Copper(I) halide complexes with 1,3-propanebis(diphenylphosphine) and heterocyclic thione ligands: Crystal and electronic structures (DFT) of [CuCl(pymtH)(dppp)], [CuBr(pymtH)(dppp)], and [Cu(μ-I)(dppp)]2

Article abstract of DOI:10.1021/ic025896i

Reaction of copper(I) chloride or bromide with equimolar amounts of the diphos ligand 1,3-propanebis-(diphenylphosphine) and a heterocyclic thione (L) in acetonitrile/methanol solvent afforded mononuclear complexes of the type [CuX(dppp)(L)] with the diphosphine ligand acting as a chelating ligand. In contrast, copper(I) iodide under the same conditions gave the dimeric complex [Cu(μ-I)(dppp)]₂, which contains doubly bridging iodo ligands. The structures of three complexes, namely, [CuCl(pymtH)(dppp)], [CuCl(pymtH)(dppp)], and [Cu(u-I)(dppp)]₂, have been established by single-crystal X-ray diffraction. Density functional calculations at the B3LYP level of theory provided a satisfactory description of the structural, bonding, electronic, and related properties of the [CuX(PH₃)₂] and [CuX(1,3-pdp)] (1,3-pdp = 1,3-propane-di-phosphine) complexes and their dimers along with their associations with the pyrimidine-2-thione (pymtH) ligand. The interaction of the pymtH ligand with the Cu(I) metal center in these complexes corresponds to loose associations, the computed interaction energies predicted to be about 20 kcal/mol for all complexes in the series. The bonding mechanism of the thione ligand with the Cu(I) metal centers involves both a σ-dative and ρ-back-bonding components. The coordination of the pymtH ligand is further stabilized by X...H-N bond formation being more pronounced in the chloro than in the iodo derivatives. The Cu-X bond was also found to be a composite bond involving σ- and ρ-dative bonding components. Most important is the presence of ρ-type MOs delocalized over the entire four-membered Cu(μ-X)₂Cu ring, which supports a ring current and could probably account for the nearly equivalent Cu-X bonds in the rhombus. Moreover, all [Cu(μ-X)(PH₃)₂]₂ dimers exhibit a σ-type MO corresponding to weak Cu...Cu interactions supporting through-ring intermetallic interactions, which seems to be responsible for the stabilization of the otherwise unstable antiaromatic Cu(μ-X)₂Cu ring.

Full text of DOI:10.1021/ic025896i

Inorg. Chem. 2002, 41, 6875−6886
Copper(I) Halide Complexes with 1,3-Propanebis(diphenylphosphine)
and Heterocyclic Thione Ligands: Crystal and Electronic Structures
(DFT) of [CuCl(pymtH)(dppp)], [CuBr(pymtH)(dppp)], and [Cu(µ-I)(dppp)]2
,†
Paraskevas Aslanidis,* Philip J. Cox,‡,§ Savvas Divanidis,† and Athanassios C. Tsipis†
Inorganic Chemistry Laboratory, Faculty of Chemistry, Aristotle UniVersity of Thessaloniki,
P.O. Box 135, GR-541 24, Thessaloniki, Greece, and School of Pharmacy,
The Robert Gordon UniVersity, Schoolhill, Aberdeen AB 10 1FR, Scotland
Received July 24, 2002
Reaction of copper(I) chloride or bromide with equimolar amounts of the diphos ligand 1,3-propanebis-
(diphenylphosphine) and a heterocyclic thione (L) in acetonitrile/methanol solvent afforded mononuclear complexes
of the type [CuX(dppp)(L)] with the diphosphine ligand acting as a chelating ligand. In contrast, copper(I) iodide
under the same conditions gave the dimeric complex [Cu(µ-I)(dppp)]2, which contains doubly bridging iodo ligands.
The structures of three complexes, namely, [CuCl(pymtH)(dppp)], [CuCl(pymtH)(dppp)], and [Cu(µ-I)(dppp)]2, have
been established by single-crystal X-ray diffraction. Density functional calculations at the B3LYP level of theory
provided a satisfactory description of the structural, bonding, electronic, and related properties of the [CuX(PH ) ]
3 2
and [CuX(1,3-pdp)] (1,3-pdp ) 1,3-propane-di-phosphine) complexes and their dimers along with their associations
with the pyrimidine-2-thione (pymtH) ligand. The interaction of the pymtH ligand with the Cu(I) metal center in
these complexes corresponds to loose associations, the computed interaction energies predicted to be about 20
kcal/mol for all complexes in the series. The bonding mechanism of the thione ligand with the Cu(I) metal centers
involves both a σ-dative and π-back-bonding components. The coordination of the pymtH ligand is further stabilized
by X‚‚‚H−N bond formation being more pronounced in the chloro than in the iodo derivatives. The Cu−X bond
was also found to be a composite bond involving σ- and π-dative bonding components. Most important is the
presence of π-type MOs delocalized over the entire four-membered Cu(µ-X)2Cu ring, which supports a ring current
and could probably account for the nearly equivalent Cu−X bonds in the rhombus. Moreover, all [Cu(µ-X)(PH ) ]
3 2 2
dimers exhibit a σ-type MO corresponding to weak Cu‚‚‚Cu interactions supporting through-ring intermetallic
interactions, which seems to be responsible for the stabilization of the otherwise unstable antiaromatic Cu(µ-X)2Cu
ring.
Introduction
phine ligands. In these complexes, the neutral thione
molecules presenting both nitrogen and sulfur as potentially
donating atoms always adopt the monodentate coordination
mode binding the metal centers exclusively through the
exocyclic sulfur atom in either a terminal or bridging mode.
Nonetheless, a remarkable variety of structures ranging from
mononuclear three- or four-coordinate species with trigonal
planar and tetrahedral Cu(I), respectively, to double bridged
dimers with asymmetric Cu(µ-S)2Cu or Cu(µ-X)2Cu units
and a pseudotetrahedral geometry were obtained even within
Owing to their relevance in biological systems, heterocy-
clic thiones have attracted considerable interest as ligands
in metal complexes.1 Because of their application in catalytic
functions, transition metal complexes with phosphine ligands
have also been studied extensively.2 In a series of studies3
we have explored the mixed-ligand copper(I) halide com-
plexes blending both heterocyclic thione and tertiary arylphos-
* Author to whom correspondence should be addressed. E-mail:
aslanidi@chem.auth.gr. Fax: +30 31 /997738.
† Aristotle University of Thessaloniki.
(2) (a) Cotton, F. A.; Dunbar, K. R.; Verbruggen, M. G. J. Am. Chem.
Soc. 1987, 109, 5498-5506. (b) Puddephatt, R. J. Chem. Soc. ReV.
1983, 12, 99. (c) Kiress, R. V.; Eisenberg, R. Inorg. Chem. 1989 28,
3372-3378.
‡ The Robert Gordon University.
§ E-mail: p.j.cox@rgu.ac.uk.
(1) Raper, E. E. Coord. Chem. ReV. 1994, 129, 91-156.
10.1021/ic025896i CCC: $22.00 © 2002 American Chemical Society
Published on Web 11/20/2002
Inorganic Chemistry, Vol. 41, No. 25, 2002 6875

