Syntheses and crystal structures of mono- and bi-metallic zinc compounds of symmetrically- and asymmetrically-substituted bis(amino)cyclodiphosph(V)azanes

Article abstract of DOI:10.1016/j.jorganchem.2004.01.007

The dioxocyclodiphosph(V)azane cis -[(^tBuHN)O=P (μ-N^tBu)₂P=O(NH^tBu)] reacted with two equivalents of diethylzinc to form the centrosymmetric dimer {[(O=PN^tBu)₂ (N^tBu)₂ZnEt](ZnEt·THF)}₂ (1) while under identical conditions, the sulfur and selenium analogues afforded only the monoethylzinc compounds {[(^tBuHN) E=P(μ-N^tBu)₂P=E(N^tBu)] (ZnEt·THF)}E=S(2), Se (3). To further probe the apparent ligand effects on coordination number and coordination site, cis -[(PhHN)S=P(μ-N^tBu)₂P=S(NH^tBu)] (5) was synthesized from cis -[ClP(μ-N^tBu) ₂P(NH^tBu)] (4) and both were characterized by single-crystal X-ray diffraction. Two equivalents of 5 reacted with diethylzinc to produce the homoleptic, trispirocyclic complex {[(^tBuHN)S=P(μ-N^tBu)₂ P=S(NPh)]₂Zn} (6). A second asymmetrically-substituted cyclodiphosph(V)azane, namely [(^tBuNH)S=P (μ-N^tBu)₂P=N p -tol (NH^tBu)] (7), was also synthesized and structurally characterized. In contrast to 5, only one equivalent of this ligand reacted with excess diethylzinc, via its N,N′, rather than its N, S side, to afford {[(^tBuHN)S=P (μ-N^tBu)₂P=N p -tolyl (N^tBu)](ZnEt)} (8).

Full text of DOI:10.1016/j.jorganchem.2004.01.007

Journal of Organometallic Chemistry 689 (2004) 1110–1121
www.elsevier.com/locate/jorganchem
Syntheses and crystal structures of mono- and bi-metallic
zinc compounds of symmetrically- and
asymmetrically-substituted bis(amino)cyclodiphosph(V)azanes
a
a
a,
b
Graham R. Lief , Daniel F. Moser , Lothar Stahl *, Richard J. Staples
a
Department of Chemistry, University of North Dakota, Box 9024, Grand Forks, ND 58202-9024, USA
b
Department of Chemistry and Chemical Biology, Harvard University, Cambridge, MA 02138, USA
Received 16 October 2003; accepted 5 January 2004
Abstract
The dioxocyclodiphosph(V)azane cis-[(tBuHN)O@P(l-NtBu)2P@O(NHtBu)] reacted with two equivalents of diethylzinc to form
the centrosymmetric dimer {[(O@PNtBu)2(NtBu)2ZnEt](ZnEt ꢀ THF)}2 (1) while under identical conditions, the sulfur and selenium
analogues afforded only the monoethylzinc compounds {[(tBuHN)E@P(l-NtBu)2P@E(NtBu)](ZnEt ꢀ THF)}E@S(2), Se (3). To
further probe the apparent ligand effects on coordination number and coordination site, cis-[(PhHN)S@P(l-NtBu)2P@S(NHtBu)]
(5) was synthesized from cis-[ClP(l-NtBu)2P(NHtBu)] (4) and both were characterized by single-crystal X-ray diffraction. Two
equivalents of
5
reacted with diethylzinc to produce the homoleptic, trispirocyclic complex {[(tBuHN)S@P(l-
NtBu)2P@S(NPh)]2Zn} (6). A second asymmetrically-substituted cyclodiphosph(V)azane, namely [(tBuNH)S@P(l-NtBu)2P@Np-
tol(NHtBu)] (7), was also synthesized and structurally characterized. In contrast to 5, only one equivalent of this ligand reacted with
excess diethylzinc, via its N,N0, rather than its N,S side, to afford {[(tBuHN)S@P(l-NtBu)2P@Np-tolyl(NtBu)](ZnEt)} (8).
Ó 2004 Elsevier B.V. All rights reserved.
Keywords: Zinc; Ethlyzinc-amides; Ambidentate ligands; Cyclodiphosph(V)azanes; Diazadiphosphetidines
1. Introduction
been systematically investigated [6]. These ambidentate
ligands chelate metal moieties as bis(amido) (N,N0) li-
gands above the heterocycle (B) [7] as bis(chalcogenido)
(E,E0) ligands below the heterocycles (C) [8], or as a
combination of both, namely as bis(amido)/bis(chalco-
genido) ligands, i.e., N,N0 and E,E0 (D) [9,10]. The, thus
far, most common coordination mode observed, how-
ever, is one in which the metal ions are coordinated
laterally, similar to aminophosphoranes in either a
monoanionic (N,E) (E) [11,12] or a dianionic (N,E)/
(N0,E0) (F) fashion [10,13]. A unique combination of
facial and lateral coordination modes was reported by
Chivers et al. [14,15] for the dilithio salt G. But most of
these reactions were done with different cyclo-
diphosph(V)azanes, in different solvents, at different
temperatures and with a variety of metallating agents,
even for the same metal, thereby clouding the role of the
ligand on the outcomes of these reactions.
Ambidentate inorganic ligands, e.g., NOꢁ2 and SCNꢁ,
have played an important role in the development of
coordination chemistry, because they have provided
important insight into electronic metal–ligand interac-
tions [1,2]. As the structural complexity of ligands in-
creases, however, it becomes more difficult to determine
the underlying reasons for coordination site preferences
and to separate electronic from steric effects. This is
particularly true for multidentate organic ligands in
which electronic and steric influences are often inter-
connected.
The bis(amino)cyclodiphosph(V)azanes cis-[(RHN)-
E@P(l-NtBu)2P@E(NHR)] (R ¼ tBu, Ar; E ¼ O, S, Se,
NR) (A, Chart 1) are well-known tetradentate molecules
[3–5], whose coordination chemistry has only recently
* Corresponding author. Tel.: +(701)7772242; fax: +(701)7772331.
E-mail address: lstahl@chem.und.edu (L. Stahl).
We hoped to obtain insight into the role of
the bis(amino)cyclodiphosph(V)azanes themselves on
0022-328X/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.01.007

G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
1111
from sodium–benzophenone–ketyl or potassium–ben-
zophenone–ketyl immediately before use. NMR spectra
were recorded on a Bruker AVANCE-500 NMR spec-
1
trometer. The H, 13C and 31P NMR spectra are refer-
enced relative to C6D5H (7.15 ppm), C6D6 (128.0 ppm)
and P(OEt)3 (137.0 ppm), respectively. Melting points
were obtained on a Mel-Temp apparatus and are un-
corrected. Elemental analyses were performed by E and
R Microanalytical Services, Parsipanny, New Jersey,
and Desert Analytics, Tucson, Arizona.
Diethylzinc was purchased from Acros and used as
a 0.91 M hexanes stock solution. The cyclodiphos-
phazanes cis-[ClP(l-NtBu)2PCl][18], cis-[(tBuHN)P
(l-NtBu)2P(NHtBu)] [19,20] and cis-[(tBuHN)E@P
(l-NtBu)2P@E(NHtBu)], E ¼ O [13], S [3], Se [9] were
prepared by previously published procedures.
2.2. Syntheses
2.2.1. {[(O@PNtBu)2(NtBu)2ZnEt](ZnEtꢀTHF)}2 (1)
To a two-necked round-bottomed flask equipped
with an addition funnel, inlet, and stir bar, was added 7
(0.792 g, 2.08 mmol). This was dissolved in THF (20
mL) and toluene (15 mL), and the ensuing solution was
cooled to )78 °C. To the vigorously stirring solution,
ZnEt2 (5.71 mL, 5.20 mmol) in THF (10 mL) was added
slowly over 30 min. Once addition was complete, the
solution was allowed to warm to RT and left to react for
24 h. The solution was then concentrated in vacuo and
placed in a freezer at )21 °C. After 16 h, 1.09 g (82.2%)
of colorless crystals was isolated.
Mp: 203 °C dec. 1H NMR (500.13 MHz, benzene-d6,
26 °C) d ¼ 3:57 (s, 8H, THF), 1.68 (br s, 9H, NtBu), 1.65
(br s, 9H, NtBu), 1.52 (s, 24H, NtBu and CH3 (Et)), 1.49
(m, 6H, CH3), 1.41 (s, 8H, THF), 1.33 (s, 18H, NtBu),
1.23 (s, 18H, NtBu), 0.49 (m, 4H, CH2), 0.35 (m, 4H,
CH2). 13C{1H} NMR (125.76 MHz, benzene-d6, 26 °C)
d ¼ 67:77 (s, THF), 55.32 (m, NtBu), 52.97 (m, NtBu),
33.31 (s, NtBu), 32.02 (m, NtBu), 31.04 (m, NtBu), 25.76
(s, THF), 20.41 (s, CH3), 12.69 (d, JPC ¼ 10:51 Hz,
CH3), 3.89 (s, CH2), 0.70 (s, CH2). 31P{1H} NMR
(202.46 MHz benzene-d6, 26 °C) d ¼ 5:66 (d,
JPP ¼ 42:58 Hz), )3.24 (d, JPP ¼ 39:01 Hz). Anal. Calcd.
for C48H108N8O6P4Zn4: C, 45.08; H, 8.51; N, 8.76.
Found: C, 45.03; H, 8.54; N, 8.90%.
Chart 1.
product distribution and product structure by doing
these reactions with the same organometallic reagent
while systematically varying the ligand. Diethylzinc
seemed a good choice as a metallating agent because it
had been used successfully in reactions with related li-
gands [16,17]. It was also hoped that its relatively low
reactivity would discriminate between the acidities of
different ligands and between different coordination sites
in the same ligand. In addition, zinc is of intermediate
size, thereby preventing the structural bias introduced
by very large or very small ions.
2.2.2. {[(tBuHN)S@P(l-NtBu)2P@S(NtBu)](ZnEt ꢀ
THF)} (2)
A
cold ()78 °C) toluene/THF solution of
2. Experimental
[(tBuHN)S@P(l-NtBu)2P@S(NHtBu)] (0.650 g, 1.58
mmol) was treated dropwise with ZnEt2 (3.40 mL, 3.10
mmol). The reaction mixture was allowed to warm to
room temperature, stirred at that temperature for 16 h
and then concentrated to 5 mL in vacuo. After the flask
had been stored at )21 °C for several days, colorless
crystals separated. Yield: 0.496 g (54.3%).
2.1. General procedures
All experiments were performed under an atmosphere
of purified nitrogen or argon, using standard Schlenk
techniques. Solvents were dried and freed of molecular
oxygen by distillation under an atmosphere of nitrogen

1112
G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
1
Mp: 71 °C dec. H NMR (500.13 MHz, benzene-d6,
26 °C) d ¼ 3:57 (t, 4H, THF), 2.87 (s, 1H, HNtBu), 1.65
(s, 18H, NtBu), 1.48 (t, 3H, CH3, JHH ¼ 8:1 Hz), 1.34 (t,
4H, THF), 1.33 (s, 9H, NtBu), 1.21 (s, 9H, NtBu), 0.78
(q, 2H, CH2, JHH ¼ 8:1 Hz). 13C{1H} NMR (125.76
MHz, benzene-d6, 26 °C) d ¼ 68:07 (s, THF), 57.63 (s,
NtBu), 55.66 (d, JPC ¼ 13:8 Hz, NtBu), 54.82 (s, NtBu),
33.26 (d, JPC ¼ 9:9 Hz, NtBu), 31.65 (d, JPC ¼ 4:2 Hz,
NtBu), 30.53 (t, JPC ¼ 4:5 Hz, NtBu), 25.93 (s, THF),
12.60 (s, CH3), 1.77 (s, CH2). 31P{1H} NMR (202.46
MHz, benzene-d6, 26 °C) d ¼ 36:84 (d, JPP ¼ 23:4 Hz),
31.85 (d, JPP ¼ 25:2 Hz). Anal. Calcd. for
C22H50N4OP2S2Zn: C, 45.71; H, 8.72; N, 9.69. Found:
C, 45.37; H, 9.10; N, 10.04%.
2.2.5. cis-[(PhHN)S@P(l-NtBu)2P@S(NHtBu)] (5)
Aniline (1.16 mL, 12.7 mmol), dissolved in 60 mL
toluene and cooled to 0°, was treated dropwise with n-
BuLi (5.10 mL, 12.7 mmol). The resulting solution was
heated to reflux for 1 h, allowed to cool to RT, and
treated with a solution of 4 (3.61 g, 11.6 mmol) in 30 mL
of toluene. The reaction-mixture was allowed to stir for
16 h at RT and filtered through a medium-porosity frit
to remove lithium chloride. Elemental sulfur (0.743 g,
23.2 mmol) was then added, and the solution was kept at
50 °C for 4 h. The solution was then concentrated in
vacuo to a volume of 20 mL and kept at )21 °C, until
several fractions of off-white crystals had yielded 3.39 g
(67.7%) of product.
Mp: 165 °C dec. 1H NMR (500.13 MHz, benzene-d6,
26 °C) d ¼ 6:98 (t, 2H, JHP ¼ 7:7 Hz, o-Ph), 6.90 (d, 2H,
JHP ¼ 7:5 Hz, m-Ph), 6.82 (t, 1H, JHP ¼ 7:4 Hz, p-Ph),
5.06 (d, JHP ¼ 13:2 Hz, 1H, HNPh), 3.00 (s, 1H,
HNtBu), 1.61 (s, 18H, NtBu), 1.25 (s, 9H, NtBu).
13C{1H} NMR (125.76 MHz, benzene-d6, 26 °C)
d ¼ 140:54 (s, i-Ph), 129.41 (s, o-Ph), 128.31 (s, m-Ph),
121.95 (d, JPC ¼ 5:0 Hz, p-Ph), 57.04 (s, NtBu), 54.46 (s,
NtBu), 31.32 (d, JPC ¼ 15:9 Hz, NtBu), 29.66 (t,
JPC ¼ 4:4 Hz). 31P{1H} NMR (202.46 MHz, benzene-
d6, 26 °C) d ¼ 38:77 (d, JPP ¼ 34:1 Hz), 37.6 (br d).
Anal. Calcd. For C18H34N4P2S2: C, 49.97; H, 7.92; N,
12.97. Found: C, 49.94; H, 7.98; N, 13.22%.
2.2.3. {[(tBuHN)Se@P(l-NtBu)2P@Se(NtBu)](ZnEt ꢀ
THF)} (3)
In a manner identical to that used for the synthesis of
2, [(tBuHN)Se@P(l-NtBu)2P@Se(NHtBu)] (0.528 g,
1.04 mmol) was allowed to react with ZnEt2 (2.85 mL,
2.60 mmol). Yield: 0.675 g (96.6%).
1
Mp: 66–69 °C. H NMR (500.13 MHz, benzene-d6,
26 °C) d ¼ 3:57 (t, 4H, THF), 3.14 (s, 1H, HNtBu), 1.69
(s, 18H, NtBu), 1.50 (t, 3H, CH3, JHH ¼ 8:1 Hz), 1.42 (t,
4H, THF), 1.34 (s, 9H, NtBu), 1.20 (s, 9H, NtBu), 0.81
(q, 2H, CH2, JHH ¼ 8:1 Hz). 13C{1H} NMR (125.76
MHz, benzene-d6, 26 °C) d ¼ 68:04 (s, NtBu), 58.39 (s,
NtBu), 56.86 (d, JPC ¼ 17:1 Hz, NtBu), 55.49 (s, NtBu),
33.07 (d, JPC ¼ 10:3 Hz, NtBu), 31.67 (d, JPC ¼ 3:9 Hz,
NtBu), 30.62 (t, JPC ¼ 4:3 Hz, NtBu), 25.94 (s, THF),
12.67 (s, CH3), 2.91 (s, CH2). 31P{1H} NMR (202.46
MHz, benzene-d6, 26 °C) d ¼ 22:45 (d, JPP ¼ 14:3 Hz),
11.88 (d, JPP ¼ 14:1 Hz). Anal. Calcd. for
C22H50N4OP2Se2Zn: C, 39.33; H, 7.50; N, 8.34. Found:
C, 38.99; H, 7.59; N, 8.28%.
2.2.6. {[(tBuHN)S@P(l-NtBu)2P@S(NPh)]2Zn} (6)
A cold ()78 °C) toluene/THF solution of 5 (0.446 g,
1.03 mmol) was treated dropwise with ZnEt2 (2.07 mL,
1.88 mmol), dissolved in a toluene/THF solvent mixture.
Once the addition was complete, the reaction mixture
was allowed to warm to room temperature and stirred
for 14 h. The solution was then concentrated in vacuo
and placed in a freezer at )78 °C. Several crops of col-
orless crystals yielded 0.456 g (95.3%) of product.
Mp: 280–282 °C. 1H NMR (500.13 MHz, benzene-d6,
26 °C) d ¼ 7:44 (d, 4H, JHP ¼ 8:0 Hz, o-Ph), 7.13 (d,
JHP ¼ 7:6 Hz, 4H, m-Ph), 6.92 (t, JHP ¼ 6:8 Hz, 2H, p-
Ph), 3.27 (s, 2H, HNtBu), 1.58 (s, 36H, NtBu), 1.29 (s,
18H, NtBu). 13C{1H} NMR (125.76 MHz, benzene-d6,
26 °C) d ¼ 147:81 (d, JPC ¼ 12:5 Hz, i-Ph), 129.87 (s, o-
Ph), 128.70 (s, m-Ph), 123.04 (d, JPC ¼ 13:8 Hz, p-Ph),
57.27 (d, JPC ¼ 3:8 Hz, NtBu), 54.71 (s, NtBu), 31.33 (d,
JPC ¼ 4:0 Hz, NtBu), 29.99 (t, JPC ¼ 4:1 Hz, NtBu).
31P{1H} NMR (202.46 MHz, benzene-d6, 26 °C)
d ¼ 38:31 (d, JPP ¼ 25:2 Hz), 37.69 (d, JPP ¼ 25:3 Hz).
Anal. Calcd. for C36H66N8P4S4Zn: C, 46.57; H, 7.16; N,
12.07. Found: C, 46.59; H, 7.27; N, 11.78%.
2.2.4. cis-[ClP(l-NtBu) P(NHtBu)] (4)
2
This is a simplified version of a previously published
synthesis. To a cold ()78 °C) solution of phosphorus
trichloride (200 mmol) and triethylamine (300 mmol) in
200 mL of hexanes was added dropwise tert-butylamine
(500 mmol). A thick, white precipitate of triethylam-
monium chloride formed almost immediately. The re-
action mixture was allowed to warm and stirred at RT
for 12 h. It was then filtered on a large, coarse frit and
the clear, colorless filtrate was concentrated in vacuo
and stored at )12 °C until colorless needles had formed.
Yield: 20.9 g (67.0%)
1
Mp: 70–74 °C. H NMR (500.13 MHz, benzene-d6,
25 °C): d ¼ 3:78 (br s, 1 H), 1.34 (s, 18 H), 1.05. (s, 9
H). 13C {1H} NMR (125.76 MHz, benzene-d6, 25 °C):
d ¼ 53:1 (t, JPC ¼ 8:4 Hz), 52.1 (d, JPC ¼ 7:3 Hz), 32.6
(d, JPC ¼ 4:9 Hz), 31.1 (t, JPC ¼ 6:5 Hz). 31P{1H}
NMR (202.46 MHz, benzene-d6, 25 °C): d ¼ 194:6 (d,
P–Cl, JPP ¼ 21:0 Hz), 135.8 (d, P–NHtBu, JPP ¼ 20:8
Hz).
2.2.7. [(tBuNH)S@P(l-NtBu)2P@Np-tolyl(NHtBu)]
(7)
A sample of [(tBuNH)P(l-NtBu)2P(NHtBu)] (4.99 g,
14.3 mmol), dissolved in 80 mL of toluene, was treated
at room temperature with a toluene solution (10 mL) of

G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
1113
p-tolyl azide (1.91 g, 14.3 mmol). After the reaction
mixture had been stirred for 16 h, it was treated with
sulfur (0.459 g, 14.3 mmol) and allowed to react at room
temperature for an additional 3 h. The solution was
concentrated to 40 mL in vacuo and stored in a freezer
()21 °C) for several days. Yield: 5.57 g (80.1%).
were measured using x scans of 0.3° per frame for 30 s
until a complete hemisphere had been collected. The first
50 frames were recollected at the end of the data col-
lection to monitor for decay. The crystals of 1 decom-
posed at room temperature and cracked at low
temperature. It was therefore necessary to rapidly collect
its intensity data at room temperature using x scans of
3.0° per frame for 30 s with a final resolution of 0.80
Mp: 163–165 °C. 1H NMR (500.13 MHz, benzene-d6,
26 °C) d ¼ 7:63 (d, JHH ¼ 7:9 Hz, 2 H, p-tolyl), 7.19 (d,
JHH ¼ 7:6 Hz, 2H, p-tolyl), 2.65 (d, JHP ¼ 5:6 Hz, 1H,
HNtBu), 2.22 (s, 3H, Me-p-tolyl), 1.93 (br s, 1H,
HNtBu), 1.51 (s, 18H, NtBu), 1.42 (s, 9H, NtBu), 1.21 (s,
9H, NtBu). 13C{1H} NMR (125.76 MHz, benzene-d6,
26 °C) d ¼ 145:69 (s, ipso-(p-tolyl)), 129.86 (d, JPC ¼ 1:3
Hz, o-(p-tolyl)), 128.70 (s, m-(p-tolyl)), 125.10 (d,
JPC ¼ 21:9 Hz, p-(p-tolyl)), 55.9 (s, NtBu), 54.2 (t,
JPC ¼ 4:3 Hz, NtBu), 32.06 (d, JPC ¼ 4:6 Hz, NtBu),
31.61 (d, JPC ¼ 4:3 Hz, NtBu), 30.79 (t, JPC ¼ 4:1 Hz,
NtBu), 21.09 (s, Me-p-tolyl). 31P{1H} NMR (202.46
MHz, benzene-d6, 26 °C) d ¼ 35:48 (d, JPP ¼ 39:7 Hz),
)23.22 (d, JPP ¼ 40:1 Hz). Anal. Calcd. for
C23H45N5P2S: C, 56.89; H, 9.34; N, 14.42. Found: C,
57.08; H, 9.58; N, 14.43%.
3
ꢀ
e/A . This data acquisition mode leads to lower preci-
sion in the derived structure parameters.
Cell parameters were retrieved using SMART [21]
software and refined with SAINT [22] on all observed
reflections. Data were reduced with SAINT, which cor-
rects for Lorentz and polarization effects and decay. An
empirical absorption correction was applied with SAD-
ABS [23]. The structures were solved by direct methods
with the SHELXS 90 [24] program and refined by full-
matrix least squares methods on F2 with SHELXL 97
[25], incorporated in SHELXTL Version 5.10 [26].
2.3.2. Compounds 2, 3 and 7 ꢀ 1.5(THF)
The crystals were obtained as indicated in Section 2.
Compound 7 did not give suitable single crystals in the
absence of THF, but single crystals could be obtained
when the colorless powder was recrystallized from THF.
In compound 7 ꢀ 1.5(THF) the inversion center at 1/2, 0,
1/2 is occupied by one-half molecule of THF. This dis-
ordered molecule was modeled using 0.5 oxygen atoms
and two methylene groups, i.e.; O0:5C2H4. This model
leads to two orientations of the THF molecule which are
related to each other by inversion symmetry. Because
the data were collected at RT and because of the low
electron-densities at these positions, we did not attempt
to split the methylene carbon atoms. The final difference
map showed no electron density greater, or less, than
2.2.8. {[(tBuHN)S@P(l-NtBu)2P@Np-tolyl(NtBu)]
(ZnEt)} (8)
In a manner identical to those used for the syntheses
of 2 and 3, a toluene/THF solution of 7 (0.484 g, 0.997
mmol) was treated with ZnEt2 (2.73 mL, 2.49 mmol).
Yield: 0.548 g (96.1%) of colorless crystals.
Mp: 115–120 °C. 1H NMR (500.13 MHz, benzene-d6,
26 °C) d ¼ 7:68 (d, JHH ¼ 8:2 Hz, 2 H, tolyl), 7.21 (d,
JHH ¼ 8:2 Hz, 2H, tolyl), 2.98 (s, 1H, HNtBu), 2.22 (s,
3H, Me-tolyl), 1.61 (t, JHH ¼ 8:1 Hz, 3H, Me), 1.44 (s,
18H, NtBu), 1.37 (s, 9H, NtBu), 1.30 (s, 9H, NtBu), 0.88
(q, JHH ¼ 8:1 Hz, CH2). 13C{1H} NMR (125.76 MHz,
benzene-d6, 26 °C) d ¼ 143:85 (s, ipso-Ph), 129.58 (s, o-
Ph), 128.24 (s, m-Ph), 121.63 (d, JPC ¼ 16:2 Hz, p-Ph),
56.15 (s, NtBu), 54.01 (s, NtBu), 51.02 (s, NtBu), 33.77
(d, JPC ¼ 8:3 Hz, NtBu), 31.34 (d, JPC ¼ 4:1 Hz, NtBu),
30.55 (t, JPC ¼ 4:1 Hz, NtBu), 20.58 (s, Me-tolyl), 12.47
(s, CH2), 1.16 (s, Me). 31P{1H} NMR (202.46 MHz,
benzene-d6, 26 °C) d ¼ 35:89 (d, JPP ¼ 36:5 Hz), 1.12 (d,
JPP ¼ 34:0 Hz). Anal. Calcd. for C25H49N5P2SZn: C,
51.85; H, 8.53; N, 12.10. Found: C, 51.87; H, 8.86; N,
12.06%.
3
ꢀ
about 0.32 e/A . The crystals were sealed inside argon-
filled glass capillaries, and the intensity data were
collected at room temperature on a Bruker P4 diffrac-
tometer. The structures were solved by direct-methods
with SHELXL NT, Version 5.10, and were refined anal-
ogously to 1, 4, 5 and 8.
3. Results
In order to ensure that the product distribution was
not affected by solubility, all reactions were done in
THF-toluene solvent mixtures, in which both reactants
and products were very soluble. We began our reactivity
studies with the deprotonation of the di(tert-butyla-
mino)-dioxo-substituted heterocycle cis-[(tBuHN)O@P
(l-NtBu)2P@O(NHtBu)], whose bis(dimethylaluminum)
derivative had shown bilateral coordination (F, Chart
1). On addition of two equivalents of diethylzinc to this
ligand we obtained (Scheme 1), again, a bimetallic
complex, namely {[(O@PNtBu)2(NtBu)2ZnEt](ZnEt ꢀ
2.3. X-ray crystallography
2.3.1. Compounds 1, 4, 5 and 8
Suitable, single crystals were coated with oil, affixed
to a glass capillary, and centered on the diffractometer.
Reflection intensities were collected with a Bruker
SMART CCD diffractometer, equipped with an LT-2
low-temperature apparatus. Intensity data for com-
pounds 1 and 5 were collected at room temperature,
while those of 4 and 8 were collected at )60 °C. Data

1114
G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
Table 2
ꢀ
Bond lengths (A) and angles (°) of 1
Bond lengths
Zn(1)–C(17)
Zn(1)–O(1)
Zn(1)–O(2)
Zn(1)–O(3)
Zn(2)–C(23)
Zn(2)–N(3)
Zn(2)–N(4)
P(1)–N(3)
P(2)–N(4)
P(1)–N(1)
P(1)–N(2)
P(2)–N(1)
P(2)–N(2)
P(1)–O(1)
1.981(7)
1.980(4)
1.981(4)
2.193(4)
2.014(9)
2.028(4)
2.047(4)
1.592(5)
1.598(5)
1.669(4)
1.700(4)
1.679(4)
1.688(4)
1.495(4)
1.497(4)
P(2)–O(2)
Bond angles
Scheme 1.
O(1)–Zn(1)–O(2)
O(1)–Zn(1)–O(3)
O(2)–Zn(1)–O(3)
C(17)–Zn(1)–O(1)
C(17)–Zn(1)–O(2)
C(17)–Zn(1)–O(3)
N(3)–Zn(2)–N(4)
C(23)–Zn(2)–N(4)
C(23)–Zn(2)–N(3)
N(1)–P(1)–N(2)
N(1)–P(2)–N(2)
P(1)–O(1)–Zn(1)
P(2)–O(2)–Zn(1)
107.08(15)
96.88(17)
92.95(17)
112.7(3)
130.6(3)
109.7(2)
109.20(13)
122.7(3)
128.1(3)
83.06(19)
83.1(2)
THF)}2 (1). Composition and NMR spectra, however,
suggested that it was not as symmetrical as the dialu-
minum species. For example, two mutually coupled
doublets in the 31P NMR spectrum indicated two non-
identical phosphorus environments. A single-crystal
X-ray analysis proved that 1 had indeed an entirely
different structure than the aluminum analogue. Selected
data collection and bond parameters of this complex are
given in Tables 1 and 2, respectively.
160.4(2)
150.8(2)
As Fig.
1
shows, this complex consists of
two monoethylzinc bis(amido)cyclodiphosph(V)azanes,
which are connected to a centrosymmetric dimer by
bridging ethylzinc moieties. A potentially higher sym-
metry is prevented by the displacement of both cyclod-
iphosphazanes with respect to each other by the
repulsive interactions of their tert-butylimino groups.
The two chemically distinct ethylzinc moieties display
Table 1
Crystal and refinement data for compounds 1–5, 7 ꢀ 1.5(THF) and 8
1
2
3
4
5
7 ꢀ 1.5(THF)
C29H57N5O1:5
8
Chemical formula
C48H108N8O6
P4Zn4
1278.91
C22H50N4OP2
S2Zn
578.09
C22H50N4OP2
Se2Zn
671.93
C12H27
ClN3P2
311.76
C12H28
ClN3P2
432.55
C25H49N5
P2SZn
579.06
P2S
593.80
Fw
Space group
ꢁ
P21=n (No. 14)
23
9.921(6)
P212121 (No. 19) P212121 (No. 19) P21 (No. 4) P21=c (No. 14) P21=n (No. 14)
P1 (No. 2)
T (°C)
a (A)
23
22
)60
20
22
)60
ꢀ
10.2675(8)
10.3381(13)
30.149(6)
10.311(2)
10.3956(17)
30.398(10)
9.6036(5)
10.8116(6)
9.6355(5)
16.8512(7)
9.9622(4)
14.6847(6)
9.5857(14)
32.361(3)
11.6507(10)
9.4251(4)
10.3103(5)
16.3432(8)
95.4570(10)
100.7820(10)
97.1370(10)
1536.58(12)
2
ꢀ
b (A)
18.202(11)
18.141(11)
ꢀ
c (A)
a (°)
b (°)
c (°)
90.782(7)
115.678(1)
103.9850(10)
102.176(9)
3
ꢀ
V (A )
3276(3)
2
3200.2(8)
4
1.200
3258.5(13)
4
1.707
901.65(8)
2
1.145
2392.12(17)
4
1.201
3532.8(7)
4
1.116
Z
qcalc (g cmꢁ1
k (A)
)
1.297
0.710 73
15.91
0.0556
1.252
ꢀ
0.710 73
10.17
0.710 73
38.38
0.710 73
3.80
0.710 73
3.66
0.710 73
2.11
0.710 73
9.93
0.0313
l (cmꢁ1
)
a
RðF Þ [I > rðIÞ]
0.0343
0.0999
0.0624
0.1746
0.0518
0.1252
0.0353
0.1088
0.0477
0.1432
wR2ðF 2Þ (all data) 0.1088
0.0838
b
2
2
w ¼ 1=½r2ðFoÞ þ ðxPÞ þ yPꢂ, where P ¼ ðFo2 þ 2Fc2Þ=3.
a R ¼ rjFo ꢁ Fcj=rjFoj.
2
2
b Rw ¼ f½rwFo2 ꢁ FcÞ ꢂ=½wRFo2Þ ꢂg
.
1=2

G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
1115
Scheme 2.
(Scheme 2) under identical conditions. In both cases the
bis(amino)cyclodiphosph(V)azanes reacted cleanly with
only one of the two equivalents of diethylzinc added,
producing the mono(ethylzinc) compounds {[(tBuHN)
E@P(l-NtBu)2P@E(NtBu)](ZnEt ꢀ THF)}, E ¼ S (2), Se
(3), respectively. Prolonged heating of these reaction
mixtures revealed the presumed formation of symmet-
rically-coordinated bimetallic species having structures
similar to D or F. These reactions were accompanied by
considerable decomposition of the ligand and resulted in
complex mixtures, and we, therefore, did not attempt to
isolate these compounds. The diminished reactivity of 2
and 3 is likely due to their lesser acidity, sulfur and se-
lenium being less electronegative than oxygen. The 31P
NMR spectra of 2 and 3 exhibited two sets of mutually-
coupled doublets, thereby confirming the lateral mono-
coordination of each ligand by one ethylzinc moiety.
Single-crystal X-ray analyses on both complexes,
whose crystal and data collection parameters are listed
Fig. 1. Thermal-ellipsoid (35% probability) plot and partial numbering
scheme of 1. Hydrogen atoms have been omitted for clarity.
different coordination environments; one zinc atom is
N,N0-chelated by the bis(amido) substituents above the
(PN)2 ring, as in B, while the second zinc atom is
coordinated by a THF molecule and two phosphoryl
oxygen atoms from different ligands.
The chelation of an ethylzinc moiety by both amido
groups is not unexpected, but the coordination by the
phosphoryl oxygen atoms is unusual and, as far as we
know, unprecedented for this ligand system. The three-
coordinate zinc atom has a perfectly trigonal-planar
coordination environment and forms two slightly
Table 3
ꢀ
ꢀ
Selected bond lengths (A) and angles (°) for 2 and 3
asymmetrical Zn–N bonds (2.028(4) and 2.047(4) A)
that are only insignificantly longer than those of related
amido-coordinated zinc ions [17,27,28]. The second zinc
atom (Zn1) is in a distorted ZnO3C environment. The
zinc–oxygen bonds to the phosphoryl groups are equi-
2
3
Bond lengths
Zn–S(e)
Zn–N(3)
2.4490(12)
2.009(3)
1.962(5)
2.241(3)
2.0037(13)
1.9272(12)
1.580(3)
1.695(3)
1.684(3)
1.688(3)
1.699(3)
1.632(3)
2.5398(18)
2.020(7)
1.992(13)
2.236(8)
2.154(2)
2.077(2)
1.581 (8)
1.692(7)
1.697(7)
1.678(7)
1.685(8)
1.602(10)
Zn-C(5)
Zn-O(1)
ꢀ
distant (1.980(4) and 1.981(4) A) and thus quite com-
parable in length to those with bridging isopropoxide
P(1)–S(e)(1)
P(2)–S(e)(2)
P(1)–N(3)
P(1)–N(1)
P(1)–N(2)
P(2)–N(1)
P(2)–N(2)
P(2)–N(4)
ꢀ
groups (1.983(3) A) [28]. The THF donor bond to zinc is
ꢀ
expectedly longer (2.193(4) A) and appears to be among
the longest bonds of this type, which normally range
ꢀ
from about 2.00 to 2.10 A [29,30]. The zinc–carbon
ꢀ
bonds (2.014(9) and 1.981(7) A), by contrast, are similar
in length to those of related complexes with three-co-
ordinate zinc ions [17,31].
Given their partial-negative charge and di-coordina-
tion, the oxygen atoms form surprisingly short phos-
ꢀ
phorus–oxygen bonds (1.495(4) A) that are not much
ꢀ
longer than those in the pristine ligand (1.467(2) A) [32].
Bond angles
S(e)–Zn–N(3)
78.22(9)
80.2(2)
105.6(3)
95.8(2)
101.8(5)
73.39(7)
102.5(4)
103.9(3)
82.3(3)
83.1(3)
112.3(3)
97.1(3)
96.6(3)
N(3)–Zn–O(1)
O(1)–Zn–S(e)(1)
O(1)–Zn–C(5)
104.60(14)
95.97(9)
102.0(2)
Zn(1)–S(e)(1)–P(1)
Zn(1)–N(3)–P(1)
N(3)–P(1)–S(e)(1)
N(1)–P(1)–N(2)
N(1)–P(2)–N(2)
N(4)–P(2)–S(e)(2)
P(1)–N(1)–P(2)
P(1)–N(2)–P(2)
76.75(4)
As the result of the asymmetry in the dimer, the P–O–Zn
bond angles are unequal, spanning 150.8(2)° and
160.4(2)°, respectively.
101.18(15)
103.85(12)
83.48(13)
83.52(14)
112.66(120)
96.24(14)
95.83(14)
To determine if this unusual coordination behavior of
the ligand was retained in the disulfur and diselenium
analogues, we repeated this reaction with cis-
[(tBuHN)E@P(l-NtBu)2P@E(NHtBu)],
E ¼ S,
Se

1116
G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
in Table 1, confirmed this structure assignment. Table 3
provides a side-by-side comparison of selected bond
parameters for these isostructural molecules, for which
the solid-structure of 2 is shown representatively in
Fig. 2. In these spirocyclic complexes one side of the
ligand coordinates the zinc atom in a distorted tetrahe-
dral fashion, the bond angles ranging from 78.22(9)° to
105.6(3)°. While the zinc–nitrogen and zinc–oxygen
bonds are identical in both compounds, the shorter
ethyl–zinc bond of the sulfur analogue may reflect the
lesser steric crowding in this compound. The Zn–S bond
Such asymmetrically-substituted species are readily
accessible from the monochlorocyclodiphosph(III)azane
cis-[ClP(l-NtBu)2P(NHtBu)] (4) [35]. This known com-
pound was synthesized by a simplified method, and its
solid-state structure is presented in Fig. 3. Crystal data
and selected bond parameters of 4 are listed in Tables 1
and 4, respectively.
The monochlorocyclodiphosph(III)azane is
a
P(l-NtBu)2P heterocycle, bearing two almost coplanar
tert-butyl groups and two cis-configured exocyclic
phosphorus substituents. The exocyclic phosphorus
substituents (chloro and tert-butylamino) form obtuse
angles with the heterocycle. As is commonly observed
for these molecules, the lone exocyclic P–N bond is
ꢀ
(2.4480(12) A) is somewhat longer than that in
{[Me2Si(l-NtBu)2P@S(NPh)]2Zn}[33], while the Zn–Se
ꢀ
(2.5398(18) A) bond is almost identical in length to that
ꢀ
in a tetracoordinate zinc complex having two acyclic
amino(seleno)phosphoranate ligands [27].
shorter (1.659(3) A) than the endocyclic ones, but the
latter are decidedly unequal, the bonds to the chlorine-
bearing phosphorus atom being much shorter (1.665(3)
The comparatively small change on going from oxy-
gen to the heavier chalcogens thus caused a rather large
change in the outcomes of these reactions, supporting
our contention that zinc is a good probe for electronic
variations in bis(amino)cyclodiphosph(V)azanes. We
therefore decided to investigate the role of the organic
amino-substituents on the outcomes of these reactions.
It may be recalled that AlMe3 reacted with all ligands,
namely [(RHN)E@P(l-NtBu)2P@E(NHR)], E ¼ O, S,
Se; R ¼ tBu, Ph, under double-deprotonation with the
formation of analogous bimetallic products [13].
ꢀ
vs. 1.744(3) A). Donation of the nitrogen lone-pair into
the rꢃ orbital of the P–Cl bond has lengthened this bond
ꢀ
(2.2047(13) A) substantially beyond the sum of covalent
ꢀ
radii of P and Cl (2.09 A) [36]. This bond is also
Upon changing the amino-substituent from tert-butyl
to phenyl we discovered that the phenyl-substituted di-
amine [(PhHN)S@P(l-NtBu)2P@S(NHPh)] reacted with
one equivalent of diethylzinc to form off-white powders,
whose insolubility made a structural analysis impossible
[34]. Although not conclusive, the insolubility suggests
the formation of a dianionic ligand, involved in bilateral
metal chelation. Because of the contrasting behavior –
tert-butyl-substituted ligands form monomers while
phenyl-substituted ligands form oligomers or polymers –
it was of interest to synthesize a ligand bearing both, a
tert-butylamino and an anilino group.
Fig. 3. Thermal-ellipsoid (35% probability) plot and partial numbering
scheme of 4. Hydrogen atoms have been omitted for clarity.
Table 4
ꢀ
Selected bond lengths (A) and angles (°) for 4 and 5
4
5
Bond lengths
P(1)–Cl(1), N(3)
P(2)–N(3), N(4)
P(1)–N(1)
2.2047(13)
1.659(3)
1.680(3)
1.665(3)
1.752(3)
1.744(3)
–
1.6480(15)
1.6289(15)
1.6743(15)
1.6767(15)
1.7042(14)
1.6902(15)
1.9317(6)
1.9247(6)
P(1)–N(2)
P(2)–N(1)
P(2)–N(2)
P(1)–S(1)
P(2)–S(2)
–
Bond angles
N(1)–P(1)–N(2)
N(1)–P(2)–N(2)
N(3)–P(1)–S(1)
N(4)–P(2)–S(2)
P(1)–N(1)–P(2)
P(1)–N(2)–P(2)
83.34(14)
79.11(13)
–
84.23(7)
82.91(7)
108.80(6)
113.69(6)
95.84(7)
96.28(8)
–
Fig. 2. Thermal-ellipsoid (35% probability) plot and partial numbering
scheme of 2. Compound 3 is isostructural. Hydrogen atoms have been
omitted for clarity.
98.1(2)
98.9(2)

G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
1117
ꢀ
significantly longer than the P–Cl bonds (2.105(9) A) in
the dichloro analogue, cis-[ClP(l-NtBu)2PCl], of this
molecule [37].
substituted phosphorus atom are longer (avg. 1.6972(15)
ꢀ
A vs. 1.6755(15) A).
ꢀ
This molecule was allowed to react with two equiv-
alents of diethylzinc with the intention of obtaining a
dizinc species (Scheme 4), containing metal ions in dif-
ferent chemical environments. The 1H NMR data,
however, showed that the product had the opposite
composition, i.e., it was a monozinc–diligand complex,
namely, {[(tBuHN)S@P(l-NtBu)2P@S(NPh)]2Zn} (6).
An analogous product was obtained by Chivers et al. [6]
when they treated [(tBuHN)Se@P(l-NtBu)2P@Se-
(NHtBu)] with one equivalent of ZnMe2.
The outcome of this reaction thus demonstrates the
much greater reactivity of the anilino- versus the tert-
butylamino-substituted side of the ligand. Because we
recently reported the X-ray structure analysis of the
related {[Me2Si(l-NtBu)2P@S(NPh)]2Zn}[33], which
shows a tetrahedral zinc atom chelated by two
P@S(NPh) moieties, it may be assumed that 6 is a
similar trispirocycle. The structure of 6 is an indication
that the product of the reaction between
[(PhHN)S@P(l-NtBu)2P@S(NHPh)] and diethylzinc is,
indeed, an oligomer or a polymer.
When 4 was treated with one equivalent of lithium
anilide (Scheme 3), the asymmetrically-substituted
cyclodiphosph(III)azane cis-[(tBuHN)P(l-NtBu)2P(NHPh)]
was obtained, which upon oxidation with sulfur fur-
nished the cyclodiphosph(V)azane cis-[(tBuHN) S@P(l-
NtBu)2P@S(NHPh)] (5); its solid-state structure is
shown in Fig. 4. This molecule is a hybrid of
cis-[(tBuHN)S@P(l-NtBu)2P@S(NHtBu)] [4] and cis-
[(PhHN)S@P(l-NtBu)2P@S(NHPh)] [13], whose struc-
tures are known. Crystal and data collection parameters
for 5 are listed in Table 1, while selected bond lengths
and angles are given in Table 4.
It is a common structural feature of bis(amino)dith-
iocyclodiphosph(V)azanes that one amino substituent
has an endo conformation, while the second one retains
its conventional exo conformation. Compound 5 is no
exception, and it comes as no surprise that the smaller
phenyl substituent occupies the more crowded endo
position. The asymmetry of 5 is much less pronounced
than that of 4, as reflected in smaller differences of
similar bonds. Thus, the P–N(H)tBu bond is shorter
Competitive reactions such as this clearly show the
different reactivities of dissimilar coordination sites in
bis(amino)cyclodiphosph(V)azanes and demonstrate the
importance of the amino substituents on the outcomes
of the reactions. Thus, by merely substituting one side of
the cyclodiphosphazane with a tert-butyl group, the
degree of metallation can be controlled and condensa-
tion prevented.
ꢀ
ꢀ
(1.6289(15) A) than the P–N(H)Ph bond (1.6480(15) A),
while the endocyclic P–N bonds of the tert-butyl-
Ligand asymmetry can also be introduced by the
addition of two nonidentical chalcogens or by mixed
(chalcogen/imino) groups on phosphorus. Such asym-
metrically-substituted cyclodiphosph(V)azanes can be
obtained even more easily than those discussed above,
by oxidizing cyclodiphosph(III)azanes sequentially with
two different reagents, as shown in Scheme 5. Thus, the
sequential treatment of [(tBuHN)P(l-NtBu)2P(NHtBu)]
with one equivalent of p-tolyl azide and one equivalent
of elemental sulfur afforded the asymmetrically
Scheme 3.
Fig. 4. Thermal-ellipsoid (35% probability) plot and partial numbering
scheme of 5. Hydrogen atoms have been omitted for clarity.
Scheme 4.

1118
G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
Table 5
ꢀ
Selected bond lengths (A) and angles (°) for 7 ꢀ 1.5(THF) and 8
7 ꢀ 1.5(THF)
8
Bond lengths
Zn(1)–C(1)
Zn(1)–N(1)
Zn(1)–N(2)
P(1)–N(1)
P(1)–N(2)
P(1)–N(3)
P(1)–N(4)
P(2)–N(3)
P(2)–N(4)
P(2)–N(5)
P(2)–S(1)
–
1.946(2)
–
2.0077(16)
1.9954(15)
1.6080(15)
1.5934(15)
1.6834(15)
16885(15)
1.7000(15)
1.6909(15)
1.6350(16)
1.9247(7)
–
1.5391(19)
1.6316(18)
1.6889(18)
1.6895(18)
1.6845(19)
1.6816(18)
1.633(2)
1.9334(8)
Scheme 5.
P-substituted [(tBuNH)S@P(l-NtBu)2P@Np-tol(NHt-
Bu)] (7). Single crystals, suitable for X-ray analysis,
could only be obtained as a sesqui-tetrahydrofuranate of
7, namely 7 ꢀ 1.5(THF).
The solid-state structure of this cyclodiphosph(V)az-
ane in 7 ꢀ 1.5(THF) is shown in Fig. 5, while crystal and
selected bond parameters are listed in Tables 1 and 5,
respectively. In contrast to the bis(amino)cyclo-
diphosph(V)azane dichalcogenides, here both tert-butyl
substituents of the amino nitrogens are in an exo posi-
tion, and the p-tolylimino group is coplanar with the
N2–P2–N1 moiety, suggestive of conjugation effects.
The molecule has three distinctly different P–N bonds,
namely four equidistant, endocyclic P–N bonds (avg.
Bond angles
N(1)–Zn(1)–N(2)
N(1)–Zn(1)–C(1)
N(2)–Zn(1)–C(1)
N(1)–P(1)–N(2)
N(3)–P(1)–N(4)
P(1)–N(3)–P(2)
P(1)–N(4)–P(2)
N(3)–P(2)–N(4)
N(5)–P(2)–S(1)
–
73.83(6)
139.15(8)
146.53(8)
97.36(8)
83.33(7)
96.83(8)
96.99(8)
82.76(7)
116.64(6)
–
–
106.58(10)
82.91(9)
96.82(9)
96.90(9)
83.28(9)
118.98(8)
t
ꢀ
1.689(2) A), two equidistant exocyclic P–N(H) Bu bonds
ꢀ
(avg. 1.634(2) A) and one unique phosphorus–tolyli-
ꢀ
mino bond (1.542(2) A). These phosphorus–nitrogen
bonds, as well as the lone phosphorus–sulfur bond
ꢀ
(1.9334(9) A), have lengths similar to those found in
related bis(amino)cyclodiphosph(V)azanes [4,13].
Because this ligand bears identical tert-butylamino
groups, its acidity and coordinative selectivity now rests
solely on the second exocyclic phosphorus substituents,
namely p-tolylimino and sulfur. While the sulfur side of
the ligand is sterically more accessible, p-electron-delo-
calization is more effective at the amino/imino side of the
ligand, presumably making it more acidic. The treat-
ment of 7 with two equivalents of diethylzinc (Scheme 6)
yielded again only a monometallic, monoligand com-
plex, similar to 2 and 3.
Scheme 6.
Both amino protons can be distinguished by 1H
NMR spectroscopy, and it was therefore evident that
the N,N0, rather than the N,S side, had reacted and
chelated the zinc atom. These solution NMR findings
were confirmed by a single crystal X-ray analysis of 8,
whose data collection and selected bond parameters are
collected in Tables 1 and 5, respectively. The ORTEP
diagram (Fig. 6) shows that compound 8 is a spirocycle
of two four-membered rings, one being terminated by a
trigonal-planar zinc atom. The ligand coordinates the
zinc atom as an almost perfectly-symmetrical
amino(imino)phosphoranate. Thus, in notable contrast
to the distinctly different phosphorus-nitrogen bonds
ꢀ
in the pristine ligand (P1–N1 ¼ 1.540(2) A and P1–
ꢀ
N2 ¼ 1.633(2) A), these bonds have almost identical
ꢀ
lengths (avg. 1.6007(15) A, as do the Zn–N1 and Zn–N2
ꢀ
bonds (avg. 2.0016(16) A. These latter bonds are slightly
shorter than those of the three-coordinate zinc atom in
1, possibly reflecting the superiority of the lateral coor-
dination of these ligands.
Fig. 5. Thermal-ellipsoid (35% probability) plot and partial numbering
scheme of 7. Hydrogen atoms have been omitted for clarity.

G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
1119
The degree of deprotonation of bis(amino)cyclo-
diphosph(V)azanes – and this is no surprise – depends
both on the reactivities of the metallating agents and the
acidity of the bis(amino)cyclodiphosph(V)azanes them-
selves. Thus, with all bis(amino)cyclodiphosph(V)azanes
the more polar and thus more reactive group 1- and
group 13-metal alkyls form bimetallic species. A similar
leveling effect is seen with the most acidic ligands, and
bis(anilino)cyclodiphosph(V)azanes invariably form
dianionic species, no matter what metallating agent is
used [6,12,13].
Fig. 6. Thermal-ellipsoid (35% probability) plot and partial numbering
scheme of 8. Hydrogen atoms have been omitted for clarity.
The less reactive diethylzinc, however, does discrim-
inate between the acidities of the various bis(amino)cy-
clodiphosph(V)azanes. Diethylzinc, for example,
dimetallates only the most acidic bis(tert-butylamino)
cyclodiphosph(V)azane, namely, cis-[(tBuHN) O@P(l-
NtBu)2P@O(NHtBu)] cleanly, while the sulfur and se-
lenium analogues react reluctantly with a second
equivalent of base. The relative reactivities of the vari-
ous bis(amino)cyclodiphosph(V)azanes are summarized
in Chart 3, and they can be interpreted entirely in terms
of the Brønsted acidities of these ligands.
Equally predictable as the degree of deprotonation is
the coordination site preference of mono-anionic li-
gands. In general, and in keeping with results from other
investigators [6,10,11], monoanionic ligands coordinate
metals laterally (Chart 4, Case 1, H), and this can be
understood in terms of the p-electron delocalization
over the N,P,E or N,P,N0 moieties at the sides of the
ligands. For these monoanionic ligands the coordination
site preference follows the acidities of the various sites
shown in Chart 3, thus ligand sides having anilino
groups are preferred over those having tert-butylamino
groups, and N,N0 coordination is preferred over N,S and
N,Se coordination. The N,N0 sites apparently also do-
nate more electron density to the zinc moieties, because
zinc ions are three-coordinate when chelated by two
nitrogen atoms, but four-coordinate when chelated by
N,S and N,Se groups, with THF occupying the fourth
coordination site.
4. Discussion
Despite their comparatively recent dates, studies on
the coordination chemistry of bis(amino)cyclo-
diphosph(V)azanes have revealed clear trends [6], re-
garding both the degree of deprotonation and the
coordination sites of these molecules. With the present
study we have tried to add to our understanding of
coordinative preferences of these ligands by focusing on
ligand-substituent effects.
Bis(amino)cyclodiphosph(V)azanes have four exocy-
clic phosphorus substituents (R1, R2, E1 and E2) which
can be altered independently of each other (Chart 2)
leading to a large variety of possible molecules. To keep
the number of ligands and their complexes manageable
we chose two primary amines, namely, tert-butylamine
and aniline, and four oxidants, namely organic perox-
ides, sulfur, selenium and p-tolyl azide. With these rea-
sonably diverse ligands we hoped to shed some light on
the inherent acidities and coordination preferences of
bis(amido)cylodiphosph(V)azanes.
Monometallic complexes of dianionic ligands exhibit
only one of two coordination modes, with the lone metal
chelated above (Case 2, I) or below the (PN)2 ring (J).
Chart 2.
Chart 3.

1120
G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
Chart 4.
Only one example for each of these modes has been
confirmed, with the respective binding site presumably
being dictated by HSAB considerations. Thus, tin(IV)
and platinum(II) were chelated by both amido or both
sulfur atoms, respectively.
Dianionic bis(amido)cyclodiphosph(V)azanes bear-
ing two metal moieties offer the greatest potential for
structural variety, and this is borne out by the diversity
of the isolated compounds. Five different coordination
modes have been authenticated by X-ray crystallogra-
phy thus far, and these are shown schematically as Case
3 in Chart 4. The by far most common coordination
mode is the bilateral one (K), with the bifacial (L) and
the heterocubic (M) modes being the next most com-
mon. Based on the data from previous studies by Chi-
€
vers [6,9–11], Scherer [39] and Noth [7], it appears that
bis(amido)cyclodiphosph(V)azanes coordinate two
metals laterally (K) as long as the covalent radius of the
ꢀ
metal is less than about 1.28 A. Thus aluminum has
shown exclusively lateral coordination, but metal atoms
with larger covalent radii (Na, K) prefer the sterically
more accessible positions above and below the (PN)2
ring (L). The coordination modes N and O are hybrids
of facial- and lateral-coordinations, and each has been
observed only once thus far. Highly sterically-encum-
bered metal moieties, i.e., metals bearing additional

G.R. Lief et al. / Journal of Organometallic Chemistry 689 (2004) 1110–1121
1121
ligands, seem to prefer the more open lateral coordi-
nation sites (K) because of excessive steric crowding
above the (PN)2 ring, while M has only been observed
in cases where the ligand was formed in situ [38,39].
[14] G.G. Briand, T. Chivers, M. Parvez, Angew. Chem., Int. Ed.
Engl. 41 (2002) 3468.
[15] G.G. Briand, T. Chivers, M. Parvez, G. Schatte, Inorg. Chem. 42
(2003) 525.
[16] G.T. Lawson, C. Jacob, A. Steiner, Eur. J. Inorg. Chem. (1999)
1881.
[17] S.K. Pandey, A. Steiner, H.W. Roesky, D. Stalke, Inorg. Chem.
32 (1993) 5444.
5. Supporting information available
[18] O.J. Scherer, P. Klusmann, Angew. Chem., Int. Ed. Engl. 8 (1969)
752.
Crystallographic data for the structural analyses have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC, Nos. 221020–221026 for com-
pounds 1–5, 7 ꢀ 1.5(THF) and 8, respectively. Copies of
this information may be obtained free of charge from:
The Director, CCDC, 12 Union Road, Cambridge CB2
EZ UK (Fax: +44-1223-336033); deposit@ccdc.cam.ac.
uk or www.http://ccdc.cam.ac.uk.
[19] I. Schranz, L. Stahl, R.J. Staples, Inorg. Chem. 37 (1998)
1493.
[20] R.R. Holmes, J.A. Forstner, Inorg. Chem. 2 (1963) 380.
[21] SMART V 4.043 Software for the CCD Detector System, Bruker
Analytical X-ray Systems, Madison, WI, 1995.
[22] SAINT V 4.035 Software for the CCD Detector System, Bruker
Analytical X-ray Systems, Madison, WI, 1995.
[23] SADABS program for absorption corrections using the Bruker
CCD Detector System. Based on: R. Blessing, Acta Crystallogr. A
51 (1995) 33.
[24] G.M. Sheldrick, SHELXS 90, Program for the Solution of Crystal
Acknowledgements
€
Structures, University of Gottingen, Gottingen, Germany,
€
1990.
[25] G.M. Sheldrick, SHELXL 97, Program for the Solution of Crystal
We thank the Chevron Phillips Chemical Company
for financial support of this research and the National
Science Foundation for a GAANN fellowship (to GRL).
€
it Structures, University of Gottingen, Gottingen, Germany,
€
1997.
[26] SHELXTL 5.10 (PC-Version), Siemens Analytical X-Ray Instru-
ments, Inc. Madison, WI, 1998.
[27] M. Bochmann, G.C. Bwembya, M.B. Hursthouse, S.J. Coles, J.
Chem. Soc., Dalton Trans. (1995) 2813.
References
[28] M. Cheng, A.B. Attygalle, E.B. Lobkovsky, G.W. Coates, J. Am.
Chem. Soc. 121 (1999) 11583.
[1] A.H. Norbury, A.I.P. Sinha, Q. Rev. 5 (1970) 69.
[2] J.L. Burmeister, Coord. Chem. Rev. 3 (1968) 225.
[3] C.-C. Chang, R.C. Haltiwanger, M.L. Thompson, H.-W. Chen,
A.D. Norman, Inorg. Chem. 18 (1979) 1899.
[29] D.J. Darensbourg, M.W. Holtcamp, G.E. Struck, S.A.N. Zim-
mer, P. Rainey, J.B. Robertson, J.D. Draper, J.H. Reibenspies, J.
Am. Chem. Soc. 121 (1999) 107.
[4] T.G. Hill, R.C. Haltiwanger, M.L. Thompson, S.A. Katz, A.D.
Norman, Inorg. Chem. 33 (1994) 1770.
[30] C. Kimblin, B.M. Bridgewater, D.G. Churchill, G. Parkin, Chem.
Commun. (1999) 2301.
[31] J. Prust, A. Stasch, W. Zheng, H.W. Roesky, E. Alexopoulos, I.
[5] M.D. Noirot, E.G. Bent, A.D. Norman, Phosphorus Sulfur 62
(1991) 177.
ꢂ
Uson, D. Bohler, T. Schuchardt, Organometallics 20 (2001)
3825.
€
[6] For a comprehensive review, see: G.G. Briand, T. Chivers, M.
Krahn, Coord. Chem. Rev. 233–234 (2002) 237.
[32] G.R. Lief, D.C. Haagenson, R.J. Staples, L. Stahl, in preparation.
[33] D.C. Haagenson, D.F. Moser, L. Stahl, R.J. Staples, Inorg.
Chem. 41 (2002) 1245.
€
[7] G. Linti, H. Noth, E. Schneider, W. Storch, Chem. Ber. 126 (1993)
619.
[8] T. Chivers, C. Fedorchuk, M. Krahn, M. Parvez, G. Schatte,
Inorg. Chem. 40 (2001) 1936.
[34] G.R. Lief, L. Stahl, unpublished results.
[35] R. Jefferson, J.F. Nixon, T.M. Painter, R. Keat, L. Stobbs, J.
Chem. Soc., Dalton Trans. (1973) 1414.
[9] T. Chivers, M. Krahn, M. Parvez, Chem. Commun. (2000) 463.
[10] T. Chivers, M. Krahn, M. Parvez, G. Schatte, Inorg. Chem. 40
(2001) 2547.
[36] L. Pauling, The Nature of the Chemical Bond, third ed., Cornell
University Press, Ithaca, 1960.
[11] G.G. Briand, T. Chivers, G. Schatte, Inorg. Chem. 41 (2002) 1958.
[12] D.F. Moser, C.J. Carrow, L. Stahl, R.J. Staples, J. Chem. Soc.,
Dalton Trans. (2001) 1246.
[37] K.W.J. Muir, J. Chem. Soc., Dalton Trans. (1975) 259.
[38] T. Chivers, M. Krahn, M. Parvez, G. Schatte, Chem. Commun.
(2001) 2547.
[13] G.R. Lief, C.J. Carrow, L. Stahl, R.J. Staples, Organometallics 20
(2001) 1629.
[39] O.J. Scherer, P. Quintus, J. Kaub, W.S. Sheldrick, Chem. Ber.
(1987) 1183.