Synthesis and characterisation of mono- and dinuclear diethyldithiocarbamatonickel(II) organochalcogenide complexes

Article abstract of DOI:10.1016/S0277-5387(98)00400-8

The reaction of Ni(dtc)(PR₃)Cl (dtc=diethyldithiocarbamate, R=Ph or Bu) with HSC₆H₄Cl-4 or HSCH₂C₆H₄Cl-4 and Et₃N gave two types of complex. For PPh₃, the products wer

Full text of DOI:10.1016/S0277-5387(98)00400-8

Polyhedron 18 (1999) 1115–1122
Synthesis and characterisation of mono- and dinuclear
diethyldithiocarbamatonickel(II) organochalcogenide complexes
b
b
James Darkwaa, , Emmanuel Y. Osei-Twum , Luis A. Litorja Jr.
*
aDepartment of Chemistry, University of the Western Cape, Private Bag X17, Bellville, 7535, South Africa
bResearch Institute, King Fahd University of Petroleum and Minerals, KFUPM, Box 1472, Dhahran 31261, Saudi Arabia
Received 24 July 1998; accepted 28 October 1998
Abstract
The reaction of Ni(dtc)(PR3)Cl (dtc5diethyldithiocarbamate, R5Ph or Bu) with HSC6H4Cl-4 or HSCH2C6H4Cl-4 and Et3N gave two
types of complex. For PPh3, the products were [Ni(dtc)(m-SC6H4Cl-4)]2 (1) and [Ni(dtc)(m-SCH2C6H4Cl-4)]2 (2); whilst PBu3 gave
Ni(dtc)(PBu3)(SC6H4Cl-4) (3). The structure of freshly prepared 3 was determined to be monomeric, as indicated by X-ray diffraction
studies. However, at room temperature in solution, 3 was observed to slowly convert to 1. Structural identification of 1 and 2 and similar
dimers, and structural identification of 3 and analogous monomers, were investigated by mass spectrometry. Electron impact mass
spectrometry (EIMS) failed to confirm the proposed structures due to extensive decomposition in the mass spectrometer. In the electron
impact (EI) mode, all complexes invariably decomposed to Ni(dtc)2; on the other hand, fast atom bombardment (FAB) ionisation gave the
expected molecular ions for all compounds.
1999 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Nickel thiolato complexes; Crystal structure; Mass spectrometry
1. Introduction
the electronic effect the organic substituent has on the
organochalcogenide ligands [3]. Those with electron-re-
leasing ligands tend to form dinuclear compounds, as the
increased electron density on the chalcogen allows the
ligand to form a dative bond with a second nickel atom.
Hence, to prepare monomeric compounds one needs to
have the right balance between all three factors; namely,
the phosphine, the chalcogen and the organic group
attached to the chalocogen. In this paper, we report on the
different types of complexes that are formed when the
phosphine changes between PPh3 and PBu3, whilst the
organic group is restricted to 4-chlorobenzyl, in an attempt
to throw some light on the factors that influence monomer
or dimer formation.
The second aspect of this report is on the use of mass
spectrometry to determine the structural types of complex-
es described above. In our recent work [3], we attempted
to establish the structures of these complexes by electron
impact mass spectrometry (EIMS), without much success.
The peaks found were mainly those of decomposition
products. This seems to be closely linked to the ability of
the compounds to withstand conditions in the ion source in
order to produce molecular ions and has thus made EIMS a
less attractive tool for establishing the structures of these
compounds. Early literature reports seem to have circum-
The nuclearity of nickel thiolato complexes appears to
be highly dependent on the electron density at the nickel
centre. The most common of these complexes are mono-
and dinuclear compounds. Generally, complexes contain-
ing alkyl phosphine ancillary ligands form mononuclear
compounds of the type Ni(SR)2L2 where R is an aryl
group and L is aphosphine [1]. The presence of other
ligands that influence the electron density on the central
metal atoms have also been found to greatly affect the
eventual structures of these thiolato complexes. For exam-
ple, thiolato complexes formed from the reactions of
Ni(dtc)(PPh3)Cl, where dtc5diethyldithiocarbamate, and
the appropriate thiol invariably leads to the formation of
dinuclear compounds, which contain bridging thiolato
ligands and the reactions are accompanied by loss of the
phosphine [2]. However, replacing the sulfur with
selenium results in either a monomeric or a dimeric
complex, depending on the organic group attached to the
chalcogen. For instance, aryl selenides form monomers
whilst alkyl ones are dimers [3]. This has been attributed to
*
Corresponding
jdarkwa@cs.uwc.ac.za
author.