Aslanidis et al.
a closely related series of thione ligands. In the course of
our investigations we extensively discussed the factors which
affect the stoichiometric and structural preferences of these
complexes. Although the interplay of several parameters like
the steric and electronic requirements of ligands and the
geometrical flexibility on the part of the metal ion make
structural prediction of these complexes extremely difficult,
we found that steric demands of the phosphine ligands are
likely to be the main factor that determines the coordination
number of the metal. In the case of monomers, for example,
triphenylphosphine but also tri-m- and tri-p-tolylphosphine
always form four-coordinate complexes with the general
formula [CuX(PR3)2(thione)], whereas the steric demands of
the o-methyl groups clearly favor three-coordination in the
copper(I) halide complexes of tri-o-tolylphosphine.3d,e On the
other hand, the question why two different types of binuclear
copper(I) halide complexes of general formula [CuX(PR3)-
(thione)]2 exist involving either a double halide or a double
thione-sulfur bridge is still waiting to be answered. On the
basis of our earlier observations on a series of dimers, the
coordination behavior (terminal or bridging mode) of each
ligand has been supposed to depend on the nature of the
halide ligand. In particular, formation of halide bridges has
been found to be favored for the “soft” iodide but not for
the “harder” chloride ligands,4 whereas both bridging and
terminal bonding modes were observed for the bromide
ligands5 since bromide lies on the borderline between soft
and hard bases. However, the information obtained so far is
not enough to establish the above supposition. To obtain a
deeper insight into this very interesting field of research we
have recently initiated a comparative investigation on
substituting the unidentate triarylphosphines with bulkier
multidentate phosphine ligands starting with acyclic bidentate
phosphines of various chain lengths. At first attempt we used
1,2-bis(diphenylphosphino)ethane in an attempt to isolate
chelated products, but this bidentate diphosphine ligand
appeared to be insufficient to span tetrahedral positions, and
it was not possible to isolate any mononuclear complexes.6
In this work we report on the synthesis of a series of the
copper(I) halide complexes formed upon reacting copper(I)
halides with 1,3-propanebis(diphenylphosphine) (dppp) and
a variety of heterocyclic thione ligands, as well as the
structures of [CuCl(pymtH)(dppp)], [CuBr(pymtH)(dppp)],
and [Cu(µ-I)(dppp)]2. Furthermore, to explore the electronic
factors tuning the striking differences in the coordination
behavior between the copper(I) halides, we performed DFT
computational studies pursuing a 3-fold objective: (i) the
determination of the structural, electronic, bonding, and
related properties of the complexes, (ii) the understanding
of the bonding mechanism of the coordination of the pymtH
ligand with Cu(I) metal centers, and (iii) the exploration of
the dimer formation only in the case of the iodide derivatives.
Experimental Section
Materials and Instruments. Copper(I) chloride, copper(I)
bromide, copper(I) iodide, 1,3-bis(diphenylphosphino)propane, and
all solvents are commercially available and were used as obtained,
while the thiones (Merck or Aldrich) were recrystallized from hot
ethanol prior to their use. IR, UV, and 1H NMR spectra, conductivi-
ties, melting points, and elemental analyses of carbon, nitrogen,
and hydrogen were performed as described previously.3b
Computational Details. The structural, electronic, and energetic
properties of all compounds were computed at the Becke’s
3-parameter hybrid functional7 combined with the Lee-Yang-Parr
correlation functional8 level of theory, abbreviated as B3LYP, using
the SDD basis set that includes Dunning/Huzinaga full double-ú
on the first row and Stuttgart/Dresden ECP’s on the copper,
phosphorus, and halogen atoms. The hybrid B3LYP functional was
used, since it gives acceptable results for molecular energies and
geometries, as well as proton donation, and weak and strong
H-bonds.9 In all computations no constraints were imposed on the
geometry. Full geometry optimization was performed for each
structure using Schlegel’s analytical gradient method,10 and the
attainment of the energy minimum was verified by calculating the
vibrational frequencies that result in absence of imaginary eigen-
values. All the stationary points have been identified for minimum
(number of imaginary frequencies NIMAG ) 0) or transition states
(NIMAG ) 1). All calculations were performed using the GAUSS-
IAN98 series of programs.11 Moreover, the qualitative concepts and
the graphs derived from the Chem3D program suite12 highlight the
basic interactions resulted from the DFT calculations.
(7) (a) Becke, A. D. J. Chem. Phys. 1992, 96, 2155-2160. (b) Becke, A.
D. J. Chem. Phys. 1993, 98, 5648-5652.
(8) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. 1988, B37, 785.
(9) (a) Nicholas, J. B. Top. Catal. 1997, 4, 157. (b) Koch, W. R.; Hertwing,
H. Chem. Phys. Lett. 1997, 186, 345-351. (c) Curtis, L. A.;
Raghavachari, K.; Redfern, P. C.; Pople, J. A. Chem. Phys. Lett. 1997,
270, 419-426. (d) Smith, D. M.; Golding, B. T.; Radom, L. J. Am.
Chem. Soc. 1999, 121, 9388-9399. (e) Ghandra, A. K.; Nguyen, M.
T. Chem. Phys. 1998, 232, 299-306. (f) Nicholas, J. B. Top. Catal.
1999, 9, 181. (g) Arnaud, R.; Adamo, C.; Cossi, M.; Millet, A.; Valle´,
Y.; Barone, V. J. Am. Chem. Soc. 2000, 122, 324-330.
(3) (a) Lecomte, C.; Skoulika, S.; Aslanidis, P.; Karagiannidis, P.;
Papastefanou, S. Polyhedron 1989, 8, 1103-1109. (b) Karagiannidis,
P.; Hadjikakou, S. K.; Aslanidis, P.; Hountas, A. Inorg. Chim. Acta
1990, 178, 27-34. (c) Karagiannidis, P.; Aslanidis, P.; Papastefanou,
S.; Mentzafos, D.; Hountas, A.; Terzis, A. Polyhedron 1990, 9, 981-
986. (d) Hadjikakou, S. K.; Aslanidis, P.; Karagiannidis, P.; Mentzafos,
D.; Terzis, A. Inorg. Chim. Acta 1991, 186, 199-204. (e) Hadjikakou,
S. K.; Aslanidis, P.; Karagiannidis, P.; Aubry, A.; Skoulika, S. Inorg.
Chim. Acta 1992, 193, 129-135. (f) Aslanidis, P.; Hadjikakou, S.
K.; Karagiannidis, P.; Gdaniec, M.; Kosturkiewicz, Z. Polyhedron
1993, 12, 2221-2226. (g) Aslanidis, P.; Hadjikakou, S. K.; Karagi-
annidis, P.; Kojic-Prodic B.; Luic, M.; Polyhedron 1994, 13, 3119-
3125.
(4) (a) Hadjikakou, S. K.; Aslanidis, P.; Karagiannidis, P.; Mentzafos D.;.
Terzis, A. Polyhedron 1991, 10, 935-940. (b) Hadjikakou, S. K.;
Aslanidis, P.; Karagiannidis, P.; Hountas, A.; Terzis, A. Inorg. Chim.
Acta 1991, 184, 161-166.
(5) (a) Karagiannidis, P.; Aslanidis, P.; Kessissoglou, D. P.; Krebs, B.;
Dartmann, M. Inorg. Chim. Acta 1989, 156, 47-56. (b) Lecomte, C.;
Skoulika, S.; Aslanidis, P.; Karagiannidis P.; Papastefanou, S.
Polyhedron 1989, 8, 1103-1109.
(6) Cox, P. J.; Aslanidis, P.; Karagiannidis, P. Polyhedron 2000, 19, 1615-
1620.
(10) Schlegel, H. B. J. Comput. Chem. 1982, 3, 214.
(11) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scusseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millan, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone,
V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Orchterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A., Piskorz, P., Komaroni, I.; Gomperts, R.; Martin, R.
L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Gonzalez, C.; Challacombe, M.; Gill, P. M.; Johnson, P.; Chen,
W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J. A. Gaussian 98, revision A.7; Gaussian Inc.: Pittsburgh,
PA, 1998.
(12) ChemOffice 97, Cambridge Scientific Computing Inc., 875 Mas-
sachusetts Ave., Suite 41, Cambridge, MA 02139.
6876 Inorganic Chemistry, Vol. 41, No. 25, 2002

1,3-Propanebis(diphenylphosphine) Complexes
N, 4.49. Found: C, 59.62; H, 4.76; N, 4.57. IR (cm-1): 3055w,
2887w, 1600vs, 1576vs, 1486vs, 1325vs, 1217s, 1175vs, 1091s,
950vs, 822s, 744vs, 696vs, 505vs, 487vs, 469s. UV-vis (λmax, log
Table 1. Crystal Data and Structure Refinements for
[CuCl(pymtH)(dppp)] (1), [CuBr(pymtH)(dppp)] (2), and [Cu(µ-I)
(dppp)]2 (3)
1
1
2
3
ꢀ) (CHCl3): 247.5 (4.11), 295.5 (4.24), 408.5 (3.22). H NMR
(CDCl3, δ ppm): 14.85 (br, 1H, NHpymtH), 7.57 (m, 10H, C6H5),
7.24 (m, 12H, C6H5 + CHpymtH), 6.51 (t, 1H, CHpymtH), 1.25-2.54
(mm, 6H, dppp).
[CuBr(dppp)(pymtH)] (2): red crystals (280 mg, 84%), mp
215-216 °C. Anal. Calcd for C31H30BrCuN2P2S: C, 55.74; H, 4.49;
N, 4.19. Found: C, 55.94; H, 4.62; N, 2.49. IR (cm-1): 3055m,
2851m, 1600vs, 1564vs, 1480vs, 1427vs, 1324vs, 1205s, 1175vs,
chem formula C31H30ClCuN2P2S C31H30BrCuN2P2S C54H52Cu2I2P4
fw
623.56
668.02
602.86
temp/K
λ(Mo KR)/Å
cryst syst
space group
a/Å
b/Å
c/Å
R/deg
â/deg
γ/deg
V/Å3
120(2)
0.71073
triclinic
120(2)
0.71073
triclinic
P1h
9.9021(2)
10.0952 (2)
16.2210 (4)
76.8225 (7)
85.0994 (7)
67.3809 (8)
1457.37 (5)
2
120(2)
0.71073
triclinic
P1h
P1h
9.86140(10)
10.09070 (10)
16.1331 (4)
76.5270 (5)
85.3940 (5)
67.3400 (6)
1440.61 (4)
2
9.9164(2)
10.5308 (2)
12.7852 (3)
74.905 (2)
74.826 (2)
82.5563 (10)
1246.73 (5)
2
1097s, 971s, 739vs, 696vs, 501vs, 490vs, 475vs. UV-vis (λmax
,
1
log ꢀ) (CHCl3): 246 (4.16), 295,5 (4.11), 411.5 (2.47). H NMR
(CDCl3, δ ppm): 14.30 (br, 1H, NHpymtH), 7.56 (m, 10H, C6H5),
7.24 (m, 12H, C6H5 + CHpymtH), 6.52 (t, 1H, CHpymtH), 1.26-2.54
(mm, 6H, dppp).
Z
F
calcd/g cm-3
1.438
1.059
1.522
2.326
1.023
1.606
2.255
0.958
[Cu(µ-I)(dppp)]2 (3): colorless crystals (256 mg, 85%), mp 238
°C. Anal. Calcd for C54H52Cu2I2P4: C, 53.79; H, 4.31. Found: C,
53.75; H, 4.18. IR (cm-1): 3048m, 2920m, 1582s, 1482vs, 1434vs,
1309s, 1185s, 1098vs, 1026s, 964s, 742vs, 694vs, 509vs, 488vs.
UV-vis (λmax, log ꢀ) (CHCl3): 245.0 (3.96), 282 (3.68). 1H NMR
(CDCl3, δ ppm): 7.47 + 7.23 (mm, 20H, C6H5), 1.26-2.44 (mm,
6H, dppp).
µ/mm-1
GOF on F2 (S) 0.979
R [(I > 2σ(I)]
R1 ) 0.0325,
R1 ) 0.0365,
wR2 ) 0.0955
R1 ) 0.0442,
wR2 ) 0.1015
b
R1 ) 0.0299,
wR2 ) 0.0730
R1 ) 0.0365,
wR2 ) 0.0763
c
wR2 ) 0.0834
R1 ) 0.0421,
wR2 ) 0.0881
a
R (all data)
final weighting
scheme
[CuCl(dppp)(py2SH)] (4): yellow crystals (271 mg, 87%), mp
170-171 °C. Anal. Calcd for C32H31ClCuNP2S: C, 61.74; H, 4.98;
N, 2.25. Found: C, 61.80; H, 4.83; N, 2.22. IR (cm-1): 3042w,
2935m, 1618s, 1582vs, 1480s, 1426vs, 1366vs, 1181s, 1133vs,
1097vs, 1001s, 822vs, 744vs, 696vs, 642s, 511vs, 481vs. UV-vis
(λmax, log ꢀ) (CHCl3): 249.5 (4.18), 294 (4.26), 393.5 (4.16);). 1H
NMR (CDCl3, δ ppm): 14.65 (br, 1H, NHpy2SH), 7.56 (m, 10H,
C6H5), 7.40 (d, 1H, CHpy2SH), 7.23 (m, 12H, C6H5 + CHpy2SH), 6.55
(t, 1H, CHpy2SH), 1.71-2.49 (mm, 6H, dppp).
[CuBr(dppp)(py2SH)] (5): yellow crystals (293 mg, 88%), mp
219-220 °C. Anal. Calcd for C32H31BrCuNP2S: C, 57.62; H, 4.65;
N, 2.10. Found: C, 56.92; H, 4.93; N, 2.25. IR (cm-1): 3052w,
293om, 1606vs, 1570s, 1488s, 1424vs, 1366s, 1181s, 1133vs,-
2
2
a Calcd w ) 1/[σ2(Fo ) + (0.0496P)2 + 0.6789P), where P ) (Fo
+
2
2
2
2Fc )/3. b Calcd w ) 1/[σ2(Fo ) + (0.0643P)2 + 0.6153P), where P ) (Fo
+ 2Fc )/3. c Calcd w ) 1/[σ2(Fo ) + (0.0468P)2 + 0.1589P), where P )
2
2
2
2
(Fo + 2Fc )/3.
X-ray Crystallographic Study. X-ray diffraction data were
collected on an Enraf-Nonius Kappa CCD area-detector diffracto-
meter. The programs DENZO13 and COLLECT14 were used in data
collection and cell refinement. Details of crystal and structure
refinement are shown in Table 1. The structure was solved using
program SIR9715 and refined with program SHELX-97.16 Molecular
plots were obtained with the program ORTEP-3.17
Crystallographic data (excluding structure factors) for the
structures reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplementary publica-
tions CCDC-183206 (1), CCDC-183207 (2), and CCDC-183208
(3). Copies of the data can be obtained free of charge on application
to CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K. [Fax,
1097vs, 996vs 750vs, 697vs, 642s, 515vs, 481vs. UV-vis (λmax
,
1
log ꢀ) (CHCl3): 247 (4.11), 294 (2.15), 391.5 (3.64). H NMR
(CDCl3, δ ppm): 14.05 (br, 1H, NHpy2SH), 7.56 (m, 10H, C6H5),
7.46 (d, 1H, CHpy2SH), 7.23 (m, 12H, C6H5 + CHpy2SH), 6.61 (t,
1H, CHpy2SH), 1.69-2.54 (mm, 6H, dppp).
international code
ccdc.cam.ac.uk].
+ 44(1223)336-033; e-mail, deposit@
[CuCl(dppp)(mtdztH)] (6): yellow crystals (264 mg, 82%), mp
216-217 °C. Anal. Calcd for C30H30ClCuN2P2S2: C, 55.98; H,
4.66; N, 2.25. Found: C, 56.19; H, 4.75; N, 2.46. IR (cm-1):
3042w, 1635m, 1570m, 1480s, 1432vs, 1271s, 1161s, 1049s, 738vs,
697vs, 603s, 511vs, 481s. UV-vis (λmax, log ꢀ) (CHCl3): 245
(4.09), 386 (4.24). 1H NMR (CDCl3, δ ppm): 14.20 (br, 1H,
NHmtdztH), 7.26-7.69 (m, 20H, C6H5), 2.42 (s, 3H, CH3-C), 1.57-
2.48 (mm, 6H, dppp).
Synthesis of the Complexes. The complexes were prepared
according to the following general procedure. A suspension of
copper(I) halide (0.5 mmol) and 1,3-propanebis(diphenylphosphine)
(206 mg, 0.5 mmol) in 20 cm3 of acetonitrile was stirred until a
white precipitation was formed. A solution of the appropriate thione
(0.5 mmol) in methanol was then added, and the new suspension
was stirred to clearness. The resulting solution was filtered off,
and the clear solution was kept at room temperature. Slow
evaporation of the solvent at room temperature gave the micro-
crystalline solid, which was filtered off and dried in vacuo.
[CuCl(dppp)(pymtH)] (1): orange crystals (255 mg 82%), mp
218-219 °C. Anal. Calcd for C31H30ClCuN2P2S: C, 59.71; H, 4.85;
[CuBr(dppp)(mtdztH)] (7): yellow crystals (162 mg, 47%), mp
218 °C. Anal. Calcd for C30H30BrCuN2P2S2: C, 52.35; H, 4.36;
N, 4.07. Found: C, 52.07; H, 4.23; N, 3.97. IR (cm-1): 3048w,
2855m, 1618m, 1570m, 1482s, 1459s, 1433vs, 1278s, 1157s, 1097s,
1053s 741vs, 694vs, 511vs, 480s. UV-vis (λmax, log ꢀ) (CHCl3):
1
245 (4.11), 307 (4.09). H NMR (CDCl3, δ ppm): 14.20 (br, 1H,
(13) Otwinowski, Z.; Minor, W. In Macromolecular Crystallography;
Carter, C. W., Jr, Sweet, R. M., Eds.; Methods in Enzymology, Vol.
276; Academic Press: New York, 1997; Part A, pp 307-326.
(14) Hooft, R. COLLECT Data collection software; Nonius B.V.: Delft,
The Netherlands, 1998.
(15) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giaco-
vazzo, C.; Guagliardi A.; Moliterni, A. G. G.; Polidori, G.; Spagna,
R. J. Appl. Crystallogr. 1994, 27, 435-436.
(16) Sheldrick, G. M.; SHELXL-97. Program for crystal structure analysis,
release 97-2; University of Go¨ttingen: Go¨ttingen, Germany, 1997.
(17) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
NHmtdztH), 7.10-7.56 (m, 20H, C6H5), 2.40 (s, 3H, CH3-C), 2.43-
1.57 (mm, 6H, dppp).
[CuCl(dppp)(mftztH)] (8): white crystals (191 mg, 55%), mp
170-171 °C. Anal. Calcd for C31H30ClCuF3N3P2S: C, 53.60; H,
4.32; N, 6.08. Found: C, 53.69; H, 4.47; N, 6.11. IR (cm-1):
3055m, 2900m, 1636m, 1588s, 1480vs, 1432vs, 1283vs, 1199vs,
1151vs, 1091vs, 822s, 744vs, 690vs, 547vs, 511vs, 487vs. UV-
vis (λmax, log ꢀ) (CHCl3): 248 (4.30), 295 (3.88). 1H NMR (CDCl3,
Inorganic Chemistry, Vol. 41, No. 25, 2002 6877

