Syntheses and quadratic hyperpolarizabilities of some (pyridylalkynyl)metal complexes: Crystal structures of [Ni{2-(C≡C)C5H3NNO2-5}(PPh 3)(η-C5H5)], [Au{2-(C≡C)C5H3NNO

Article abstract of DOI:10.1039/a702249b

The complexes [Ru{2-(C≡C)C₅H₃NR-5}(PPh₃) ₂(η-C₅H₅)] (R = NO₂ 1 or H 2), [Ni{2-(C≡C)C₅H₃NR-5}(PPh₃)-(η-C ₅H₅)] (R = NO₂

Full text of DOI:10.1039/a702249b

View Article Online / Journal Homepage / Table of Contents for this issue
Syntheses and quadratic hyperpolarizabilities of some
(pyridylalkynyl)metal complexes: crystal structures of [Ni{2-
᎐
᎐
(C᎐C)C H NNO -5}(PPh )(ç-C H )], [Au{2-(C᎐C)C H NNO -
᎐
᎐
5
3
2
3
5
5
5
3
2
᎐
5}(PPh )] and [Au{2-(C᎐C)C H N}(PPh )]‡
᎐
3
5
4
3
Raina H. Naulty,a Marie P. Cifuentes,a Mark G. Humphrey,*,†,a Stephan Houbrechts,b
Carlo Boutton,b André Persoons,b Graham A. Heath,c David C. R. Hockless,c
Barry Luther-Daviesd and Marek Samocd
a Department of Chemistry, Australian National University, Canberra, ACT 0200, Australia
b Centre for Research on Molecular Electronics and Photonics, Laboratory of Chemical and
Biological Dynamics, University of Leuven, Celestijnenlaan 200D, B-3001 Leuven, Belgium
c Research School of Chemistry, Australian National University, Canberra, ACT 0200, Australia
d Australian Photonics Cooperative Research Centre, Laser Physics Centre, Research School of
Physical Sciences and Engineering, Australian National University, Canberra, ACT 0200,
Australia
᎐
᎐
᎐
The complexes [Ru{2-(C᎐C)C H NR-5}(PPh ) (η-C H )] (R = NO 1 or H 2), [Ni{2-(C᎐C)C H NR-5}(PPh )-
᎐
5
3
3
2
5
5
2
5
3
3
᎐
(η-C H )] (R = NO 3 or H 4) and [Au{2-(C᎐C)C H NR-5}L] (L = PPh , R = NO 5 or H 6; L = PMe , R = NO 7)
᎐
5
5
2
5
3
3
2
3
2
have been synthesized and 3, 5 and 6 structurally characterized; no significant increase in quinoidal vinylidene
contribution to the acetylide ground-state structure is apparent on progression from structurally characterized
phenylacetylide complexes to the new pyridylacetylide complexes, or upon replacement of 5-H by 5-NO2 in
progressing from 6 to 5. The molecular quadratic optical non-linearities of 1–7 have been determined by
hyper-Rayleigh scattering (HRS). The HRS measurements at 1064 nm are consistent with an increase in β upon
replacement of phenyl by an N-heterocyclic ring (replacing a nitrophenylacetylide by a nitropyridylacetylide ligand)
for the ruthenium and gold systems, but with no change for the nickel complexes, and with an increase in non-
linearity upon replacement of PMe3 by PPh3 in progressing from 7 to 5. The bulk second-order susceptibilities
of the series have been determined by Kurtz powder measurements at 1064 nm, with the only significant
response (about eight times that of urea) being that of 3; this complex was the only one of the three structurally
characterized to pack non-centrosymmetrically in the crystal lattice. Electrochemical data for 1–4 have been
obtained; comparison to analogous nitrophenylacetlylide complexes reveals that replacing nitrophenylacetylide
by nitropyridylacetylide leads to a significant increase in MII/III oxidation potential for the ruthenium complexes,
II/III
Ϫ
but to no change for the nickel examples. The parameter EЊM Ϫ EЊNO /NO was evaluated for 1–4, results for
M = Ru vs. Ni being consistent with experimentally determined non-linear2ities, i.e. smaller ∆EЊ and larger non-
2
linearities for ruthenium vs. nickel.
The optical non-linearities of organometallic complexes have
been investigated intensively recently.2 Our focus in this area has
been on metal σ-acetylide complexes with the donor–bridge–
acceptor composition shown to produce enhanced non-linear
optical (NLO) responses in organic molecules.1,3 We have probed
the effect of varying metal and coligand at the ligated metal
donor group, the significance of chain lengthening of the
σ-acetylide ligand by various bridging functionalities, and the
importance of the nitro acceptor group on both quadratic and
cubic NLO performance, and have recently commented on the
significance of metal valence electron count and oxidizability
on quadratic NLO merit, in the process defining ‘figures of
merit’ for several ligand metal centres.
Our studies have thus far focused on σ-phenylacetylide com-
plexes. However, organic compounds where the phenyl rings
are replaced by heterocyclic groups such as thiophene-2,5-diyl
or furyl have been reported to give increased second-order
responses. For example, Dirk et al.4 examined chromophores
in which the phenyl ring of the bridge was replaced with a
thiazolyl group, noting an improved quadratic non-linearity,
β, relative to that of the prototypical organic NLO material
4-(dimethylamino)-4Ј-nitrostilbene (dans); this was attributed
to the decreased aromaticity of the heterocyclic ring compared
to that of the phenyl group. Cheng et al.5 examined chromo-
phores in which a phenyl ring of the bridge was replaced with a
furyl or thiophenyl ring, observing bathochromic shifts in the
optical absorption spectra and enhanced hyperpolarizabilities
compared to those of the aromatic analogues. Replacement
of both of the phenyl rings in an analogue of dans with thio-
phene-2,5-diyl groups resulted in a two-fold increase in the
product of the dipole moment and quadratic non-linearity, µβ.6
Given these precedents in organic chemistry, it seems logical to
examine the NLO merit of heterocyclic σ-acetylide complexes.
This paper reports syntheses, spectroscopic and, in some
instances, X-ray structural and electrochemical characteriz-
ation of new σ-pyridylacetylide complexes, their molecular
quadratic optical non-linearities at 1064 nm and two-level-
corrected values as determined by hyper-Rayleigh scattering
(HRS), and their bulk second-order susceptibilities at 1064 nm
measured by the Kurtz powder technique, together with com-
parisons with previously reported data for the analogous σ-
phenylacetylide complexes.
Experimental
General
† E-Mail: mark.humphrey@anu.edu.au
‡ Organometallic complexes for non-linear optics. Part 13.1
All reactions were carried out under an atmosphere of nitrogen
J. Chem. Soc., Dalton Trans., 1997, Pages 4167–4174
4167

View Article Online
3
3
4
[1 H, d, J(HH) 2, H4], 7.65 [1 H, td, J(HH) 2, J(HH) 1, H5]
and 8.58 [1 H, d, 3J(HH) 1 Hz, H7]. δC(CDCl3) 82.7 (C2), 103.0
(C1), 127.5 (C4), 128.4 (C5), 136.2 (C3), 142.3 (C7) and 150.0
(C6).