Fax:
127-21-959-3055;
e-mail:
0277-5387/99/$ – see front matter
PII: S0277-5387(98)00400-8
1999 Elsevier Science Ireland Ltd. All rights reserved.

1116
J. Darkwa et al. / Polyhedron 18 (1999) 1115–1122
vented this by using alternative means to establish struc-
tures; for example, the dimeric structures of the chalcogen-
bridged dithiobenzoatenickel(II), [(C6H5CS2)Ni(m-S)]2
and dithiocarbamatonickel(II), [(C6H5NHCS2)Ni(m-S)]222
compounds were deduced from their molecular weight
measurements [4]. However, molecular weight determi-
nations are generally a less accurate structural technique.
In instances where direct evidence was lacking, structural
information had been deduced from that of the related
compounds as in the thioxanthate compounds,
[(C6H5SCS2)Ni(m-SCH2C6H5)]2 [5] and [(EtSCS2)Ni(m-
SEt)]2 [6], which are readily converted to the corre-
sponding dithiocarbamate analogs by reacting them with
triethylamine. Based on the X-ray structures of the thiox-
anthates [5,6], the dithiocarbamate complexes were in-
ferred to be dimeric as well. When our attempts to establish
the monomeric and dimeric structures of our compounds
by EIMS were unsuccessful, we changed to softer ioniza-
tion techniques like fast atom bombardment (FAB) and
field desorption (FD) to show that both the monomeric and
dimeric dithiocarbamate complexes that we have syn-
thesised and those previously reported [2,3] can be routine-
ly characterised, using the right ionization technique.
These two ionization methods, which did not require high
ion source temperatures, readily gave the molecular ions
for all of the compounds investigated.
mL) was reacted with Et3N (0.20 mL). The solution
immediately turned green and was stirred at room tempera-
ture for 2 h. The resultant mixture was filtered to remove
Et3NHCl and an equal volume of hexane was added. After
cooling at 2158C overnight, green crystals of [Ni(dtc)(m-
SC6H4Cl-4)]2 were obtained. Yield50.40 g, 81%. Anal.
calc. for C22H28Cl2N2S6Ni2: C, 37.69; H, 4.06; N, 4.00.
1
Found: C, 38.04; H, 4.49; N, 4.36. H NMR (CDCl3): d
7.82 (d, 4H, JHH58.55 Hz, SC6H4Cl-4); 7.12 (d, 4H,
J
HH58.55 Hz, SC6H4Cl-4); 3.48 (q, 8H, dtc); 1.15 (t, 12H,
dtc). 13Ch1Hj NMR: d 136.4(s), 132.8(s), 131.3(s), 128.0(s)
(SC6H4Cl-4); 43.8(s), 12.4(s) (dtc). IR (KBr pellet cm21):
2976 w, 2929 w, 2677 w, 1514 vs, 1472 m, 1439 s, 1387
m, 1277 s, 1209 m, 1151 s, 1086 s, 1010 s, 913 w, 815 s,
801 s, 739 w.
2.3. Reaction of Ni(dtc)(PPh3 )Cl with 4-
chlorobenzylmercaptan: formation of [Ni(dtc)(m-
SCH2C6H4Cl-4)]2 (2)
The reaction was performed and worked up in a manner
similar to that in Section 2.2 using Ni(dtc)(PPh3)Cl (0.50
g, 1.00 mmol) and 4-chlorobenzylmercaptan (0.20 g, 1.07
mmol) to give [Ni(dtc)(m-SCH2C6H4Cl-4)]2 (2). Yield5
0.20 g, 55%. Anal. calc. for C24H32Cl2N2S6Ni2: C, 39.53;
1
H, 4.44; N, 3.84. Found: C, 39.46; H, 4.73; N, 4.21. H
NMR (CDCl3):
d
7.15 (d, 4H,
J
HH58.36 Hz,
SCH2C6H4Cl-4); 7.07 (d, 4H,
J
HH58.36 Hz,
SCH2C6H4Cl-4); 3.64 (q, 8H, dtc); 2.72 (s, 4H,
SCH2C6H4Cl-4); 1.26 (t, 12H, dtc). 13Ch1Hj NMR: d
137.8(s), 132.2(s), 130.4(s), 128.3(s) (SCH2C6H4Cl-4);
43.9(s) (dtc), 32.3(s), (SCH2C6H4Cl-4); 12.5(s) (dtc). IR
(KBr pellet cm21): 2972 w, 2927 w, 1521 vs, 1489 m,
1457 m, 1436 m, 1405 w, 1379 m, 1356 m, 1282 s, 1261
m, 1206 m, 1153 m, 1092 s, 1014 s, 910 w, 850 m, 804 s,
724 w, 668 w, 646 w, 504 m.
2. Experimental
2.1. Materials and instrumentation
All solvents were of analytical grade and were used
without further purification. Tributylphosphine, tri-
phenylphosphine, 4-chlorothiophenol, 4-chlorobenzylmer-
captan and nitrobenzyl alcohol were purchased from
Aldrich and used as received. The starting materials,
Ni(dtc)(PPh3)Cl and Ni(dtc)(PBu3)Cl, were prepared by
the literature procedures [7]. All reactions were performed
under a nitrogen atmosphere but the air-stable products
were worked-up in air.
2.4. Reaction of Ni(dtc)(PBu3 )Cl with 4-
chlorothiophenol: formation of Ni(dtc)(PBu3 )(SC6H4Cl-4)
To a solution of Ni(dtc)(PBu3)Cl (0.55 g, 1.00 mmol)
and HSC6H4Cl-4 (0.15 g, 1.00 mmol) in toluene (30 mL)
was added excess Et3N (1 mL). The purple colour
immediately changed to greenish–brown, accompanied by
precipitation of a white solid. The mixture was stirred at
room temperature for 4 h and filtered to remove the white
solid of Et3NHCl, which was washed with 235 mL of
hexane. After the addition of hexane (20 mL), the filtrate
was cooled at 2158C overnight to give green, X-ray
quality crystals of Ni(dtc)(PBu3)(SC6H4Cl-4). Yield50.37
g, 71%. Anal. calc. for C23H41ClNPS3Ni: C, 49.97; H,
7.46; N, 2.53. Found: C, 49.50; H, 7.35; N, 2.75. 1H NMR
(CDCl3): d 7.39 (d, 2H, JHH58.40 Hz, SC6H4Cl-4); 7.00
(d, 2H, JHH58.40 Hz, SC6H4Cl-4); 3.50 (q, 4H, dtc); 1.55
(m, 18H, PBu3); 1.18 (t, 6H, dtc) 1.00 (t, 9H, PBu3). IR
IR spectra were recorded on a Nicolet 205 FT-IR
spectrometer. 1H and 13C NMR were recorded on a Varian
Gemini 2000 spectrometer at 200 and 50.28 MHz, respec-
1
tively, and referenced to residual CHCl3 for H (d 7.26)
and 13C (d 77.0). Elemental analyses were performed by
the Microanalytical Laboratory at the University of Cape
Town, South Africa, as a service.
2.2. Reaction of Ni(dtc)(PPh3 )Cl with 4-
chlorothiophenol: formation of [Ni(dtc)(m-SC6H4Cl-4)]2
(1)
A solution of Ni(dtc)(PPh3)Cl (0.70 g, 1.40 mmol) and
HSC6H4Cl-4 (0.20 g, 1.40 mmol) in degassed toluene (50

J. Darkwa et al. / Polyhedron 18 (1999) 1115–1122
1117
(nujol mull cm21): 1515 vs, 1405 s, 1357 m, 1301 m, 1278
2.6. Mass spectrometry
s, 1206 m, 1148 m, 1088 vs, 1072 s, 1009 s, 911 w, 850 w,
819 m, 799 w, 734 w, 724 w, 695 w, 545 w, 497 m, 468
w.
Mass spectrometry of all of the compounds was per-
formed on a JMS-HX100 EBE spectrometer (JEOL,
Japan). EI spectra were obtained at an electron energy of
70 eV, an emission current of 100 mA and an acceleration
voltage of 5 kV. The ion source temperature was 2508C and
the temperature of the EI direct insertion probe (DIP) was
programmed from 50 to 2508C at 328C/min. Fast atom
bombardment (FAB) spectra of all of the compounds were
obtained with nitrobenzyl alcohol (NBA) as the matrix.