Aslanidis et al.
δ ppm): 12.02 (br, 1H, NHmftztH), 6.93-7.73 (m, 20H, C6H5), 3.54
(s, 3H, CH3-N), 2.62-1.71 (mm, 6H, dppp).
[CuBr(dppp)(mftztH)] (9): white crystals (196 mg, 53%), mp
231 °C. Anal. Calcd for C31H30BrCuF3N3P2S: C, 50.37; H, 4.06;
N, 5.68. Found: C, 50.23; H, 4.60; N, 6.20. IR (cm-1): 3055w,
2899w, 1618m, 1584m, 1480vs, 1432vs, 1157vs, 1270s, 1187s,
1097vs 822s, 738vs, 690vs, 505vs, 503vs, 585s. UV-vis (λmax
,
log ꢀ) (CHCl3): 246 (4.13), 292.5 (3.70). 1H NMR (CDCl3, δ
ppm): 7.05-7.74 (m, 20H, C6H5), 3.46 (s, 3H, CH3N), 1.65-2.78
(mm, 6H, dppp).
[CuCl(dppp)(pu6SH)] (10): pale yellow crystals (206 mg, 62%),
mp 185-186 °C. Anal. Calcd for C32H30ClCuN4P2S: C, 56.38; H,
4.69; N, 8.22. Found: C, 64.56; H, 5.34; N, 7.65. IR (cm-1):
3042m, 2899w, 1612s, 1582s, 1493vs, 1474s, 1426s, 1408s, 1181s,
1097vs, 822s, 744vs, 697vs, 643s, 511vs, 481s. UV-vis (λmax, log
ꢀ) (CHCl3): 245 (3.98), 283.5 (3.92), 396 (2.20). 1H NMR (CDCl3,
δ ppm): 7.06-7.68 (m, 22H, C6H5 + CHpur6SH), 1.60-2.55 (mm,
6H, dppp).
Figure 1. The heterocyclic thione ligands used with their abbreviations.
metal centers. Among the acyclic diphosphines of the type
Ph2P(CH2)nPPh2, we have expected 1,2-ethanebis(diphenyl-
phosphine) (dppe) to be the best canditate for chelating
copper(I). However, dppe appears to act as a bridging rather
than as a chelating ligand, as it was demonstrated at least in
the case of copper(I) bromide.6 Contrary to this previous
observation we succeeded in chelating Cu(I) using the larger
1,3-propanebis(diphenylphosphine) (dppp). It should be noted
that only a few copper(I) complexes containing dppp as a
chelating ligand have been structurally characterized so far,
most of them being polynuclear aggregates, but also two
examples of mononuclear cations of the type [Cu(dppp)2]+.
Treatment of equimolar quantities of copper(I) chloride
or copper(I) bromide and 1,3-propanebis(diphenylphosphine)
(dppp) followed by the addition of 1 equiv of the appropriate
thione (L) [L ) pyridine-2-thione (py2SH), pyrimidine-2-
thione (pymtH), 5-methyl-1,3,4-thiazole-2-thione (mtdztH),
4-methyl-5-trifluoromethyl-4H-1,2,4-triazoline-3(2H)-thi-
one (mftztH), 6-mercaptopurine (pur6SH), benz-1,3-oxazole-
2-thione (bzoxtH), 5-methylbenz-1,3-imidazole-2-thione (mb-
zimtH) (Figure 1)] in dry acetonitrile/methanol solution yields
monomeric mixed ligand complexes of the type [CuX-
(dppp)L] with the neutral heterocyclic thiones acting in an
S-monodentate coordination mode, despite their potentially
ambidentate character. On the contrary, reaction of copper-
(I) iodide under the same conditions proceeds in a different
way to give the dimeric complex [Cu(µ-I)(dppp)]2. Single
crystals of 1-3 suitable for X-ray analysis are obtained from
the initial products deposited from the reactions by slow
evaporation of their acetonitrile solutions. Compounds 1-15
are microcrystalline solids slightly soluble in chloroform,
methanol, ethanol, and acetone and moderately soluble in
acetonitrile. They are stable to air and moisture and can be
manipulated in air without appreciable decomposition. The
composition of the compounds has been confirmed by
elemental analysis. Their solutions in acetone and chloroform
are nonconducting. Room temperature magnetic measure-
ments confirm the diamagnetic nature of the compounds.
Spectroscopy. The electronic absorption spectra of the
complexes in chloroform solutions appear to be mainly of
intraligand character, presenting two broad bands with
maxima in the 245-255 and 280-310 nm regions, respec-
tively. The first one can be attributed to intraligand π* r π
transitions on the phenyl groups of the phosphine ligand,
since the uncoordinated 1,3-propanebis(diphenylphosphine)
reveals a strong absorption at 253.5 nm which usually remain
[CuBr(dppp)(pu6SH)] (11): orange crystals (290 mg, 82%), mp
226-227 °C. Anal. Calcd for C32H30BrCuN4P2S: C, 52.92; H, 4.41;
N, 7.72. Found: C, 52.89; H, 4.38; N, 7.83. IR (cm-1): 3055m,
2910w, 1612vs, 1540vs, 1480s, 1432vs, 1397vs, 1205s, 1094s, 822s,
738vs, 691vs, 649s, 505vs, 475s. UV-vis (λmax, log ꢀ) (CHCl3):
1
245 (4.07), 284.5 (3.99), 397.5 (2.30). H NMR (CDCl3, δ ppm):
7.08-7.75 (m, 22H, C6H5 + CHpur6SH), 1.66-2.52 (mm, 6H, dppp).
[CuCl(dppp)(bzoxtH)] (12): white crystals (295 mg, 89%), mp
159-160 °C. Anal. Calcd for C34H31ClCuNP2OS: C, 61.63; H,
4.45; N, 2.11. Found: C, 59.30; H, 5.68; N, 2.19. IR (cm-1):
3052w, 2886w, 1612m, 1588m, 1504vs, 1438s, 1253vs, 1187s,
1139vs, 1097vs, 929s, 816s, 738vs, 697vs, 511vs, 481s. UV-vis
(λmax, log ꢀ) (CHCl3): 255 (4.14), 298.5 (4.39). 1H NMR (CDCl3,
δ ppm): 13.30 (br, 1H, NHbzoxtH), 7.20-7.72 (m, 24H, C6H5 +
CHbzoxtH), 1.78-2.47 (mm, 6H, dppp).
[CuBr(dppp)(bzoxtH)] (13): white crystals (180 mg (51%), mp
218 °C. Anal. Calcd for C34H31BrCuNP2OS: C, 57.74; H, 4.38;
N, 1.98. Found: C, 58.32; H, 4.64; N, 1.97. IR (cm-1): 3054w,
2887w, 1612s, 1584m, 1504m, 1480vs, 1432vs, 1270s, 1187s,
1097vs, 1139vs, 1097vs, 959s, 822s, 744vs, 690vs, 505vs, 480s.
UV-vis (λmax, log ꢀ) (CHCl3): 245 (4.11), 285.0 (3.65). 1H NMR
(CDCl3, δ ppm): 13.25 (br, 1H, NHbzoxtH), 7.20-7.73 (m, 24H,
C6H5 + CHbzoxtH), 1.56-2.47 (mm, 6H, dppp).
[CuCl(dppp)(mbzimtH)] (14): pale yellow crystals (304 mg,
90%), mp 159-160 °C. Anal. Calcd for C35H34ClCuN2P2S: C,
62.22; H, 5.03; N, 4.34. Found: C, 63.05; H, 5.12; N, 4.42. IR
(cm-1): 3055w, 2885w, 1612s, 1588m, 1480vs,1432vs, 1378s,
1324s, 1175vs, 1097vs, 966s, 798s, 744vs, 697vs, 630s, 547vs,
505vs, 481s. UV-vis (λmax, log ꢀ) (CHCl3): 250 (4.23), 303 (4.38).
1H NMR (CDCl3, δ ppm): 11.71 (br, 2H, NHmbzimtH), 6.85-7.72
(m, 23H, C6H5 + CHmbzimtH), 2.24 (s, 3H, CH3-CPh), 1.71-2.71
(mm, 6H, dppp).
[CuBr(dppp)(mbzimtH)] (15): pale yellow crystals (331 mg,
92%), mp 218 °C. Anal. Calcd for C35H34BrCuN2P2S: C, 57.62;
H, 4.65; N, 2.25. Found: C, 56.02; H, 4.93; N, 2.10. IR (cm-1):
3052w, 2882w, 1612vs, 1586s, 1514vs, 1464vs, 1325vs, 1181vs,
1097vs, 966s, 800vs, 738vs, 696vs, 629s, 544s, 510vs, 482vs. UV-
vis (λmax, log ꢀ) (CHCl3): 246 (4.04), 313.5 (4.20). 1H NMR
(CDCl3, δ ppm): 11.95 (br, 2H, NHmbzimtH), 6.80-7.71 (m, 23H,
C6H5 + CHmbzimtH), 2.25 (s, 3H, CH3-CPh), 1.71-2.59 (mm, 6H,
dppp).
Results and Discussion
Synthesis. Diphosphines have been frequently used for
bridging or chelation of two low oxidation state transition
6878 Inorganic Chemistry, Vol. 41, No. 25, 2002

