Lewis-base copper(I) formates: Synthesis, reaction chemistry, structural characterization and their use as spin-coating precursors for copper deposition

Article abstract of DOI:10.1016/j.ica.2010.05.048

Consecutive synthesis methodologies for the preparation of a series of copper(I) formates [L_mCuO₂CH] (L = ⁿBu ₃P: 4a, m = 1; 4b, m = 2; 5, L = [Ti](CCSiMe₃) ₂, m = 1, [Ti] = (η⁵-C₅H ₄SiMe₃)₂Ti) and [L_mCuO ₂CH·HO₂CR] (L = ⁿBu₃P: 7a, m = 1, R = H; 7b, m = 2, R = H; 7c, m = 2, R = Me; 7d, m = 2, R = CF₃; 7e, m = 2, R = Ph. L = (^cC₆H₁₁)₃P, R = H: 8a, m = 2; 8b, m = 3. L = (CF₃CH₂O)₃P, R = H: 9a, m = 2; 9b, m = 3. L = (CH₃CH₂O)₃P, R = H: 10a, m = 2; 10b, m = 3. L = [Ti](CCSiMe₃)₂; m = 1: 11a, R = H; 11b, R = Ph) is reported using [CuO₂CH] (1) and L (2a, L = ⁿBu₃P; 2b, L (^cC₆H ₁₁)₃P; 2c, L = (CF₃CH₂O) ₃P; 2d, L = (CH₃CH₂O)₃P; 3, L = [Ti](CCSiMe₃)₂) as key starting materials. Addition of formic acid (6a) or carboxylic acid HO₂CR (6b, R = Me; 6c, R = CF₃; 6d, R = Ph) to the afore itemized copper(I) formates 4 and 5 gave metal-organic or organometallic 7-11. The molecular structures of 8a and 11a in the solid state are reported showing a threefold coordinated copper(I) ion, setup by either two coordinatively-bonded phosphorus atoms and one formate oxygen atom (8a) or two π-bonded alkyne ligands and one oxygen atom (11a). A formic acid molecule is additionally hydrogen-bonded to the CuO₂CH moiety. The use of 7b as suitable precursor for the deposition of copper onto TiN-coated oxidized silicon wafers by the spin-coating process below 300 °C is described. Complex 7b offers an appropriate transformation behavior into metal phase by an elimination-decarboxylation mechanism. The morphology of the copper films strongly depends on the annealing conditions. A closed grain network densified by a post-treatment is obtained (8 °C min^-1, N₂/H₂ carrier gas). Hydrogen post-anneal to 420 °C after film deposition gave a copper film showing resistivities from 2.5 to 3.7 μΩ cm. This precursor was also used for gap-filling processes.

Full text of DOI:10.1016/j.ica.2010.05.048

Inorganica Chimica Acta 365 (2011) 10–19
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Lewis-base copper(I) formates: Synthesis, reaction chemistry, structural
characterization and their use as spin-coating precursors for copper deposition
André Tuchscherer a, Yingzhong Shen a, Alexander Jakob a, Robert Mothes a, Mohammed Al-Anber a,
Bernhard Walfort a, Tobias Rüffer a, Swantje Frühauf b, Ramona Ecke b, Stephan E. Schulz b,
Thomas Gessner b, Heinrich Lang a,
*
a Technische Universität Chemnitz, Fakultät für Naturwissenschaften, Institut für Chemie, Lehrstuhl für Anorganische Chemie, Straße der Nationen 62, 09111 Chemnitz, Germany
b Technische Universität Chemnitz, Fakultät für Elektrotechnik und Informationstechnik, Zentrum für Mikrotechnologien, Reichenhainer Straße 70, 09126 Chemnitz, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Consecutive synthesis methodologies for the preparation of a series of copper(I) formates [LmCuO2CH]
(L = nBu3P: 4a, m = 1; 4b, m = 2; 5, L = [Ti](C„CSiMe3)2, m = 1, [Ti] = ( 5-C5H4SiMe3)2Ti) and
g
Received 9 February 2010
Accepted 20 May 2010
Available online 1 June 2010
[LmCuO2CHꢀHO2CR] (L = nBu3P: 7a, m = 1, R = H; 7b, m = 2, R = H; 7c, m = 2, R = Me; 7d, m = 2, R = CF3;
7e, m = 2, R = Ph. L = (cC6H11)3P, R = H: 8a, m = 2; 8b, m = 3. L = (CF3CH2O)3P, R = H: 9a, m = 2; 9b, m = 3.
L = (CH3CH2O)3P, R = H: 10a, m = 2; 10b, m = 3. L = [Ti](C„CSiMe3)2; m = 1: 11a, R = H; 11b, R = Ph) is
reported using [CuO2CH] (1) and
L L
(2a, L = nBu3P; 2b, (cC6H11)3P; 2c, L = (CF3CH2O)3P; 2d,
Keywords:
Copper
Formate
Carboxylic acid
Hydrogen-bonding
Spin coating
L = (CH3CH2O)3P; 3, L = [Ti](C„CSiMe3)2) as key starting materials. Addition of formic acid (6a) or carbox-
ylic acid HO2CR (6b, R = Me; 6c, R = CF3; 6d, R = Ph) to the afore itemized copper(I) formates 4 and 5 gave
metal–organic or organometallic 7–11. The molecular structures of 8a and 11a in the solid state are
reported showing a threefold coordinated copper(I) ion, setup by either two coordinatively-bonded phos-
phorus atoms and one formate oxygen atom (8a) or two p-bonded alkyne ligands and one oxygen atom
(11a). A formic acid molecule is additionally hydrogen-bonded to the CuO2CH moiety. The use of 7b as
suitable precursor for the deposition of copper onto TiN-coated oxidized silicon wafers by the spin-coat-
ing process below 300 °C is described. Complex 7b offers an appropriate transformation behavior into
metal phase by an elimination–decarboxylation mechanism. The morphology of the copper films strongly
depends on the annealing conditions. A closed grain network densified by a post-treatment is obtained
(8 °C minꢁ1, N2/H2 carrier gas). Hydrogen post-anneal to 420 °C after film deposition gave a copper film
showing resistivities from 2.5 to 3.7
l
X
cm. This precursor was also used for gap-filling processes.
Ó 2010 Published by Elsevier B.V.
1. Introduction
the above mentioned transition metal compounds only less is
known about the use of copper(I) formate in metal deposition pro-
cesses [5].
Copper(I) formate in its non-coordinated form is a reactive me-
tal–organic compound and is sensitive to moisture and tempera-
ture [1]. Nevertheless, it can be stabilized either by Lewis-base
ligands L (L = neutral or ionic 2–4-electron donor) and/or hydro-
gen-bonded formic acid or carboxylic acids [2]. Depending on the
nature and number of L in [LmCuO2CR] (m = 1, 2, 3; R = H, single-
bonded organic group) different aggregated species are formed,
for example, tetra-nuclear cubane- or staircase-like structured
molecules, dinuclear or monometallic copper(I) complexes [3].
Recently, it could be shown that volatile phosphane, phosphite,
or alkyne copper(I) carboxylates can successfully be used as CVD
(Chemical Vapor Deposition) precursors in the deposition of thin
copper layers on different substrate materials [4]. In contrast to
This prompted us to synthesize coordination complexes of gen-
eral composition [LmCuO2CH] (L = nBu3P, (cC6H11)3P, (CF3CH2O)3P,
(CH3CH2O)3P; m = 1, 2, 3; L = [Ti](C„CSiMe3)2, m = 1, [Ti] = (g5
-
C5H4SiMe3)2Ti). Their reaction behavior towards formic acid and
carboxylic acids RCO2H (R = Me, CF3, Ph) is discussed. The use of
[(nBu3P)2CuO2CHꢀCH2O] as spin-coating precursors for the deposi-
tion of pure copper layers is reported as well.
2. Experimental section
2.1. Materials and methods
Complexes [LmCuO2CH] (L = P(cC6H11)3, P(OCH2CH3)3, P(OCH2-
CF3)3; m = 2, 3) were prepared by published procedures [5]. All
* Corresponding author. Tel.: +49 (0) 371 531 21210; fax: +49 (0) 371 531 21219.
E-mail address: heinrich.lang@chemie.tu-chemnitz.de (H. Lang).
0020-1693/$ - see front matter Ó 2010 Published by Elsevier B.V.
doi:10.1016/j.ica.2010.05.048

A. Tuchscherer et al. / Inorganica Chimica Acta 365 (2011) 10–19
11
other chemicals were purchased by commercial providers and
were used as received.
(CDCl3): d 0.74 (t, 9 H, (CH2)3CH3, JHH = 14.3 Hz), 1.20–1.30 (m, 18
H, (CH2)3CH3), 8.27 (s, 1 H, HCO2ꢁ). 13C{1H} NMR (CDCl3): d 13.9
(CH2CH2CH2CH3), 24.6 (CH2CH2CH2CH3, JPC = 5.8 Hz), 24.9 (CH2-
CH2CH2CH3, JPC = 12.0 Hz), 27.2 (CH2CH2CH2CH3, JPC = 3.4 Hz),
167.2 (CO). 31P{1H} NMR (CDCl3): d ꢁ16.7.
All reactions were carried out under an atmosphere of purified
nitrogen (O2 traces: copper oxide catalyst, BASF AG, Ludwigshafen;
H2O: molecular sieve 4 Å, Aldrich Company) using standard
Schlenk techniques. Dichloromethane was purified by distillation
from P2O5; diethyl ether was purified by distillation from so-
dium/benzophenone ketyl; n-pentane was purified by distillation
from calcium hydride. Infrared spectra were recorded with a Perkin
Elmer FT-IR spectrometer (Spectrum 1000). 1H NMR spectra were
recorded with a Bruker Avance 250 spectrometer operating at
250.130 MHz in the Fourier transform mode; 13C{1H} NMR spectra
were recorded at 62.895 MHz. Chemical shifts are reported in d
units (parts per million) downfield from tetramethylsilane
(d = 0.0 ppm) with the solvent as the reference signal (1H NMR,
CDCl3 d = 7.26; 13C{1H} NMR, CDCl3 d = 77.55). 31P{1H} NMR spectra
were recorded at 101.255 MHz in CDCl3 with P(OMe)3 as external
standard (d = 139.0, rel. to H3PO4 (85%) with d = 0.0). Thermogravi-
metric studies were carried out with the Perkin Elmer System Pyris
TGA 6 with a constant heating rate of 8 K minꢁ1 under N2
(1.0 cm3 minꢁ1). For the film preparation a spin-coating track
RC8 with the Gyrset system from Suss Microtec Corp. was used.