Argon fast atoms were used to bombard the samples,
dissolved in the matrix, on a stainless steel probe tip. The
ion source temperature was kept below 1008C to prevent
decomposition of the compounds. The mass spectrometer
was calibrated using CsI. Field ionization (FI) and field
desorption (FD) spectra were obtained using the FAB1
acquisition parameters. The FD spectra were acquired at
50, 100 and 1508C. The optimum cathode voltage was 4.5
kV and the emitter current ramp program was 0–40 mA at
4 mA/min.
2.5. Crystal structure determination of
Ni(dtc)(PBu3 )(SC6H4Cl-4)
Single crystals for structural determination were grown
by slow diffusion of hexane into a CH2Cl2 solution of 3,
which was cooled at 2158C. A green cube-like crystal was
mounted in a sealed capillary tube on a Siemens SMART
diffractometer, equipped with a CCD detector for geomet-
ric and intensity data collection. The structure was solved
by the Patterson method for primary atom sites and by the
difference map for atoms in secondary sites. All non-
hydrogen atoms were refined anisotropically and hydrogen
atoms were refined without restraints. The structure was
refined by full matrix least-squares on F2 with a weighting
scheme w215w2(Fo2)1(0.100P)210.00P, where P5(Fo2
12Fc2)/3, using SHELXL-93 [8]. An empirical absorption
correction was applied to the data [9]. Crystal data and
structure refinement details are listed in Table 1 and other
pertinent information is deposited at the Cambridge Crys-
tallographic Data Centre (deposition number 102412).
3. Results and discussion
3.1. Synthesis of complexes
The reaction of 4-chloro-substituted thiols with
Ni(dtc)(PR3)Cl (R5Ph or Bu) gave two types of products,
depending on the nature of the phosphine. When the
phosphine was PPh3, the reaction occurred as in Eq. (1)
and was accompanied by the loss of PPh3.
Table 1
Crystal data and structural refinement for 3
Formula
Fw
Temperature (K)
Wavelength (A)
C23H41ClNPS3Ni
552.88
293(2)
0.71073
Triclinic
˚
Et3N
Crystal system
Space group
1
2 PPh3]
]
Ni(dtc)(PPh3)Cl 1 HSR9
2 [Ni(dtc)(m-SR9)]2 1
P1 bar
˚
a (A)
10.348(9)
10.761(9)
15.086(13)
102.93(10)
106.10(10)
104.17(10)
1486.4(2), 2
1.235
Et3NHCl
(1)
˚
b (A)
˚
c (A)
a (deg)
b (deg)
g (deg)
[R9 5 C6H4Cl-4 (1), CH2C6H4Cl-4 (2)]
1
The H NMR spectra of compounds 1 and 2 showed no
PPh3 peaks. The only phenyl protons were the two
doublets assignable to the thiolato ligands. Confirmation of
the absence of PPh3 in the nickel complexes was obtained
when PPh3 was isolated from the filtrate after the nickel
compounds had been filtered off. Our earlier work has
shown that when R9 is a phenyl group, a similar dimeric
compound, [Ni(dtc)(m-SC6H5)]2 [2], is formed when the
same reaction as that in Eq. (1) is performed with
thiophenol (HSC6H5). However, the selenium analogue
forms Ni(dtc)(PPh3)(SeC6H5) [3]. By replacing the hydro-
gen in the para-position with a halide, 4-chlorothiophenol
formed [Ni(dtc)(m-SC6H4Cl-4)]2, whilst the selenium ana-
logue formed Ni(dtc)(PPh3)(SeC6H4Cl-4) [3]. The propen-
sity towards dimer formation appears to be associated with
two factors. First is the ability of the chalcogen to donate
an electron pair to a second nickel atom. The second is the
3
˚
Volume (A ), Z
density (calculated) (Mg m23
)
Abs coeff (mm21
)
1.018
588
F(000)
Crystal size (mm)
u-range for data collected (deg)
Limiting indices
0.6430.4230.24
1.48–25.53
211#h#12
213#k#10
216#l#18
5995
No. of reflections collected
No. of independent reflections
R(int)
448
0.0622
Refinement method full-matrix least-squares on F2
Data/restraints/parameters
Goodness of fit on F2
Final R indices [I.2s(I)]
R indices (all data)
4446/0/276
0.893
R1 50.0977 wR2 50.2650
R1 50.1224 wR2 50.2926
1.212 and 20.748
23
˚
Largest difference peak and hole (e A
)

1118
J. Darkwa et al. / Polyhedron 18 (1999) 1115–1122
steric bulk of the phosphine. Thus, if the chalcogen has a
strong donor ability like sulfur, the process of binding a
second metal atom results in the loss of PPh3. On the other
hand, when aryl groups are attached to selenium, they form
monomers, while their alkyl analogues give dimers. The
former have less electron density on the selenium, while
the latter are more electron-rich.
We tested this hypothesis by changing the phosphine in
Ni(dtc)(PR3)Cl to PBu3. On performing the reaction above
with Ni(dtc)(PBu3)Cl, we were able to isolate and char-
acterise Ni(dtc)(PBu3)(SC6H4Cl-4) (3) (Eq. (2)) by a
combination of NMR spectroscopy and X-ray crystal-
lography.
set of doublets appeared at 7.81 and 7.12 ppm, which
increased in intensity over a period of three days, sug-
gesting that the monomeric compound decomposes in
solution (Fig. 1). The new set of doublets is the same as
those of 1 and indicates a gradual conversion of 3 to 1, via
a loss of PBu3. Hence, even the less bulky PBu3 seem to
gradually give way to forming dimers in the presence of
the electron-donating sulfur. Hence, it appears that the
most important factor in maintaining the monomeric form
is the ability of the chalcogen to donate electrons and that
the steric factor might be secondary.
3.2. Structure of 3
Ni(dtc)(PBu3)Cl 1 HSC6H4Cl-4
The structure of 3 in the solid state was determined by
X-ray crystallography. The molecular structure of 3 is
shown in Fig. 2 and some of the relevant bond distances
and angles are given in Table 2. The final R (0.0977) value
is quite high, particularly when compared to the R(int)
(0.0622). This might be due to some disorder and thus
should account for the poor refinement. The high final R
Et
N
3
→
Ni(dtc)(PBu3)(SC6H4Cl-4) (2)
2Et NHCl
3
Freshly prepared solutions of this product showed a 1H
NMR spectrum with two doublets in the phenyl region at
7.39 and 7.00 ppm. However, after one day, an additional
Fig. 1. 1H NMR spectrum of Ni(dtc)(PBu3)(SC6H4Cl-4) after 24 h and the spectrum of freshly prepared sample (insert).

J. Darkwa et al. / Polyhedron 18 (1999) 1115–1122
1119
value limits the extent to which the structural data can be
discussed. Nevertheless, the crystal structure confirms what
could be deduced from the analytical data. The geometry
around the nickel is a distorted square plane. The distorted
square planar environment of the nickel is similar to that of
Ni(dtc)(PPh3)(SeC6H5), if one considers the S(1), S(2),
S(3) and the P atoms as occupying the corners of the
square plane. The largest deviation from the square planar
angle is the S(2)–Ni(1)–S(1) angle of 78.72(7)8 and the
least is the P(1)–Ni(1)–S(3) of 87.928. The former appears
to be the constraint the bidentate diethyldithiocarbamato
ligand imposes on the nickel when bound to it. This angle
was found to be the least in Ni(dtc)(PPh3)(SeC6H5)
[78.0(1)8 [3] and largest in Ni(dtc)2 (79.4(1)8] [10]. The
˚
Ni–S bond distances Ni(1)–S(1) [2.228(2) A] and Ni(1)–
˚
S(2) [2.210(2) A] of the Ni–dtc interaction in 3 are
slightly longer than those in Ni(dtc)(PPh3)(SeC6H5)
˚
[2.219(1) and 2.208(1) A, respectively]. The Ni–S(3)
bond to the thiolato ligand in 3 has a bond distance of
5
˚
2.190(2) A and is similar to that of (h -C5H5)Ni(P-
˚
Bu3)(SC6H4Cl-4) (2.194(9) A) [11]. The Ni–P distance
[2.169(2) A] though is slightly longer than that in (h -
C5H5)Ni(PBu3)(SC6H4Cl-4) [2.140(8) A], it is within
5
˚
˚
acceptable distances for Ni–P bonds for Ni in the 12
Fig. 2. Molecular structure of Ni(dtc)(PBu3)(SC6H4Cl-4)
oxidation state, for example (h5-C5H5)Ni(PPh3)-
˚
(CH2SO2Ph) [2.150(1) A] [12].
Table 2
Selected bond distances and angles for 3
3.3. Mass spectra
˚
Bond distances (A)
Ni(1)–P(1)
Ni(1)–S(2)
P(1)–C(12)
P(1)–C(20)
S(1)–C(1)
C(1)–N(1)
N(1)–C(10)
2.169(2)
2.210(2)
1.842(6)
1.832(6)
1.733(7)
1.330(10)
1.630(2)
Ni(1)–S(1)
Ni(1)–S(3)
P(1)–C(16)
S(3)–C(2)
S(2)–C(1)
N(1)–C(8)
Cl(1)–C(5)
2.228(1)
2.190(2)
1.830(7)
1.746(8)
1.687(9)
1.493(13)
1.736(9)
In order to identify a mass spectrometric technique that
could establish the structures of both the monomeric and
dimeric complexes, we used electron impact (EI), fast
atom bombardment (FAB) and field ionization/field de-
sorption (FI/FD) ionization modes to study the following
compounds: [Ni(dtc)(m-SC6H4R9-4)]2 (R95Cl (1), H (4),
Me (5)), [Ni(dtc)(m-SCH2C6H4Cl-4)]2 (2), [Ni(dtc)(m-
SeCH2C6H5)]2 (6), [Ni(dtc)(m-SeMe)]2 (7), [Ni(dtc)(m-
SeC6H4Cl-4)]2 (8) (Table 3) and Ni(dtc)(PBu3)(SC6H4Cl-
Bond angles (8)
P(1)–Ni(1)–S(1)
P(1)–Ni(1)–S(3)
S(1)–Ni(1)–S(3)
Ni(1)–P(1)–C(12)
Ni(1)–P(1)–C(20)
C(1)–S(2)–Ni(1)
N(1)–C(1)–S(1)
173.7(7)
87.9(8)
97.9(9)
115.8(2)
114.0(2)
86.1(3)
P(1)–Ni(1)–S(2)
S(2)–Ni(1)–S(3)
S(1)–Ni(1)–S(2)
Ni(1)–P(1)–C(16)
C(1)–S(1)–Ni(1)
Ni(1)–S(3)–C(2)
N(1)–C(1)–S(2)
95.6(8)
175.0(9)
78.7(7)
112.4(3)
84.4(3)
4)
(3),
Ni(dtc)(PPh3)(SeC6H5)
(9)
and
Ni(dtc)(PPh3)(SeC6H4Cl-4) (10) (Table 4). When the
mass spectra of all of the compounds were investigated by
EI ionization, none of them gave a molecular ion. Two
prominent peaks, [Ni(dtc)2]1 (354) and those of the
109.6(2)
125.6(6)
123.7(7)
Table 3
Correlation of m/z values for the important ionsa in the FAB mass spectra of [Ni(dtc)(m-ER)]2 (E5S, Se)
Compound
Molecular ion
a
b
c
d
e
(CH3)2NCS
1
2
4
5
6
7
8
a
704
728
632
660
754
602
794
n.d.
603
555
569
663
586
n.d.
557
571
523
537
585
506
605
n.d.
446
446
492
492
492
492
414
414
414
414
414
414
414
354
354
354
354
354
354
354
116
116
116
116
116
116
116
Refer to Scheme 1 for ion assignment, n.d.5not detected.

1120
J. Darkwa et al. / Polyhedron 18 (1999) 1115–1122
Table 4
FAB mass spectra of Ni(dtc)(ER9)(PR03)
Compound
Molecular ion (m/z)
Other prominent fragments
3
551
[Ni(dtc)(m-SC6H4Cl-4]12 (704),
[Ni2(dtc)2]1 (414)
9
625
[Ni(dtc)(m-SeC6H5]12 (726)
[Ni2(dtc)2(SeC6H5)]1 (572),
[Ni2(dtc)2]1 (414), Ni(dtc)2]1 (354),
[PPh3]1 (262)
10
659
[Ni(dtc)(m-SeC6H4Cl-4]12 (796),
[Ni(dtc)(PPh3)]1 (468),
[Ni2(dtc)2]1 (414), [Ni(dtc)2]1 (354),
[PPh3]1 (262)
phosphines, [PPh3]1 (m/z 262) or [PBu3]1 (m/z 202),
were found. Since all of the compounds investigated had
previously been fully characterized by a combination of
IR, NMR, elemental analyses and, in the case of
Ni(dtc)(PPh3)(SeC6H5) [3] and Ni(dtc)(PBu3)(SC6H4Cl-
4), by X-ray crystallography (Fig. 2), the absence of
molecular ions and the formation of [Ni(dtc)2]1 (m/z 354)
by EIMS is an indication of thermal decomposition in the
ion source of the mass spectrometer. However, it is
difficult to deduce a structure for any of the complexes by
putting together the fragments of the EI spectrum. This
prompted us to use the softer ionization technique of FAB.
Even here, the choice of matrix proved to be quite crucial
to the successful acquisition of a meaningful mass spec-
trum showing the molecular ion. When glycerol was used
as the matrix, only ions due to the matrix were observed.
The most successful matrix was nitrobenzyl alcohol. Fig. 3
shows the FAB mass spectrum of [Ni(dtc)(m-SC6H5)]2
and Table 3 lists all of the dimeric compounds examined.
All of these compounds showed molecular ions in their
respective spectra. That these compounds are all less stable
relative to bis(diethyldithiocabarmato)nickel(II), Ni(dtc)2,
could clearly be established; in most instances, the ion due
to Ni(dtc)2 formed the base peak. This was particularly
true for the EI spectra where all compounds produced
Ni(dtc)2 as the parent ion. The FAB mass spectral frag-
mentation pathway appears to be similar for all of the
dimeric compounds and is proposed to occur in a sequen-
tial fashion, involving the loss of the organochalcogenide
ligands, to produce the various ions in the spectra. The
fragmentation pathways are depicted for [Ni(dtc)(m-
SCH2C6H4Cl-4]2 in Scheme 1, which shows the stepwise
loss of SCH2C6H4Cl-4 and Ni, to give Ni(dtc)2. In contrast
to the EI spectra, the fragments from the FAB spectra
could be seen as sequential loss of units from the molecu-
lar ions, thus, confirming the structure suggested by the
molecular peaks.
It was not clear whether the EI spectrum was the result
of thermal decomposition in the ion source or due to
electron impact fragmentation. However, we suspected that
thermal decomposition could be the major process respon-
sible for the spectra we were observing. To test this
hypothesis, both FI and FD of three compounds (1, 2 and
3) were studied. In the FIMS analysis, the compounds
needed to be heated to evaporate them into the ion source,
while the FDMS did not require heating to enable their
ionization. It was observed that the FI spectra of these
compounds were different from those of the EI and FD
spectra. Unlike the FDMS, the compounds appeared to
decompose with heating, thus giving Ni(dtc)2 as a cluster
of ions at m/z 354 in both FIMS and EIMS. However,
unlike the EIMS, the FIMS seemed to have different
ionization pathways. Ions that were prominent in the EI
spectra were observed to be minor in the FI spectra and
vice versa. A cluster of ions were observed at m/z 710 in
all of the FI spectra and these were interpreted to be
coming from the dimerization of Ni(dtc)2, which leads to
the formation of [Ni(dtc)(m-dtc)]2.
When the same crystalline samples from which crystals
were selected for the X-ray structural determination of
Ni(dtc)(PPh3)(SeC6H5) and Ni(dtc)(PBu3)(SC6H4Cl-4)
were subjected to FABMS analysis, molecular ions corre-
sponding to the solid state structures were found (Table 4).
In addition, ions corresponding to the dimers [Ni(dtc)(m-
SeC6H5)]12 (m/z 726) and [Ni(dtc)(m-SC6H4Cl-4)]12 (m/z
704) were observed. A typical spectrum showing both the
molecular ions of the monomer and the decomposed dimer
of Ni(dtc)(PPh3)(SeC6H5) is shown in Fig. 4. Even
Fig. 3. FAB spectrum of [Ni(dtc)(m-SC6H5)]2.
though,
in
solution,
the
1H
NMR
of

J. Darkwa et al. / Polyhedron 18 (1999) 1115–1122
1121
Scheme 1.
Ni(dtc)(PBu3)(SC6H4Cl-4) shows
a
monomer/dimer
SeC6H4Cl-4)]12 . There are two viable decomposition path-
ways for the monomers, all of which have identifiable
fragments in the spectra. The first is the simultaneous loss
of the phosphino and the organochalcogenide ligands to
form Ni(dtc)2. The second route involves the loss of only
the phosphino ligand, which generates the electron-de-
ficient species Ni(dtc)(ER) and its subsequent dimerization
to [Ni(dtc)(m-ER)]2 (E5S, Se) (Scheme 2). Further de-
composition of the dimers follows the same pathway as in
Scheme 1. The strength of the dithiocarbamate as a ligand
in the decomposition is quite clear. It forms most of the
identifiable species in the spectra. The stability of dithio-
carbamate as a ligand for Pt(II) in mass spectral studies
has been noted by Colton and Tedesco [13,14]. This allows
the formation of stable ions in the gaseous phase, which
helps in identifying complexes.
equilibrium with time, the time frame required for the
monomer/dimer equilibrium is much longer than that
needed to obtain the mass spectrum. The dimer formation
here is also believed to take place in the mass spectrometer
but via a process that is completely different from that
observed in solution by NMR. A similar observation was
made from the spectrum of a microcrystalline sample of
Ni(dtc)(PPh3)(SeC6H4Cl-4), which has a monomeric peak
of m/z 659 and a dimeric peak of m/z 796 for [Ni(dtc)(m-
Acknowledgements
Financial support by the Foundation for Research De-
velopment, South Africa (JD) and the Research Institute,
King Fahd University of Petroleum and Minerals is
gratefully acknowledged. We thank A.E. Thage for her
assistance with the synthesis. We are also grateful to
Fig. 4. FAB spectrum of Ni(dtc)(PPh3)(SeC6H5).

1122
J. Darkwa et al. / Polyhedron 18 (1999) 1115–1122
Scheme 2.
[7] P.L. Maxfield, Inorg. Nucl. Chem. Lett. 6 (1970) 693.
[8] G.M. Sheldrick, SHELTXL-93, Program for the refinement of
crystal structures. University of Gottingen, 1993.
[9] G.M. Sheldrick, SADABS, An empirical absorption correction.
Private communication to subscribers of the Siemens CCD e-mail
list.
Professor J.C.A. Boeyens and Ms. Leanne Cook, Structural
Chemistry Department, University of the Witwatersrand,
South Africa, for determining the crystal structure as a
service.
[10] R. Selvarau, K. Panchanatheswaran, A. Thiruvalluvar, V. Parth-
asarathi, Acta Cryst. C51 (1995) 606.
[11] R.M. Moutloali, J. Darkwa, T. Nyokong, J. Organomet. Chem. 569
(1998) 37.
References
[1] T.B. Rauchfuss, J.S. Shu, D.M. Roundhill, Inorg. Chem. 15 (1976)
2096.
[12] M. Julia, H. Lauron, J.N.Vrepeaux, Y. Jeanin, C. Bois, J. Organomet.
Chem. 358 (1988) C11.
[2] J. Darkwa, L.M. Koczon, J.M. Rost, Synth. React. Inorg. Met.-Org.
Chem. 23 (1993) 509.
[3] S. Babikanyisa, J. Darkwa, Inorg. Chim. Acta. 256 (1997) 15.
[13] R. Colton, V. Tedesco, Inorg. Chem. 30 (1991) 2451.
[14] A.M. Bond, R. Colton, Y.A. Nah, J.C. Traeger, Inorg. Chem. 33
(1994) 2548.
¨
[4] W. Hieber, R. Bruck, Z. Anorg. Allg. Chem. 269 (1952) 13.
[5] J.P. Fackler Jr., W.J. Zegarski, J. Am. Chem. Soc. 95 (1973) 8666.
[6] A.C. Villa, A.G. Manfredotti, M. Narledi, C. Pelizzi, J. Chem. Soc.
Chem. Commun. (1970) 1322.