1,3-Propanebis(diphenylphosphine) Complexes
Table 2. Selected Bond Distances (Å) and Bond Angles (deg) for
[CuCl(pymtH)(dppp)] (1)
Cu(1)-P(1)
Cu(1)-P(2)
Cu(1)-S(1)
Cu(1)-Cl(1)
S(1)-C(1)
2.2698(5)
2.2701(5)
2.2924(5)
2.4071(5)
1.6946(19)
1.8208(19)
P(1)-C(26)
P(1)-C(19)
P(2)-C(5)
P(2)-C(11)
P(2)-C(17)
1.8348(19)
1.8487(18)
1.8253(19)
1.8314(19)
1.8432(19)
P(1)-C(20)
P(1)-Cu(1)-P(2)
P(1)-Cu(1)-S(1)
P(2)-Cu(1)-S(1)
99.649(19) C(26)-P(1)-Cu(1)
124.393(19) C(19)-P(1)-Cu(1)
114.891(19) C(5)-P(2)-C(11)
122.75(6)
107.75(6)
104.55(8)
104.68(9)
P(1)-Cu(1)-Cl(1) 103.256(18) C(5)-P(2)-C(17)
P(2)-Cu(1)-Cl(1)
S(1)-Cu(1)-Cl(1) 112.387(18) C(5)-P(2)-C(1)
98.526(18) C(11)-P(2)-Cu(11) 101.46(8)
116.94(6)
119.81(6)
107.37(6)
113.61(13)
C(1)-S(1)-Cu(1)
112.35(6)
C(11)-P(2)-C(1)
C(17)-P(2)-Cu(1)
C(18)-C(17)-P(2)
C(17)-C(18)-C(19) 116.20(16)
C(18)-C(19)-P(1) 117.79(13)
C(20)-P(1)-C(26) 102.55(8)
C(20)-P(1)-C(19) 105.68(9)
C(26)-P(1)-C(19)
99.25(8)
C(20)-P(1)-Cu(1) 116.50(6)
Figure 2. Thermal ellipsoid drawing of the crystal structure of 1 at the
50% probability level with the labeling scheme.
unshifted upon coordination to Cu(I). The lower energy band
lies in the region where the free thiones absorb, expressing
a small red shift as a consequence of the coordination to
Cu(I) and should be, therefore, considered as a thione-
originating intraligand transition which possess some MLCT
character.18,19 The infrared spectra of compounds 1-15,
recorded in the range 4000-250 cm-1, show all the expected
strong phosphine bands, which remain practically unshifted
upon coordination. Moreover, with the exception of the
spectrum of the iodo compound (3), they contain all of the
bands due to the presence of the heterocyclic thione ligands.
In detail, they exhibit the usual four “thioamide bands” in
the regions ∼1510, 1320, 1000, and 750 cm-1, although in
some cases partly masked by strong phosphine absorptions,
as well as the characteristic NH stretching vibrations
observed in the ∼3150 cm-1 region. The shifts observed for
these bands due to coordination in conjunction with the lack
of ν(SH) bands at ca. 2500-2600 cm-1 signify the exclusive
S-coordination mode of the thione ligands. Room temperature
1H NMR spectra recorded in deuterated chloroform display,
apart from the signals expected for the phosphine and thione
ligands, a single broad resonance at ∼14 ppm attributed to
the NH proton reflecting the prevalence of the thione
tautomer in the complexes.
Description of the Structures. [CuCl(dppp)(pymtH)] (1)
and [CuBr(dppp)(pymtH)] (2). Selected bond lengths and
angles for compounds 1 and 2 are given in Tables 2 and 3,
and perspective drawings showing the atom numbering are
shown in Figures 2 and 3, respectively. In both complexes
copper(I) exhibits a highly distorted tetrahedral coordination
geometry, with the metal atom being surrounded by two P,
one S, and one halide donor atoms. The largest deviation
from the ideal geometry is reflected on the P(1)-Cu(1)-
S(1) bond angles whose values (124.393(19)o in 1 and
125.52(2)o in 2, respectively) are markedly higher than the
tetrahedral value of 109.4°. It is interesting to note that, for
a series of monomeric copper(I) halide complexes involving
one heterocyclic thione and two monodentate triphenylphos-
phine ligands, exhibiting the same CuSP2X core,3,5 the largest
Figure 3. Thermal ellipsoid drawing of the crystal structure of 2 at the
50% probability level with the labeling scheme.
Table 3. Selected Bond Distances (Å) and Bond Angles (deg) for
[CuBr(pymtH)(dppp)] (2)
Cu(1)-P(1)
Cu(1)-P(2)
Cu(1)-S(1)
Cu(1)-Br(1)
S(1)-C(1)
2.2688(6)
2.2715(7)
2.2903(6)
2.5369(4)
1.690(3)
1.820(2)
P(1)-C(26)
P(1)-C(19)
P(2)-C(5)
P(2)-C(11)
P(2)-C(17)
1.835(3)
1.849(2)
1.826(2)
1.834(2)
1.838(2)
P(1)-C(20)
P(1)-Cu(1)-P(2)
P(1)-Cu(1)-S(1)
P(2)-Cu(1)-S(1)
99.59(2)
125.52(2)
114.65(2)
C(26)-P(1)-Cu(1)
C(19)-P(1)-Cu(1)
C(5)-P(2)-C(11)
123.53(8)
107.77(8)
104.38(11)
104.62(11)
101.31(11)
116.71(8)
119.99(8)
107.80(8)
113.71(16)
P(1)-Cu(1)-Br(1) 102.553(19) C(5)-P(2)-C(17)
P(2)-Cu(1)-Br(1) 96.775(19) C(11)-P(2)-C(17)
S(1)-Cu(1)-Br(1) 113.245(19) C(5)-P(2)-Cu(1)
C(1)-S(1)-Cu(1)
113.13(9)
C(11)-P(2)-Cu(1)
C(17)-P(2)-Cu(1)
C(18)-C(17)-P(2)
C(17)-C(18)-C(19) 115.9(2)
C(18)-C(19)-P(1) 118.26(16)
C(20)-P(1)-C(26) 102.61(11)
C(20)-P(1)-C(19) 105.62(11)
C(26)-P(1)-C(19)
C(20)-P(1)-Cu(1) 115.69(8)
99.28(11)
angle was found, as expected, between the two most bulky
phosphine ligands. In the two complexes under investigation,
however, the chelating dppp ligand is coordinated to the
copper(I) center with a P-Cu-P angle of 99.649(19)o for 1
and 99.59(2)o for 2, which is well below what is ideal for a
tetrahedrally coordinated metal ion. The strong intramolecular
(18) Del Paggio, A. A.; McMillin, D. R. Inorg. Chem. 1983, 22, 691-
692.
(19) Kutal, C. Coord. Chem. ReV. 1990, 90, 213-253.
Inorganic Chemistry, Vol. 41, No. 25, 2002 6879

