
[5] a) C. Nuckolls, R. F. Shao, W. G. Jang, N. A. Clark, D. M. Walba,
T. J. Katz, Chem. Mater. 2002, 14, 773–776; b) A. J. Lovinger, C.
Nuckolls, T. J. Katz, J. Am. Chem. Soc. 1998, 120, 264–268.
found in addition to the neutral coordination complex. This solvent mole-
cule was found to be disordered and modeled over two positions. Hydro-
gen atoms were been introduced in this model. As a result, rather high
anisotropic displacement parameters for some of its atoms were ob-
tained. In addition, the cyclohexyl ring of the phosphole ligand in [Pd-
[6] a) M. Miyasaka, A. Rajca, M. Pink, S. Rajca, J. Am. Chem. Soc.
2005, 127, 13806–13807; b) J. Misek, F. Teply, I. G. Stara, M. Tichy,
D. Saman, I. Cisarova, P. Vojtisek, I. Stary, Angew. Chem. 2008, 120,
3232–3235; Angew. Chem. Int. Ed. 2008, 47, 3188–3191; c) D. C.
Harrowven, I. L. Guy, L. Nanson, Angew. Chem. 2006, 118, 2300–
2303; Angew. Chem. Int. Ed. 2006, 45, 2242–2245; d) E. Murguly, R.
McDonald, N. R. Branda, Org. Lett. 2000, 2, 3169–3172; e) J. Ichika-
wa, M. Yokota, T. Kudo, S. Umezaki, Angew. Chem. 2008, 120,
4948–4951; Angew. Chem. Int. Ed. 2008, 47, 4870–4873; f) T. Ras-
musson, L. J. P. Martyn, G. Chen, A. Lough, M. Oh, A. K. Yudin,
Angew. Chem. 2008, 120, 7117–7120; Angew. Chem. Int. Ed. 2008,
47, 7009–7012; g) B. Laleu, P. Mobian, C. Herse, B. W. Laursen, G.
Hopfgartner, G. Bernardinelli, J. Lacour, Angew. Chem. 2005, 117,
1913–1917; Angew. Chem. Int. Ed. 2005, 44, 1879–1883.
ACHTUNGTRENNUNG(5c)Cl2] was found to be disordered over two positions, whose relative
occupancies were refined. As a consequence, final agreement factor (R)
has modest values (Table 2). Nevertheless, anisotropic displacement pa-
rameters associated to the atoms of the neutral coordination complexes
that are not involved in the disorder are satisfactory. Atomic scattering
factors for all atoms were taken from International Tables for X-ray
Crystallography.[20] CCDC-736269 ((ꢀ)-4c), CCDC-736270 (5a), and
CCDC-736271 ([PdACHTUNGTRENNUNG(5c)Cl2]) contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[7] a) J.-M. Lehn, Supramolecular Chemistry: Concepts and Perspectives,
VCH, Weinheim, 1995; b) J. P. Sauvage, Transition Metals in Supra-
molecular Chemistry, Wiley, New York, 1994.
[8] a) K. Sꢀnꢀchal-David, A. Hemeryck, N. Tancrez, L. Toupet, J. A. G.
Williams, I. Ledoux, J. Zyss, A. Boucekkine, J. P. Guegan, H. Le Bo-
zec, O. Maury, J. Am. Chem. Soc. 2006, 128, 12243–12255; b) T.
Le Bouder, O. Maury, A. Bondon, K. Costuas, E. Amouyal, I.
Ledoux, J. Zyss, H. Le Bozec, J. Am. Chem. Soc. 2003, 125, 12284–
12299; c) O. Maury, H. Le Bozec, Acc. Chem. Res. 2005, 38, 691–
704.
[9] a) C. M. Hartshorn, P. J. Steel, Inorg. Chem. 1996, 35, 6902–6903;
b) M. Munakata, G. L. Ning, Y. Suenaga, T. Kurodo-Sowa, M. Mae-
kawa, T. Ohta, Angew. Chem. 2000, 112, 4729–4731; Angew. Chem.
Int. Ed. 2000, 39, 4555–4557; c) M. Yoshizawa, J. Nakagawa, K. Ku-
mazawa, M. Nagao, M. Kawano, T. Ozeki, M. Fujita, Angew. Chem.
2005, 117, 1844–1847; Angew. Chem. Int. Ed. 2005, 44, 1810–1813;
d) Q. Chu, D. C. Swenson, L. R. MacGillivray Angew. Chem. 2005,
117, 3635–3638; Angew. Chem. Int. Ed. 2005, 44, 3569–3572;
Angew. Chem. Int. Ed. 2005, 44, 3569–3572.
Acknowledgements
Dr. Karine Costuas is warmly thanked for fruitful discussions. This work
is supported by the Ministꢄre de l’Education Nationale, de la Recherche
et de la Technologie, the Centre National de la Recherche Scientifique
(CNRS), the Rꢀgion Bretagne, the Agence Nationale de la Recherche
(ANR) (project PHOSHELIX-137104), and the National Science Foun-
dation (CHE 0447321). J.A. thanks Professor Fillip Furche for helpful in-
formation regarding one of the Turbomole data files.
[1] a) T. J. Katz, Angew. Chem. 2000, 112, 1997–1999; Angew. Chem.
Int. Ed. 2000, 39, 1921–1923; b) A. Urbano, Angew. Chem. 2003,
115, 4116–4119; Angew. Chem. Int. Ed. 2003, 42, 3986–3989;
c) R. H. Martin, Angew. Chem. 1974, 86, 727–738; Angew. Chem.
Int. Ed. Engl. 1974, 13, 649–660; d) A. Rajca, M. Miyasaka in Func-
tional Organic Materials (Eds.: T. J. J. Mꢅller, U. H. F. Bunz), Wiley-
VCH, Weinheim, 2007, pp. 543–577.
[2] a) F. Furche, R. Ahlrichs, C. Wachsmann, E. Weber, A. Sobanski, F.
Vogtle, S. Grimme, J. Am. Chem. Soc. 2000, 122, 1717–1724; b) S.
Grimme, J. Harren, A. Sobanski, F. Vogtle, Eur. J. Org. Chem. 1998,
[10] a) G. Rapenne, C. Dietrich-Buchecker, J. P. Sauvage, J. Am. Chem.
Soc. 1996, 118, 10932–10933; b) C. J. Meskers, H. P. J. M. Dekkers,
G. Rapenne, J. P. Sauvage, Chem. Eur. J. 2000, 6, 2129–2134.
[11] a) A. Sudhakar, T. J. Katz, J. Am. Chem. Soc. 1986, 108, 179–181;
b) A. Sudhakar, T. J. Katz, B. W. Yang, J. Am. Chem. Soc. 1986, 108,
2790–2791; c) F. Pammer, Y. Sun, M. Pagels, D. Weismann, H. Sitz-
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6003

R. Rꢀau, J. Autschbach, J. Crassous et al.
mann, W. R. Thiel, Angew. Chem. 2008, 120, 3315–3318; Angew.
Chem. Int. Ed. 2008, 47, 3271–3274; d) T. J. Katz, A. Sudhakar,
M. F. Teasley, A. M. Gilbert, W. E. Geiger, M. P. Robben, M.
Wuensch, M. D. Ward, J. Am. Chem. Soc. 1993, 115, 3182–3198;
e) A. M. Gilbert, T. J. Katz, W. E. Geiger, M. P. Robben, A. L.
Rheingold, J. Am. Chem. Soc. 1993, 115, 3199–3211; f) Z. Gomez-
Sandoval, E. Pena, C. Fonseca Guerra, F. M. Bickelhaupt, M. A.
Mendez-Rojas, G. Merino, Inorg. Chem. 2009, 48, 2714–2716; g) R.
El-Abed, F. Aloui, J.-P. Genet, B. Ben-Hassine, A. Marinetti, J. Or-
ganomet. Chem. 2007, 692, 1156–1160; h) M. H. Garcia, P. Florindo,
M. F. M. Piedade, M. T. Duarte, Polyhedron 2009, 28, 621–629.
[12] a) S. Graule, M. Rudolph, N. Vanthuyne, J. Autschbach, C. Roussel,
J. Crassous, R. Rꢀau, J. Am. Chem. Soc. 2009, 131, 3183–3185;
b) W. Shen, S. Graule, J. Crassous, C. Lescop, H. Gornitzka, R.
Rꢀau, Chem. Commun. 2008, 850–852; c) L. Norel, M. Rudolph, N.
Vanthuyne, J. A. G. Williams, C. Lescop, C. Roussel, J. Auchtsbach,
J. Crassous, R. Rꢀau, Angew. Chem. Int. Ed. 2010, 49, 99–102.
[13] a) U. Knof, A. von Zelewsky, Angew. Chem. 1999, 111, 312–333;
Angew. Chem. Int. Ed. 1999, 38, 302–322; b) H. Brunner, Angew.
Chem. 1999, 111, 1248; Angew. Chem. Int. Ed. 1999, 38, 1194; c) J.
Crassous, Chem. Soc. Rev. 2009, 38, 830–845; d) A. Amouri, M.
Gruselle, Chirality in Transition Metal Chemistry: Molecules, Supra-
molecular Assemblies and Materials, Wiley-VCH, Weinheim, 2009;
e) D. Amabilino, Chirality at the Nanoscale: Nanoparticles, Surfaces,
Materials and More, Wiley-VCH, Weinheim, 2009.
[14] a) P. Braunstein, J. Organomet. Chem. 2004, 689, 3953–3967;
b) R. G. Pearson, Inorg. Chem. 1973, 12, 712–713; c) O. Mamula, A.
von Zelewsky, J. Chem. Soc. Dalton Trans. 2001, 219–231; d) J. N.
Harvey, K. M. Heslop, A. G. Orpen, P. G. Pringle, Chem. Commun.
2003, 278–279.
[15] a) R. Rꢀau, P. Dyer, Comprehensive Heterocyclic Chemistry, Elsevi-
er, Oxford, 2008; b) F. Mathey, Phosphorus-carbon Heterocyclic
Chemistry: The Rise of a New Domain, Elsevier, Oxford, 2001; c) F.
Mathey, Angew. Chem. 2003, 115, 1616–1643; Angew. Chem. Int.
Ed. 2003, 42, 1578–1604; d) T. Baumgartner, R. Rꢀau, Chem. Rev.
2006, 106, 4681–4727.
2003, 9, 3785–3795; f) F. Leca, C. Lescop, E. Rodriguez, K. Costuas,
J.-F. Halet, R. Rꢀau, R. Angew. Chem. 2005, 117, 4436–4439;
Angew. Chem. Int. Ed. 2005, 44, 4362–4365; Angew. Chem. Int. Ed.
2005, 44, 4362–4365; g) B. Nohra, S. Graule, C. Lescop, R. Rꢀau, J.
Am. Chem. Soc. 2006, 128, 3520.
[21] a) S. Grimme, Chem. Phys. Lett. 2001, 339, 380–388; b) F. Furche,
R. Ahlrichs, C. Wachsmann, E. Weber, A. Sobanski, F. Vçgtle, S.
Grimme, J. Am. Chem. Soc. 2000, 122, 1717–1724.
[22] a) Turbomole, Version 5.7.1, Quantum Chemistry Group, University
of Karlsruhe, Karlsruhe; b) K. Eichkorn, O. Treutler, H. ꢈhm, M.
Hꢇser, R. Ahlrichs, Chem. Phys. Lett. 1995, 242, 652–660; c) K.
Eichkorn, F. Weigend, O. Treutler, R. Ahlrichs, Theor. Chem. Acc.
1997, 97, 119–124.
[23] A. Schꢇfer, H. Horn, R. Alrichs, J. Chem. Phys. 1992, 97, 2571–
2577.
[24] a) P. A. M. Dirac, Proc. R. Soc. London Ser. A 1929, 123, 714–733;
b) J. C. Slater, Phys. Rev. 1951, 81, 385–390; c) S. Vosko, L. Wilk,
M. Nusair, Can. J. Phys. 1980, 58, 1200–1211; d) A. D. Becke, Phys.
Rev. A 1988, 38, 3098–3100; e) J. P. Perdew, Phys. Rev. B 1986, 77,
3865–3868.
[25] A. D. Becke, J. Chem. Phys. 1992, 96–97, 5648.
[26] A. Klamt, G. Schꢅꢅrmann, J. Chem. Soc. Perkin Trans. 2 1993, 799–
805.
[27] NBO 5.0, http://www.chem.wisc.edu/~nbo5/, E. D. Glendening, J. K.
Badenhoop, A. E. Reed, J. E. Carpenter, J. A. Bohmann, C. M. Mo-
rales, F. Weinhold, Theoretical Chemistry Institute, University of
Wisconsin, Madison, 2001.
[28] Gaussian 03, Revision C.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford, CT, 2004.
[29] “Natural Bond Orbital Methods”: F. Weinhold, in Encyclopedia of
Computational Chemistry (Ed.: P. von R. Schleyer), Wiley, New
York, 1998, pp. 1792–1811.
[30] a) J. Autschbach, T. Ziegler, J. Chem. Phys. 2002, 116, 891–896; b) J.
Autschbach, T. Ziegler, S. J. A. van Gisbergen, E. J. Baerends, J.
Chem. Phys. 2002, 116, 6930–6940; c) J. Autschbach, T. Ziegler, S.
Patchkovskii, S. J. A. van Gisbergen, E. J. Baerends, J. Chem. Phys.
2002, 117, 581–592.
[16] M. K. CyraÇski, T. M. Krygowski, A. R. Katritzky, P. von R. Schley-
er, J. Org. Chem. 2002, 67, 1333–1338.
[17] a) H.-C. Su, O. Fadhel, C.-J. Yang, T.-Y. Cho, C. Fave, M. Hissler,
C.-C. Wu, R. Rꢀau, J. Am. Chem. Soc. 2006, 128, 983–995; b) M. Se-
bastian, M. Hissler, C. Fave, J. Rault-Berthelot, C. Odin, R. Rꢀau,
Angew. Chem. 2006, 118, 6298–6301; Angew. Chem. Int. Ed. 2006,
45, 6152–6155; c) V. Lemau de Talancꢀ, M. Hissler, L.-Z. Zhang, ;
T. Kꢆrpꢆti, L. Nyulꢆszi, D. Caras-Quintero, P. Bꢇuerle, R. Rꢀau,
Chem. Commun. 2008, 2200–2202; d) Y. Matano, T. Miyajima, T.
Fukushima, H. Kaji, Y. Kimura, H. Imahori, Chem. Eur. J. 2008, 14,
8102–8115; e) Y. Matano, T. Miyajima, H. Imahori, Y. Kimura, J.
Org. Chem. 2007, 72, 6200–6205; f) Y. Dienes, S. Durben, T. Karpa-
ti, T. Neumann, U. Englert, L. Nyulaszi, T. Baumgartner, Chem.
Eur. J. 2007, 13, 7487–7500; g) A. Fukazawa, M. Hara, T. Okamoto,
E.-C. Son, C. Xu, K. Tamao, S. Yamaguchi, Org. Lett. 2008, 10, 913–
916; h) A. Fukazawa, H. Yamada, S. Yamaguchi, Angew. Chem.
2008, 120, 5664–5667; Angew. Chem. Int. Ed. 2008, 47, 5582–5585;
i) H. T. Ngoc Huy, D. Donnadieu, F. Mathey, A. Muller, K. Colby,
C. J. Bardeen, Organometallics 2008, 27, 5521–5524; j) C. Hay, M.
Hissler, C. Fischmeister, J. Rault-Berthelot, L. Toupet, L. Nyulaszi,
R. Rꢀau, Chem. Eur. J. 2001, 7, 4222–4236.
[18] a) P. J. Fagan, W. A. Nugent, J. C. Calabrese, J. Am. Chem. Soc.
1994, 116, 1880–1889; b) P. J. Fagan, W. A. Nugent, J. Am. Chem.
Soc. 1988, 110, 2310–2312.
[19] L. B. Liu, B. W. Yang, T. J. Katz, M. K. Poindexter, J. Org. Chem.
1991, 56, 3769–3775.
[20] a) J. Crassous, R. Rꢀau, Dalton Trans. 2008, 6865–6876; b) C. Fave,
M. Hissler, K. Sꢀnꢀchal, I. Ledoux, J. Zyss, R. Rꢀau, Chem.
Commun. 2002, 1674–1675; c) M. Sauthier, F. Leca, L. Toupet, R.
Rꢀau, Organometallics 2002, 21, 1591–1602; d) M. Sauthier, B. Le
Guennic, V. Deborde, L. Toupet, J.-F. Halet, R. Rꢀau, Angew.
Chem. 2001, 113, 234; Angew. Chem. Int. Ed. 2001, 40, 228; e) F.
Leca, M. Sauthier, V. Deborde, L. Toupet, R. Rꢀau, Chem. Eur. J.
[31] “Time-dependent Density Functional Response Theory for Mole-
cules”: M. E. Casida in Recent Advances in Density Functional
Methods, Vol. 1, World Scientific, Singapore, 1995, pp. 155–192.
[32] R. Bauernschmitt, R. Ahlrichs, Chem. Phys. Lett. 1996, 256, 454–
464.
[33] A. D. McLachlan, M. A. Ball, Rev. Mod. Phys. 1964, 36, 844–855.
[34] a) S. F. Boys, J. M. Foster, Rev. Mod. Phys. 1960, 32, 300–302; b) C.
Edmiston, K. Ruedenberg, Rev. Mod. Phys. 1963, 35, 457–465;
c) W. von Niessen, J. Chem. Phys. 1972, 56, 4290–4297; d) J. Pipek,
P. G. Mezey, J. Chem. Phys. 1989, 90, 4916–4926.
[35] A. Schꢇfer, C. Huber, R. Alrichs, J. Chem. Phys. 1994, 100, 5829–
5835.
[36] When not specified 5c1 corresponds to (P,SP)-s-trans configuration.
[37] I. J. Tinoco, R. W. Woody, J. Chem. Phys. 1964, 40, 160–165.
6004
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 5976 – 6005

Chiral Coordination Complexes of Phosphole Azahelicenes
FULL PAPER
[38] a) O. Mamula, A. von Zelewsky, J. Chem. Soc. Dalton Trans. 2000,
3, 219–231; b) J. N. Harvey, K. M. Heslop, A. G. Orpen, P. G. Prin-
gle, Chem. Commun. 2003, 278–279.
[42] A. Altomare, M. C. Burla, M. Camalli, G. Cascarano, C. Giacovaz-
zo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R. Spagna, J.
Appl. Cryst. 1999, 32, 115.
[39] J. Autschbach, Chirality 2010, in press.
[43] SHELX97, Program for the refinement of crystal structures, G. M.
Sheldrick, University of Gçttingen, Gçttingen, 1997.
[44] International Tables for X-ray Crystallography, Vol. C, Kluwer, Dor-
drech, 1992.
[40] a) L. D. Quin, J. P. Gratz, T. P. Barket, J. Org. Chem. 1968, 33, 1034–
1041; b) D. A. Lightner, D. T. Hefelfinger, T. W. Powers, G. W.
Frank, K. N. Trueblood, J. Am. Chem. Soc. 1972, 94, 3492–3497.
[41] Z. Otwinowski, W. Minor in Methods in Enzymology (Eds.: C. W.
Carter, Jr., R. M. Sweet), Academic Press, New York, 1997, p. 307.
Received: November 26, 2009
Published online: April 15, 2010
Chem. Eur. J. 2010, 16, 5976 – 6005
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6005