In a typical spin-coating experiment 2 mL of the appropriate pre-
cursor was dispensed under a stream of argon on the wafer mate-
rial, which rotated with a spin-off speed between 300–2000 rpm
adjusted for the desired film thickness. After the deposition was
finished, the wafer was transferred to a Vacutherm VT 6060 oven
from Heraeus Corp. and was heated under an oxygen-free argon
atmosphere to 400 °C for 30 min with a variable heating ramp
(1–15 °C minꢁ1) for optimized transformation of the precursor
material into metallic copper. Thus, copper films could be obtained
on silicon wafers priory isolated by silicon oxide and coated with a
metal diffusion barrier layer, for example, TiN or TaN/Ta. Specific
resistivity measurements were carried out as follows: sheet resis-
tance was measured by current/voltage measurements at 81 points
distributed over the whole wafer surface with a four-point probe
using a sheet resistivity mapping system OmniMapÒ RS50/efrom
Prometrix Corp. The film thicknesses were determined by etching
at least three dots and surveying steps in the film using a TENCOR
profilometer. The chemical composition profile was detected by
AES and RBS (Rutherford Backscattering Spectrometry). TEM and
EFTM were used to investigate the grain structure and composi-
tion. SEM micrographs were recorded with a SEM instrument type
DSM 982, Gemini from LEO Corp. EDX analysis was performed with
the EDX system EDWIN (Röntec Corp.) attached to the SEM. Micro-
analyses were performed by the Institute of Organic Chemistry,
Chemnitz University of Technology (Foss Heraeus Vario EL C, H,
N-Analyzer), and by the Institute of Organic Chemistry, University
of Heidelberg (Heraeus C, H, N-Analyser). Melting points were
determined using analytically pure samples, sealed off in nitrogen
purged capillaries on a Gallenkamp MFB 595 010 M melting point
apparatus.
2.3. Synthesis of [(nBu3P)2CuO2CH] (4b)
The title compound was synthesized by applying the same pro-
cedure as used in the preparation of 4a (Section 2.2). Thus, nBu3P
(2a) (4.04 g, 20.0 mmol) was reacted with 1 (1.08 g, 10.0 mmol).
After appropriate work-up, complex 4b could be isolated as a col-
orless liquid. Yield: 3.95 g (7.7 mmol, 77% based on 1).
Elemental analysis: Anal. Calc. for C25H55CuO2P2 (51ꢂ3.20): C,
58.51; H, 10.80. Found: C, 58.34; H, 10.77%. IR (NaCl):
m ]
[cmꢁ1
2956 (vs), 2929 (vs), 2869 (vs), 1607 (vs) (CO, asym), 1562 (s),
1460 (s), 1418 (m), 1377 (m), 1335 (m) (CO, sym), 1207 (m),
1127 (m), 1092 (m), 1053 (m), 1007 (m), 904 (s), 775 (m), 724
(s), 693 (m). 1H NMR (CDCl3):
d 0.77 (t, 18 H, (CH2)3CH3,
JHH = 6.9 Hz), 1.30–1.40 (m, 36 H, (CH2)3CH3), 8.44 (s, 1 H, HCO2ꢁ).
13C{1H} NMR (CDCl3): d 14.0 (CH2CH2CH2CH3), 25.0 (CH2CH2-
CH2CH3, JPC = 11.5 Hz), 25.2 (CH2CH2CH2CH3, JPC = 11.0 Hz), 27.0
(CH2CH2CH2CH3, JPC = 2.4 Hz), 166.8 (HCO2ꢁ). 31P{1H} NMR
(CDCl3): d ꢁ20.8. TG: Tbegin = 60 °C, Tend = 136 °C,
Tbegin = 136 °C, Tend = 284 °C, m = 82.95%; Tbegin = 284 °C, Tend
400 °C, m = 2.26%.
Dm = 3.64%;
D
=
D
2.4. Synthesis of [{[Ti](l-r,p-C„CSiMe3)2}CuO2CH] (5)
Complex 5 was accessible by the reaction of [Ti](C„CSiMe3)2
(3) (165.3 mg, 0.32 mmol) dissolved in 30 mL of dichloromethane
with 1 (34.72 mg, 0.32 mmol) dissolved in 30 mL of dichlorometh-
ane, whereby the dichloromethane solution of 1 is slowly added to
3 at ꢁ20 °C. After 1 h of stirring at this temperature, the reaction
solution was filtered through a pad of Celite and all volatiles were
removed in oil-pump vacuum to afford a deep brown solid mate-
rial. Yield: 170 mg (0.27 mmol, 85% based on 1).
M.p.: 145 °C (dec.). Elemental analysis: Anal. Calc. for C27H45
CuO2Si4Ti (626.42): C, 51.77, H, 7.24. Found: C, 51.43; H, 7.12%. IR
ꢂ
(NaCl):
m
m
[cmꢁ1] 1920 (s) (
m
C„C), 1611 (s) (CO,
m
as), 1454 (s) (CO,
s). 1H NMR (CDCl3): d 0.23 (s, 18 H, SiMe3), 0.37 (s, 18 H, SiMe3),
5.99 (pt, 4 H, C5H4, JHH = 2.2 Hz), 6.14 (pt, 4 H, C5H4, JHH = 2.2 Hz),
8.55 (bs, 1 H, HCOꢁ2 ).
2.5. Synthesis of [(nBu3P)CuO2CHꢀHO2CH] (7a)
Method a: Formic acid (6a) (0.46 g, 10.0 mmol) dissolved in
20 mL of dichloromethane was added to 4a (3.11 g, 10.0 mmol)
dissolved in 30 mL of dichloromethane at ꢁ20 °C. After 1 h of stir-
ring at this temperature the reaction solution was filtered through
a pad of Celite. Afterwards, all volatiles were removed from the fil-
trate in oil-pump vacuum to produce a colorless liquid: Yield:
2.78 g (7.8 mmol, 78% based on 4a).
Method b: Tri-n-butylphosphane (2a) (2.02 g, 10 mmol) was dis-
solved in 10 mL of dichloromethane and this solution was drop-
wise added to [CuO2CH] (1) (1.09 g, 10.0 mmol) and formic acid
(6a) (0.46 g, 10.0 mmol) dissolved in 30 mL of dichloromethane
at ꢁ20 °C. After 1 h of stirring at this temperature, the reaction
solution was filtered through a pad of Celite. Afterwards, all vola-
tiles were removed in oil-pump vacuum to produce a colorless li-
quid. Yield: 2.89 g (8.1 mmol, 81% based on 1).
2.2. Synthesis of [(nBu3P)CuO2CH] (4a)
[CuO2CH] (1) (1.0 g, 10.0 mmol) was suspended in 30 mL of
dichloromethane at ꢁ20 °C and nBu3P (2.02 g, 10.0 mmol) was
dropwise added. After 1 h of stirring at this temperature, the reac-
tion mixture was filtered through a pad of Celite. All volatiles were
removed in oil-pump vacuum to afford a colorless liquid. Yield:
2.29 g (7.4 mmol, 74% based on 1).
Elemental analysis: Anal. Calc. for C13H28CuO2P (31ꢂ0.88): C,
Elemental analysis: Anal. Calc. for C14H30CuO4P (35ꢂ6.91): C,
50.23; H, 9.08. Found: C, 50.10; H, 8.96%. IR (NaCl):
m
[cmꢁ1
]
47.11; H, 8.47. Found: C, 47.09; H, 8.32%. IR (NaCl): m ]
[cmꢁ1
2957 (vs), 2930 (vs), 2870 (vs), 1633 (s) (CO, asym), 1592 (vs),
1461 (s), 1414 (m), 1378 (m), 1346 (m) (CO, sym), 1278 (m),
1209 (m), 1093 (m), 1054 (m), 907 (s), 779 (m), 728 (s). 1H NMR
3419 (w) (OH), 2958 (vs), 2930 (vs), 2871 (s), 1715 (s) (C@O,
HCO2H), 1647 (s), 1593 (s) (CO, asym), 1461 (m), 1378 (m), 1343
(m) (CO, sym), 1257 (m), 1208 (m) (C–O, HCO2H), 1123 (m),

12
A. Tuchscherer et al. / Inorganica Chimica Acta 365 (2011) 10–19
1093 (m), 1063 (m), 1006 (m), 963 (m), 905 (m), 791 (m), 720 (m),
(m), 1287 (s) (CF3, str.), 1140 (s), 1094 (m), 1050 (m), 1003 (m),
967 (m), 906 (s), 795 (m), 720 (m), 703 (m), 670 (m), 516 (m).
1H NMR (CDCl3): d 0.69 (t, 18 H, (CH2)3CH3, JHH = 6.1 Hz), 1.2–1.5
(m, 36 H, (CH2)3CH3), 8.17 (s, 1 H, HCO2ꢁ), 14.81 (s, 1 H, CF3CO2H).
13C{1H} NMR (CDCl3): d 13.7 (CH2CH2CH2CH3), 24.5 (d, CH2CH2
CH2CH3, JPC = 2.4 Hz), 24.6 (d, CH2CH2CH2CH3, JPC = 2.4 Hz), 26.9
(CH2CH2CH2CH3), 116.6 (q, CF3, JCF = 291 Hz), 160.8 (q, CF3CO2,
JFC = 35.9 Hz), 163.6 (HCO2). 31P{1H} NMR (CDCl3): d ꢁ19.0. TG: Tbe-
690 (m), 666 (m). 1H NMR (CDCl3): d 0.54 (t, 9 H, (CH2)3CH3,
ꢁ
JHH = 6.2 Hz), 1.00–1.20 (m, 18 H, (CH2)3CH3), 8.10 (s, 2 H, HCO2
/
HCO2H), 13.7 (s, 1 H, HCO2H). 13C{1H} NMR (CDCl3): d 13.7
(CH2CH2CH2CH3), 24.4 (CH2CH2CH2CH3, JPC = 6.2 Hz), 24.7
(CH2CH2CH2CH3, JPC = 13.0 Hz), 27.1 (CH2CH2CH2CH3, JPC = 2.4 Hz),
165.7 (CO, HCO2ꢁ/HCO2H). 31P{1H} NMR (CDCl3): d ꢁ14.6. TG: Tbe-
gin = 70 °C, Tend = 111 °C,
D
m = 17.4%; Tbegin = 111 °C, Tend = 233 °C,
m = 10.5%.
D
m = 53.6%; Tbegin = 233 °C, Tend = 300 °C,
D
gin = 85 °C, Tend = 232 °C,
m = 39.50%.
Dm = 50.20%; Tbegin = 232 °C, Tend = 300 °C,
D
2.6. Synthesis of [(nBu3P)2CuO2CHꢀHO2CH] (7b)
2.9. Synthesis of [(nBu3P)2CuO2CHꢀHO2CPh] (7e)
Formic acid HO2CH (6a) (0.46 g, 10.0 mmol) dissolved in 20 mL
of dichloromethane was dropwise added to 4b (5.13 g, 10.0 mmol)
dissolved in 30 mL of dichloromethane at ꢁ20 °C. After stirring for
2 h at this temperature, the reaction mixture was filtered through a
pad of Celite and all volatiles were evaporated in oil-pump vacuum
to produce a colorless liquid: Yield: 4.53 g (8.1 mmol, 81% based on
4b).
Complex 7e was prepared as described for 7b (Section 2.6):
[(nBu3)2PCuO2CH] (4b) (5.13 g, 10.0 mmol), PhCO2H (6d) (1.22 g,
10.0 mmol). After appropriate work-up, complex 7e could be iso-
lated as a colorless liquid. Yield: 4.64 g (7.3 mmol, 73% based on
4b).
Elemental analysis: Anal. Calc. for C32H61CuP2O4 (635.33): C,
ꢂ
Elemental analysis: Anal. Calc. for C26H57CuO4P2 (55ꢂ9.23): C,
60.50; H, 9.68. Found: C, 60.32; H, 9.46%. IR (NaCl):
m
[cmꢁ1
]
55.84; H, 10.27. Found: C, 55.67; H, 10.12%. IR (NaCl):
m
[cmꢁ1
]
3060 (w), 2956 (vs), 2927 (vs), 2871 (s), 2737 (m), 1713 (s) (CO,
asym), 1650 (m) (CO, asym), 1597 (m) (CO, sym), 1583 (m) (CO,
sym), 1556 (m) (CO, sym), 1463 (s) (CO, sym), 1417 (m), 1379
(s), 1312 (m), 1264 (s), 1172 (m), 1116 (m), 1094 (m), 1067 (m),
1025 (m), 970 (m), 904 (m), 789 (m), 714 (s), 690 (m), 656 (m),
518 (m). 1H NMR (CDCl3): d 0.82 (t, 18 H, (CH2)3CH3, JHH = 6.4 Hz),
1.2–1.5 (m, 36 H, (CH2)3CH3), 7.2–7.4 (m, 3 H, Ph), 7.98 (s, 1 H, Ph),
8.02 (s, 1 H, Ph), 8.63 (s, 1 H, HCO2ꢁ), 14.22 (s, 1 H, PhCO2H).
13C{1H} NMR (CDCl3): d 14.1 (CH2CH2CH2CH3), 24.9 (d, CH2CH2
CH2CH3, JPC = 9.1 Hz), 25.1 (d, CH2CH2CH2CH3, JPC = 10.1 Hz), 27.5
(d, CH2CH2CH2CH3, JPC = 2.9 Hz), 128.1 (Ph), 130.1 (Ph), 132.1
(Ph), 133.1 (Ph), 167.4 (HCO2ꢁ), 170.0 (PhCO2H). 31P{1H} NMR
2956 (vs), 2920 (vs), 2872 (s), 2737 (m), 2693 (m), 2503 (m),
1738 (s), 1710 (s) (C@O, HO2CH), 1575 (s) (CO, asym), 1465 (s),
1418 (s), 1379 (m) (CO, sym), 1275 (m), 1206 (s) (C–O, HO2CH),
1128 (m), 1094 (s), 1068 (m), 1006 (m), 969 (m), 904 (s), 777
(m), 725 (m), 694 (m), 668 (m). 1H NMR (CDCl3): d 0.64 (t, 18 H,
(CH2)3CH3, JHH = 6.5 Hz), 1.10–1.30 (m, 36 H, (CH2)3CH3), 8.10 (s,
2 H, HCO2ꢁ/HCO2H), 14.40 (s, 1 H, HCO2H). 13C{1H} NMR (CDCl3):
d 14.1 (CH2CH2CH2CH3), 24.9 (CH2CH2CH2CH3, JPC = 1.9 Hz), 25.1
(CH2CH2CH2CH3, JPC = 3.4 Hz), 27.2 (CH2CH2CH2CH3), 165.7 (CO,
HCO2ꢁ/HCO2H). 31P{1H} NMR (CDCl3): d ꢁ17.7. TG: Tbegin = 90 °C,
Tend = 130 °C,
Dm = 12.63%; Tbegin = 130 °C, Tend = 280 °C, Dm =
75.16%.
(CDCl3): d ꢁ19.3. TG: Tbegin = 70 °C, Tend = 200 °C,
Dm = 36.36%;
Tbegin = 200 °C, Tend = 354 °C, m = 47.30%.
D
2.7. Synthesis of [(nBu3P)2CuO2CHꢀHO2CMe] (7c)
2.10. Synthesis of [((cC6H11)3P)2CuO2CHꢀHO2CH] (8a)
Complex 7c was synthesized in the same manner as 7b (Section
2.6). In this respect, [(nBu3)2PCuO2CH] (4b) (5.13 g, 10.0 mmol) was
reacted with CH3CO2H (6b) (0.60 g, 10.0 mmol). After appropriate
work-up, complex 7c could be isolated as a colorless liquid. Yield:
4.64 g (8.1 mmol, 81% based on 4b).
Formic acid (6a) (0.46 g, 10.0 mmol) was dissolved in 20 mL of
dichloromethane and was dropwise added to [CuO2CH] (1)
(1.08 g, 10 mmol) dissolved in 20 mL of dichloromethane at
ꢁ20 °C. After stirring the reaction mixture for 30 min at this tem-
perature, (cC6H11)3P (2b) (5.61 g, 20 mmol) dissolved in 20 mL of
dichloromethane was drop-wise added to the reaction solution.
After stirring for another 1 h at this temperature, the obtained
reaction mixture was filtered through a pad of Celite and all vola-
tiles were evaporated in oil-pump vacuum to give a colorless solid
of 8a. Yield: 6.22 g (8.7 mmol, 87% based on 1). Crystallization from
a dichloromethane solution containing 8a at ꢁ20 °C afforded color-
less single crystals of 8a.
Elemental analysis: Anal. Calc. for C27H59CuP2O4 (57ꢂ3.26): C,
56.57; H, 10.37. Found: C, 56.43; H, 10.22%. IR (NaCl):
m ]
[cmꢁ1
3407 (w) (OH), 2957 (vs), 2931 (vs), 2871 (s), 1748 (s) (CO, asym),
1720 (s) (CO, asym), 1587 (s) (CO, sym), 1463 (s) (CO, sym), 1415
(m), 1378 (m), 1261 (m), 1210 (m), 1130 (w), 1093 (m), 1068
(m), 1007 (m), 969 (m), 904 (s), 791 (m), 723 (m), 666 (w), 603
(m). 1H NMR (CDCl3): d 0.85 (t, 18 H, (CH2)3CH3, JHH = 6.7 Hz),
1.3–1.5 (m, 36 H, (CH2)3CH3), 1.97 (s, 3 H, CH3), 8.47 (s, 1
H=HCO2ꢁ), 12.19 (s, 1 H, CH3CO2H). 13C{1H} NMR (CDCl3): d 14.0
(CH2CH2CH2CH3), 21.0 (CH3CO2ꢁ), 24.9 (CH2CH2CH2CH3), 25.1
(CH2CH2CH2CH3, JPC = 2.4 Hz), 27.1 (CH2CH2CH2CH3, JPC = 2.4 Hz),
175.1 (CH3CO2), 162.7 (HCO2). 31P{1H} NMR (CDCl3): d ꢁ19.0. TG:
M.p.: 85 °C. Elemental analysis: Anal. Calc. for C38H69CuO4Pꢂ2
(715.46): C, 63.79; H, 9.72. Found: C, 63.58; H, 9.64%. IR (KBr):
m
[cmꢁ1] 3144 (m), 2926 (s), 2850 (s) (
mCH), 2668 (m) (mCH), 2515
(m), 2267 (m), 1943 (m) (broad, OH), 1737 (s) (CO, asym, HCO2H),
1576 (vs) (CO, sym, HCO2ꢁ), 1447 (s) (CO, sym, HCO2H), 1365 (m),
(CO, sym, HCO2ꢁ), 1295 (m), 1269 (m), 1177 (s), 1128 (m), 1073
(m), 1049 (m), 1004 (m), 916 (m), 888 (m), 852 (s), 819 (m), 781
(m), 750 (m), 693 (s), 669 (m), 512 (m), 489 (m). 1H NMR (CDCl3):
d 1.20–2.27 (m, 66 H, C6H11), 8.47 (s, 2 H, CH), 12.75 (s, 1 H, OH).
13C{1H} NMR (CDCl3): d 26.5 (C6H11), 27.8 (C6H11), 30.8 (C6H11),
32.9 (C6H11), 165.7 (HCO2H/HCO2ꢁ). 31P{1H} NMR (CDCl3): d 11.29.
Tbegin = 68 °C, Tend = 135 °C,
328 °C, m = 74.56%.
Dm = 11.25%; Tbegin = 135 °C, Tend =
D
2.8. Synthesis of [(nBu3P)2CuO2CHꢀHO2CF3] (7d)
Complex 7d was synthesized in the same manner as 7b (Section
2.6): [(nBu3)2PCuO2CH] (4b) (5.13 g, 10.0 mmol), CF3CO2H (6c)
(1.14 g, 10 mmol). After appropriate work-up, 7d could be isolated
as a colorless liquid. Yield: 4.77 g (7.6 mmol, 76% based on 4b).
Elemental analysis: Anal. Calc. for C27H56CuF3P2O4 (627.23): C,
51.70; H, 9.00. Found: C, 51.56; H, 8.97%. IR (NaCl):
2.11. Synthesis of [((cC6H11)3P)3CuO2CHꢀHO2CH] (8b)
ꢂ
m
[cmꢁ1
]
Complex 8b was synthesized in the same manner as 8a (Sec-
tion 2.10): [CuO2CH] (1) (1.08 g, 10.0 mmol), (cC6H11)3P (2b)
(8.42 g, 30 mmol), HO2CH (6a) (0.46 g, 10.0 mmol). After appropri-
2960 (vs), 2967 (vs), 2873 (s), 1737 (s) (CO, asym), 1683 (s) (CO,
asym), 1666 (s) (CO, sym), 1463 (s) (CO, sym), 1416 (m), 1380

A. Tuchscherer et al. / Inorganica Chimica Acta 365 (2011) 10–19
13
ate work-up, complex 8b could be obtained as a colorless solid.
Yield: 8.86 g (8.9 mmol, 89% based on 1).
12 H, CH2), 8.72 (s, 2 H, HCO2/ꢁO2CH), 13.84 (s, 1 H, CO2H).
13C{1H} NMR (CDCl3): d 16.7 (CH3), 60.8 (d, CH2, JPC = 24.8 Hz),
165.4 (s, HCO2H), 174.2 (s, HCO2ꢁ). 31P{1H} NMR (CDCl3): d 117.0.
M.p.: 76 °C. Elemental analysis: Anal. Calc. for C56H102CuO4Pꢂ3
(995.89): C, 67.54; H, 10.32. Found: C, 67.43; H, 9.97%. IR (KBr):
m
[cmꢁ1] 2921 (vs), 2848 (s), 2665 (m) (
m
CH), 2605 (w) (
m
CH), 1942
2.15. Synthesis of [((EtO)3P)3CuO2CHꢀHO2CH] (10b)
(m) (broad, OH), 1743 (s) (CO, asym, HCO2H), 1702 (m), (CO, asym,
HCO2ꢁ), 1576 (s) (CO, sym, HCO2H), 1446 (s) (CO, sym, HCO2ꢁ),
1353 (m), 1294 (m), 1207 (m), 1175 (m), 1129 (w), 1107 (m),
1048 (w), 1003 (m), 916 (m), 888 (m), 850 (s), 819 (m), 781 (m),
750 (m), 693 (s), 669 (m), 512 (m), 489 (m). 1H NMR (CDCl3): d
1.23–1.82 (m, 99 H, C6H11), 8.44 (s, 2 H, CH), 12.75 (s, 1 H, OH).
Complex 10b was synthesized in the same manner as 8a (Sec-
tion 2.10): [CuO2CH] (1) (1.08 g, 10.0 mmol), (CH3CH2O)3P (2d)
(4.99 g, 30.0 mmol), HO2CH (6a) (0.46 g, 10 mmol). After appropri-
ate work-up, complex 10b was obtained as a colorless liquid. Yield:
5.81 g (8.9 mmol, 89% based on 1).
13C{1H} NMR (CDCl3):
d
26.73 (C6H11), 27.98 (C6H11), 30.92
Elemental analysis: Anal. Calc. for C20H48CuO13P3 (653.06): C,
ꢂ
(C6H11), 32.81 (C6H11), 166.56 (HCO2H/HCO2ꢁ). 31P{1H} NMR
(CDCl3): d 10.19.
36.78; H, 7.41. Found: C, 36.49; H, 7.21%. IR (NaCl):
2983 (vs), 2930 (s), 2900 (s), 2708 (m) (
m
[cmꢁ1
CH),
]
m
CH), 2509 (m) (
m
1952 (m) (broad, OH), 1723 (vs) (CO, asym, HCO2H), 1585 (vs),
(CO, asym, HCO2ꢁ), 1477 (m) (CO, sym, HCO2H), 1443 (m) (CO,
sym HCO2ꢁ), 1391 (s), 1292 (m), 1188 (m), 1163 (s), 1093 (s),
1023 (vs), 946 (vs), 777 (s), 691 (m), 538 (m). 1H NMR (CDCl3): d
1.19 (t, 27 H, CH3, JHH = 7.1), 3.90 (tq, 18 H, CH2, JHH = 7.0 Hz),
8.73 (s, 2 H, HO2CH/HCO2ꢁ), 13.38 (s, 1 H, CO2H). 13C{1H} NMR
2.12. Synthesis of [((CF3CH2O)3P)2CuO2CHꢀHO2CH] (9a)
Complex 9a was synthesized as described earlier (Section 2.10):
[CuO2CH] (1) (1.08 g, 10.0 mmol), (CF3CH2O)3P (2c) (6.56 g,
20.0 mmol), HO2CH (6a) (0.46 g, 10 mmol). Yield of colorless 9a:
6.90 g (8.5 mmol, 85% based on 1).
(CDCl3):
d 16.5 (CH3), 60.4 (d, CH2, JPC = 25.6 Hz), 165.0 (s,
Elemental analysis: Anal. Calc. for C14H15CuF18O10P2 (810.73): C,
HCO2H/HCO2ꢁ). 31P{1H} NMR (CDCl3): d 116.9.
ꢂ
20.74; H, 1.86. Found: C, 20.75; H, 1.63%. IR (NaCl):
(NaCl): 2962 (m), 2893 (w) ( CH), 2803 (w) (
m
[cmꢁ1
]
m
mCH), 1915 (m) (broad,
2.16. Synthesis of [{[Ti](
l
-r
,p-C„CSiMe3)2}CuO2CHꢀHO2CH] (11a)
OH), 1698 (s) (CO, asym, HCO2H), 1456 (m) (CO, sym, HCO2ꢁ), 1417
(s) (CO, sym, HCO2H), 1373 (m), (CO, sym, HCO2ꢁ), 1284 (vs), 1173
(s), 1107 (s), 1067 (vs), 992 (m), 963 (vs), 856 (s), 656 (s), 579 (m),
547 (m), 517 (m). 1H NMR (CDCl3): d 4.37 (m, 12 H, CF3CH2O), 6.80
(s, 2 H, CH), 9.70 (s, 1 H, OH). 13C{1H} NMR (CDCl3): d 61.88
To [CuO2CH] (1) (32.57 mg, 0.30 mmol) in 30 mL of dichloro-
methane was added formic acid (6a) (14.12 mg, 0.30 mmol) in a
single portion at 20 °C. The reaction mixture was stirred for
20 min at this temperature. Afterwards, [Ti](C„CSiMe3)2 (3)
(155 mg, 0.30 mmol) dissolved in 20 mL of dichloromethane was
dropwise added to the reaction solution and stirring was continued
for 1 h. Then the reaction mixture was filtered through a pad of
Celite and all volatiles were removed in oil-pump vacuum to give
a deep brown solid. Yield: 200 mg (0.295 mmol, 98% based on 1).
Crystallization from a dichloromethane solution containing 11a
at ꢁ20 °C gave deep brown single crystals of 11a.
(CF3CH2O, JFC = 37.9 Hz, JPC = 4.8 Hz), 123.18 (CF3CH2O, JFC
=
277.4 Hz, JPC = 8.6 Hz), 159.21 (HCO2H/HCO2ꢁ). 31P{1H} NMR
(CDCl3): d 117.33.
2.13. Synthesis of [((CF3CH2O)3P)3CuO2CHꢀHO2CH] (9b)
Complex 9b was synthesized as described earlier (Section 2.10):
[CuO2CH] (1) (1.08 g, 10.0 mmol), (CF3CH2O)3P (2c) (9.84 g,
30.0 mmol), HO2CH (6a) (0.46 g, 10 mmol). After appropriate
work-up, complex 9b was obtained as a colorless liquid. Yield:
9.33 g (8.2 mmol, 82% based on 1).
M.p.: 133 °C (dec.). Elemental analysis: Anal. Calc. for C28H47
-
CuO4Si4Ti (671.44): C, 50.10; H, 7.06. Found: C, 50.40; H, 7.53%.
ꢂ
IR (NaCl):
m mOH), 1918 (m) (mC„C), 1727 (s)
[cmꢁ1] 3396 (broad) (
(CO, asym), 1621 (s) (CO, asym), 1455 (s) (CO, sym), 1406 (s)
(CO, sym), 1380 (m), 1295 (m), 1261 (s), 1183 (m). 1H NMR
(CDCl3): d 0.19 (s, 18 H, SiMe3), 0.24 (s, 18 H, SiMe3), 6.11 (pt, 4
H, C5H4, JHH = 2.34 Hz), 6.69 (pt, 4 H, C5H4, JHH = 2.34 Hz), 8.53 (s,
2 H, CO2H), 11.95 (s, 1 H, HO2CH). 13C{1H} NMR (CDCl3): d 0.9
(SiMe3), 1.4 (SiMe3), 114.1 (CH/C5H4), 117.0 (CH/C5H4), 122.8 (iC/
C5H4), 135.0 (C„CSi), 167.9 (TiC„C), 170.1 (CO2), 170.8 (CO2).
ESI-MS ([m/z (rel. inten.)] (Thf)): [[Ti](C„CSiMe3)2]+, 581 (50).
Elemental analysis: Anal. Calc. for C20H21CuF27O13P3 (1138.80): C,
ꢂ
21.09; H, 1.86. Found: C, 20.97; H, 1.73%. IR (NaCl):
(NaCl): 2970 (s), 2885 (s) ( CH), 2715 (m) (
m
[cmꢁ1
]
m
m
CH), 2524 (m), 1920
(m) (broad, OH), 1706 (s) (CO, asym, HCO2H), 1455 (s) (CO, sym,
HCO2ꢁ), 1417 (s), 1284 (s), 1171 (vs), 1063 (vs), 963 (vs), 854 (s),
657 (m), 572 (m), 511 (m), 433 (m). 1H NMR (CDCl3): d 4.36 (m,
18 H, CF3CH2O), 6.81 (s, 2 H, CH), 10.25 (s, 1 H, OH). 13C{1H}
NMR (CDCl3):
d
159.0 (HCO2H/HCO2ꢁ), 122.7 (CF3CH2O,
61.7 (CF3CH2O, JFC = 37.8 Hz,
JFC = 277.3 Hz,
JPC = 5.8 Hz),
2.17. Synthesis of [{[Ti](
l
-r
,p-C„CSiMe3)2}CuO2CHꢀHO2CPh] (11b)
JPC = 4.7 Hz). 31P{1H} NMR (CDCl3): d 120.0.
To [CuO2CH] (1) (30.00 mg, 0.276 mmol) dissolved in 30 mL of
dichloromethane was added HO2CPh (6d) (33.75 mg, 0.276 mmol)
in 30 mL of dichloromethane at ambient temperature. The reaction
solution was stirred for 30 min and then [Ti](C„CSiMe3)2 (3)
(142.83 mg, 0.276 mmol) was added in a single portion. The
work-up procedure according to the preparation of 6 gave a deep
brown solid. Yield: 170 mg (0.2276 mmol, 86% based on 1).
M.p.: 140 °C (dec.). Elemental analysis: Anal. Calc. for C34H51
2.14. Synthesis of [((CH3CH2O)3P)2CuO2CHꢀHO2CH] (10a)
Complex 10a was synthesized as described earlier (Section
2.10): [CuO2CH] (1) (1.08 g, 10.0 mmol), (CH3CH2O)3P (2d)
(3.33 g, 20.0 mmol), HO2CH (6a) (0.46 g, 10 mmol). After appropri-
ate work-up, 10a was obtained as a colorless liquid. Yield: 4.24 g
(8.7 mmol, 87% based on 1).
Elemental analysis: Anal. Calc. for C14H33CuO10P2 (486.90): C,
CuO4Si4Ti (747.543): C, 54.63; H, 6.88. Found: C, 54.56; H, 6.53%.
ꢂ
ꢂ
34.54; H, 6.83. Found: C, 34.43; H, 6.67%. IR (NaCl):
3439 (w), 2982 (vs), 2930 (s), 2904 (s), 2708 (m) (
m
[cmꢁ1
]
IR (NaCl):
m
[cmꢁ1] 3486 (broad), 1916 (m) (
mC„C), 1724 (vs)
m
CH), 2509 (m)
(C@O, HO2CPh), 1696 (vs) (CO, asym), 1404 (s) (C–O, sym), 1374
(s) (CO, sym). 1H NMR (D6-acetone): d 0.20 (s, 18 H, SiMe3), 0.25
(s, 18 H, SiMe3), 6.27 (pt, 4 H, C5H4, JHH = 2.2 Hz), 6.35 (pt, 4 H,
C5H4, JHH = 2.2 Hz), 6.77 (m, 1 H, Ph), 7.62 (bs, 2 H, Ph), 8.10 (bs,
2 H, Ph), 9.61 (bs, 1 H, ꢁO2CH). 13C{1H} NMR (D6-acetone): d 0.8
(SiMe3), 1.0 (SiMe3), 114.4 (CH/C5H4), 117.1 (CH/C5H4), 123.1
(mCH), 1956 (m) (broad, OH), 1722 (vs) (CO, asym, HCO2H), 1589
(vs) (CO, asym, HCO2ꢁ), 1477 (m) (CO, asym, HCO2H), 1444 (m)
(CO, sym, HCO2ꢁ), 1391 (s), 1347 (m), 1296 (m), 1185 (m), 1163
(s), 1097 (s), 1025 (vs), 947 (vs), 777 (s), 691 (m), 665 (w), 538
(m). 1H NMR (CDCl3): d 1.14 (t, 18 H, CH3, JHH = 7.0 Hz), 3.88 (m,

14
A. Tuchscherer et al. / Inorganica Chimica Acta 365 (2011) 10–19
Table 1
2
Crystal data, collection parameter and refinement parameter for complexes 8a and
11a.
8a
11a
2
m
Formula
C
38H69CuO4P2
C28H47CuO4Si4Ti
671.46
dark brown
0.4 ꢃ 0.3 ꢃ 0.2
triclinic
Formula weight
Crystal color
Crystal dimensions (mm)
Crystal system
Space group
a (Å)
715.41
colorless
0.7 ꢃ 0.3 ꢃ 0.3
monoclinic
P21/n
10.095(2)
27.427(6)
14.605(4)
90.0
99.062(6)
90.0
3993.6(15)
4
m
2
ꢀ
P1
11.068(4)
12.451(4)
14.752(5)
96.287(6)
110.109(6)
98.767(7)
1857.7(11)
2
b (Å)
c (Å)
Scheme 1. Synthesis of 7–11 by reacting 1, 4, and 5 with 2, 3, and 6, respectively. (i)
Dichloromethane, 1 h, ꢁ20 °C; (ii) dichloromethane, 0.5–2 h, ꢁ20 °C.
a
(°)
b (°)
c
(°)
V (Å3)
The structures were solved by direct methods using SHELXS 97 [6].
The structures were refined by full-matrix least-squares proce-
dures on F2 using SHELXL 97 [7]. All non-hydrogen atoms were re-
fined anisotropically. In complex 11a the hydrogen atom
positions at the formiate and the formiat acid have been taken
from the difference Fourier map and refined isotropically. All other
hydrogen atom positions were refined using a riding model. In
complex 8a the hydrogen atom positions less H200 have been re-
fined using the atom corresponding riding model. The positions of
H200 have been taken from the difference Fourier map and refined
freely. Two cyclohexyl rings and the formate molecule are disor-
dered and have been refined to split occupancies of 0.63/0.37
(C19 to C24), 0.65/0.35 (C31 to C36) and 0.57/0.43 (O3 to O4),
respectively.
Z value
Dcalc (g cmꢁ3
)
1.190
1.199
Index ranges
ꢁ12 6 h 6 12,
0 6 k 6 35,
0 6 l 6 18
1552
0.662
1.48–27.16
k = 0.71073
ꢁ13 6 h 6 12,
ꢁ15 6 k 6 15,
0 6 l 6 18
706
0.943
1.68–26.68
k = 0.71073
F(0 0 0)
l
(Mo K
a
) (mmꢁ1
)
h Range for data collection
0
Radiation, Mo K
Temperature (K)
Number of reflections measured
Independent reflections observed
[I > 2r(I)]
a (AÅ)
298
49 462
9030
298
14 611
7184
Data/restraints/parameters
Tmaximum/minimum
8827/493/548
0.99999/
0.43452
0.0406/0.0691
0.0903/0.1005
1.028
7184/0/367
0.99999/
0.75234
0.0447/0.0635
0.1209/0.1311
1.029
R1 [I > 2
wR2 [I > 2
Goodness of fit (GOF) on F2c
r
(I)]/alla
r
(I)]/allb
3. Results and discussion
D
D
qmax (e Åꢁ3
qmin (e Åꢁ3
)
)
0.278
ꢁ0.248
0.398
ꢁ0.459
3.1. Synthesis
a
b
c
R1
wR2 = [
S = [
=
R
|Fobs ꢁ Fcalc|/
RFcalc.
2
R
w(Fobs ꢁ Fcalc2)]2/[
R
w(Fobs2)]2.
A straightforward synthesis methodology for the high-yield
w(Fobs ꢁ Fcalc2)]/(n ꢁ p)1/2
, n = number of reflections; p = parameters
2
R
preparation of the copper(I) formates [LmCuO2CH] (L = nBu3P: 4a,
used.
m = 1; 4b, m = 2. 5, L = [Ti](C„CSiMe3)2, m = 1, [Ti] = (g
5-C5H4Si-
Me3)2Ti) is given in Reaction 1 and includes the treatment of
[CuO2CH] (1) with L (2a, L = nBu3P; 3, L = [Ti](C„CSiMe3)2) in the
ratio of 1:1 (synthesis of 4a and 5) or 1:2 (4b) in dichloromethane
as solvent. The synthesis of the appropriate copper(I) formates
with L = P(cC6H11)3, P(OCH2CH3)3 and P(OCH2CF3)3 is described
elsewhere [5].
(iC(C5H4), 123.9 (Ph), 128.6 (Ph), 130.3 (Ph), 131.9 (Ph), 135.0
(C„CSi), 169.7 (TiC„C), 171.9 (CO2), 172.6 (CO2).
2.18. Structure determination of 8a and 11a
Crystal data crystal data, collection parameter and refinement
parameter of 8a and 11a are presented in Table 1. Data were col-
lected with a Bruker Smart CCD diffractometer at 298 K using
After appropriate work-up, complexes 4a and 4b could be isolated
as colorless liquids, while heterobimetallic 5 forms a dark brown
solid with a decomposition point of 145 °C (Experimental Part).
Mo Ka radiation (k = 0.71073 Å). Reflections were collected in the
h scan modus in 0.3° steps and an exposure time of 20 s per frame.

A. Tuchscherer et al. / Inorganica Chimica Acta 365 (2011) 10–19
15
Most likely, 4a and 4b are tetramers or dimers [3], though 5 fea-
tures a low-valent copper(I) formate moiety stabilized by the che-
lating effect of the organometallic p-tweezer [Ti](C„CSiMe3)2 [8].
Organometallic 5 is fairly stable both in the solid state and in solu-
tion, while 4a and 4b tend to decompose on prolonged storage even
in the solid state upon formation of greenish products, i.e. copper
oxide-containing materials.
for ‘‘weak” to ‘‘medium” hydrogen bonds of type A–Hꢀ ꢀ ꢀB (
m
(AH))
an absorption in the range of 2000–3500 cmꢁ1 occurs, for com-
pounds with ‘‘strong” to ‘‘very strong” hydrogen bonds very broad
vibrations are found below 1600 cmꢁ1 [11], which is illustrated by,
for example, hydrogen-bridged [HCO2H]2 [12,13]. There a Oꢀ ꢀ ꢀO dis-
tance of 2.703 Å [13] and a m(OH) vibration with the maximum at ca.
3100 cmꢁ1 [12] is observed, while for the potassium salt
Reaction of 4a, 4b, and 5 with an excess of formic acid (6a), ace-
tic acid (6b), trifluoroacetic acid (6c) or benzoic acid (6d) produced
the more stable metal–organic complexes [LmCuO2CHꢀHO2CR] and
K[H(O2CH)2] an Oꢀ ꢀ ꢀO separation of 2.45 Å and
a m(OH) at
1400 cmꢁ1 is characteristic [14] (for comparison see also, single so-
lid state structure of 8a, Fig. 1).
[{[Ti](l-
r,p-C„CSiMe3)2}CuO2CHꢀHO2CR] ([Ti] = (
g
5-C5H4SiMe3)2-
The 1H NMR spectra of all newly synthesized transition metal
complexes (Table 2) show well-resolved resonance signals with
the expected coupling patterns and are consistent with the empir-
ical formula given in the Experimental Part. Most characteristic for
all complexes is the appearance of a very characteristic signal be-
tween 8.1 and 8.7 ppm (9a,b: 6.8 ppm) which can be assigned to
the formate CH hydrogen atom of the non-solvated complexes.
For the appropriate mono-solvated formic acid and carboxylic acid
adducts (Table 2) a second signal is observed between 9.6 and
14.8 ppm typical for the HO hydrogen atom. In addition to the for-
mate, formic acid and carboxylic acid resonances, signals for the
auxiliary ligands L = PnBu3, P(cC6H11)3, P(OCH2CH3)3, P(OCH2CF3)3
and the chelating [Ti](C„CSiMe3)2 building block are observed in
the range characteristic for this type of coordinating ligands
(Experimental Part) [5].
Ti; R = H, C6H5) (synthesis of 7a–7e, 11a, and 11b) (Scheme 1, Table
2) in which a formic acid or a carboxylic acid molecule is addition-
ally hydrogen-bonded to the keto-oxygen atom of the copper for-
mate building block. Another possibility to synthesize the formic
or carboxylic acid adducts 7–11 (Table 2) is given by the consecu-
tive reaction of [CuO2CH] (1) with m equivalents of L (2a, L = nBu3P;
2b, L (cC6H11)3P; 2c, L = (CF3CH2O)3P; 2d, L = (CH3CH2O)3P; 3,
L = [Ti](C„CSiMe3)2) followed by addition of HO2CR (6a, R = H;
6b, R = Me) (Scheme 1). The yield of individual 7–11 is 73–98%
(Experimental Part, Table 2).
Complexes 7–11 are, compared with their non-coordinated,
non-solvated counter parts 4 and 5, significantly more stabile and
can, for example, be handled even in air for a short period of time.
The non-solvated as well as formic acid and carboxylic acid
mono-solvated adducts were characterized by elemental analysis
and IR and NMR (1H, 13C{1H}) spectroscopy. The molecular struc-
tures of 8a and 11a in the solid state were determined by single
X-ray structure analysis.
As the 1H NMR spectra of 4, 5, and 7–11, also the 13C{1H} NMR
spectra of these molꢁecules show well-resolved signals with reso-
nances for the HCO2 anions and RCO2H solvates (R = H, CH3, CF3,
C6H5) being most typical (Experimental Part). For the non-solvated
copper(I) formats [CuO2CH] one signal for the formate carbon atom
is found at ca. 167 ppm, while in the corresponding mono-solvated
species this signal is somewhat shifted to higher field. For the
hydrogen-bonded formic and carboxylic acid molecules reso-
nances are found between 159 and 175 ppm depending on the
organic groups. The signals of the formic and carboxylic acid are
observed as expected (Experimental Part).
Most characteristic in the IR spectra of all complexes are the
strong
1454 (
m
m
(CO2) vibrations of the carboxylato units between 1284–
s(CO2)) and 1455–1702 cmꢁ1
(m
a(CO2)) ( s = symmetric,
m
m
a = anti-symmetric O–C–O stretching mode) (Experimental Part),
which is typical in transition metal chemistry [9]. Whereas the
anti-symmetric (CO2) bands can always be detected, the respective
symmetric vibrations only at times, which is attributed to additional
ligand absorptions in this region. The frequencies of the (CO2)
modes of the hydrogen-bonded solvated RO2CH acids (R = H, CH3,
CF3, C6H5) are observed between 1698–1748 cmꢁ1 for
(C@O) and
1163–1278 cmꢁ1 for
(C–O). The differences between these values
m
m
m
m
range from 450 to 570 cmꢁ1 and show that the formic and carboxylic
acids are situated between the data for the free acid molecules (e.g.,
HCO2H, 671 cmꢁ1) and, for example, Na[H(HCO2)2] (439 cmꢁ1),
which is similar to the hydrogen-bonded dimer of formic acid
(524 cmꢁ1) [10a]. These data are also characteristic for other sol-
vated transition complexes including [(Ph3P)mMO2CHꢀHO2CH]
(M = Cu, Ag; m = 2, 3) [2a], [(Ph3P)2CuO2CHꢀ1/2EtOH] [2a], and
[((C6H4-2-NMe2)3P)CuO2CHꢀHO2CR] (R = H, CH3) [2b]. In general,
Table 2
Synthesis of complexes [LmCuO2CHꢀHO2CR] and [{[Ti](
l
-
r,
p
-C„CSiMe3)2}
CuO2CHꢀHO2CR] (7–11).
Compounds
L
M
R
Yielda
7a
7b
7c
7d
7e
8a
8b
9a
nBu3P
1
2
2
2
2
2
3
2
3
2
3
1
1
H
H
Me
CF3
Ph
H
H
H
H
H
78
81
81
76
73
87
89
85
82
87
89
98
86
nBu3P
nBu3P
nBu3P
nBu3P
(cC6H11)3P
(cC6H11)3P
(CF3CH2O)3P
(CF3CH2O)3P
(CH3CH2O)3P
(CH3CH2O)3P
[Ti](C„CSiMe3)2
[Ti](C„CSiMe3)2
Fig. 1. ORTEP diagram (30% probability level) of the molecular structure of 8a. All
cyclohexyl-bonded hydrogen atoms have been omitted for clarity. Only one atomic
position is displayed of disordered atoms (C19–C24, C31–C36, and O3, O4, C38,
respectively). Selected bond distances (Å) and bond angles (°): Cu(1)–O(1)
2.0657(19), Cu(1)–P(1) 2.2381(6), Cu(1)–P(2) 2.2431(7), C(37)–O(1) 1.218(4)
C(37)–O(2) 1.210(4), C(38)–O(3) 1.297(15), C(38)–O(4) 1.175(13); O(1)–Cu(1)–
P(1) 114.34(6), P(1)–Cu(1)–P(2) 143.31(2), C(37)–O(1)–Cu(1) 111.7(2), O(1)–Cu(1)–
P(2) 102.07(6), O(2)–C(37)–O(1) 124.7(3), O(4)–C(38)–O(3) 122.0(12). Standard
deviations are given as the last significant figure(s) in parentheses.
9b
10a
10b
11a
11b
H
H
Ph
a
Based on 1 (Scheme 1, reaction (i)).

16
A. Tuchscherer et al. / Inorganica Chimica Acta 365 (2011) 10–19
The coordination of the phosphane and phosphite ligands to
ligand (ꢁ32.3 ppm) [15] to copper a typical downfield-shift is ob-
served (4a, ꢁ16.7; 4b, ꢁ20.8 ppm). The phosphorus signals of the
appropriate mono-solvated adducts are found at somewhat lower
field (7a: ꢁ14.6; 7b: ꢁ17.7). Similar observations were made for
all the other phosphane or phosphite copper(I) species (Experi-
mental Part) [5].
copper(I) can nicely been verified from 31P{1H} NMR studies. For
example, upon coordination of the free, non-coordinated nBu3P
3.2. Molecular structures of 8a and 11a in the solid state
The molecular structures of 8a and 11a in the solid state were
elucidated by single crystal X-ray diffraction studies (complex 8a,
Fig. 1; complex 11a, Fig. 2). Single crystals were grown by slowly
cooling a saturated dichloromethane solution containing either
8a or 11a to ꢁ20 °C, whereby colorless (8a) or brown (11a) crystals
were obtained. Selected bond distances (Å) and bond angles (°) are
given in the captions of Figs. 1 and 2 and Table 1. Crystal data and
collection and refinement parameters of both compounds are sum-
marized in Table 1 (Experimental Part).
Mono solvated 8a crystallized in the monoclinic space group
P21/n. The [(cC6H11)3P)2CuO2CH] molecular unit in 8a comprises a
tri-coordinate copper atom with the crystallographically indepen-
dent P(cC6H11 3
sites and the formate anion is
)
ligands occupying two of the three coordination
-bonded through one oxygen atom
r
to the third site forming a somewhat distorted trigonal planar
environment around copper(I) (Fig. 1). The atoms Cu1, P1, P2,
and O1 are thereby in-plane bound (RMS deviation 0.028 Å, highest
deviation from planarity observed for Cu1: 0.0479(5) Å). The range
of the angles around Cu1 is 359.72(9)°, displaying in addition its
distorted trigonal planar environment.
Fig. 2. ORTEP diagram (30% probability level) of the molecular structure of 11a. All
cyclopentadienyl- and trimethylsilyl-bonded hydrogen atoms have been omitted
for clarity. Selected bond distances (Å) and angles (°): Cu(1)–Ti(1) 2.9938(10),
Cu(1)–O(1) 1.949(2), C(1)–Ti(1) 2.131(3), C(6)–Ti(1) 2.116(3), Cu(1)–C(1) 2.084(2),
Cu(1)–C(6) 2.098(3), Cu(1)–C(2) 2.156(3), Cu(1)–C(7) 2.165(3), C(27)–O(1) 1.242(4),
C(27)–O(2) 1.244(5), C(28)–O(3) 1.285(5), C(28)–O(4) 1.176(5), Ti(1)–D(1) 2.067(1),
Ti(1)–D(2) 2.074(1); C(6)–Ti(1)–C(1) 88.59(10), C(2)–C(1)–Ti(1) 166.6(2), C(7)–
C(6)–Ti(1) 166.1(2), C(6)–C(7)–Si(2) 163.2(2), C(1)–C(2)–Si(1) 162.5(2), C(27)–
O(1)–Cu(1) 129.6(2), O(2)–C(27)–O(1) 128.7(4), O(4)–C(28)–O(3) 128.4(5), D(1)–
Ti(1)–D(2) 47.9(1), O(1)–Cu(1)–Ti(1) 172.68(7). D(1) = centroid of C11–C15;
D(2) = centroid of C19–C23. The estimated standard deviations of the last signif-
icant digits are shown in parentheses.
Of complex 8a, the Cu1–O1 distance is 2.0657(19) Å while the
Cu–P distances are 2.2381(6) and 2.2431(7) Å. Comparing these
separations with a uni-dentate [2a] and chelate-bonded formate
anion confirms this finding [2c]. The Cu–O/Cu–P distances, respec-
tively, decreases from ca. 2.2/2.3 Å for a chelate-bonded format li-
ꢁ
gand [2c] to 2.04–2.08/2.24 Å for a
r
-bonded HCO2 anion [2a]. It
should be noted, that 8a represents the first monomeric Cu(I) car-
boxylato complex with the cupper(I) atom possessing an almost
trigonal planar CuP2O setup.
The formic acid solvate present in 8a is hydrogen-bonded to the
copper(I) formate unit to form a building block which best can be
described as a bi-formate anion [H(HCO2)2]ꢁ. This moiety is the
simplest form of a family in which formic or carboxylic acid mole-
cules are bonded to metal formate building blocks by molecular
recognition, and have been the subject of a number of theoretical
studies [16]. For 8a, the hydrogen bond with a Oꢀ ꢀ ꢀO distance of
2.598(10) and 2.583(14) Å (Table 3) of the [H(HCO2)2]ꢁ unit falls
into the category of ‘‘very strong”, which is also verified and typical
for the potassium formate salt K[H(HCO2)2] [14a] (dO..O = 2.45 Å).
This finding, when compared with the appropriate free bi-formate
Table 3
Selected bond distances (Å) and angles (°) of the intermolecular hydrogen bonds of
metal–organic 8a.
D–Hꢀ ꢀ ꢀAa
D–H
Hꢀ ꢀ ꢀA
Dꢀ ꢀ ꢀA
D–Hꢀ ꢀ ꢀA
O(2)–H(200)ꢀ ꢀ ꢀO(3)
0.95(6)
0.95(6)
1.65(2)
1.82(6)
2.598(10)
2.583(14)
174(8)
132(8)
O(2)–H(200)ꢀ ꢀ ꢀO(30)b
a
D = donor atom; A = acceptor atom
b
0
Label refers to the atom O30 of the disordered formate anion.
Fig. 3. TG studies of complexes 7a (left) and 7b (right) (nitrogen, flow rate 20 L hꢁ1, heating rate 8 K minꢁ1).

A. Tuchscherer et al. / Inorganica Chimica Acta 365 (2011) 10–19
17
Fig. 4. Surface SEM image showing the morphology of the copper film obtained from 7b deposited on a TiN-coated oxidized silicon substrate at 400 °C (left), spin-off speed vs.
film thickness (right).
ion, can be resulting from the competition of the formate ion
electron density between the copper ion and the formic acid
adduct part. A similar situation was recently found for [(Ph3P)3
AgO2CHꢀ(HCO2H)2] [10a].
400 °C (Experimental Part). The weight losses were accompanied
by residues in the TG pan, the amount of which was comparable
to the theoretical percentage of metallic copper present in the
complex, though small amounts (0.5–3%) of additional material
are also formed (Experimental Part). Evidently, these complexes
eliminate first the solvated acid, followed by the ligands nBu3P or
(RO)3P upon heating without significant volatilization of the intact
copper formiate. Thus, at atmospheric pressure the majority of the
studied copper species showed a high utility for spin-coating. Fol-
lowing TG studies of complexes [(nBu3P)mCuO2CHꢀCH2O] (7a,
m = 1; 7b, m = 2) are discussed in detail, i.e. the thermal gravimetric
curves of both molecules are depicted in Fig. 3.
The titanium–copper organometallic p-tweezer compound 11a
ꢀ
(Fig. 2) crystallized in the triclinic space group P1.
The molecular solid state structure of 11a resembles to the
structures typically found for heterobimetallic organometallic
p-
tweezer complexes of general type {[M](
l-r,p
-C„CR)2}M0L
([M] = (g5-C5H4R0)2Ti (Zr, Hf); M0L = 10–12 valence electron com-
plex fragment, such as Ni(CO), Co(CO), CuX, AgX, AuX, FeCl2, NiCl2,
CoCl2; R = single-bonded organic or organometallic ligand; R0 = H,
Me, SiMe3; X = halide, pseudo-halide, inorganic, organic or organo-
metallic ligand) with M0 in a planar (M0 = Ni, Co, Cu, Ag, Au) or
pseudo-tetrahedral coordination sphere (M0 = Fe, Ni, Co) [17], and
hence the bis(alkynyl)titanocene framework is not discussed here
in detail. Concerning the [CuO2CHꢀHCO2H] building block following
trends can be seen: (i) the copper–oxygen bond is with 1.949(2) Å
shorter than the corresponding bond distances found in phopshine
or phosphite copper formate systems (vide supra and Ref. [2a]) but
is in the range of similar Ti–Cu tweezer molecules [3f,8a]. This
shows that in 11a the bi-formate unit is somewhat stronger
bonded to copper, compared with the appropriate copper phos-
phane complexes (Refs. [3a–g]). The oxygen–oxygen distance in
the bi-formate building block is 2.571(5) Å and can be considered
as ‘‘very strong” (vide supra). Selected bond distances (Å) and an-
gles (°) of the intermolecular hydrogen bond D–Hꢀ ꢀ ꢀA (O2–
H29ꢀ ꢀ ꢀO3) of 11a are: D–H (1.14(8)), Hꢀ ꢀ ꢀA (1.49(8)), Dꢀ ꢀ ꢀA
(2.571(5)); and D–Hꢀ ꢀ ꢀA (156(7)), whereby D = donor atom and
A = acceptor atom.
For both metal–organic complexes it was found that the mass
loss of 81.49% (7a) and 88.64% (7b) is almost in accordance with
the theoretical percentage calculated for the formation of elemen-
tal copper, the difference is less then 1% (C, O impurities). Similar
results were recently reported for the decomposition of phosphane
and phosphite copper(I) and silver(I) carboxylates, respectively
[4a,4b,10]. Organic products formed during the redox process of
the internal formiate/formic acid entities are in general volatile
and vaporize during the decomposition process. These investiga-
tions prove that both complexes should be suitable as spin-coating
precursors. Since 7b with its two tri-n-butyl phosphane ligands is
more stable towards air and moisture then 7a this molecule was
chosen for further detailed TG and finally spin-coating studies. At-
tempts to remove impurities (vide supra) we repeated the TG mea-
surement of 7b in presence of a reducing agent and hence, as
carrier gas nitrogen containing 2% of hydrogen was applied. It
was found that 7b decomposes in a similar decomposition pattern
as without addition of H2. An overall decomposition starting at 84
and ending at 360 °C is observed (Experimental Part). The mass
loss of 88.85% is now in accordance with the theoretical percentage
(88.64%) calculated for the formation of elemental copper from 7b.
This shows that sintering of the spin-coated wafer should best be
carried out under reducing conditions (vide infra). From the data
obtained (Experimental Part) it is most likely that the mono-sol-
vated complexes (7–11) eliminate at first the hydrogen-bonded
formic acid or carboxylic acid units giving the appropriate non-sol-
vated species [(R3P)mCuO2CH] (R = nBu, cC6H11, OCH2CF3, OCH2CH3;
m = 1, 2, 3). On further heating these molecules eliminate in the
temperature range of 100–400 °C, depending on m and R, ligands
R3P or (RO)3P to afford [CuO2CH]. Decarboxylation finally leads to
elementary copper. Comparable decomposition pathways are
found for dicopper and disilver dicarboxylates (e.g., oxalate, squa-
rate) as it could be demonstrated by TG-MS studies [20].
3.3. TG- and spin-coating-studies
Metal thin film deposition by using the spin-coating process is a
technique for coating a thin metal layer of non-volatile precursors
within a few tens up to several hundred of nanometers onto a sub-
strate or onto previously deposited layers [18]. On heating, a chem-
ical change at the surface takes place leaving a metal film. This
method is, when compared with other deposition techniques
[19], a very cost effective method because it does not require com-
plicated technical and hence, expensive equipment.
TG (ThermoGravimetry) studies were carried out to obtain first
information on the decomposition temperature and on the relative
stability of some of the newly synthesized complexes. Experiments
were conducted at atmospheric pressure under a nitrogen purge.
Exemplary, the TG studies of 4, 7 and 9 are discussed. It was found
that these complexes show a 2–4 step decomposition. In general,
the thermal degradation starts at 80 °C and is completed above
Based on the TG studies (vide supra) metal–organic 7b was ap-
plied as spin-coating precursor in the deposition of copper onto
TiN-coated oxidized silicon wafer materials. The spin-coating

18
A. Tuchscherer et al. / Inorganica Chimica Acta 365 (2011) 10–19
optimized towards the key parameters for mass flow (viscosity),
film formation (surface tension, wetting, etc.) and film thickness.
Since the films always contained traces of solvent molecules
resulting from the precursor material the films are not completely
solidifying during the spinning process. Better results were ob-
tained by higher spin speeds (P2000 rpm), however, at a low film
thickness the precursor films were destroyed during wetting pro-
cesses. In addition, the low copper yield obtained during thermol-
ysis (vide supra) causes open grain network structures, especially,
when solidification takes place before the decomposition process is
finished and hence, the deposited films still contain low boiling or-
ganic molecules. As summary, a film thickness of ca. 300 nm is nec-
essary in order to guarantee comparable sheet resistance results.
The chemical composition versus film thickness relation was
studied by AES (Auger Electron Spectroscopy) (Fig. 5). As result,
small amounts of organic materials as well as phosphorus-contain-
ing residues could be detected. Nevertheless, these deposits could
nicely be removed by hydrogen post-treatment (see above).
Fig. 5. AES spectrum of the copper film obtained by the spin-coating process using
7b as precursor material at 400 °C (deposition occurred on a TiN-coated SiO2 wafer
material).
technique was chosen over physical or Chemical Vapor Deposition
processes (for example, PVD (Physical Vapor Deposition), CVD
(Chemical Vapor Deposition), ALD (Atomic Layer Deposition)) and
electroplating because next to their benefits they do possess some
drawbacks. PVD has, for example, limited gap-fill capabilities [21],
CVD may yield uniformity issues with high areas, especially on
large wafers and is not suitable for films thicker than 100 nm,
and finally the costs for the appropriate precursors as well as
equipment is high (see above) [22]. In contrast, electroplating is
much more effective and hence, more widespread in microelec-
tronics technology but there are also some challenges like control
of metal crystal growth from nano-scale (feature filling) up to the
macro-scale (uniformity across the wafers) or the introduction of
bath additives into the deposited films increasing their resistivity.
Thus, we favor the low cost spin-coating process as an alternative
technique for metal film deposition. The process benefits of supe-
rior thickness uniformity, due to non-conformal coatings, the high
potential to fill small gaps (<100 nm) with high aspect ratios, and
atmospheric-like ambient process conditions.
Mono-solvated highly viscous 7b (vide supra) was used as a
non-volatile material in first spin-coating studies for the deposition
of copper films. Film deposition were verified as follows: In a typ-
ical deposition process 7b dissolved in tetrahydrofuran was dis-
pensed under argon from a pipette onto a pre-cleaned and
afterwards oxidized silicon wafer coated with a double layer of
sputter deposited TiN diffusion barrier and Cu with a thickness of
40/100 nm with 150 rpm rotation speed. A spin-off speed of
2000 rpm was chosen in order to get a thin homogeneous liquid
film on the wafer after 30 s of dispersion time. An industrial
spin-coating system (Suss Microtec) with a closed deposition
chamber and a protecting nitrogen stream was used for these stud-
ies. The treated wafers were transferred to an oven and have then
been heated with a defined ramp rate of 8 K minꢁ1 up to 400 °C.
The decomposition temperature followed the thermal gravimetric
curve shown in Fig. 3. The film was annealed at 400 °C for 30 min
in an N2/H2 (ratio of 98:2) atmosphere. A SEM image of the ob-
tained copper film is shown in Fig. 4 (left). As it can be seen from
this figure, the copper film is not completely homogeneous and
shows some defects. This can be explained by the different film
thickness as result of the spin-coating process, which upon heating
produce defects due to unequal evaporation of the in situ formed
reaction products leading to partially cracked copper films.
Supplementary material
CCDC 765261 and 765262 contain the supplementary crystallo-
graphic data for 8a and 11a. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgement
Financial support from the Deutsche Forschungsgemeinschaft
(International Research Training Group GRK 1215, ‘‘Materials and
Concepts for Advanced Interconnects”), the Fonds der Chemischen
Industrie, and the BASF AG, Ludwigshafen is gratefully ack-
nowledged.
References
[1] (a) T. Ogura, Q. Fernando, Inorg. Chem. 12 (1973) 2611;
(b) D.A. Edwards, R. Richards, J. Chem. Soc., Dalton Trans. (1973) 2463.
[2] (a) G.A. Bowmaker, Effendy, J.V. Hanna, P.C. Healy, J.C. Reid, C.E.F. Rickard, A.H.
White, J. Chem. Soc., Dalton Trans. (2000) 753;
(b) H. Lang, Y. Shen, T. Rüffer, B. Walfort, Inorg. Chim. Acta 361 (2008) 95;
(c) N. Marsich, A. Camus, G. Nardin, J. Inorg. Chem. 239 (1982) 429.
[3] (a) G. La Monica, M.A. Angaroni, F. Cariati, S. Cenini, Inorg. Chim. Acta 143
(1988) 239;
(b) D.J. Darensbourg, M.W. Holtcamp, E.M. Longridge, K.K. Klausmeyer, J.H.
Reibenspies, Inorg. Chim. Acta 227 (1994) 223;
(c) D.J. Darensbourg, M.W. Holtcamp, B. Khandelwal, K.K. Klausmeyer, J.H.
Reibenspies, Inorg. Chem. 34 (1995) 2389;
(d) F. Olbrich, U. Behrens, G. Gröger, E. Weiss, J. Organomet. Chem. 448 (1993)
C10;
(e) S.P. Neo, Z.-Y. Zhou, T.C.W. Mak, T.S.A. Hor, J. Chem. Soc., Dalton Trans.
(1994) 3451;
(f) Y. Sevryugina, M.A. Petrukhina, Eur. J. Inorg. Chem. (2008) 219;
(g) Y.-P. Tian, X.-J. Zhang, J.-Y. Wu, H.-K. Fun, M.-H. Jiang, Z.-Q. Xu, A. Usman, S.
Chantrapromma, L.K. Thompson, New J. Chem. 26 (2002) 1468;
(h) W. Frosch, S. Back, G. Rheinwald, K. Köhler, L. Zsolnai, G. Huttner, H. Lang,
Organometallics 19 (2000) 5769;
(i) Y.-Y. Wang, Q. Shi, Q.-Z. Shi, Y.-C. Gao, X. Hou, Polyhedron 19 (2000) 891;
(j) H. Lang, K. Köhler, L. Zsolnai, Chem. Ber. 128 (1995) 519;
(k) Y. Sevryugina, A.Y. Rogachev, M.A. Petrukhina, Inorg. Chem. 46 (2007) 7870.
[4] (a) A. Grodzicki, I. Łakomska, P. Piszczek, I. Szymánska, E. Szłyk, Coord. Chem.
Rev. 249 (2005) 2232;
(b) N. Roth, A. Jakob, T. Waechtler, S.E. Schulz, T. Gessner, H. Lang, Surf. Coat.
Technol. 201 (2007) 9089;
(c) A. Jakob, Y. Shen, T. Wächtler, S.E. Schulz, T. Gessner, R. Riedel, C. Fasel, H.
Lang, Z. Anorg. Allg. Chem. 634 (2008) 2226.
[5] L. Wittenbecher, H. Lang, Y. Shen, PCT Int. Appl. (2005) WO 20057058789 A2.
[6] G.M. Sheldrick, Acta Crystallogr., Sect. A 46 (1990) 467.
[7] G.M. Sheldrick, SHELXL 97, Program for Crystal Structure Refinement, University
of Göttingen, 1997.
Additionally, the behavior between the spin speed and film
thickness was studied (Fig. 4, right). As it can be seen from this fig-
ure the exponent of ꢁ3 obtained from the spin speed curve is high
as compared with other spin-coating metal–organic precursors
including polyimide complexes (0.2–0.8) [23], which is explainable
by the fact that the deposition process described (vide supra) is not
ˇ
ˇ
ˇ
[8] (a) J. Kühnert, M. Lamac, T. Rüffer, B. Walfort, P. Štepnicka, H. Lang, J.
Organomet. Chem. 692 (2007) 4303;
(b) W. Frosch, S. Back, H. Lang, J. Organomet. Chem. 625 (2001) 140;
(c) W. Frosch, S. Back, G. Rheinwald, K. Köhler, H. Pritzkow, H. Lang, Organo-
metallics (2000) 4016.

A. Tuchscherer et al. / Inorganica Chimica Acta 365 (2011) 10–19
19
[9] (a) E.G. Palaciosa, G. Juárez-López, A.J. Monhemius, Hydrometallurgy 72 (2004)
139;
(b) M. Nara, M. Tanokura, Biochem. Biophys. Res. Commun. 369 (2008) 225;
(c) R.J. Deeth, Inorg. Chem. 47 (2008) 6711.
[10] (a) G.A. Bowmaker, Effendy, J.V. Hanna, P.C. Healy, G.J. Millar, B.W. Skelton,
A.H. White, J. Phys. Chem. 99 (1995) 3909;
(b) A.P. Ehiasarian, J.G. Wen, I. Petrov, J. Appl. Phys. 101 (2007) 054301;
(c) W. Zhang, L. Yi, J. Tu, P. Chang, D. Mao, J. Wu, J. Elec. Mater. 34 (2005) 1307;
(d) O. Zimmer, Surf. Coat. Technol. 200 (2005) 440;
(e) P. Kumara, F. Zhua, A. Madan, Int. J. Hydrogen Energy 33 (2008) 3938;
(f) L. Zeng, H. Peng, W. Wang, Y. Chen, W. Qi, D. Lei, J. Liang, J. Zhao, X. Kong, J.
Phys. Chem. C 112 (2008) 10735;
(b) A. Jakob, H. Schmidt, B. Walfort, G. Rheinwald, S. Frühauf, S. Schulz, T.
Gessner, H. Lang, Z. Anorg. Allg. Chem. 631 (2005) 1079;
(c) H. Schmidt, Y. Shen, M. Leschke, T. Haase, K. Kohse-Höinghaus, H. Lang, J.
Organomet. Chem. 669 (2003) 25;
(g) Q. Cheng, S. Xua, J.W. Chai, S.Y. Huang, Y.P. Ren, J.D. Long, P.P. Rutkevych, K.
Ostrikov, Thin Solid Films 516 (2008) 5991;
(h) S. Mathur, T. Ruegamer, N. Donia, H. Shen, J. Nanosci. Nanotechnol. 8
(2008) 2597;
(d) C. Oldham, W.F. Sandford, J. Chem. Soc., Dalton Trans. (1977) 2068;
(e) D.A. Edwards, M.F. Mahon, K.C. Molloy, V. Ogrodnik, J. Mater. Chem. 13
(2003) 563.
(i) J. Prokop, G. Finnah, J. Lorenz, V. Piotter, R. Ruprecht, J. Hausselt, Microsyst.
Technol. 14 (2008) 1669;
(j) N. Uchiya, Y. Furuta, H. Nishikawa, T. Watanabe, J. Haga, T. Satoh, M.
Oikawa, Y. Ishii, T. Kamiya, Microsyst. Technol. 14 (2008) 1537;
(k) B.-J. Kim, S.-J. Park, J. Colloid Interface Sci. 325 (2008) 297;
(l) K. Klosová, J. Hubálek, Phys. Stat. Sol. A 205 (2008) 1435;
(m) L. Yuan, A. Goldbach, H. Xu, J. Membr. Sci. 322 (2008) 39;
(n) C. Domínguez-Ríos, M.V. Moreno, R. Torres-Sánchez, Wilber Antúnez, A.
Aguilar-Elguézabal, J. González-Hernández, Surf. Coat. Technol. 202 (2008)
4848;
[11] P. Gilli, V. Bertolasi, V. Ferretti, G. Gilli, J. Am. Chem. Soc. 116 (1994) 909.
[12] T. Wachs, D. Borchardt, S.H. Bauer, Spectrochim. Acta, Part A 43 (1987) 965.
[13] A. Almenningen, O. Bastiensen, T. Motzfeldt, Acta Chem. Scand. 23 (1969)
2848.
[14] (a) E. Spinner, J. Am. Chem. Soc. 105 (1983) 756;
(b) G. Larsson, I. Nahringbauer, Acta Crystallogr., Sect. B 24 (1968) 666.
[15] J. Tong, S. Liu, S. Zhang, S.Z. Li, Spectrochim. Acta Part A 67 (2007) 837.
[16] H. Basch, W.J. Stevens, J. Am. Chem. Soc. 113 (1991) 95.
[17] (a) W. Frosch, S. Back, H. Müller, K. Köhler, A. Driess, B. Schiemenz, G. Huttner,
H. Lang, J. Organomet. Chem. 619 (2001) 99;
(o) H. Lang, R. Buschbeck, in: J. Zabicky (Ed.), The Chemistry of Metal Enolates,
John Wiley & Sons Limited, Chichester, 2009.
[20] (a) T. Haase, K. Kohse-Hoeinghaus, N. Bahlawane, P. Djiele, A. Jakob, Heinrich
Lang, Chem. Vap. Deposition 11 (2005) 195;
(b) H. Lang, N. Mansilla, G. Rheinwald, Organometallics 20 (2001) 1592;
S. Back, R.A. Gossage, M. Lutz, I. del Río, A.L. Spek, H. Lang, G. van Koten,
Organometallics 19 (2000) 3296;
(b) T. Haase, K. Kohse-Hoeinghaus, B. Atakan, H. Schmidt, H. Lang, Chem. Vap.
Deposition 9 (2003) 144;
(c) S. Back, G. Rheinwald, H. Lang, J. Organomet. Chem. 601 (2000) 93;
(d) H. Lang, D.S.A. George, G. Rheinwald, Coord. Chem. Rev. 206–207 (2000)
101.
(c) I. Szyman´ ska, P. Piszczek, R. Szcze˛sny, E. Szłyk, Polyhedron 26 (2007) 2440.
[21] (a) P.P. Lau, C.C. Wong, L. Chan, Appl. Surf. Sci. 253 (2006) 2357;
(b) J. Reid, S. Mayer, E. Broadbent, E. Klawuhn, K. Ashtiani, Solid State Technol.
43 (2000) 86.
[22] X. Hou, K.-L. Choy, Chem. Vap. Deposition 12 (2006) 583.
[23] R.K. Yonkoski, D.S. Soane, J. Appl. Phys. 72 (1992) 725.
[18] (a) L. Rast, A. Stanishevsky, Appl. Phys. Lett . 87 (2005) 223118;
(b) S. Schaefers, L. Rast, A. Stanishevsky, Mater. Lett. 60 (2006) 706;
(c) J. He, T. Kunitake, Langmuir 22 (2006) 7881.
[19] (a) C. Niu, F. Wang, G. Magsamen, B. Le, B. Stephenson, Thin Solid Films 516
(2008) 1826;