Aslanidis et al.
Table 4. Selected Bond Distances (Å) and Bond Angles (deg) for
[Cu(µ-I)(dppp)]2 (3)
complexes with two iodine atoms serving as bridges between
the two copper atoms, although there are also some repre-
sentatives of mononuclear species with a CuIP2S core among
the structures characterized by us so far.3f,21 This is not
unexpected for a typically soft acceptor such as the mono-
valent copper which strongly prefers soft ligands such as
iodide. It is remarkable that the strong tendency of copper-
(I) to bind iodide ligands together with the formation of stable
six-membered chelate rings results in an exclusion of the
thione ligand from the coordination sphere. The same holds
also true for the 1,5-pentanebis(diphenylphosphine) (dpppe)
ligand, which also reacts with copper(I) iodide to give the
dimeric [Cu(µ-I)(dpppe)]2, even in the presence of 1 equiv
of a heterocyclic thione ligand.22 In contrast in the dppp
complex, each of the two diphosphine ligands of the dpppe
molecule does not chelate at a single copper center but
bridges intramolecularly the two metal centers, probably
because of the much weaker stabilizing effect of the eight-
membered ring. Considering the geometrical parameters of
the dppp ligand one would expect only minor differences
between the two monomeric complexes 1 and 2 and the
dimeric complex 3. The six-membered chelate ring in all
three complexes has the expected chair conformation, while
the distances of the two individual Cu-P bonds are almost
equal in the three complexes and show a small asymmetry
of the same magnitude, with the bite angle of the phosphine
ligand in 3 being enlarged only by ca. 3° when compared
with that in 1 and 2. Within the Cu2I2 backbone, the two
Cu-I bond distances of 2.6470(3) and 2.6823(3) Å are
clearly longer than those found for tetrahedral Cu(I) in
monomeric complexes but only slightly shorter than the
corresponding bond distances observed in bis[µ-iodo(pyri-
dine-2-thione)(tri-m-tolylphosphine)copper(I)].4b Moreover,
there is a quite large asymmetry in the Cu2I2 core of 3 with
regard to these bond distances, much larger than in the above
menthioned mixed-ligand dimer. Surprisingly, the I-Cu-I
and Cu-I-Cu angles, 77.23° and 102.77°, respectively, are
far from the predicted geometry for Y2MX2MY2 dimers
assumed by Summerville and Hoffmann for d10 systems
(Cu-I-Cu ) 94°), but also disagree with the ideal values
(109.5° and 70.5°) calculated for a symmetric dimer.23
Computational Studies Using DFT. Equilibrium Ge-
ometry, Electronic Structure, and Bonding Mechanism
in Mononuclear [CuX(PH3)2] and [CuX(1,3-pdp)] (X )
Cl, Br, or I) Compounds. Let us first explore the details of
the structural, energetic, and electronic properties of the
mononuclear [CuX(PH3)2] and [CuX(1,3-pdp)] compounds
(X ) Cl, Br, I; 1,3-pdp ) 1,3-propylene-diphosphine ligand
formulated as H2PCH2CH2CH2PH2). Selected geometrical
parameters of the [CuX(PH3)2] and [CuX(1,3-pdp)] com-
plexes along with their total electronic energies, dipole
moments, and some electronic properties are given in Tables
5 and 6, respectively. The [CuX(PH3)2] and [CuX(1,3-pdp)]
Cu(1)-I(1)
Cu(1)#1-I(1)a
Cu(1)-P(1)
Cu(1)-P(2)
Cu(1)-I(1)#(1)
P(1)-C(1)
2.6470(3)
2.6823(3)
2.2658(7)
2.2704(7)
2.6823(3)
1.836(3)
P(1)-C(13)
P(2)-C(16)
P(2)-C(22)
P(2)-C(15)
C(13)-C(14)
C(14)-C(15)
1.846(2)
1.832(3)
1.831(3)
1.841(3)
1.537(4)
1.542(4)
P(1)-C(7)
1.822(3)
Cu(1)-I(1)-Cu(1)#1
P(1)-Cu(1)-P(2)
P(1)-Cu(1)-I(1)
P(2)-Cu(1)-I(1)
P(1)-Cu(1)-I(1)#1
P(2)-Cu(1)-I(1)#1
I(1)-Cu(1)-I(1)#1
C(7)-P(1)-C(1)
C(7)-P(1)-C(13)
C(1)-P(1)-C(13)
C(7)-P(1)-Cu(1)
77.231(11) C(1)-P(1)-Cu(1)
125.14(9)
107.17(9)
103.64(11)
103.13(12)
102.73(2)
124.63(2)
112.42(2)
110.94(2)
101.13(2)
C(13)-P(1)-Cu(1)
C(22)-P(2)-C(16)
C(22)-P(2)-C(15)
C(16)-P(2)-Cu(15) 103.00(12)
C(22)-P(2)-Cu(1)
118.40(8)
118.92(8)
107.66(9)
116.23(17)
102.769(10) C(16)-P(2)-Cu(1)
102.35(12)
103.95(12)
98.59(11)
116.50(8)
C(15)-P(2)-Cu(1)
C(14)-C(13)-P(1)
C(13)-C(14)-C(15) 115.1(2)
C(14)-C(16)-P(2)
114.62(17)
a # indicates coordinates transposed by 1 - x, -y, 2 - z.
Figure 4. Thermal ellipsoid drawing of the crystal structure of 3 at the
50% probability level with the labeling scheme.
hydrogen bond between the halogen and the N(1)H can
hardly be responsible for these rather large distortions since
the same bite angle for the six-membered ring was also found
in [Cu(dppp)2]BF4.20 The two individual Cu-P distances are
somewhat shorter than those found in [Cu(dppp)2]BF4 or in
other monomeric copper(I) complexes with chelating diphos-
phines, e.g., in [Cu(dppe)2]ClO4. Finally, the Cu-S, Cu-
Cl, and Cu-Br bond lengths lie in the range normally
observed for tetrahedrally coordinated copper(I) complexes
with terminal chlorine, bromine, and thione-sulfur donors.
[Cu(µ-I)(dppp)]2 (3). Selected interatomic distances and
angles are given in Table 4. This complex consists of discrete
centrosymmetric dimers in which the two copper atoms are
doubly bridged by the iodide ligands to form a four-
membered Cu2I2 ring with a Cu‚‚‚Cu separation of 3.3261-
(6) Å (Figure 4). The strongly distorted tetrahedral coordi-
nation around each copper(I) central atom is completed by
the two P donor atoms of the chelating dppp ligand. It should
be noted that copper(I) iodide usually tends to form dimeric
(21) Aslanidis, P.; Cox, P. J.; Karagiannidis, P.; Hadjikakou, S. K.;
Antoniadis, C. D. Eur. J. Inorg. Chem. 2002, 2216-2222.
(22) Kokkou, S. C.; Voutsas, G.; Aslanidis, P.; Karagiannidis, P. Submitted
to Acta Crystallogr.
(20) Comba, P.; Katsichtis, C.; Nuber B.; Pritzkow, H. Eur. J. Inorg. Chem.
1999, 777-788.
(23) Summerville, R. H.; Hoffmann, R. J. Am. Chem. Soc. 1976, 98, 7240-
7254.
6880 Inorganic Chemistry, Vol. 41, No. 25, 2002

1,3-Propanebis(diphenylphosphine) Complexes
Table 5. Selected Structural Parameters, Total Energies, and Electronic
Properties of CuX(PH3)2 (X ) Cl, Br, I) Compounds Computed at the
B3LYP/SDD Level of Theory
CuCl(PH3)2
CuBr(PH3)2
CuI(PH3)2
E (hartrees)
-1343.79263
2.240
2.324
-896.99300
2.364
2.328
113.2113.3
133.5133.4
7.95
-895.03252
2.532
2.339
113.9
132.2
8.09
2.213
0.51 (0.20)
Re(Cu-X) (Å)
Re(Cu-P) (Å)
θe(P-Cu-X) (deg)
θe(P-Cu-P) (deg)
µe (D)
7.30
η (eV)
2.362
2.329
0.56 (0.24)
a
qCu
0.56 (0.19)b
qX
qP
-0.73 (-0.36) -0.75 (-0.44) -0.70 (-0.40)
-0.07 (-0.11) -0.07 (-0.10) -0.08 (-0.10)
bop(Cu-X)c
bop(Cu-P)
nec(Cu)d
nec(X)
0.362
0.302
0.314
0.020
0.042
0.044
4s0.493d9.88
4s1.973d5.76
4s0.493d9.90
4s1.973d5.78
4s0.533d9.91
4s1.973d5.73
a Natural charges. b Figures in parentheses are the Mulliken net atomic
charges. c Mulliken bond overlap population. d Natural electron configura-
tion.
Table 6. Selected Structural Parameters, Total Energies, and Electronic
Properties of CuX(1,3-pdp) (X ) Cl, Br, I) Compounds Computed at
the B3LYP/SDD Level of Theory
CuCl(1,3-pdp) CuBr(1,3-pdp)
CuI(1,3-pdp)
E (hartrees)
-1360.44684
2.204
2.355
-1013.64676
2.326
2.365
131.5
96.8
11.46
2.177
0.56 (0.14)
-1010.68645
2.496
2.366
131.3
97.1
11.65
2.079
0.51 (0.10)
Figure 5. Equilibrium structures of the CuX(PH3)2 (a), CuX(1,3-pdp) (b),
and [Cu(µ-X)(PH3)2]2 (c) complexes computed at the B3LYP/SDD level
of theory.
Re(Cu-X) (Å)
Re(Cu-P) (Å)
θe(P-Cu-X) (deg) 131.2
θe(P-Cu-P) (deg) 97.3
only with respect to the chloro derivative, the elongation
being 0.011 Å. On the other hand, the Cu-X bond lengths
are affected by the phosphine ligand. Interestingly, the Cu-
Cl, Cu-Br, and Cu-I bonds in [CuX(PH3)2] complexes are
lengthened by about 0.036 Å compared to the corresponding
bonds of the [CuX(1,3-pdp)] complexes. Moreover, the
Cu-X bond lengths in both series of complexes following
the expected trend Cu-Cl < Cu-Br < Cu-I are in excellent
agreement with available experimental data.24 The experi-
mentally determined Cu-I bond length of 2.591 Å in the
analogous [CuI(PCy3)2] complex compares well with the
computed Cu-I bond length of the [CuI(PH3)2] compound.12a
The observed elongation of the Cu-I bond (by 0.059 Å) in
the [CuI(PCy3)2] complex is the result of the stronger
inductive effect of the PCy3 than the PH3 ligand (PCy3 is a
stronger σ donor than PH3). Noteworthy is the excellent
agreement of the computed P-Cu-P bond angle in the
[CuI(PH3)2] compound (132.2°) and the experimentally
determined one in the the [CuI(PCy3)2] complex (134.06°)
in spite of the bulky PCy3 ligands. The experimentally
determined P-Cu-P bond angles in [CuI(PPh3)2] and
[CuI(PPh2)(o-tolyl)] compounds12b,c are 126.9° and 126.4°,
respectively. The ν(Cu-X) stretching vibrational frequencies
appearing at 320 (351) cm-1, 238 (274) cm-1, and 182 (198)
cm-1, for the chloro, bromo, and iodo derivatives of the
[CuX(PH3)2] and [CuX(1,3-pdp)] complexes, respectively,
are compatible with the computed Cu-X bond lengths.
µe (D)
η (eV)
10.77
2.195
a
qCu
0.55(0.05)b
qX
qP
-0.70 (-0.38) -0.73 (-0.46) -0.68 (-0.43)
0.15 (0.16)
0.341
0.15 (0.15)
0.273
0.15 (0.15)
0.275
bop(Cu-X)c
bop(Cu-P)
nec(Cu)d
nec(X)
0.110
0.110
0.108
4s0.513d9.87
4s1.963d5.74
4s0.513d9.89
4s1.963d5.76
4s0.553d9.90
4s1.963d5.71
a Natural charges. b Figures in parentheses are the Mulliken net atomic
charges. c Mulliken bond overlap population. d Natural electron configura-
tion.
complexes are three-coordinated with a trigonal planar
coordination environment shown in Figure 5. In the [CuX-
(PH3)2] complexes the three donor and the copper central
atoms are found in a plane. However, in the [CuX(1,3-pdp)]
analogues there is a deviation of the copper(I) atom from
the plane defined by the three donor atoms amounting to
0.026, 0.026, and 0.019 Å for the chloro, bromo, and iodo
derivatives, respectively. The metallacycle chelate ring adopts
a chair configuration (Figure 5b) with a P-Cu-P bond
angle of about 97° and exhibits a symmetry plane defined
by X, Cu, and the central C atom of the bidentate pdp ligand.
This bond angle is larger by about 36° in the [CuX(PH3)2]
complexes involving the unidentate phosphine ligand. It can
also be seen that there are no significant structural changes
in [CuX(PH3)2] and [CuX(1,3-pdp)] complexes due to the
influence of the identity of the halide ligand X. Noteworthy
is the small elongation of the Cu-P bond in the iodo
compound by 0.015 and 0.011 Å with respect to the chloro
and bromo analogues, respectively, followed by a concomi-
tant narrowing of the P-Cu-P bond angle by about 1.2°.
A small elongation of the Cu-P bond is also observed in
the iodo derivative of the [CuX(1,3-pdp)] complexes, but
(24) (a) Soloveichik, G. L.; Eisenstein, O.; Poulton, J. T.; Streib, W. E.;
Huffman, J. C.; Caulton, K. G. Inorg. Chem. 1992, 31, 3306-3312.
(b) Bowmaker, G. A.;. Dyason, J. C.; Healy, P. C.; Engelhardt, L.
M.; Pakauatchai, C.; White, A. H. J. Chem. Soc., Dalton Trans. 1987,
1089-1091. (c) Bowmaker, G. A.; Engelhardt, L. M.; Healy, P. C.;
Kildea, J. D.; Papasergio, R. I.; White, A. H. Inorg. Chem. 1987, 26,
3533-3538.
Inorganic Chemistry, Vol. 41, No. 25, 2002 6881

Aslanidis et al.
Table 7. Selected Structural Parameters, Total Energies, and Electronic
Properties of [Cu(µ-X)(PH3)2]2 (X ) Cl, Br, I) Dimers Computed at the
B3LYP/SDD Level of Theory
[Cu(µ-Cl)-
(PH3)2]2
[Cu(µ-Br)-
(PH3)2]2
[Cu(µ-I)-
(PH3)2]2
E (hartrees)
Re(Cu-X) (Å)
-1343.85528 -1794.01037 -1790.08777
2.464
2.600 (2.580)
2.570 (2.590)a
2.345
3.094
106.5
2.745
Re(Cu-P) (Å)
2.339
3.130
101.1
2.369
2.895
116.4
Re(Cu‚‚‚Cu) (Å)
θe(X-Cu-X) (deg)
θe(Cu-X-Cu) (deg) 78.9
73.7 (73.3)
126.1
63.7
122.2
θe(P-Cu-P) (deg)
127.0
τ(P-P-Cu-X) (deg) 118.5
122.0 (125.2)
0.06
126.2 (126.5)
0.02
µe (D)
η (eV)
0.01
Figure 6. Representative FMOs of the CuX(PH3)2 (a) and CuX(1,3-pdp)
(b) molecules, along with the MOs describing the Cu-X interactions.
2.264
2.213
0.57 (0.29)
2.167
0.49 (0.23)
a
qCu
0.58 (0.28)c
qX
qP
-0.72 (-0.35) -0.73 (-0.40) -0.64 (-0.35)
Finally, the computed dipole moments of the [CuX(PH3)2]
and [CuX(1,3-pdp)] complexes illustrate that their polarity
follows the order I > Br > Cl.
-0.05 (-0.12) -0.05 (-0.11) -0.06 (-0.10)
bop(Cu-X)d
bop(Cu-P)
bop(Cu‚‚‚Cu)
nec(Cu)e
nec(X)
0.188
0.157
0.159
-0.072
-0.020
-0.017
-0.367
-0.107
-0.366
4s0.463d9.91
4s1.963d5.76
4s0.473d9.91
4s1.963d5.76
4s0.533d9.93
4s1.963d5.68
The population analysis (both Mulliken and natural bond
orbital, NBO) indicated that the Cu(I) metal center upon
coordination accepts electron density from both the halide
and phosphine ligands, thus acquiring a positive natural
charge of about 0.5-0.6 charge unit (0.19-0.24 and 0.05-
0.14 charge unit from the Mulliken population analysis of
the wave functions of the [CuX(PH3)2] and [CuX(1,3-pdp)]
complexes, respectively). Notice that the natural electron
configuration (nec) on the Cu central atom and the halide
ligands assigned by the NBO analysis is not affected
significantly by the nature of the halide and phosphine
ligands. Interestingly, the computed bond overlap populations
(bop) of the Cu-X and Cu-P bonds do not correlate with
the computed bond lengths; thereby the metal-ligand
interactions in the [CuX(PH3)2] and [CuX(1,3-pdp)] com-
plexes should exhibit a strong covalent component. This is
substantiated by the respective molecular orbitals (MOs)
describing the corresponding coordination bonds. Represen-
tative frontier molecular orbitals (FMOs) of the [CuX(PH3)2]
and [CuX(1,3-pdp)] molecules, along with the MOs describ-
ing the Cu-X interactions, are depicted schematically in
Figure 6. It can be seen that the Cu-X bond is a composite
bond involving both σ and π components, thus acquiring a
partially double bond character. The highest occupied MO
(HOMO) corresponding to a π* antibonding Cu-X interac-
tion is strongly polarized toward the halide ligand. On the
other hand, the lowest unoccupied MO (LUMO) corresponds
to a delocalized MO over the H atoms of the phosphine
ligands with a small contribution from a d orbital of the Cu-
(I) central atom, as well. On the basis of the nature of the
FMOs it can be envisaged that all monomers could be
dimerized through the favorable HOMO-LUMO interac-
tions. The tendency for dimerization is expected to be greater
for the iodo derivative exhibiting the lower HOMO-LUMO
energy gap (less hardness) than for the chloro and bromo
derivatives in both series of complexes. Compare the
hardness values η of the [CuX(PH3)2] and [CuX(1,3-pdp)]
complexes shown in Tables 5 and 6, respectively. The
hardness η was defined by Parr and Pearson25 as η ) 1/2(IP
- EA) ) 1/2(ꢀLUMO - ꢀHOMO), where IP and EA are the first
vertical ionization potential and electron affinity of the
a These four values correspond to four inequivalent Cu-Br bonds.
b Natural charges. c Figures in parentheses are the Mulliken net atomic
charges. d Mulliken bond overlap population. e Natural electron configura-
tion.
species, which in the framework of Koopman’s theorem can
be approximated by ꢀHOMO and ꢀLUMO with opposite sign,
respectively.
Equilibrium Geometry, Electronic Structure, and Bond-
ing Mechanism in Binuclear [Cu(µ-X)(PH3)2]2 (X ) Cl,
Br, or I) Model Compounds. To obtain a computationally
convenient size for the dinuclear [Cu(µ-X)(1,3-dppp)]2 (X
) Cl, Br, or I) compounds, we used only model compounds
involving the unsubstituted phosphine ligands. The use of
such models does not alter significantly the description of
the “core” region of the compounds and is ultimately the
most efficient and productive route to modeling the electronic
structure and related properties of the real large-sized
dinuclear [Cu(µ-X)(1,3-dppp)]2 complexes. The [Cu(µ-X)-
(PH3)2]2 dimers are predicted to be thermodynamically more
stable than the monomers, the computed heats ∆RH of the
dimerization processes being equal to 20.9, 15.3, and 14.3
kcal mol-1, for the chloro, bromo, and iodo derivatives,
respectively, at the B3LYP/SDD level of theory. Selected
geometrical parameters of the [Cu(µ-X)(PH3)2]2 dimers along
with their total electronic energies, dipole moments and some
electronic properties are given in Table 7. The equilibrium
geometry of the [Cu(µ-X)(PH3)2]2 dimers, except of the
bromo derivative, exhibiting D2h symmetry, corresponds to
a diamond-like structure (Figure 5c) with the Cu-X bond
distances defining the Cu(µ-X)2Cu rhombus being equivalent.
Surprisingly, in the bromo derivative there are no equivalent
Cu-Br bond distances in the Cu(µ-Br)2Cu rhombus, the Cu-
Br bond lengths being 2.600, 2.580, 2.590, and 2.570 Å.
The Cu-X bond lengths in the dimers are longer by about
0.22 Å relative to the corresponding bonds of the monomers.
The intermetallic Cu‚‚‚Cu distance is tuned by the identity
(25) (a) Parr, R. G.; Pearson, R. G. J. Am. Chem. Soc. 1983, 105, 7512-
7516. (b) Pearson, R. G. J. Chem. Educ. 1987, 64, 561-565. Pearson,
R. G. Acc. Chem. Res. 1993, 26, 250-255.
6882 Inorganic Chemistry, Vol. 41, No. 25, 2002

1,3-Propanebis(diphenylphosphine) Complexes
of the halide ligand, X, following paradoxically the trend
Cu(µ-I)Cu < Cu(µ-Br)Cu < Cu(µ-Cl)Cu. The shortest Cu‚‚‚
Cu distance was observed for the best electron-donating
bridged iodide ligand. A similar observation has previously
been made by Soloveichik et al.12a for the [LnCuI(µ-X)CuILn]
(n ) 1 or 2, L ) PR3) complexes. The variation in the Cu‚
‚‚Cu distance in these complexes was interpreted in the
framework of the EHMO theory in terms of a direct electron
density transfer from the orbitals of the bridging halide atoms,
X, into the 4s and 4p orbitals of Cu, which is traced to weak
σ and π Cu‚‚‚Cu interactions. More recently, Alvarez et al.26
have analyzed the effect of the halide ligand on Cu-Cu
bonding in a series of complexes [CuXL]2 (X ) Cl, Br, I; L
) NH3, PH3,CNMe) using MP2 and B3LYP computational
techniques. They found that the Cu-Cu distance decreases
upon increasing the size of the bridging halide ligand. The
computed X-Cu-X bond angles found in the range
101.4-116.4° increases by 5.4° and 9.9° along the series of
the chloro, bromo, and iodo derivatives, respectively. On the
other hand, the Cu-X-Cu bond angles found in the range
78.9-63.7° follow the reverse order, decreasing by 5.4° and
9.8° along the series of the chloro, bromo, and iodo
derivatives, respectively. Finally, the τ(P-P-Cu-X) dihe-
dral angle defined by the P-Cu-P and X-Cu-X planes in
the tetrahedral coordination sphere of each Cu(I) central atom
found in the range 118.5-126.5° increases successively by
about 4° along the series of the chloro, bromo, and iodo
derivatives. The computed νsym(X-Cu-X) stretching vibra-
tional frequencies appearing at 180, 125, and 95 cm-1, for
the [Cu(µ-Cl)(PH3)2]2, [Cu(µ-Br)(PH3)2]2, and [Cu(µ-I)-
(PH3)2]2 complexes, respectively, are compatible with the
elongated Cu-X bonds in the dimers. The same holds true
also for the νsym(Cu-X-Cu) stretching vibrational frequen-
cies appearing at 225, 146, and 136 cm-1, for the chloro,
bromo, and iodo derivatives of the dimers, respectively.
Obviously, both these bands could be used as a guide for
distinguishing the dimers from their monomers.
Interestingly the dimerization of CuX(PH3)2 monomers
does not tune the natural and net atomic charges on the Cu-
(I) metal centers and the bridging halide ligands as well. The
Cu(I) metal centers acquire a positive natural charge of about
0.5-0.6 charge unit (0.2-0.3 charge unit of Mulliken net
atomic charge), while the bridging halide ligands acquire a
negative natural (Mulliken net atomic) charge of -0.70
(-0.36) to -0.75 (-0.44) charge unit. Notice that the natural
electron configuration on the Cu(I) central atom and the
halide ligands assigned by the NBO analysis (Table 7) is
not affected significantly by the nature of the halide ligands.
The most noticeable differences concern the iodo derivative
and could probably be accounted for by the larger overlap
between the orbitals used to built the Cu-X bonds in the
rhombus of the [Cu(µ-I)(PH3)2]2 dimer compared to that of
the other counterparts.
Figure 7. The most important MOs of the [Cu(µ-X)(PH3)2]2 dimers:
HOMO, π- and σ-type MOs (a), σ-type Cu-Cu MO of the chloro (b),
bromo (c), and iodo derivatives (d), and π-type MO of C4H4 (e).
bromo and to iodo derivatives. This is consistent with the
higher stability of the [Cu(µ-I)(PH3)2]2 dimer in the series.
The most important MOs of the [Cu(µ-X)(PH3)2]2 dimers
are collected in Figure 7. Perusal of Figure 7 reveals that
the Cu(µ-X)Cu moiety exhibits a composite bonding mode
involving both σ and π components. The HOMO corre-
sponding to a π* antibonding Cu-X interaction is strongly
polarized toward the bridging halide ligands. On the other
hand, the LUMO corresponds to a delocalized MO over the
H atoms of the phospine ligands with a small contribution
from a d orbital of the Cu(I) central atoms, as well.
Noteworthy is the presence of π-type MOs delocalized over
the entire four-membered Cu(µ-X)2Cu ring similar to the
π-type MOs of the cyclobutadiene molecule, which support
a ring current. However, the Cu(µ-X)2Cu ring, having a total
of 8 framework electrons, does not obey the (4n + 2) Hu¨ckel
rule to be an aromatic systems and thereby corresponds to
an antiaromatic system. The delocalized π electron density
in the Cu(µ-X)2Cu rhombus could probably account for the
nearly equivalent Cu-X bonds in the rhombus. Surprisingly,
all [Cu(µ-X)(PH3)2]2 dimers exhibit a σ-type MO corre-
sponding to weak intermetallic Cu‚‚‚Cu interactions (Figure
7), which seems to be inconsistent with the relatively long
Cu‚‚‚Cu separation distances and the FEC (framework
electron counting) rule introduced by Alvarez et al.27
According to this rule it is the number of framework electrons
that account for the through-ring bonding in edge-sharing
dimers of square planar complexes. Thus, a short through-
ring distance is predicted for compounds with FEC ) 4 or
6, but a distance that is not short is expected for those with
FEC ) 8. Notice that the electron density in the region
between the two bonded Cu atoms increases successively
along the series of the chloro, bromo, and iodo derivatives
in line with the computed Cu‚‚‚Cu separation distances. On
All [Cu(µ-X)(PH3)2]2 dimers exhibit a similar pattern of
molecular orbital level diagram with a successive lowering
of the respective eigenvalues on going from the chloro to
(26) Carvajal, A.; Liu, X.-Y.; Alemany, P.; Novoa J. J.; Alvarez, S. Int. J.
Quantum Chem. 2002, 86, 100-105.
(27) Alemany, P.; Alvarez, S. Inorg. Chem. 1992, 31, 4266-4275.
Inorganic Chemistry, Vol. 41, No. 25, 2002 6883

Aslanidis et al.
Table 8. Selected Structural Parameters, Total Energies, and Electronic
Properties of CuX(PH3)2(pymtH) (X ) Cl, Br, I) Compounds Computed
at the B3LYP/SDD Level of Theory
CuCl(PH3)2-
(pymtH)
CuBr(PH3)2-
(pymtH)
CuI(PH3)2-
(pymtH)
E (hartrees)
-2006.18170 -1559.37896 -1557.41601
Re(Cu-X) (Å)
2.417
2.377
2.441
113.0
102.7
108.9
120.2
2.550
2.383
2.435
114.9
103.5
108.0
119.2
114.1
1.33
2.731
2.386
2.431
118.9
101.9
108.1
118.5
110.7
2.04
Re(Cu-P) (Å)
Re(Cu-S) (Å)
θe(S-Cu-X) (deg)
θe(P-Cu-X) (deg)
θe(P-Cu-S) (deg)
θe(P-Cu-P) (deg)
τ(P-P-Cu-X) (deg) 113.5
µe (D)
η (eV)
1.13
1.614
1.619
0.55 (0.17)
1.537
0.52 (0.14)
a
qCu
0.55 (0.15)b
qX
qS
-0.72 (-0.44) -0.74 (-0.51) -0.71 (-0.50)
-0.26 (0.08)
0.235
-0.26 (0.08)
0.166
-0.25 (0.09)
0.166
bop(Cu-X)c
bop(Cu-S)
nec(Cu)d
nec(X)
nec(S)
0.224
0.245
0.244
4s0.493d9.88
4s1.973d5.76
4s1.973d5.76
4s0.493d9.90
4s1.973d5.78
4s1.973d5.78
4s0.533d9.91
4s1.973d5.73
4s1.973d5.73
a Natural charges. b Figures in parentheses are the Mulliken net atomic
charges. c Mulliken bond overlap population. d Natural electron configura-
tion.
Figure 8. Equilibrium structures of the CuX(1,3-pdp)(pymtH) complexes
computed at the B3LYP/SDD level of theory.
Table 9. Selected Structural Parameters, Total Energies, and Electronic
Properties of CuX(1,3-pdp)(pymtH) (X ) Cl, Br, I) Compounds
Computed at the B3LYP/SDD Level of Theory
the other hand the computed negative bop values (Table 7)
for the Cu‚‚‚Cu interactions are indicative of antibonding
interactions, but this might be an artifact of the Mulliken
population analysis with its well-known shortcomings. In
summary, the present computational results indicate that
weak intermetallic Cu‚‚‚Cu interactions are indeed present
in [Cu(µ-X)(PH3)2]2 dimers probably as a correlation effect,
in the framework of the B3LYP level of theory, being
responsible for the stabilization of the otherwise unstable
antiaromatic Cu(µ-X)2Cu ring.
Equilibrium Geometry, Electronic Structure, and Bond-
ing Mechanism in [CuX(PH3)2(pymtH)] and [CuX(1,3-
pdp)(pymtH)] (X ) Cl, Br or I) Compounds. The details
of the structural, energetic, and electronic properties of the
[CuX(PH3)2(pymtH)] and [CuX(1,3-pdp)(pymtH)] (X ) Cl,
Br, or I; pymtH ) pyrimidine-2-thione) complexes have also
been explored at the B3LYP/SDD level of theory. Repre-
sentative fully optimized equilibrium geometries of the [CuX-
(1,3-pdp)(pymtH)] compounds along with the atom-labeling
scheme are shown in Figure 8. Selected structural parameters
for the [CuX(PH3)2(pymtH)] and [CuX(1,3-pdp)(pymtH)]
compounds along with their total electronic energies, dipole
moments, and some electronic properties are compiled in
Tables 8 and 9, respectively. In general terms the optimized
geometries of the [CuX(1,3-pdp)(pymtH)] compounds re-
semble those determined experimentally by X-ray crystal-
lography. All these compounds involve a four-coordinated
Cu(I) central atom in a tetrahedral coordination environment
with the plane of the thione ring coinciding with the line
bisecting the P-Cu-P angle. Interestingly, the conforma-
tion of the thione ligand is further stabilized through an X‚‚‚
H interaction (H-bond formation). The computed X‚‚‚H
distances in CuX(PH3)2(pymtH) compounds are 2.026, 2.200,
and 2.451 Å for the chloro, bromo, and iodo derivatives,
respectively, whereas those in [CuX(1,3-pdp)(pymtH)] com-
CuCl(1,3-pdp)- CuBr(1,3-pdp)- CuI(1,3-pdp)-
(pymtH)
(pymtH)
(pymtH)
E (hartrees)
-2123.05869
2.408
2.386
2.408
114.0
107.1
115.1
96.9
-1676.25633
2.556
2.390
2.395
116.6
104.5
116.0
96.8
106.9
-1674.29264
2.722
2.395
2.405
119.4
105.9
113.1
96.8
108.7
Re(Cu-X) (Å)
Re(Cu-P) (Å)
Re(Cu-S) (Å)
θe(S-Cu-X) (deg)
θe(P-Cu-X) (deg)
θe(P-Cu-S) (deg)
θe(P-Cu-P) (deg)
τ(P-P-Cu-X) (deg) 110.3
µe (D)
η (eV)
1.60
1.77
1.533
0.55 (0.04)
2.32
1.456
0.52 (0.02)
1.538
a
qCu
0.56 (0.01)b
qX
qS
-0.72 (-0.45) -0.74 (-0.52) -0.71 (-0.51)
-0.25 (0.07)
0.231
-0.24 (0.09)
0.141
-0.24 (0.09)
0.146
bop(Cu-X)c
bop(Cu-S)
nec(Cu)d
nec(X)
nec(S)
0.217
0.239
0.237
4s0.493d9.90
4s1.963d5.76
4s1.773d4.46
4s0.503d9.90
4s1.973d5.77
4s1.773d4.46
4s0.523d9.91
4s1.973d5.74
4s1.773d4.46
a Natural charges. b Figures in parentheses are the Mulliken net atomic
charges. c Mulliken bond overlap population. d Natural electron configura-
tion.
pounds are 2.024, 2.200, and 2.405 Å. The H-bond formation
is also reflected in the computed bop values for the respective
interactions as well as in the existence of a molecular orbital
having a strong σ component corresponding to the X‚‚‚H
interactions (Figure 9). The strength of the H-bonds formed
is tuned by the identity of the halide ligand following the
expected order: Cl > Br > I. The H-bond formation
introduces a small lengthening of the N-H bond in the
pymtH ligand, its length being 1.054, 1.048, and 1.043 Å
for the chloro, bromo, and iodo derivatives, respectively. The
X‚‚‚H-N moiety is nearly linear, with the computed X‚‚‚
H-N bond angle found in the range 172.7-176.6°. In both
series of complexes the linearity of the X‚‚‚H-N moiety
6884 Inorganic Chemistry, Vol. 41, No. 25, 2002

1,3-Propanebis(diphenylphosphine) Complexes
Figure 10. The HOMO and LUMO of the pymtH ligand (a) along the
MOs corresponding to the π bonding component of the Cu-S (b) and Cu-X
(c) interactions.
derivatives in conjunction with the weaker H-bond formation
could account well for the preference of the iodo monomers
to form dimers rather than to be coordinated with the thione
ligand.
Figure 9. The σ-type MOs corresponding to the X‚‚‚H-N interactions in
the chloro (a), bromo (b), and iodo derivatives (c) of the CuX(1,3-pdp)-
(pymtH) complexes.
The Cu-X bond also exhibits a partially double bond
character analogous to that found in the [CuX(PH3)2] and
[CuX(1,3-pdp)] monomers. A representative π-type MO
localized on the Cu-X bond is shown in Figure 7c. It can
be seen that the Cu-X bonds are elongated by approximately
0.2 Å upon coordination of the thione ligand to the [CuX-
(PH3)2] and [CuX(1,3-pdp)] complexes. The elongation of
the Cu-X bond is reflected on the computed bond overlap
populations found in the range 0.141-0.235 in the complexes
as compared to the bop(Cu-X) values of the [CuX(PH3)2]
and [CuX(1,3-pdp)] species (Tables 5 and 6). Similarly, the
elongation of the CdS bond is reflected on the computed
bond overlap populations found in the range 0.269-0.285
in the complexes as compared to the bop(C-S) value of
0.422 in the free thione ligand. Noteworthy is the transfer
of electron density from the S donor atom of the thione ligand
to the halide ligand, assisted by the Cu(I) central atom.
Overall, there are no remarkable changes in both the natural
and net atomic charges of the donor (P, X) and central (Cu)
atoms in the [CuX(PH3)2(pymtH)] and [CuX(1,3-pdp)-
(pymtH)] compounds as compared to those of the precursor
[CuX(PH3)2] and [CuX(1,3-pdp)] complexes. The same holds
true also for the natural electron configurations of the
respective atoms, which seem to be insensitive to the identity
of both the phosphine and halide ligands. This strongly
supports our finding that the interaction of the thione ligand
with the closed-shell d10 Cu(I) metal center in a variety of
coordination environments is very weak, the computed
interaction energy predicted to be approximately 20 kcal/
mol. It seems that H-bond formation and other types of
decreases from the chloro to bromo and to iodo compounds.
Moreover, in both series of complexes the Cu-X-H bond
angle, found in the range 74.4-110.1°, decreases succes-
sively by about 5° from the chloro to bromo and to iodo
derivatives. However, the opposite holds true for the S-
Cu-X and Cu-S-C bond angles.
Upon coordination of the thione ligand with the Cu(I)
metal center in the [CuX(PH3)2] and [CuX(1,3-pdp)] com-
plexes, no significant structural changes are observed in the
heterocycle ring of thione. However, in both series of
complexes the coordination of thione through the S donor
atom of the CdS group results in an elongation of the CdS
bond by 0.037 Å. This might be the result of the transfer of
electron density from the CdS group to the Cu(I) metal
center upon formation of the coordination Cu-S bond
through the interaction of the HOMO of thione, being a
π-type MO localized on the CdS group (Figure 10), with a
vacant orbital of the central atom. There is also a possibility
for π-back-bonding through the interaction of an occupied
orbital of the central Cu(I) atom with the LUMO of the thione
ligand, being a π*-type MO delocalized over the entire
nuclear skeleton of the thione ligand. The π-back-bonding
effect is clearly illustrated by the presence of the appropriate
MO (Figure 10b) in all complexes of the two series. In spite
of the composite nature of the Cu-S coordination bond,
exhibiting both a σ and a π component, the bond is relatively
weak, the computed bond dissociation energy predicted to
be 22.7, 20.8, and 19.2 kcal/mol for the chloro, bromo, and
iodo derivatives, respectively, at the B3LYP/SDD level of
theory. The weaker Cu-S coordination bond in the iodo
Inorganic Chemistry, Vol. 41, No. 25, 2002 6885

Aslanidis et al.
intermolecular forces might be of key importance in stabiliz-
ing the coordination of thione ligands with the Cu(I) central
atom.
been isolated, even under the same reaction conditions.
Although within this class of mixed-ligand complexes we
often observed such a dimerization, it is remarkable that the
strong tendency of copper(I) to bind the “soft” iodide ligands
enforces, in the present case, the exclusion of the thione
ligand from the coordination sphere. This could be supported
by the presence of the diphosphine ligand and its strong
chelating effect. However, similar results are obtained using
1,5-pentanebis(diphenylphosphine), which does not act as a
chelating agent but bridges the two metal centers in the
dimer. Consequently, this phenomenon needs a more exten-
sive study; therefore further examples of this series of
compounds will be examined in the future.
According to the results of the DFT calculations the
interaction of the thione ligand with the Cu(I) central atom
corresponds to loose associations, the interaction energies
predicted to be about 20 kcal/mol for all complexes in the
series. The bonding mechanism of the thione ligand with
the Cu(I) central atom involves both σ-dative and π-back-
bonding components. The contribution of each component
to the bonding mechanism is not affected by the identity of
the phosphine and the halide ligands. The coordination of
the thione ligand is further stabilized by X‚‚‚H-N bond
formation being more pronounced in the chloro than in the
iodo derivatives. Similarly, the Cu-X bond in all complexes
of the series involves both σ- and π-dative bonding com-
ponents. However, most important is the presence of π-type
MOs delocalized over the entire four-membered Cu(µ-X)2Cu
ring, supporting a ring current and probably accounting for
the nearly equivalent Cu-X bonds in the rhombus. More-
over, all [Cu(µ-X)(PH3)2]2 dimers exhibit a σ-type MO
corresponding to weak through-ring intermetallic Cu‚‚‚Cu
interactions (d-d interactions), which seems to be respon-
sible for the stabilization of the relatively unstable antiaro-
matic Cu(µ-X)2Cu ring. Finally, the preference of the iodo
monomers to afford dimers rather than to be coordinated with
the thione ligand could be attributed both to the weaker
Cu-S coordination bond in conjunction with the weaker X‚‚‚
H-N interactions as compared to those of the chloro and
bromo monomers and to the higher stability of the [Cu(µ-
I)(PH3)2]2 dimers as compared to the chloro and bromo
counterparts.
Finally, the computed ν(Cu-X) vibrational frequencies
appearing at 208, 160, and 127 cm-1, for the chloro, bromo,
and iodo derivatives of the [CuX(PH3)2(pymtH)] complexes,
respectively, are compatible with the computed Cu-X bond
lengths. The ν(Cu-S) vibrational frequencies were found
at 170, 162, and 161 cm-1 for the chloro, bromo, and iodo
derivatives of [CuX(PH3)2(pymtH)] complexes, respectively.
Notice that the previous tentative assignment of the band of
medium intensity in the region of 340-370 cm-1 to the ν-
(Cu-S) vibrational mode28 seems to be incorrect, as this band
is predicted to occur at much lower frequencies. The ν(Cu-
P) vibrational modes are predicted to absorb in the region
of 205-208 cm-1. The ν(C-S) vibrational frequency
calculated to be 455 cm-1 for all complexes of the series is
shifted to much lower frequencies than the corresponding
band of the free pymtH ligand found at 734 cm-1. The
computed ν(C-S) vibrational frequencies are in line with
the computed C-S bond lengths, being exactly the same
for all complexes in the series, but remarkably elongated
with respect to the free pymtH ligand. Of particular
importance is the ν(N-H) stretching vibrational mode
appearing at 2974, 3033, 3110, and 3591 cm-1 in the [CuCl-
(PH3)2(pymtH)], [CuBr(PH3)2(pymtH)], and [CuI(PH3)2-
(pymtH)] complexes and the free pymtH, ligand, respec-
tively. It can be seen that there is a strong shift of the ν(N-
H) vibrational frequencies to much lower values upon
coordination of the ligand with the Cu(I) metal center. The
shifting of the ν(N-H) vibrational frequencies to lower
frequencies is indicative of the involvement of the H atom
in H-bond formation with the halide ligand and reflects the
relative strength of the X‚‚‚H-N bond formed.
Conclusions
In this work we report on the coordination behavior of
Cu(I) chloride, bromide, and iodide in the presence of a series
of heterocyclic thiones and 1,3-propanebis(diphenylphos-
phine) as ligands. In all the complexes, 1,3-propanebis-
(diphenylphosphine) acts as a bidentate ligand forming stable
six-membered chelate rings, whereby the neutral thione
ligands adopt the monodentate coordination mode binding
the Cu(I) center through the exocyclic sulfur atom. Neverthe-
less, depending on the nature of the halide ligands, two
different types of compounds, namely, monomeric with
tetrahedrally coordinated Cu(I) for X ) Cl and Br and
dimeric with double-bridging iodo ligands for X ) I, have
Acknowledgment. We thank the EPSRC X-ray crystal-
lography service in the University of Southampton for
collecting the X-ray data.
Supporting Information Available: Crystallographic data in
CIF format. This material is available free of charge via the Internet
at http://pubs.acs.org.
(28) Belamy, L. J. The Infrared spectra of Complex molecules; Wiley: New
York, 1966.
IC025896I
6886 Inorganic Chemistry, Vol. 41, No. 25, 2002