H5
C6
C7
H7
H4
C4 C5
Ni C1 C2 C3
R
N
PPh2
Ci
Co Ho
Cp Cm
Hp
Syntheses
᎐
2-(Me SiC᎐C)C H NNO -5 II. 2-Bromo-5-nitropyridine
᎐
3
5
3
2
(1.00 g, 4.54 mmol), trimethylsilylacetylene (0.60 cm3, 0.42 g,
4.3 mmol), CuI (5 mg, 0.03 mmol) and [PdCl2(PPh3)2] (73 mg,
0.22 mmol) were added to deoxygenated triethylamine (20 cm3).
The reaction mixture was stirred for 3 h, after which time the
triethylamine was removed, and the crude product extracted
into dichloromethane and adsorbed onto alumina. Column
chromatography, performed using 20% dichloromethane–80%
light petroleum as eluent, afforded 830 mg (77%) of white
crystals (Found: C, 54.90; H, 5.46; N, 12.32. C10H12N2O2Si
requires C, 54.52; H, 5.49; N, 12.72%). ν max/cmϪ1 (CH2Cl2) =
Hm
Scheme 1 Numbering scheme for NMR spectral assignments of com-
plexes
3
(R = NO2) and
4
(R = H6). Analogous schemes for 2-
᎐
᎐
᎐
(HC᎐C)C5H4N I, 2-(Me3SiC᎐C)C5H3NNO2-5 II, [Ru{2-(C᎐C)C5H3-
᎐
᎐
᎐
᎐
NR-5}(PPh3)2(η-C5H5)] and [Au{2-(C᎐C)C5H3NR-5}(PPh3)]
᎐
H5
H4
C3
C4 C5
Me
Me
Me
C2
C6 NO2
C7
H7
C1
Au
P
N
3
᎐
2061 (C᎐C). δ (CDCl ) 1.56 (9 H, s, Me), 7.60 [1 H, d, J(HH)
᎐
H
3
9, H4], 8.43 [1 H, dd, 3J(HH) 9, 4J(HH) 3, H5] and 9.37 [1 H, d,
4J(HH) 3 Hz, H7]. δC(CDCl3) 0.6 (s, Me), 82.3 (C2), 101.8 (C1),
127.2 (C4), 131.3 (C5), 142.6 (C3), 145.3 (C7) and 148.1 (C6). m/z
220 (Mϩ, 17), 205 (100, [M Ϫ Me]ϩ) and 159 (23%, [M Ϫ
Me Ϫ NO2]ϩ).
Scheme 2 Numbering scheme for NMR spectral assignments of
᎐
[Au{2-(C᎐C)C5H3NNO2-5}(PMe3)] 7
᎐
with the use of standard Schlenk techniques;7 no attempt
was made to exclude air during work-up of products. Solvents
were dried as follows: dichloromethane by distilling over
CaH2, methanol by distilling over magnesium activated with I2,
toluene by distilling over sodium–benzophenone, and diethyl
ether by distilling over sodium–benzophenone; other solvents
were used as received. Column chromatography was carried out
using Merck aluminium oxide 90 active basic (activity stage III,
70–230 mesh ASTM, with 10 cm columns). Light petroleum
refers to a fraction of boiling range 60–80 ЊC.
The following compounds were prepared by literature
methods: 2-ethynylpyridine,8 [NiCl(PPh3)(η-C5H5)],9 [RuCl-
(PPh3)2(η-C5H5)],10 [PdCl2(PPh3)2]11 and [AuCl(PPh3)];12
[AuCl(PMe3)]13 was provided by Dr. Mark Brown, Research
School of Chemistry, Australian National University. Cop-
per() iodide (Aldrich), 2-bromo-5-nitropyridine (Aldrich), tri-
methylsilylacetylene (Aldrich) and triethylamine (BDH) were
used as received.
᎐
[Ru{2-(C᎐C)C H NNO -5}(PPh ) (ç-C H )] 1. A mixture of
᎐
5
3
2
3
2
5
5
[RuCl(PPh3)2(η-C5H5)] (300 mg, 0.41 mmol) and compound II
(132 mg, 0.60 mmol) was refluxed in a methanolic solution of
NaOMe (30 cm3 methanol, 15 mg sodium) for 1 h, and the
resultant solution allowed to cool to room temperature. The
solvent was removed in vacuo and the residue adsorbed onto
alumina. Elution with 50% acetone–50% dichloromethane
afforded a red solution; light petroleum was added and the
product precipitated by reducing the solvent volume on a rotary
evaporator. Upon filtering, 155 mg of a red powder (44%) was
isolated (Found: C, 68.02; H, 4.89; N, 3.40. C48H38N2-
O2P2Ruؒ0.5C3H6O requires C, 68.58; H, 4.77; N, 3.23%). λmax
/
nm (thf) 468 (ε/dm3 molϪ1 cmϪ1 17 400). ν max/cmϪ1 (CH2Cl2) =
2029 cmϪ1 (C᎐C). δ (CDCl ) 2.14 (3 H, s, Me CO), 4.42 (5 H, s,
᎐
᎐
H
3
2
C5H5), 6.50 [1 H, d, J(HH) 9, H4], 7.07 [12 H, t, J(HH) 7 Hz,
3
Microanalyses were performed at the Research School of
Chemistry, Australian National University. Infrared spectra
were recorded as solutions in 0.1 mm CaF2 cells using a Perkin-
Elmer System 2000 FT-IR spectrometer, and UV/VIS spectra
as tetrahydrofuran (thf) solutions using a Cary 5 spectro-
photometer. Mass spectra were recorded using a VG ZAB
2SEQ instrument (30 kV Csϩ ions, current 1 mA, accelerating
potential 8 kV, 3-nitrobenzyl alcohol matrix) at the Research
School of Chemistry, Australian National University; peaks
are reported as m/z (assignment, relative intensity). Electro-
chemical measurements were carried out using a Princeton
Applied Research model 170 potentiostat. The supporting
Hm], 7.19 (6 H, m, Hp), 7.33 (12 H, m, Ho), 8.02 [1 H, dd,
4
4
3J(HH) 9, J(HH) 3, H5] and 9.25 [1 H, d, J(HH) 3 Hz, H7].
δC(CDCl3) 86.3 (C5H5), 121.5 (C2), 124.8 (C4), 127.4 (Cm), 128.8
(Cp), 129.8 (C5), 133.5 (Co), 137.8 (m, Ci), 145.9 (C7); carbon
atoms C1, C3 and C6 were not detected. δP(CDCl3) 51.2. m/z
ϩ
ϩ
᎐
838 (M , 37), 691 {9, [M Ϫ 2-(C᎐C)C H NNO -5] }, 576
᎐
5
3
2
ϩ
᎐
(43, [M Ϫ PPh ] ) and 429 {100%, [M Ϫ 2-(C᎐C)C H NNO -
᎐
3
5
3
2
5 Ϫ PPh3]ϩ}.
᎐
[Ru{2-(C᎐C)C H N}(PPh ) (ç-C H )] 2. A solution of com-
᎐
5
4
3
2
5
5
pound I in ether (2 cm3, 0.1 mol dmϪ3) was transferred to a
Schlenk tube and the ether removed in vacuo. Methanol (20 cm3)
was added immediately via a cannula. The complex [RuCl-
(PPh3)2(η-C5H5)] (200 mg, 0.28 mmol) was added to the
resultant solution and the mixture heated at reflux for 30 min.
The chloro complex gradually dissolved to give a bright orange
solution, which was then allowed to cool. Sodium (10 mg, 0.43
mmol) was added, whereupon a yellow precipitate formed,
which was filtered off and washed with methanol and light
petroleum. Yield 188 mg (83%) (Found: C, 71.65; H, 5.05;
N, 1.64. C48H39NP2Ruؒ0.5CH4O requires C, 72.02; H, 5.11; N,
electrolyte was [NBun ][PF6] (0.5 mol dmϪ3) in distilled deoxy-
4
genated dichloromethane. Solutions (1 × 10Ϫ3 mol dmϪ3) were
made under a purge of nitrogen and measured versus an Ag–
AgCl reference electrode at Ϫ50 ЊC, such that the ferrocene–
ferrocenium redox couple was located at ϩ0.55 V.
Proton, 13C and 31P NMR spectra were recorded using a
Varian Gemini-300 FT NMR spectrometer and are referenced
to residual CHCl3 (δ 7.24), CDCl3 (δ 77.0) and external 85%
H3PO4 (δ 0.0), respectively. All carbon and phosphorus NMR
spectra are broad-band proton decoupled, and coupling
constants are quoted as absolute values. The NMR spectral
assignments follow the numbering schemes shown in Schemes
1 and 2.
1.73%). λmax/nm (thf) 331 (ε/dm3 molϪ1 cmϪ1 14 500). ν max/cmϪ1
᎐
(CH Cl ) 2070 (C᎐C). δ (CDCl ) 3.48 (1.5 H, s, MeOH), 4.35
᎐
2
2
H
3
(5 H, s, C5H5), 6.65 [1 H, d, 3J(HH) 7, H4], 6.82 [1 H, t, 3J(HH)
7, H5], 7.07 [12 H, t, J(HH) 7, Hm], 7.18 (6 H, m, Hp), 7.43 (12
H, m, Ho), 8.41 [1 H, d, J(HH) 4 Hz, H7]; resonance for H6
3
8
᎐
Characterization of 2-(HC᎐C)C H N I
᎐
5
4
is obscured by those of Hm and Hp. δC(CDCl3) 85.6 (C5H5),
117.8 (C2), 125.6 (C4), 127.3 (Cm), 128.4 (Cp), 133.7 (Co), 138.7
(Ci); carbon atoms C1, C3, C5, C6 and C7 were not detected
due to poor signal/noise. δP(CDCl3) 51.4. m/z 793 (Mϩ, 35), 691
This compound was found to be highly unstable in air; it was
stored as a solution in diethyl ether. ν max/cmϪ1 (CH2Cl2) = 2108
6
᎐
᎐
(C᎐C). δ (CDCl ) 3.13 (1 H, s, ᎐CH), 7.25 (1 H, m, H ), 7.46
᎐
᎐
H
3
4168
J. Chem. Soc., Dalton Trans., 1997, Pages 4167–4174

View Article Online
ϩ
ϩ
᎐
{6, [M Ϫ 2-(C᎐C)C H N] }, 531 (47, [M Ϫ PPh ] ) and 429
(200 mg, 0.40 mmol) and sodium (ca. 40 mg, excess). The solu-
᎐
5
4
3
ϩ
᎐
{100%, [M Ϫ 2-(C᎐C)C H N Ϫ PPh ] }.
tion was allowed to stir for 16 h and the solvent was then
removed in vacuo. Addition of a small amount of methanol
afforded 130 mg (58%) of a pale microcrystalline solid (Found:
C, 53.46; H, 3.27; N, 2.38. C25H19AuNP requires C, 53.49; H,
3.41; N, 2.50%). λmax/nm (thf) 300 (ε/dm3 molϪ1 cmϪ1 20 800).
᎐
5
4
3
᎐
[Ni{2-(C᎐C)C H NNO -5}(PPh )(ç-C H )] 3. A mixture of
᎐
5
3
2
3
5
5
[NiCl(PPh3)(η-C5H5)] (120 mg, 0.29 mmol), compound II (66
mg, 0.30 mmol) and CuI (5 mg, 0.03 mmol) was stirred in tri-
ethylamine (20 cm3) containing NaOMe in methanol (3.0 cm3,
0.22 mol dmϪ3) for 16 h. The solvent was removed under
reduced pressure, the residue extracted with dichloromethane
(40 cm3), and the extract adsorbed onto alumina. Column
chromatography with dichloromethane as eluent afforded a
brown band; addition of light petroleum (10 cm3) and ethanol
(2 cm3) to the eluate and removal of the dichloromethane pre-
cipitated the microcrystalline product which was filtered off (90
mg, 58%) (Found: C, 67.56; H, 4.21; N, 5.00. C30H23N2NiO2P
requires C, 67.58; H, 4.35; N, 5.25%). λmax/nm (thf) 456 (ε/dm3
4
6
ν max/cmϪ1 2123 (C᎐C). δ (CDCl ) 7.07 (2 H, m, H , H ), 7.41–
᎐
᎐
H
3
7.55 (16 H, m, Ph, H5) and 8.50 (1 H, m, H7). δC(CDCl3) 103.1
[d, J(CP) 26, C1], 121.3 (C2), 126.9 (C4), 129.1 [d, J(CP) 11,
Cm], 130.0 (C5), 131.6 (Cp), 134.3 [d, J(CP) 14 Hz, Co], 135.4
(C3), 144.5 (C7), 149.6 (C6); Ci partially obscured by Cm.
δP(CDCl3) 42.7. m/z 1020 {37, [M ϩ Au(PPh3)]ϩ}, 721 {35,
[Au(PPh3)2]ϩ}, 562 (61, [M ϩ H]ϩ) and 459 (100, [M Ϫ
ϩ
᎐
C᎐CC H N] ).
᎐
5
3
᎐
[Au{2-(C᎐C)C H NNO -5}(PMe )] 7. Copper() iodide (5
᎐
5
3
2
3
molϪ1 cmϪ1 11 000). ν max/cmϪ1 2086 (C᎐C). δ (CDCl ) 5.24 (5
mg, 0.03 mmol) and a solution of NaOMe in methanol (8 cm3,
0.3 mol dmϪ3) were added to a mixture of [AuCl(PMe3)] (85
mg, 0.27 mmol) and compound II (60 mg, 0.27 mmol) in
dichloromethane (10 cm3), and the resultant mixture allowed to
stir for 16 h. The solvent was removed under reduced pressure
and the yellow product crystallized from methanol, yield 83 mg
(72%) (Found: C, 28.64; H, 2.75; N, 6.88. C10H12AuN2O2P
requires C, 28.59; H, 2.88; N, 6.67%). λmax/nm (thf) 340 (ε/dm3
᎐
᎐
H
3
H, s, C5H5), 6.02 [1 H, d, 3J(HH) 9, H4], 7.40 (9 H, m, Hm, Hp),
7.69 (6 H, m, Ho), 7.88 [1 H, dd, J(HH) 9, J(HH) 2, H5] and
3
4
9.08 [1 H, d, J(HH) 2 Hz, H7]. δC(CDCl3) 93.1 (C5H5), 113.6
4
[d, J(CP) 45, C1], 120.7 (C2), 125.6 (C4), 128.3 [d, J(CP) 10, Cm],
129.9 (C5), 130.5 (Cp), 133.5 [d, J(CP) 58, Ci], 133.9 [d, J(CP) 11
Hz, Co], 139.9 (C3), 145.1 (C7) and 150.4 (C6). δP(CDCl3) 43.2.
ϩ
ϩ
᎐
m/z 533 (M , 26), 385 {19, [M Ϫ 2-(C᎐C)C H NNO -5] } and
320 {16, [M Ϫ 2-(C᎐C)C H NNO -5 Ϫ C H ] }.
᎐
5
3
ϩ
2
molϪ1 cmϪ1 16 500). ν max/cmϪ1 2121 (C᎐C). δ (CDCl ) 1.54 (9
᎐
᎐
᎐
᎐
5
3
2
5
5
H
3
H, m, Me), 7.48 [1 H, d, 3J(HH) 9, H4], 8.31 [1 H, dd, 3J(HH) 9,
4J(HH) 3, H5] and 9.30 [1 H, d, 4J(HH) 3 Hz, H7]. δC(CDCl3) 15.6
[d, J(CP) 36 Hz, Me], 103.1 (br, C1), 126.9 (C4), 130.8 (C5),
141.6 (C3), 145.3 (C7), 149.9 (C6); C2 not observed. δP(CDCl3)
1.3. m/z 693 {15, [Au(PMe3) ϩ M]ϩ}, 421 (58, [M ϩ H]ϩ), 349
᎐
[Ni{2-(C᎐C)C H N}(PPh )(ç-C H )] 4. A solution of com-
᎐
5
4
3
5
5
pound I in ether (6.5 cm3, 0.1 mol dmϪ3) was collected and the
ether removed. Triethylamine (10 cm3) was added immediately
via a cannula, [NiCl(PPh3)(η-C5H5)] (200 mg, 0.47 mmol) and
CuI (5 mg, 0.03 mmol) were added and the resultant mixture
stirred for 16 h. The reaction mixture was then adsorbed onto
alumina and subjected to column chromatography with 50%
dichloromethane–50% light petroleum as eluent. The green
solution collected was reduced to dryness. Addition of ether
and light petroleum afforded a green microcrystalline solid,
yield 99 mg (42%). This complex decomposes slowly in solution
over a period of hours, preventing recrystallization; satisfactory
microanalyses could not be obtained. λmax/nm (thf) 413 (ε/dm3
ϩ
᎐
(100, [Au(PMe ) ] ) and 273 {87, [M Ϫ 2-(C᎐C)C H NNO -
᎐
3
2
5
3
2
5]ϩ}.
Crystallography
Crystals of complexes 3, 5 and 6 suitable for diffraction anal-
yses were grown by slow diffusion of hexane into dichloro-
methane at room temperature (5, 6) or 268 K (3). Unique
diffractometer data sets were obtained using an ω–2θ scan
mode at 296 K and yielded N independent reflections, No of
these with I у 3.00σ(I) being considered ‘observed’ and used in
full-matrix least-squares refinement. An empirical ψ-type
absorption correction was applied in each case; no decay cor-
rections were required. Anisotropic thermal parameters were
refined for the non-hydrogen atoms; (x, y, z, Uiso)H were
included constrained at estimated values. Conventional resid-
uals R and RЈ on |F | are given; the weighting function w = 1/
molϪ1 cmϪ1 1400). ν max/cmϪ1 2101 (C᎐C). δ (CDCl ) 5.22 (5 H,
᎐
᎐
H
3
s, C5H5), 6.03 [1 H, d, 3J(HH) 8 Hz, H4], 6.75 (m, H5), 7.12 (m,
H6), 7.38 (9 H, m, Hm, Hp), 7.72 (6 H, m, Ho) and 8.25 (m, H7).
δC(CDCl3) 92.8 (C5H5), 119.3 (C2), 125.9 (C4), 128.2 [d, J(CP)
10, Cm], 130.1 (C5), 130.2 (Cp), 133.9 [d, J(CP) 11, Co], 134.0 [d,
J(CP) 48 Hz, Ci], 134.9 (C3), 146.0 (C7) and 148.7 (C6).
ϩ
᎐
δ (CDCl ) 42.3. m/z 487 (M , 79), 385 {44, [M Ϫ 2-(C᎐C)-
C H N] } and 320 {28%, [M Ϫ 2-(C᎐C)C H N Ϫ C H ] }.
᎐
P
3
ϩ
ϩ
᎐
᎐
2
σ2(Fo) [where σ2(Fo) = σc2(Fo) ϩ (p2/4)Fo , σc(Fo) = estimated
5
4
5
4
5
5
standard deviation (e.s.d.) based on counting statistics and p = p
factor determined experimentally from standard reflections]
was employed. Computation used the TEXSAN package.14
Specific data collection, solution and refinement parameters are
given in Table 1. Pertinent results are given in the figures and
tables.
᎐
[Au{2-(C᎐C)C H NNO -5}(PPh )] 5. A mixture of [AuCl-
᎐
5
3
2
3
(PPh3)] (120 mg, 0.24 mmol) and compound II (70 mg, 0.32
mmol) in a solution of NaOMe in methanol (10 cm3, 0.10 mol
dmϪ3) was stirred for 16 h, after which time a pale yellow solid
had precipitated and was filtered off. Recrystallization was
effected by extraction into dichloromethane, addition of ether,
partial removal of solvent on a rotary evaporator and filtration
to yield 93 mg (64%) (Found: C, 48.97; H, 2.69; N, 4.67.
C25H18AuN2O2P requires C, 49.52; H, 2.99; N, 4.62%). λmax/nm
CCDC reference number 186/708.
HRS Measurements
(thf) 339 (ε/dm3 molϪ1 cmϪ1 25 600). ν max/cmϪ1 2123 (C᎐C).
An injection-seeded Nd–YAG laser (Q-switched Nd–YAG
Quanta Ray GCR5, 1064 nm, 8 ns pulse, 10 Hz) was focused
into a cylindrical cell (7 cm3) containing the sample. The inten-
sity of the incident beam was varied by rotation of a half-wave
plate placed between crossed polarizers. Part of the laser pulse
was sampled by a photodiode to measure the vertically polar-
ized incident light intensity. The frequency-doubled light was
collected by an efficient condenser system and detected by a
photomultiplier. The harmonic scattering and linear scattering
were distinguished by appropriate filters; gated integrators were
used to obtain intensities of the incident and harmonic scat-
tered light. All measurements were performed in thf using
᎐
᎐
δH(CDCl3) 7.45–7.52 (16 H, m, Ph, H4), 8.33 [1 H, dd, 3J(HH)
9, 4J(HH) 3, H5] and 9.32 [1 H, d, 4J(HH) 3 Hz, H7]. δC(CDCl3)
102.4 [d, J(CP) 26, C1], 126.7 (C4), 129.3 [d, J(CP) 60, Ci], 129.4
[d, J(CP) 11, Cm], 130.9 (C5), 131.7 (Cp), 134.3 [d, J(CP) 14 Hz,
Co], 141.6 (C3), 145.3 (C7), 149.8 (C6); C2 not observed.
δP(CDCl3) 42.3. m/z 721 {41, [Au(PPh3)2]ϩ}, 606 (21, [M ϩ H]ϩ)
ϩ
᎐
and 459 {100, [M Ϫ 2-(C᎐C)C H NNO -5] }.
᎐
5
3
2
᎐
[Au{2-(C᎐C)C H N}(PPh )] 6. A solution of compound I in
᎐
5
4
3
ether (4 cm3, 0.1 mol dmϪ3) was transferred to a Schlenk tube
and the ether removed. Methanol (20 cm3) was added immedi-
ately via a cannula. To this solution was added [AuCl(PPh3)]
p-nitroaniline (β = 21.4 × 10Ϫ30 esu; 1 esu = 2.7 × 1020 C m3 VϪ1
)
J. Chem. Soc., Dalton Trans., 1997, Pages 4167–4174
4169

View Article Online
Ni Cl
Ru Cl
Ph3P
Ph3P
Ph3P
O2N
O2N
N
C
CSiMe3
N
C
CH
N
C
CSiMe3
N
C
CH
NEt3, CuI, NaOMe–MeOH
stir 16 h
NEt3, CuI
stir 16 h
NaOMe–MeOH
reflux 1 h
(i) MeOH, reflux 1/2 h
(ii) Na
Ni
C
C
NO2
Ni
C
C
Ru
C
C
NO2
Ru
C
C
Ph3P
Ph3P
N
N
Ph3P
Ph3P
N
N
Ph3P
Ph3P
3 (58%)
4 (42%)
1 (44%)
2 (83%)
Scheme 4
Scheme 3
Ph3P Au Cl
as a reference.15 Further details of the experimental procedure
have been reported elsewhere.16
O2N
N
C
CSiMe3
N
C
CH
Kurtz powder measurements
NaOMe–MeOH
stir 16 h
NaOMe–MeOH
stir 16 h
Samples were unsized microcrystalline powders placed in the
circular cavity (10 mm diameter × 0.5 mm depth) of a micro-
scope slide with a cover slip. The fundamental output of a
Quanta-Ray GC-130 Nd–YAG laser was directed onto the
sample (spot size ≈ 5 mm, power ≈ 20 mJ). A collecting lens
(orthogonally placed with respect to the fundamental beam)
focused the backscattered second harmonic light through an
infrared absorbing filter and a 532 nm interference filter onto
a photodiode detector which was connected to a Hewlett-
Packard digital 54510A oscilloscope. Measurements thus made
were compared with those for a urea powder sample.
Ph3P Au
C
C
NO2
Ph3P Au
C
C
N
N
5 (64%)
6 (58%)
Scheme 5
O2N
Results and Discussion
Syntheses of acetylenes
Me3P Au Cl +
N
C
CSiMe3
2-Pyridylacetylene I was prepared by the literature procedure,8
namely [PdCl2(PPh3)2]/CuI-catalysed coupling of 2-bromo-
pyridine with trimethylsilylacetylene, followed by deprotection
of the intermediate pyridyl(trimethylsilyl)acetylene with base
[we also investigated reaction of 2-bromopyridine with (the less
CuI, NaOMe–MeOH
CH2Cl2
Me3P Au
C
C
NO2
N
7 (72%)
Scheme 6
expensive) 2-methylbut-3-yn-2-ol which proceeded to afford
᎐
Me C(OH)C᎐CC H N-2, but we were unable to deprotect this
᎐
2
5
4
decomposing over a period of hours; we consequently investi-
gated shorter reaction times for its synthesis, but these did not
improve its yield. The (triphenylphosphine)gold σ-pyridyl-
acetylide complexes 5 and 6 were prepared similarly to their
σ-phenylacetylide analogues (Scheme 5), but the yield of the
trimethylphosphine complex 7 was low by this methodology;
the yield of 7 was increased substantially by addition of
catalytic amounts of CuI to the reaction mixture (Scheme 6).
The new σ-pyridylacetylide complexes were characterized
successfully]. Analogous palladium–copper catalysed reaction
of 2-bromo-5-nitropyridine with trimethylsilylacetylene affor-
᎐
ded 2-(Me SiC᎐C)C H NNO -5 II in excellent yield. However,
᎐
3
5
3
2
attempted deprotection of II with base afforded a dark oil, the
analysis of which revealed that none of the desired 2-
᎐
(HC᎐C)C H NNO -5 was present. We then found that deprotec-
᎐
5
3
2
tion of II in the presence of the metal chloride afforded fair to
good yields of the desired acetylide complexes, so isolation of
᎐
2-(HC᎐C)C H NNO -5 was not pursued further.
᎐
1
by H, 13C, 31P NMR, UV/VIS and IR spectroscopies, mass
5
3
2
spectrometry, satisfactory microanalyses (with the exception of
4 which decomposes in solution over a period of hours prevent-
Syntheses of ó-acetylide complexes
The synthetic methodologies employed for the preparation of
the new σ-pyridylacetylide complexes are adaptations of those
successfully utilized for the preparation of the correspond-
ing σ-phenylacetylides.3a,b,g,i,j The (cyclopentadienyl)bis(tri-
phenylphosphine)ruthenium σ-pyridylacetylide complexes were
prepared in fair (1) or excellent (2) yield by reaction of the pyri-
dylacetylene with [RuCl(PPh3)2(η-C5H5)] and deprotonation
of the intermediate vinylidene complex (Scheme 3). The
(cyclopentadienyl)(triphenylphosphine)nickel σ-pyridylacetyl-
ide complexes 3 and 4 were prepared in good yields by reaction
of [NiCl(PPh3)(η-C5H5)] with the pyridylacetylenes (Scheme 4).
The pyridylacetylide complex 4 was unstable in solution,
ing recrystallization: see above) and, in the case of 3, 5 and 6,
᎐
single-crystal X-ray diffraction studies. Characteristic ν(C᎐C)
᎐
move to lower frequency for the ruthenium complexes on intro-
duction of the nitro group [2070 (2), cf. 2029 cmϪ1 (1)], but this
effect is attenuated in the nickel complexes [2101 (4), cf. 2086
cmϪ1 (3)], and disappears with the gold examples [2123 cmϪ1 (5,
1
6)]. The H, 13C and 31P NMR spectra are similar to those
1
of previously reported phenylacetylide analogues, with the H
NMR spectra of 1–7 containing additional resonances assigned
to pyridyl hydrogens between δ 5.2 and 9.1. The electron-
withdrawing nitro and pyridyl N deshield H7 significantly, lead-
ing to a resonance with a characteristic downfield shift and a
4170
J. Chem. Soc., Dalton Trans., 1997, Pages 4167–4174

View Article Online
Table 1 Crystallographic data for complexes 3, 5 and 6
3*
5
6
Empirical formula
M
C30H23N2NiO2P
533.20
C25H18AuN2O2P
606.37
C25H19AuNP
561.37
Crystal colour, habit
Crystal dimensions/mm
Crystal symmetry
Space group
a/Å
b/Å
c/Å
α/Њ
Orange, plate
0.28 × 0.10 × 0.03
Orthorhombic
Pna21 (no. 33)
16.770(1)
8.661(2)
17.324(1)
Yellow, needle
0.72 × 0.10 × 0.12
Triclinic
Orange, plate
0.28 × 0.18 × 0.08
Monoclinic
P21/n (no. 14)
12.392(2)
17.696(4)
19.491(3)
¯
P1 (no. 2)
8.885(3)
17.346(3)
22.809(4)
78.86(2)
78.87(2)
81.98(2)
3365(2)
6
β/Њ
93.34(2)
γ/Њ
U/Å3
2516.4(8)
4
1.407
19.38 (Cu-Kα)
1.541 78
120
4267(1)
8
1.748
69.56 (Mo-Kα)
0.710 69
50.1
Z
Dc/g cmϪ3
1.795
µ/cmϪ1
66.75 (Mo-Kα)
0.710 69
50.1
λ/Å
2θmax/Њ
Transmission factors
N
No [I > 3.00σ(I)]
0.93–1.00
2181
1626
0.68–1.00
11 941
5604
0.39–1.00
7824
4160
No. variables
p Factor
325
0.003
838
0.004
505
0.004
R
RЈ
0.028
0.022
0.038
0.030
0.033
0.024
* The absolute structure was checked; preferred residuals are quoted.
Table 2 Important bond lengths (Å) and angles (Њ) for complex 3
Ni᎐C(1)
1.832(5)
1.215(7)
1.413(7)
1.407(7)
1.355(7)
1.361(7)
1.377(7)
C(7)᎐N(8)
N(8)᎐C(3)
C(6)᎐N(1)
N(1)᎐O(1)
N(1)᎐O(2)
Ni᎐P
1.329(6)
1.345(7)
1.458(7)
1.235(6)
1.211(6)
2.139(2)
C(1)᎐C(2)
C(2)᎐C(3)
C(3)᎐C(4)
C(4)᎐C(5)
C(5)᎐C(6)
C(6)᎐C(7)
Ni᎐C(1)᎐C(2)
175.6(5)
C(1)᎐C(2)᎐C(3)
169.5(6)
Fig. 1 Molecular structure and atomic labelling scheme for [Ni{2-
᎐
(C᎐C)C5H3NNO2-5}(PPh3)(η-C5H5)] 3. 20% Thermal ellipsoids are
᎐
shown for the non-hydrogen atoms; hydrogen atoms omitted for clarity
Fig. 3 Molecular structure and atomic labelling scheme for [Au{2-
᎐
(C᎐C)C5H4N}(PPh3)] 6. Details as in Fig. 1
᎐
such studies of σ-pyridylacetylide complexes; crystallographic
data are collected in Table 1 and selected bond lengths and
angles in Tables 2 (3) and 3 (5, 6). The ORTEP17 plots are
displayed in Figs. 1 (3), 2 [5: one independent molecule only, the
other two being very similar in geometry (Table 3)], and 3 [6:
one independent molecule only, the other being very similar in
geometry (Table 3)].
Fig. 2 Molecular structure and atomic labelling scheme for [Au{2-
᎐
᎐
(C᎐C)C5H3NNO2-5}(PPh3)] 5. Details as in Fig. 1
small ‘W’ coupling to H5. The UV/VIS spectra for 1–6 contain
absorption maxima to lower energy of those of the phenyl-
acetylide analogues. These absorption maxima proceed to lower
energy on introduction of a nitro substituent [331 (2), cf. 468
(1); 413 (4), cf. 456 (3); 300 (6), cf. 339 nm (5)].
The lattice packing of these materials is relevant to analysis
of the Kurtz powder measurements below. Complex 3 crystal-
lized in the non-centrosymmetric space group Pna21. In con-
trast, 5 and 6 crystallized in centrosymmetric space groups,
ensuring that no bulk response would be observed for them. For
3 the Ni᎐P(1) bond length is normal, being very similar to those
Crystal structural studies
᎐
᎐
reported for [Ni(C᎐CPh)(PPh )(η-C H )] and [Ni(C᎐CC H -
᎐
᎐
3
5
5
6
4
We have confirmed the identities of complexes 3, 5 and 6 by
NO2-4)(PPh3)(η-C5H5)].3j The Ni᎐C(1)᎐C(2)᎐C(3) unit is
carrying out single-crystal X-ray diffraction studies, the first
almost linear, the slight deviation probably arising from pack-
J. Chem. Soc., Dalton Trans., 1997, Pages 4167–4174
4171

View Article Online
Table 3 Important bond lengths (Å) and angles (Њ) for complexes 5 and 6
5
6
Molecule 1
Molecule 2
Molecule 3
Molecule 1
Molecule 2
Au(1)᎐P(1)
Au(1)᎐C(11)
C(11)᎐C(12)
C(12)᎐C(13)
C(13)᎐C(14)
C(14)᎐C(15)
C(15)᎐C(16)
C(16)᎐C(17)
C(17)᎐N(18)
N(18)᎐C(13)
C(16)᎐N(1)
N(1)᎐O(11)
N(1)᎐O(12)
2.281(4)
2.01(1)
1.08(2)
1.50(2)
1.41(2)
1.38(2)
1.32(2)
1.37(2)
1.33(2)
1.33(2)
1.46(2)
1.22(2)
1.17(2)
2.276(3)
1.97(1)
1.21(2)
1.43(2)
1.39(2)
1.38(2)
1.38(2)
1.38(2)
1.35(1)
1.36(2)
1.49(2)
1.23(1)
1.20(1)
2.273(3)
1.99(1)
1.18(1)
1.48(2)
1.36(2)
1.38(2)
1.34(2)
1.37(2)
1.34(1)
1.35(2)
1.46(2)
1.17(2)
1.26(2)
2.274(3)
1.97(1)
1.21(1)
1.45(1)
1.36(1)
1.38(2)
1.38(2)
1.36(2)
1.31(1)
1.35(1)
—
2.263(3)
1.95(1)
1.21(1)
1.47(1)
1.38(1)
1.37(1)
1.35(1)
1.38(1)
1.31(1)
1.33(1)
—
—
—
—
—
P(1)᎐Au(1)᎐C(11)
Au(1)᎐C(11)᎐C(12)
C(11)᎐C(12)᎐C(13)
175.0(4)
172(2)
177(2)
175.8(4)
174(1)
176(2)
178.0(4)
175(1)
175(2)
173.9(3)
174.9(9)
177(1)
177.3(3)
177(1)
177(1)
Table 4 Cyclic voltammetric data for complexes 1–4
Complex
EЊM a/V
ipc/ipa
EЊNO b/V
ipa/ipc
2
᎐
1 [Ru{2-(C᎐C)C5H3NNO2-5}(PPh3)2(η-C5H5)]
0.83
0.73
0.69
0.55
0.92
0.90
0.80
0.81
1.0
1.0
0.6
0.7
0.3
0.4
0.4
0.4
Ϫ0.89
Ϫ1.08
—
0.9
1.0
—
᎐
c
᎐
[Ru(C᎐CC6H4NO2-4)(PPh3)2(η-C5H5)]
᎐
᎐
2 [Ru{2-(C᎐C)C5H4N}(PPh3)2(η-C5H5)]
᎐
c
᎐
[Ru(C᎐CPh)(PPh3)2(η-C5H5)]
—
—
᎐
᎐
3 [Ni{2-(C᎐C)C5H3NNO2-5}(PPh3)(η-C5H5)]
Ϫ0.84
Ϫ0.97
—
0.9
0.7
—
᎐
d
᎐
[Ni(C᎐CC6H4NO2-4)(PPh3)(η-C5H5)]
᎐
᎐
4 [Ni{2-(C᎐C)C5H4N}(PPh3)(η-C5H5)]
᎐
d
᎐
[Ni(C᎐CPh)(PPh3)(η-C5H5)]
—
—
᎐
Experiments performed at Ϫ50 ЊC vs. Ag–AgCl in dichloromethane with scanning rate 100 mV sϪ1. Typical peak-to-peak separations: RuII/III
,
60–70; NiII/III, 80–90 mV. a Switching potential 1.2 V. b Switching potential Ϫ1.8 V. c Ref 3(a). d Ref. 3(j).
᎐
ing forces. The bond distances in the Ni᎐C(1)᎐C(2)᎐C(3) unit
no difference with the nickel compounds. Complexes 1 and 3
᎐
are possibly indicative of slightly more Niϩ᎐C(1)᎐C(2)᎐C(3)Ϫ
undergo one-electron reduction at Ϫ0.89 and Ϫ0.84 V, assigned
to reduction of the nitro substituent. As the oxidation waves are
associated with metal-centred highest occupied molecular
orbitals (HOMOs) and the reduction waves with nitro-centred
lowest unoccupied molecular orbitals (LUMOs), ease of elec-
tron removal from the donor and addition to the acceptor
group may be relevant to the non-linear optical response. We
᎐
᎐
᎐
contribution in the ground-state geometry of 3 [Ni᎐C(1)
1.832(5), C(1)᎐C(2) 1.215(7), C(2)᎐C(3) 1.413(7) Å] compared
to the analogous phenylacetylide [1.850(3), 1.856(3); 1.193(4),
1.191(4); 1.445(4), 1.437(4) Å for two independent molecules]
and 4-nitrophenylacetylide complexes [1.842(6), 1.207(7),
1.422(7) Å], but differences in distances are within 3σ. Gold–
phosphine distances in 5 and 6 are close to 2.27 Å, the same
as observed for related (triphenylphosphine)gold phenylacetyl-
ides.3g The variation of the Au(1)᎐C(11)᎐C(12)᎐C(13) bond
length data for 5 and 6 allows a direct comparison in the
ground-state structures when 5-H is replaced by the strong
acceptor 5-NO2; differences are small, and not consistent with a
vinylidene contribution to the 5-NO2 compound 5.
II/III
Ϫ
have examined EЊM Ϫ EЊNO /NO for these complexes. For
2
2
ruthenium, this parameter is 1.72 V for the nitropyridyl-
acetylide complexes and 1.81 V for the nitrophenylacetylide
compounds while, for nickel, the values are 1.76 and 1.87 V,
respectively. Our comparison of these data is necessarily
cautious, given the fact that the RuII/III and NiII/III couples are
not equally reversible. Nevertheless, if these electrochemical
data have predictive merit one should expect to see larger
non-linearities for ruthenium than for nickel.
Electrochemical studies
As the donor strength of the ligated metal centre is a factor
influencing the quadratic non-linearity of these acetylide com-
plexes, ease of oxidation may be related to large hyperpolariz-
ability. The results of cyclic voltammetric investigations of the
ruthenium and nickel complexes 1–4 are summarized in Table
4, together with data from the analogous phenylacetylide com-
plexes. All complexes undergo one-electron oxidation assigned
to the RuII/III (1, 2) or NiII/III (3, 4) couple. Replacing pyridyl-
acetylide by the more electron-withdrawing 4-nitropyridyl-
acetylide results in the expected increase in oxidation potential
(0.14 V for the ruthenium complexes, 0.12 V for the nickel
complexes). Replacing the nitrophenylacetylide ligand by the
nitropyridylacetylide ligand leads to an increase of 0.10 V in
oxidation potential for the ruthenium complexes, but no signifi-
cant difference for the nickel complexes. Similarly, replacing
phenylacetylide by pyridylacetylide leads to an increase of
0.14 V in oxidation potential for the ruthenium complexes, but
Quadratic hyperpolarizabilities
We have previously determined the molecular quadratic optical
non-linearities of ruthenium, nickel and gold σ-phenylacetylide
complexes, and have now extended these studies to the σ-
pyridylacetylide complexes 1–7. Experimentally obtained data
at 1064 nm by hyper-Rayleigh scattering and two-level-
corrected values are collected in Table 5, together with related
data for analogous σ-phenylacetylide complexes. [Note: the
two-state model has been utilized for calculating frequency-
independent values, but it may not be adequate for donor–
acceptor organometallic systems; it was developed for a
restricted class of organic compounds where structural modifi-
cations are directed at the charge-transfer band thought to con-
tribute to the hyperpolarizability, and may not be useful where
there are several dominant optical transitions close to 2ω.] For
1–7 the relative ordering for observed and two-level-corrected β
4172
J. Chem. Soc., Dalton Trans., 1997, Pages 4167–4174

View Article Online
Table 5 Experimental linear optical spectroscopic and quadratic non-linear optical response parameters
Complex
λ/nm (ε/104 dm3 molϪ1 cmϪ1
)
β1064 a/10Ϫ30 esu
24c
221
25
186
16
468
18
622
6
22
7
38
βcorr b/10Ϫ30 esu
c
᎐
[Ni(C᎐CPh)(PPh3)(η-C5H5)]
307 (2.5) (br)
439 (0.9), 368 (1.3)
15
59
8
41
10
96
10
113
4
12
4
20
6
᎐
c
᎐
[Ni(C᎐CC6H4NO2-4)(PPh3)(η-C5H5)]
᎐
᎐
4 [Ni{2-(C᎐C)C5H4N}(PPh3)(η-C5H5)]
415 (0.1), 327 (1.4) (sh), 305 (2.0)
456 (1.1), 363 (1.1), 308 (1.4)
310 (2.0)
460 (1.1), 382 (1.1)
331 (1.4), 308 (1.3)
468 (1.7), 286 (1.2) (sh)
296 (1.3), 282 (3.0), 268 (2.7)
338 (2.5)
᎐
᎐
3 [Ni{2-(C᎐C)C5H3NNO2-5}(PPh3)(η-C5H5)]
᎐
d
᎐
[Ru(C᎐CPh)(PPh3)2(η-C5H5)]
᎐
d
᎐
[Ru(C᎐CC6H4NO2-4)(PPh3)2(η-C5H5)]
᎐
᎐
2 [Ru{2-(C᎐C)C5H4N}(PPh3)2(η-C5H5)]
᎐
᎐
1 [Ru{2-(C᎐C)C5H3NNO2-5}(PPh3)2(η-C5H5)]
᎐
e
᎐
[Au(C᎐CPh)(PPh3)]
᎐
e
᎐
[Au(C᎐CC6H4NO2-4)(PPh3)]
᎐
᎐
6 [Au{2-(C᎐C)C5H4N}(PPh3)]
300 (2.1), 292 (2.2), 286 (1.9) (sh)
339 (2.6)
340 (1.6)
᎐
᎐
5 [Au{2-(C᎐C)C5H3NNO2-5}(PPh3)]
᎐
᎐
7 [Au{2-(C᎐C)C5H3NNO2-5}(PMe3)]
12
᎐
All measurements in thf solvent. All complexes are optically transparent at 1064 nm. a HRS at 1064 nm; values 10%. b HRS at 1064 nm corrected
for resonance enhancement at 532 nm using the two-level model with βcorr = βexp[1 Ϫ (2λmax/1064)2][1 Ϫ (λmax/1064)2]; damping factors not included.
c Ref. 3(j). d Ref. 3(a). e Ref. 3(g).
are almost the same. The only deviation is small: β1064 for 2 is
less than that for 4, while this ordering is reversed for the cor-
rected data. This is possibly due to significantly larger reson-
ance enhancement for 4 (λmax = 415 nm) than for 2 (λmax = 331
nm), accounted for in the two-level correction. The similarity in
the trends of the corrected and experimental data suggests that
the effects of structural modification on observed non-linearity
reflect their effect on intrinsic non-linearity.
Earlier studies with related ruthenium complexes suggested
᎐
that replacing L = PMe by PPh in [Ru(C᎐CC H NO -4)L -
᎐
3
3
6
4
2
2
3b
(η-C5H5)] leads to a doubling in βHRS
. This datum was sub-
stantially resonance enhanced, although the relative ordering
was maintained with two-level-corrected values. The gold com-
plexes have λmax much further removed from 2ω (ω = irradiating
frequency) and data for 5 and 7 provide an opportunity to
assess with more confidence the effect of phosphine replace-
ment. Replacement of PMe3 with PPh3 in proceeding from 7 to
5 results in a tripling of the β value for both experimental and
two-level-corrected data. Quadratic non-linearity depends on
the donor strength and the length of the π-electron system. The
electron-donating strength of PMe3 is greater than that of
PPh3. Conversely, the phenyl groups of the PPh3 ligand provide
increased π-electron delocalization possibilities compared to
that of PMe3. Given the differing merits of each phosphine, it
seems that π system length is more important than variation in
the donor strength of the coligand for second-order NLO merit
in this system.
Our earlier experimental and two-level-corrected data for
donor–bridge–acceptor ruthenium, nickel and gold σ-phenyl-
acetylide complexes suggested that the 18-electron, more easily
oxidizable ruthenium is a better donor than the 18-electron, less
easily oxidizable nickel, which is in turn a better donor than the
14-electron, less oxidizable gold.3g,i,j The present studies extend
this comparison to the heterocyclic acetylide domain. Com-
parison of β (both experimentally observed and two-level
corrected) across 1–7 for an invariant acetylide ligand reveals
that the first hyperpolarizability decreases for the ligated metals
according to Ru > Ni > Au, consistent with the phenyl-
acetylide results.
᎐
Fig. 4 Cell packing diagram for [Ni{2-(C᎐C)C5H3NNO2-5}(PPh3)-
᎐
(η-C5H5)] 5 (viewed down the b axis and across the ca plane)
by nitropyridylacetylide results in an increased non-linearity for
the ruthenium and gold systems, and possibly to a slight
decrease with the nickel complexes (it should be emphasized,
though, that values for the nickel complexes 3 and 4 do not
differ significantly). Molecular quadratic non-linearities for
the ruthenium and nickel nitropyridylacetylides 1 and 3 are
substantially resonance-enhanced. In contrast, the λmax (far
removed from 2ω) and oscillator strengths of the gold com-
plexes 5 and 6 are very similar, which permits comparison of
and comment on the experimentally obtained non-linearities;
values are consistent with an increase in βHRS upon replacing
phenyl by a heterocyclic bridging unit in acetylide complexes.
Further studies with, for example, thienylacetylides (thiophene
aromatic stabilization energy: 122 kJ molϪ1 18) would be helpful
in this regard.
The rationale behind the replacement of the phenyl ring
by the pyridyl ring in these acetylide complexes is that hetero-
cycles have lower aromatic stabilization energies (benzene, 150;
pyridine, 117 kJ molϪ1 18); it should thus be easier energetic-
ally to access the important charge-separated excited state
which contributes to the non-linearity. Comparison of the
experimentally obtained and two-level-corrected values of the
phenylacetylide complexes with those of the pyridylacetylide
complexes 2, 4 and 6 reveals no dramatic difference between
these systems. In contrast, replacement of nitrophenylacetylide
Powder second harmonic generation (SHG) measurements
Kurtz powder measurements were performed on all σ-
pyridylacetylides. A significant response was observed for
᎐
[Ni{2-(C᎐C)C H NNO -5}(PPh )(η-C H )] 3 only, with SHG
᎐
5
3
2
3
5
5
eight times that of urea; complex 3 is the only one of
those synthesized known to have a non-centrosymmetric lattice
J. Chem. Soc., Dalton Trans., 1997, Pages 4167–4174
4173

View Article Online
array. The cell-packing diagram for it is given in Fig. 4. The
acetylide chromophores are at an angle of approximately 45Њ to
the adjacent molecule. As the non-linearity observed is the
vectorial sum of contributions from each individual molecule,
and this is not the alignment which maximizes the bulk
response, attention needs to be paid to crystal-engineering
strategies to optimize bulk non-linearity. Given favourable
crystal engineering, it is likely that other examples from this
series of N-heterocyclic acetylide complexes may afford
materials with significant bulk non-linearities.
References
1 Part 12, I. R. Whittall, M. G. Humphrey, M. Samoc, B. Luther-
Davies and D. C. R. Hockless, J. Organomet. Chem., in the press.
2 H. S. Nalwa, Appl. Organomet. Chem., 1991, 5, 349; S. R. Marder, in
Inorganic Materials, eds. D. W. Bruce and D. O’Hare, Wiley,
Chichester, 1992, p. 115; N. J. Long, Angew. Chem., Int. Ed. Engl.,
1995, 34, 21; I. R. Whittall, A. M. McDonagh, M. G. Humphrey
and M. Samoc, Adv. Organomet. Chem., in the press.
3 (a) I. R. Whittall, M. G. Humphrey, D. C. R. Hockless, B. W. Skelton
and A. H. White, Organometallics, 1995, 14, 3970; (b) I. R. Whittall,
M. G. Humphrey, A. Persoons and S. Houbrechts, Organometallics,
1996, 15, 1935; (c) I. R. Whittall, M. G. Humphrey, M. Samoc,
J. Swiatkiewicz and B. Luther-Davies, Organometallics, 1995,
14, 5493; (d) A. M. McDonagh, I. R. Whittall, M. G. Humphrey,
B. W. Skelton and A. H. White, J. Organomet. Chem., 1996, 519, 229;
(e) A. M. McDonagh, I. R. Whittall, M. G. Humphrey, D. C. R.
Hockless, B. W. Skeleton and A. H. White, J. Organomet. Chem.,
1996, 523, 33; ( f ) A. M. McDonagh, M. P. