Functionalized phosphorus derivatives of Salpen-like compounds: Synthesis and preliminary complexation studies

Article abstract of DOI:10.1016/j.jorganchem.2005.12.022

A new class of Salpen analogues based on phosphorus derivatives where the classical alkylene backbone has been replaced by a N-P-N linkage is described. Such linkage both affords a very good stability in water and an additional (fifth) potentially complexing site. The classical ortho-OH groups have been also replaced by various ortho-substituents, including diphenylphosphino groups. The synthesis of these compounds is easy and their structure can be varied at will at several levels. Several ways of synthesis can be used to combine the various fragments constituting these Salpen analogues. The structure of one of these fragments, an azide, was determined by X-ray crystallography. A preliminary study of the complexation ability of some of these new ligands was carried out with groups 10 (Ni) and 11 (Au) elements. Depending on the type of substituents and the type of metals used, these compounds can act as mono-, or tetra-dentate ligands.

Full text of DOI:10.1016/j.jorganchem.2005.12.022

Journal of Organometallic Chemistry 691 (2006) 1333–1340
www.elsevier.com/locate/jorganchem
Functionalized phosphorus derivatives of Salpen-like
compounds: Synthesis and preliminary complexation studies
´
Alexandrine Maraval, Germinal Magro, Valerie Maraval, Laure Vendier,
*
*
Anne-Marie Caminade , Jean-Pierre Majoral
Laboratoire de Chimie de Coordination CNRS, 205 route de Narbonne, 31077 Toulouse Cedex 4, France
Received 26 October 2005; received in revised form 6 December 2005; accepted 6 December 2005
Available online 19 January 2006
Abstract
A new class of Salpen analogues based on phosphorus derivatives where the classical alkylene backbone has been replaced by a N–P–
N linkage is described. Such linkage both affords a very good stability in water and an additional (fifth) potentially complexing site. The
classical ortho-OH groups have been also replaced by various ortho-substituents, including diphenylphosphino groups. The synthesis of
these compounds is easy and their structure can be varied at will at several levels. Several ways of synthesis can be used to combine the
various fragments constituting these Salpen analogues. The structure of one of these fragments, an azide, was determined by X-ray crys-
tallography. A preliminary study of the complexation ability of some of these new ligands was carried out with groups 10 (Ni) and 11
(Au) elements. Depending on the type of substituents and the type of metals used, these compounds can act as mono-, or tetra-dentate
ligands.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Salpen; Phosphorus; Hydrazone; Phosphorhydrazide; Complexation
1. Introduction
ides of phosphorus constitute a variety of dihydrazine
derivatives which are easily available and react readily
with benzaldehydes. In particular we [5,6] and others
[7] have shown that the reaction of phosphorus dihydraz-
ides with salicylaldehyde affords the expected Salpen ana-
logues, in which the traditional C–C–C linkage between
both imines is replaced by a N–P–N linkage. Beside
the complexation ability of the OH and N@CH group,
well-known for Salpen derivatives, the presence of phos-
phorus should bring new complexation properties, using
either the lone pair of phosphorus for phosphines deriv-
atives, or the lone pairs of oxygen or sulfur for phos-
phoryl and thiophosphoryl derivatives. We have already
shown that the complexation ability of a P@S group is
enhanced when it is included in a P@N–P@S linkage
[8], therefore we introduced such linkage in all the com-
pounds we prepared in this paper. Furthermore, salicyl-
aldehyde can be easily replaced by other ortho-
functionalized benzaldehydes (Fig. 1). In this paper, we
The extraordinary simplicity of the reaction of diam-
ines with salicylaldehyde gave rise to a tremendous
amount of work dedicated to the synthesis and study
of complexation properties of the corresponding diimine
derivatives, the so-called Salen compounds [1]. Indeed,
the presence of two covalent and two coordinating sites
in these ligands lead in most cases to planar complexes
of transition metals, in which two free axial sites on
the metal are available, which are particularly useful
for catalytic experiments [2–4]. In contrast to the diimine
derivatives, the analogous dihydrazone derivatives are
relatively rare, despite the known better stability toward
hydrolysis of hydrazones compared to imines. Dihydraz-
*
Corresponding authors. Tel.: +33 5 61 33 31 25; fax: +33 5 61 55 30 03.
E-mail addresses: caminade@lcc-toulouse.fr (A.-M. Caminade),
majoral@lcc-toulouse.fr (J.-P. Majoral).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.12.022

1334
A. Maraval et al. / Journal of Organometallic Chemistry 691 (2006) 1333–1340
benzaldehyde 3c, diphenylthiophosphino benzaldehyde
3d, and helicin 3e (Scheme 1). In all cases, the condensation
induces a shielding of the signal corresponding to the P@S
group in 31P NMR from 71.0 to ca. 56 ppm, as well as a
shielding in 13C NMR of the doublet corresponding to
the Me groups from 39.1 ppm to ca. 31 ppm. In all cases
excepted for 4e, the condensation affords a single isomer
for the C@N bonds, presumably the E isomer in view of
the X-ray structure of compound 5d (see later). However,
the 13C NMR spectrum of compound 4e displays two sig-
nals instead of one for several atoms, approximately in a
1/4 ratio. This splitting is easily detectable on the NMe
groups and on several atoms of the helicin function, and
the signal of the CH@N group is broadened. These data
are indicative of the presence of two isomers for the CH@N
groups (E and Z). We have already observed the same phe-
nomenon for helicin linked to the surface of dendrimers [6].
Due to the presence of two b-D-glucoside groups, com-
pound 4e is fairly soluble in water/THF mixtures (v/v); this
allowed us to test the stability of this compound. No trace
of decomposition is observed even after several weeks in
this mixture, illustrating the very good stability toward
hydrolysis induced by the phosphorhydrazone linkages
compared to imines in classical Salpen [9,13].
Fig. 1. Potential complexation sites in Salpen and phosphorus Salpen
analogues (P-Salpen).
report the synthesis of a number of new phosphorus
derivatives of Salpen analogues (P-Salpen), having N–
PR(S)–N linkages ðR ¼ R03P@NÞ, and preliminary exper-
iments concerning the complexation properties of some
of these compounds.
2. Results and discussion
The synthesis of the phosphorus Salpen analogues
necessitates first the synthesis of phosphodihydrazides
RP(S)(NMeNH2)2 [9], to be used instead of 1,3-diamines.
The P@N–P@S linkage is easily obtained using the Stau-
dinger reaction [10] between a phosphine and a thiophos-
phoryl azide, thus the R substituent must be N3. The
reaction of triphenylphosphine with the azide 1 [11] affords
slowly but cleanly the dihydrazide 2 (Scheme 1). The slow-
ness of the reaction (5 days at room temperature) can be
ascribed to the absence of strong electron-withdrawing
groups on the azide, the PS group being not sufficient
[12]. Compound 2 is characterized in 31P NMR by a set
of two doublets for the P@N–P@S linkage at d = 12.4
This method of synthesis is very simple, adaptable, and
each component can be varied. For instance, both steps
can be inverted, as shown by the condensation of 1 with
3d leading to the azide 5d (Scheme 2). In view of the small
number of compounds possessing a k4,r5-phosphorus azide
linkage characterized by X-ray diffraction (only five exam-
ples) [14–18], it appeared interesting to obtain single crystals
of 5d. The ORTEP drawing of 5d is shown in Fig. 2. The X-
ray structure determination (Table 1) gives a N6–N7 bond
2
˚
and 71.0 ppm, respectively, with JPP = 16.7 Hz.
length value of 1.121(5) A (Table 2), one of the shortest
The next step for obtaining the Salpen analogues 4a–e
consists in the condensation of both NH2 groups of 2 with
benzaldehyde 3a and various ortho-functionalized benzal-
dehydes: hydroxybenzaldehyde 3b, diphenylphosphino-
reported up to now for a k4,r5-phosphorus azide linkage,
˚
close to the value of 1.098 A known for a triple bond [19].
˚
The N5–N6 bond length (1.228(5) A) is significantly shorter
than a single bond, for instance when compared with
Scheme 1.
Scheme 2.

A. Maraval et al. / Journal of Organometallic Chemistry 691 (2006) 1333–1340
1335
Table 2
Selected bond distances (A) and bond angles (ꢁ) of 5d (esd in parentheses)
˚
P1–N1
P1–N2
P1–N5
P1–S1
P2–S2
P3–S3
N1–N4
N2–N3
N5–N6
N6–N7
1.658(3)
1.665(3)
1.685(4)
1.9081(16)
1.9533(15)
1.9545(15)
1.380(5)
1.392(4)
1.228(5)
1.121(5)
N1–P1–N2
N1–P1–N5
N2–P1–N5
N1–P1–S1
N2–P1–S1
N5–P1–S1
N7–N6–N5
108.64(16)
100.7(2)
102.94(7)
113.62(14)
113.54(14)
116.17(13)
174.3(5)
Obviously, the synthesis of P-Salpen is not limited to tri-
phenylphosphine, as shown for instance by the use of diph-
enylphosphinoferrocene (Scheme 3). The Staudinger
reaction of the azide 1 is completed in two days, more rap-
idly than with Ph3P, and affords cleanly the expected com-
pound 6, characterized by a set of two doublets at 71.6
(P@S) and 15.7 (P@N) ppm (2JPP = 18.8 Hz). On the other
hand, the condensation reaction with helicin is slow and
needs 7 days to go to completion. Here again, two isomers
are observed for the CH@N bonds in a 1/2 ratio, as shown
by 13C NMR.
Having in hand several new Salpen analogues, it was
interesting to test the complexation properties of some of
them. As shown in Fig. 1, five sites of complexation are
potentially available in compounds 4. Since a previous
experiment with a different but related compound [5], and
nickel acetate afforded a complex in which nickel was
linked to four sites (2N and 2O), we tried the same exper-
iment with compound 4b. Surprisingly, no reaction
occurred with nickel acetate, presumably because of the
steric hindrance induced by the PPh3 group. In order to
avoid this problem, we tried in this case also to invert the
order of the reactions, i.e., to use the preformed Ni(Sale-
n)2 Æ 2H2O complex [20] in a condensation reaction with
2. After 5 days at 60 ꢁC, the dark green paramagnetic solu-
tion of 8 and 2 becomes a yellow diamagnetic solution of
9b (Scheme 4). The color change is very characteristic of
a modification of the geometry around nickel, from octahe-
dral in 8 to square planar in 9b. The reaction is also
Fig. 2. ORTEP diagram for 5d.
Table 1
Crystal data and structure refinement for 5d Æ CH2Cl2
Asymmetric unit
Formula weight
Wavelength (A)
C41H38Cl2N7P3S3
888.77
0.71073
Monoclinic
P1c1
˚
Crystal system
Space group
˚
a (A)
12.9358(8)
10.3014(7)
16.2095(12)
96.070(6)
2147.9(3)
2
1.374
˚
b (A)
˚
C (A)
b (deg)
3
˚
V (A )
Z
qcalc (g cmꢀ3
)
Crystal size (mm)
Absorption coefficient (mmꢀ1
F(000)
0.37752 · 0.16324 · 0.0193
0.448
920
)
h Range for data collection (ꢁ)
3.21–26.37
Limiting indices
ꢀ16 6 h 6 15,
ꢀ12 6 k 6 12, ꢀ12 6 l 6 20
15157/5956 [Rint = 0.0411]
99.8
Empirical (DIFABS)
0.988 and 0.856
Full-matrix least-squares on F2
5956/2/507
Reflections collected/unique
Completeness to h = 26.37 ꢁ (%)
Absorption correction
Max. and min. transmission
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
0.934
Final R indices [I > 2r(I)]
R indices (all data)
Absolute structure parameter
R1 = 0.0414, wR2 = 0.0943
R1 = 0.0546, wR2 = 0.1011
ꢀ0.01(7)
ꢀ3
˚
Largest diff. peak and hole (e A
)
0.372 and ꢀ0.461
˚
˚
N2–N3 (1.392(4) A) or N1–N4 (1.380(5) A). As expected,
both CH@N bonds have the E configuration. Compound
5d is then used in a Staudinger reaction with PPh3, to afford
again compound 4d (Scheme 2).
Scheme 3.

1336
A. Maraval et al. / Journal of Organometallic Chemistry 691 (2006) 1333–1340
replaces the traditional C–C–C linkage. The presence of
hydrazone groups induces a high stability of these deriva-
tives toward hydrolysis, in contrast to classical Salpen. In
view of the large number of ortho-functionalized benzalde-
hydes available, these compounds can be easily tailored to
afford precisely the compound desired for specific complex-
ations. We have shown that these P-Salpen can act as
mono-, or tetra-dentate ligands. Preliminary tests have
shown that metals pertaining to several columns and rows
of the periodic table can be complexed. Thus, this new
family of ligands should open the way to the obtaining of
a variety of complexes, potentially usable as catalysts.
Scheme 4.
4. Experimental
4.1. General
All manipulations were carried out with standard high
vacuum and dry-argon techniques. 1H, 13C, and 31P
NMR spectra were recorded with Bruker AC 200, AC
250, DPX 300 or AMX 400 spectrometers. References for
NMR chemical shifts are 85% H3PO4 for 31P NMR, SiMe4
for 1H and 13C NMR. The attribution of 13C NMR signals
has been done using Jmod experiments. The numbering
used for 13C NMR is Ci,o,m,p for the PPh3 part (ipso, ortho,
Scheme 5.
0
0
0
0
meta, para to P), Ci ,o ,m ,p for the PPh2 part when it exists
(compounds 4c, 4d), C1 for the substituted C6H4 carbon
linked to the hydrazone, C2 for the substituted C6H4 in
ortho-position, then C3,4,5,6 for the others C6H4 carbons,
characterized in 31P NMR by the shielding of the signal
corresponding to the P@S group from 71.0 ppm in 2 to
56.5 ppm in 9b. The later value is close but slightly different
from that of the corresponding free ligand 4b (56.9 ppm),
C1g for the glucoside carbon linked to two O, then C2g;3;4;5
.
but the coupling constant differs significantly (2JPP
=
Mass spectrometry (FAB) was carried out with Finnogan-
mat TSQ 700. Solvents were dried and distilled prior to use
(THF and ether over sodium/benzophenone, pentane and
CH2Cl2 over phosphorus pentoxide) and degazed when
phosphines are used. Compounds 1 [11], 6 [20] and
AuCl(tht) [21] were prepared as described previously.
Caution. In our hands, the N3 derivatives appear stable,
however, many azides are explosives, thus maximum care
must be taken.
28.2 Hz for 4b and 19.8 Hz for 9b). Beside multinucleus
NMR, complex 9b is also characterized by FAB mass spec-
trometry, which confirms in particular the absence of
dimers; only the expected 1[2]/1[8] condensation occurred.
The 31P NMR data and the color of the solution indi-
cate that the P@S group is not involved in the complex-
ation of nickel in 9b, only the O and N atoms are used.
In order to check the ability of the P@S group of com-
pounds 4 for complexation, we used compound 4e. The
P@S group of this compound is the sole site usable for
complexation, since the phenol group is protected by a
b-glycoside group, which should sterically hinder the
nitrogen atoms. The reaction of AuCl(tht) (tht = tetrahy-
drothiophene) with 4e occurs readily on the P@S group
(Scheme 5). The complexation induces as expected [8] a
shielding of the signal corresponding to P@S in 31P
NMR from 57.0 for 4e to 49.2 for 10e. A deshielding
of the signal corresponding to the Ph3P@N group from
11.2 for 4e to 18.3 for 10e is also observed. Both phe-
nomenon are indicative of a partial electronic transfer
along the P@N–P@S linkage.
4.2. Ph3P@NP(S)[NMe-NH2]2 (2)
Powdered triphenylphosphine (0.166 g, 0.633 mmol)
and azide 1 (0.123 g, 0.633 mmol) were dissolved in dichlo-
romethane (3 mL) and stirred for 5 days at room tempera-
ture. Then, the solvent was removed under vacuum, and
the resulting powder was washed three times with pentane,
to afford compound 2 as a white powder in 98% yield.
2
31P {1H} NMR (CH2Cl2): d = 12.4 (d, JPP = 16.7 Hz,
2
1
P@N), 71.0 (d, JPP = 16.7 Hz, P@S). H NMR (CDCl3):
3
d = 2.71 (d, JHP = 11.9 Hz, 6H, CH3), 3.52 (br s, 4H,
3
NH2), 7.43–7.57 (m, 9H, H-Cm, H-Cp), 7.76 (dd, JHH
=
7.2 Hz, JHP = 12.9 Hz, 6H, H-Co). 13C {1H} NMR
3
2
3. Conclusion
(CDCl3): d = 39.1 (d, JCP = 7.3 Hz, CH3), 128.4 (d,
3JCP = 13.1 Hz, Cm), 129.7 (dd, 1JCP = 103.5 Hz,
4
We have described the easy synthesis of new Salpen ana-
3JCP = 2.7 Hz, Ci), 132.2 (d, JCP = 2.9 Hz, Cp), 132.6 (d,
logues, in which a N–PR(S)–N linkage ðR ¼ R03P@ NÞ
2JCP = 10.3 Hz, Co). Anal. Calc. for C20H25N5P2S (429.5):

A. Maraval et al. / Journal of Organometallic Chemistry 691 (2006) 1333–1340
1337
C, 55.94; H, 5.87; N, 16.31. Found: C, 55.81; H, 5.83; N,
16.19%.
romethane (5 mL) at room temperature. The resulting solu-
tion was stirred overnight, then evaporated to dryness. Then,
the resulting powder was washed three times with pentane.
Compound 4b was obtained as a white powder in 98% yield.
4.3. o-Ph2P(S)(C6H4-CHO) (3d)
2
31P {1H} NMR (CH2Cl2): d = 15.2 (d, JPP = 28.4 Hz,
2
P@N), 56.9 (d, JPP = 28.4 Hz, P@S). 31P {1H} (CDCl3):
Powdered S8 (0.400 g, 1.56 mmol) was added to a solu-
tion of (2-formyl-phenyl)(diphenyl)phosphine 3c (0.450 g,
1.55 mmol) in THF (20 mL). The resulting mixture was
stirred for 5 h at room temperature, then centrifuged.
The resulting solution was evaporated to dryness to afford
oil, which was purified by column chromatography on sil-
ica (eluent: Et2O/pentane 3/7). Compound 3d was obtained
as pale yellow oil in 90% yield.
d = 16.2 (d, 2JPP = 28.2 Hz, P@N), 56.9 (d, 2JPP = 28.2 Hz,
1
3
P@S). H NMR (CDCl3): d = 3.20 (d, JHP = 8.2 Hz, 6H,
CH3), 6.78–7.10 (m, 8H, H-C3, H-C4, H-C5, H-C6), 7.35–
7.82 (m, 17H, HC@N, H-Ar), 11.63 (s, 2H, OH). 13C
2
{1H} NMR (CDCl3): d = 31.4 (d, JCP = 8.4 Hz, CH3),
116.4 (s, C3), 118.6 (s, C5), 119.4 (s, C1), 128.4 (d,
3JCP = 13.1 Hz, Cm), 128.6 (dd, 1JCP = 106.0 Hz,
3JCP = 4.3 Hz, Ci), 129.0 (s, C4), 129.3 (s, C6), 132.3 (d,
31P {1H} NMR (CDCl3): d = 40.4 (s, P(S)Ph2). 1H
2
4JCP = 2.7 Hz, Cp), 132.7 (d, JCP = 11.0 Hz, Co), 138.3
3
3
NMR (CDCl3): d = 7.01 (ddd, JHH = 7.6 Hz, JHP
=
3
(d, JCP = 13.8 Hz, HC@N), 157.2 (s, C2). Anal. Calc. for
14.6 Hz, JHH = 1.0 Hz, 1H, H-C3), 7.36–7.55 (m, 6H, H-
4
C4, H-C5, H-C0 ), 7.59 (br t, JHH = 7.5 Hz, 2H, H-C0 ),
3
C34H33N5O2P2S (637.7): C, 64.04; H, 5.22; N, 11.98.
Found: C, 63.91; H, 5.15; N, 11.86%.
m
p
3
3
4
7.80 (ddd, JHH = 7.7 Hz, JHP = 13.6 Hz, JHH = 1.4 Hz,
4H, H-C0o), 7.95 (ddd, JHH = 7.5 Hz, JHP = 4.2 Hz,
4JHH = 1.3 Hz, 1H, H-C6), 10.73 (s, 1H, CHO). 13C {1H}
NMR (CDCl3): d = 125.4 (s, C5), 128.7 (d, 3JCP = 12.0 Hz,
3
4
4.6. Ph3P@NP(S)[NMe–N@CH(C6H4)-o-PPh2]2 (4c)
C0 ), 129.6 (d, 3JCP = 8.8 Hz, C4 or C6), 130.9 (s, C1), 131.9
A solution of (2-formyl-phenyl)(diphenyl)phosphine 3c
(0.305 g, 1.051 mmol) in dichloromethane (4 mL) was
added dropwise to a solution of 2 (0.215 g, 0.501 mmol)
in dichloromethane (5 mL) at room temperature and stir-
red overnight. Then the solvent was evaporated and the
resulting powder was washed three times with CH2Cl2/pen-
tane (1/30). Compound 4c was obtained as pale yellow
powder in 95% yield.
m
(d, 4JCP = 2.8 Hz, C0 ), 132.2 (d, 2JCP = 11.5 Hz, C0 ), 132.3
p
o
2
2
(d, JCP = 8.0 Hz, C3), 132.7 (d, JCP = 9.1 Hz, C4 or C6),
136.5 (d, JCP = 92.9 Hz, C2), 136.7 (d, JCP = 84.3 Hz,
1
1
C0). Anal. Calc. for C19H15OPS (322.4): C, 70.79; H,
i
4.69. Found: C, 71.15; H, 4.77%.
31P {1H} NMR (CH2Cl2): d = ꢀ14.4 (s, PPh2), 11.3 (d,
4.4. Ph3P@NP(S)[NMe–N@CH(C6H5)]2 (4a)
2
1
2JPP = 27.5 Hz, P@N), 56.9 (d, JPP = 27.5 Hz, P@S). H
3
Benzaldehyde 3a (0.035 mL, 0.391 mmol) was added
dropwise to a solution of 2 (0.08 g, 0.186 mmol) in dichlo-
romethane (5 mL) at room temperature. The resulting solu-
tion was stirred overnight, then evaporated to dryness.
Then, the resulting powder was washed three times with
CH2Cl2/pentane (1/50). Compound 4a was obtained as a
white powder in 98% yield.
NMR (CDCl3): d = 2.97 (d, JHP = 9.1 Hz, 6H, CH3),
6.84 (dd, JHH = 8.8 Hz, JHP = 5.0 Hz, 2H, H-C3), 7.07–
3
3
7.16 (m, 4H, H-C4, H-C5), 7.24–7.39 (m, 28H, HC@N,
H-Cm, H-C0 ; H-C0 ; H-C0 ), 7.48 (br t, JHH = 7.6 Hz,
3
o
m
p
3
3
3H, H-Cp), 7.81 (dd, JHH = 7.3 Hz, JHP = 13.1 Hz, 6H,
H-Co), 7.91 (‘‘t’’, J = 4.8 Hz, 2H, H-C6). 13C {1H} NMR
2
(CDCl3): d = 32.4 (d, JCP = 10.1 Hz, CH3), 125.6 (br s,
2
31P {1H} NMR (CH2Cl2): d = 11.2 (d, JPP = 26.7 Hz,
C0p), 127.5 (s, C6 or C5), 128.1 (s, C5 or C6), 128.4 (d,
2
1
3JCP = 16.5 Hz, C0 ), 128.5 (d, JCP = 11.9 Hz, Cm), 129.6
3
P@N), 56.8 (d, JPP = 26.7 Hz, P@S). H NMR (CDCl3):
m
3
(dd, JCP = 105.9 Hz, JCP = 2.9 Hz, Ci), 132.0 (br s, C4,
1
3
d = 3.22 (d, JHP = 9.8 Hz, 6H, CH3), 7.18–7.22 (m, 6H,
H-C2, H-C4), 7.30 (br s, 2H, HC@N), 7.38–7.48 (m, 10H,
2
3
Cp), 132.8 (d, JCP = 10.8 Hz, Co), 132.9 (d, JCP
=
H-C3, H-Cm), 7.53 (ttd, JHH = 7.4 Hz, JHH = 1.6 Hz,
3
4
13.5 Hz, HC@N), 133.6 (s, C3), 133.8 (br s, C1), 133.8 (d,
5JHP = 1.6 Hz, 3H, H-Cp), 7.86 (ddd, JHH = 8.4 Hz,
3
2JCP = 19.5 Hz, C0 ), 136.1 (d, JCP = 9.0 Hz, C0), 139.9
1
o
i
3
4JHH = 1.2 Hz, JHP = 13.2 Hz, 6H, H-Co). 13C {1H}
(d, JCP = 17.7 Hz, C2). Anal. Calc. for C58H51N5P4S
(974.0): C, 71.52; H, 5.28; N, 7.19. Found: C, 71.47; H,
5.34; N, 7.23%.
1
2
NMR (CDCl3): d = 32.5 (d, JCP = 9.8 Hz, CH3), 126.3
(s, C2), 127.5 (s, C4), 128.2 (s, C3), 128.4 (d, 3JCP = 13.8 Hz,
1
3
Cm), 129.9 (dd, JCP = 106.0 Hz, JCP = 4.0 Hz, Ci), 132.1
2
(br s, Cp), 133.0 (d, JCP = 9.9 Hz, Co), 134.6 (d,
4.7. Ph3P@NP(S)[NMe–N@CH(C6H4)-o-P(S)Ph2]2
(4d)
3JCP = 14.2 Hz, HC@N), 136,8 (s, C1). Anal. Calc. for
C34H33N5P2S (605.7): C, 67.42; H, 5.49; N, 11.56. Found:
C, 67.50; H, 5.57; N, 11.49%.
Way from 3d.
(diphenyl)thiophosphine 3d (0.188 g, 0.583 mmol) in
dichloromethane (4 mL) was added dropwise to
A solution of (2-formyl-phenyl)-
4.5. Ph3P@NP(S)[NMe–N@CH(C6H4)-o-OH]2 (4b)
a
solution of 2 (0.114 g, 0.235 mmol) in dichloromethane
(5 mL) at room temperature and stirred overnight. Then
the solvent was evaporated and the resulting powder was
Salicylaldehyde 3b (0.128 mL, 1.2 mmol) was added
dropwise to a solution of 2 (0.250 g, 0.582 mmol) in dichlo-

1338
A. Maraval et al. / Journal of Organometallic Chemistry 691 (2006) 1333–1340
washed three times with CH2Cl2/pentane (1/50). Com-
pound 4d was obtained as white powder in 95% yield.
Way from 5d. Powdered triphenylphosphine (0.01 g,
0.038 mmol) and azide 5d (0.031 g, 0.038 mmol) were dis-
solved in dichloromethane (3 mL) and stirred for 3 days
at room temperature. Then, the solvent was removed under
vacuum, and the resulting powder was washed three times
with pentane, to afford compound 4d as a white powder in
98% yield.
135.2
(d,
2JCP = 11.8 Hz,
Co),
157.6
(s, C2). Anal. Calc. for C46H53N5O12P2S (962.0): C, 57.44;
H, 5.55; N, 7.28. Found: C, 57.23; H, 5.68; N, 7.14%.
4.9. N3-P(S)[NMe–N@CH(C6H4)-o-P(S)Ph2]2 (5d)
A solution of (2-formyl-phenyl)(diphenyl) thiophos-
phine 3d (0.290 g, 0.900 mmol) in dichloromethane
(4 mL) was added dropwise to a solution of 1 (0.080 g,
0.410 mmol) in dichloromethane (5 mL) at room tempera-
ture and stirred overnight. The resulting solution was then
evaporated under vacuum. The product was purified by sil-
ica gel chromatography using CH2Cl2 as eluent. The pure
product was obtained as a white powder in 80% yield.
Crystals suitable for X-ray diffraction were obtained in a
mixture CH2Cl2/hexane 1:4 by slow evaporation at room
temperature.
2
31P {1H} NMR (CH2Cl2): d = 11.4 (d, JPP = 28.8 Hz,
P@N), 41.1 (s, P(S)Ph2), 55.8 (d, 2JPP = 28.8 Hz, P@S). 1H
NMR (CDCl3): d = 2.55 (d, 3JHP = 8.9 Hz, 6H, CH3), 6.95
3
3
(dd, JHH = 7.8 Hz, JHP = 15.0 Hz, 2H, H-C3), 7.12 (br t,
3JHH = 7.2 Hz, 2H, H-C4), 7.27 (t, JHH = 7.2 Hz, 2H, H-
3
C5), 7.33–7.45 (m, 17H, H-Cp, H-Cm, H-C0 ), 7.53 (br t,
m
3JHH = 7.5 Hz, 4H, H-C0 ), 7.72–7.86 (m, 14H, H-C0 , H-
p
o
3
Co), 7.89 (s, 2H, HC@N), 7.95 (dd, JHH = 7.5 Hz,
4JHP = 7.5 Hz, 2H, H-C6). 13C {1H} NMR (CDCl3):
d = 32.8 (d, 2JCP = 9.7 Hz, CH3), 127.1 (d, 3JCP = 12.7 Hz,
C4), 127.4 (d, 3JCP = 9.5 Hz, C6), 128.9 (d, 3JCP = 13.0 Hz,
31P {1H} NMR (CDCl3): d = 41.1 (s, PPh2), 66.0 (s, N3-
1
3
P); H NMR (CDCl3): d = 2.62 (d, JHP = 10.3 Hz, 6H,
3
3
CH3), 6.89 (dd, JHH = 6.7 Hz, JHP = 14.8 Hz, 2H,
Cm), 129.0 (d, 3JCP = 12.4 Hz, C0 ), 129.1 (d, 3JCP = 12.4 Hz,
H-C3), 7.22 (br t, JHH = 6.8 Hz, 2H, H-C4), 7.41–7.53
3
m
C0 ), 130.1 (dd, JCP = 106.8 Hz, JCP = 3.1 Hz, Ci), 130.7
(m, 14H, H-C0p; H-Cm0 , H-C5), 7.80 (m, 8H, H-C0o), 8.04
1
3
m
(d, 1JCP = 82.8 Hz, C2), 131.9 (d, 4JCP = 2.4 Hz, Cp), 132.0
(dd, JHH = 6.8 Hz, JHP = 4.5 Hz, 2H, H-C6), 8.35 (s,
3
3
4
(d, JCP = 2.5 Hz, C0 ), 132.7 (br s, C5, C3), 132.7 (d,
2H, CH@N); 13C {1H} NMR (CDCl3): d = 31.6 (d,
p
1JCP = 84.0 Hz, C0), 132.7 (d, JCP = 10.6 Hz, C0 ), 132.9
2JCP = 11.0 Hz, CH3), 127.6 (d, JCP = 9.4 Hz, C4), 128.3
2
3
i
o
2
1
3
3
(d, JCP = 10.6 Hz, C0 ), 133.0 (d, JCP = 84.0 Hz, C0),
(d, JCP = 12.1 Hz, C6), 128.7 (d, JCP = 11.5 Hz, C0 ),
o
i
m
133.3 (d, JCP = 11.0 Hz, Co), 133.9 (dd, JCP = 14.5 Hz,
3JCP = 8.7 Hz, HC@N), 139.7 (d, 2JCP = 6.7 Hz, C1). Anal.
Calc. for C58H51N5P4S3 (1038.2): C, 67.10; H, 4.95; N,
6.75. Found: C, 67.21; H, 4.93; N, 6.68%.
131.9 (s, C0p), 132.0 (d, JCP = 75.7 Hz, C0), 132.0 (d,
2
3
1
i
1JCP = 88.5 Hz, C2), 132.4 (d, JCP = 11.3 Hz, C0 ), 132.4
2
o
(s, C3, C5), 137.4 (d, JCP = 6.3 Hz, C1), 138.7 (dd,
2
3JCP = 8.4 Hz, 3JCP = 15.2 Hz, CH@N). IR (KBr):
2149 cmꢀ1 ðꢀm N3Þ. Anal. Calc. for C40H36N7P3S3 (803.9):
C, 59.77; H, 4.51; N, 12.20. Found: C, 59.72; H, 4.48; N,
12.23%.
4.8. Ph3P@NP(S)[NMe–N@CH(C6H4)-o-(O-b-D-
glucoside)]2 (4e)
4.10. CpFeC5H4PPh2@NP(S)[NMe-NH2]2 (6)
Powdered 2 (0.10 g, 0.232 mmol) and helicin 3e (0.132 g,
0.465 mmol) are dissolved in DMF (5 mL) at room temper-
ature, and stirred overnight. The resulting solution was
then evaporated to dryness, and the powder was washed
with CH2Cl2/pentane (1/30). Compound 4e was obtained
as a yellow powder in 99% yield.
Powdered diphenylphosphinoferrocene (0.066 g, 0.178
mmol) and azide 1 (0.035 g, 0.178 mmol) were dissolved in
dichloromethane (5 mL) and stirred for 2 days at room tem-
perature. Then, the solvent was removed under vacuum, and
the resulting powder was washed three times with CH2Cl2/
pentane, to afford compound 6 as a yellow powder in 83%
yield.
2
31P {1H} NMR (DMF): d = 11.2 (d, JPP = 28.5 Hz,
2
P@N), 57.0 (d, JPP = 28.5 Hz, P@S). 31P {1H} NMR
2
(THF-d8): d = 13.2 (d, JPP = 28.4 Hz, P@N), 60.3 (d,
2
31P {1H} NMR (CH2Cl2): d = 15.7 (d, JPP = 18.8 Hz,
2JPP = 28.4 Hz, P@S). 1H NMR (THF-d8): d = 3.34 (d,
2
1
P@N), 71.6 (d, JPP = 18.8 Hz, P@S). H NMR (CDCl3):
3
3JHP = 9.1 Hz, 3H, CH3), 3.36 (d, JHP = 9.1 Hz, 3H,
3
d = 2.74 (d, JHP = 11.8 Hz, 6H, CH3), 3.52 (br s, 4H,
CH3), 3.30–4.00 (m, 12H, CH2, HCOH), 4.77–5.18 (m,
3
NH2), 3.98 (s, 5H, C5H5), 4.47 (dd, JHP = 3.8 Hz,
3
10H, H-C1, -OH), 6.93 (‘‘t’’, JHH = 7.4 Hz, 2H, H-C5),
3JHH = 1.9 Hz, 2H, C5H4), 4.65 (dd, JHP = 3.8 Hz,
3
g
7.19–7.33 (m, 4H, H-C3, H-C4), 7.54–8.10 (m, 19H,
3JHH = 1.9 Hz, 2H, C5H4), 7.40–7.74 (m, 10H, H-Ar).
HC@N, H-Ar). 13C {1H} NMR (THF-d8): d = 33.8
13C {1H} NMR (CDCl3): d = 39.1 (d, JCP = 5.8 Hz,
2
2
2
(d, JCP = 13.0 Hz, CH3) and 34.0 (d, JCP = 11.3 Hz,
CH3), 69.1 (dd, JCP = 103 Hz, JCP = 4.3 Hz, C1 5H4),
1
3
CH3), 64.0 (s, CH2OH), 72.1 (s, C4), 75.9 and 76.1 (2s,
69.5 (s, C5H5), 72.1 (d, JCP = 11.6 Hz, C3;4H4), 72.7 (d,
3
g
C5g), 79.0 (s, Cg2 or Cg3), 79.4 and 79.6 (2s, Cg3 or C2g),
104.3 and 104.5 (2s, Cg1), 119.7 and 119.8 (2s, C3),
5
2JCP = 13.3 Hz, C2;5H4), 128.1 (d, JCP = 13.3 Hz, Cm),
3
5
1
3
131.4 (dd, JCP = 106.0 Hz, JCP = 4.3 Hz, Ci), 132.02 (s,
Cp), 132.04 (d, JCP = 10.6 Hz, Co). Anal. Calc. for
2
124.5 (s, C5), 127.8 (s, C6), 129.8 (s, C1), 130.0 (s, C4), 130.3
(d, 3JCP = 13.3 Hz, Cm), 132.4 (br d, 3JCP = 13 Hz, CH@N),
132.8 (dd, 1JCP = 106.1 Hz, 3JCP = 4.0 Hz, Ci), 134.0 (s, Cp),
C24H29N5P2SFe (537.4): C, 53.64; H, 5.44; N, 13.03.
Found: C, 53.78; H, 5.43; N, 13.09%.

A. Maraval et al. / Journal of Organometallic Chemistry 691 (2006) 1333–1340
1339
4.11. CpFeC5H4PPh2@NP(S)[NMe–N@CH(C6H4)-o-
(O-b-Ds-glucoside)]2 (7e)
4.13. Ph3P@NP(S)[NMe–N@CH(C6H4)-o-(O-b-D-
glucoside)]2[AuCl] (10e)
Powdered 6 (0.015 g, 0.0279 mmol) and helicin 3e
(0.016 g, 0.0558 mmol) are dissolved in DMF (1 mL) at
room temperature, and stirred for 7 days. The resulting
solution was then evaporated to dryness, and the powder
was washed with THF/pentane. Compound 4e was
obtained as a yellow powder in 95% yield.
A solution of AuCl(tht) (0.030 g, 0.094 mmol) in CH2Cl2
(3 mL) was added dropwise to a solution of 4e (0.090 g,
0.094 mmol) in THF (2 mL), and stirred overnight at room
temperature. The solvent was evaporated to afford a solid,
which was washed three times with CH2Cl2/pentane/Et2O,
to afford compound 10e as a white powder in 85% yield.
2
2
31P {1H} NMR (THF-d8): d = 17.2 (d, JPP = 34.3 Hz,
31P {1H} NMR (THF-d8): d = 18.3 (d, JPP = 18.9 Hz,
2
2
1
P@N), 62.1 (d, JPP = 34.3 Hz, P@S). 1H NMR (THF-
P@N), 49.2 (d, JPP = 18.9 Hz, P@S). H NMR (THF-d8):
3
3
d8): d = 3.39 (d, JHP = 8.7 Hz, 6H, CH3), 3.40–4.06 (m,
d = 3.49 (d, JHP = 9.8 Hz, 6H, CH3), 3.52–3.97 (m, 12H,
12H, CH2, HCOH), 4.22 (s, 5H, C5H5), 4.64–5.10 (m,
CH2, HCOH), 4.67–5.03 (m, 10H, H-C1, -OH), 7.01 (m,
g
14H, C5H4, H-C1, –OH), 6.92 (m, 2H, H-C5), 7.22 (m,
2H, H-C5g), 7.35 (m, 4H, H-Cg3, H-C4g), 7.63–8.24 (m, 19 H,
HC@N, H-Ar). 13C {1H} NMR (THF-d8): d = 34.5 (d,
2JCP = 13.6 Hz, CH3), 34.6 (d, 2JCP = 13.6 Hz, CH3), 63.67
g
4H, H-C3, H-C4), 7.48–8.04 (m, 14H, HC@N, H-Ar). 13C
2
{1H} NMR (THF-d8): d = 34.3 (d, JCP = 11.5 Hz, CH3)
2
and 34.4 (d, JCP = 11.5 Hz, CH3), 63.8 (s, CH2OH), 71.8
and 63.74 (2s, CH2OH), 72.13 and 72.18 (2s, C4), 75.85
g
3
(s, C5H5), 72.2 and 72.3 (s, C4), 73.8 (d, JCP = 13.0 Hz,
and 75.91 (2s, C5g), 79.1 (s, Cg2 or C3g), 79.38 and 79.44 (2s,
C3g or C2g), 104.54 and 104.78 (2s, C1g), 119.7 and 120.1 (2s,
C3), 124.39 and 124.49 (2s, C5), 127.5 (s, C6), 128.10 and
g
C35;4H4), 75.4 (d, JCP = 14.0 Hz, C2;5H4), 75.9 and 76.0
2
5
(2s, C5g), 79.1 (s, Cg2 or Cg3), 79.4 and 79.6 (2s, Cg3 or C2g),
104.5 and 104.7 (2s, C1g), 119.5 and 119.7 (2s, C3), 124.3
and 124.4 (2s, C5), 127.7 (s, C6), 128.1 (s, C1), 129.8 (s,
128.16 (2s, C1), 130.2 (dd, JCP = 104.6 Hz, JCP = 4.0 Hz,
1
3
3
Ci), 130.7 (d, JCP = 12.1 Hz, Cm), 131.2 (s, C4), 134.7
C4), 130.0 (d, JCP = 13.0 Hz, Cm), 130.5 (dd, JCP
=
(s, Cp), 135.0 (d, JCP = 106.0 Hz, Co), 136.7 (d,
3
1
2
3
3
105.0 Hz, JCP = 4.3 Hz, Ci), 132.2 (br d, JCP = 13 Hz,
3JCP = 15.8 Hz, CH@N), 158.0 (s, C2). Anal. Calc. for
C46H53AuClN5O12P2S (1194.4): C, 46.26; H, 4.47; N, 5.86.
Found: C, 46.03; H, 4.52; N, 5.78%.
2
CH@N), 133.6 (s, Cp), 134.6 (d, JCP = 13.3 Hz, Co),
157.6 (s, C2). Anal. Calc. for C50H57N5O12P2SFe
(1069.9): C, 56.13; H, 5.37; N, 6.55. Found: C, 56.32; H,
5.46; N, 6.48%.
5. X-ray structure analysis
4.12. Ph3P@NP(S)[NMe–N@CH(C6H4)-o-O]2[Ni] (9b)
Data collection were collected at low temperature
(180(2) K) on an Xcalibur Oxford Diffraction diffractometer
Powdered 2 (0.133 g, 0.31 mmol) and Ni(sal)2 Æ 2H2O 8
(0.105 g, 0.31 mmol) and THF (5 mL) are mixed in a sealed
Schlenck and heated at 60 ꢁC for 5 days under stirring.
Upon heating, the suspension became a green solution,
which turned into a dark yellow solution with time. The
solvent was evaporated to dryness, and the resulting pow-
der was extracted with dichloromethane. The solution was
evaporated and compound 9b was obtained as yellow pow-
der 99% yield.
using a graphite-monochromated Mo Ka radiation
˚
(k = 0.71073 A) and equipped with an Oxford Cryosystems
Cryostream Cooler Device. The final unit cell parameters
have been obtained by means of least-squares refinement
performed on a set of 5000 well measured reflections, and
crystal decay has been monitored during the data collections,
no significant fluctuations of intensities have been observed.
The structures have been solved by Direct Methods using
SIR92 [22], and refined by means of least-squares procedures
on a F2 with the aid of the program SHELXL97 [23] include in
the software package WINGX version 1.63 [24]. The atomic
scattering factors were taken from International tables for
X-ray crystallography [25]. All hydrogen atoms were located
on a difference Fourier maps, and refined by using a riding
model. All non-hydrogen atoms were anisotropically
refined, and in the last cycles of refinement a weighting
scheme was used, where weights are calculated from the
2
31P {1H} NMR (THF): d = 13.9 (d, JPP = 21.4 Hz,
2
P@N), 56.5 (d, JPP = 21.4 Hz, P@S). 31P {1H} NMR
2
(CDCl3): d = 14.5 (d, JPP = 19.8 Hz, P@N), 55.6 (d,
2JPP = 19.8 Hz, P@S). 1H NMR (CDCl3): d = 3.50 (d,
3
3JHP = 10.2 Hz, 6H, CH3), 6.47 (‘‘t’’, JHH = 6.9 Hz, 2H,
3
H-C5), 6.80 (d, JHH = 7.4 Hz, 2H, H-C3), 6.98 (d,
3
3JHH = 8.5 Hz, 2H, H-C6), 7.20 (‘‘t’’, JHH = 7.0 Hz, 2H,
H-C4), 7.48–7.61 (m, 11H, HC@N, H-Cm, H-Cp), 7.73
2
3
3
(dd, JHH = 7.3 Hz, JHP = 13.0 Hz, 6H, H-Co). 13C {1H}
following formula: w ¼ 1=½r2ðF o2Þ þ ðaPÞ þ bPꢁ; where
NMR (CDCl3): d = 44.5 (br s, CH3), 115.4 (s, C3), 118.3
P ¼ ðF 2o þ 2F 2cÞ=3. Drawing of molecule is performed with
the program ORTEP32 [26] with 50% probability displacement
ellipsoids for non-hydrogen atoms.
(s, C5), 123.0 (br s, C1), 128.8 (d, JCP = 11.8 Hz, Cm),
3
1
3
128.9 (dd, JCP = 106.4 Hz, JCP = 3.8 Hz, Ci), 131.7 (d,
4JCP = 2.7 Hz, Cp), 132.8 (s, C4), 132.8 (d, JCP = 11.7 Hz,
2
Co), 134.8 (s, C6), 164.7 (s, C2), 166.5 (d, JCP = 5.6 Hz,
Acknowledgement
3
HC@N). Mass (FAB): m/z = 694 [MH]+. Anal. Calc. for
C34H31N5NiO2P2S (694.3): C, 58.81; H, 4.50; N, 10.09.
Found: C, 58.67; H, 4.55; N, 10.01%.
We thank Jean-Pierre Costes for a generous gift of com-
pound 8.

1340
A. Maraval et al. / Journal of Organometallic Chemistry 691 (2006) 1333–1340
[9] C. Galliot, A.M. Caminade, F. Dahan, J.P. Majoral, Angew. Chem.
Int. Ed. Engl. 32 (1993) 1477–1479.
Appendix A. Supporting information available
[10] H. Staudinger, J. Meyer, Helv. Chim. Acta 2 (1919) 635–646.
[11] J. Mitjaville, A.M. Caminade, R. Mathieu, J.P. Majoral, J. Am.
Chem. Soc. 116 (1994) 5007–5008.
[12] J.E. Leffler, R.D. Temple, J. Am. Chem. Soc. 89 (1967) 5235–5246.
[13] C. Galliot, A.M. Caminade, F. Dahan, J.P. Majoral, W. Schoeller,
Inorg. Chem. 33 (1994) 6351–6356.
[14] U. Muller, Chem. Ber. 110 (1977) 788–791.
¨
[15] H.W. Roesky, M. Noltemeyer, G.M. Sheldrick, Z. Naturforsch. B
41B (1986) 803–807.
[16] V. Maraval, R. Laurent, B. Donnadieu, M. Mauzac, A.M. Caminade,
J.P. Majoral, J. Am. Chem. Soc. 122 (2000) 2499–2511.
[17] R.J. Wehmschulte, M.A. Khan, S.I. Hossain, Inorg. Chem. 40 (2001)
2756–2762.
Crystallographic data for the structural analysis has
been deposited with the Cambridge Crystallographic Data
Centre, CCDC No. 287345 for compound 5d. Copies of
this information may be obtained free of charge from:
The Director CCDC 12 Union Road, Cambridge, CB2
1EZ UK, fax (int. code): +44 1223 336 033, or e-mail:
deposit@ccdc.cam.ac.uk or www:http://www.ccdc.cam.a-
c.uk. Supplementary data associated with this article can
be found, in the online version, at doi:10.1016/j.
jorganchem.2005.12.022.
[18] S. Kumaraswamy, P. Kommana, N.S. Kumar, K.C.K. Swamy,
Chem. Commun. (2002) 40–41.
References
[19] C.F. Campana, F.Y.K. Lo, L.F. Dahl, Inorg. Chem. 18 (1979) 3060–
3064.
[20] J.M. Stewart, E.C. Lingafelter, J.D. Breazeale, Acta Crystallogr. B 14
(1961) 888–891.
[21] R. Uson, A. Laguna, M. Laguna, Inorg. Synth. 26 (1989) 85–91.
[22] A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, J. Appl.
Crystallogr. 26 (1993) 343–350.
[1] D.A. Atwood, M.J. Harvey, Chem. Rev. 101 (2001) 37–52.
[2] E.N. Jacobsen, in: E.W. Abel, F.G.A. Stone, G. Wilkinson (Eds.),
Comprehensive Organometallic Chemistry II, vol. 12, Pergamon,
New York, 1995 (Chapter 11).
[3] N. Canali, D.C. Sherrington, Chem. Soc. Rev. 28 (1999) 85–93.
[4] E.N. Jacobsen, Acc. Chem. Res. 33 (2000) 421–431.
[5] D. Colombo-Kather, A.M. Caminade, R. Kraemer, B. Raynaud, J.
Jaud, J.P. Majoral, Bull. Soc. Chim. Fr. 131 (1994) 733–741.
[6] R.M. Sebastian, G. Magro, A.M. Caminade, J.P. Majoral, Tetrahe-
dron 56 (2000) 6269–6277.
[23] G.M. Sheldrick, SHELX97 [Includes SHELXS97, SHELXL97, CIFTAB] –
Programs for Crystal Structure Analysis (Release 97-2), Institut fur
¨
¨
Anorganische Chemie der Universita¨t, Tammanstrasse 4, D-3400
Go¨ttingen, Germany, 1998.
[7] K.V. Katti, P.R. Singh, C.L. Barnes, Inorg. Chem. 31 (1992) 4588–
4593.
[8] C. Larre´, B. Donnadieu, A.M. Caminade, J.P. Majoral, Chem. Eur.
J. 4 (1998) 2031–2036.
[24] L. Farrugia, J. Appl. Crystallogr. 32 (1999) 837–838.
[25] International tables for X-ray crystallography, Vol. IV, Kynoch
Press, Birmingham, England, 1974.
[26] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.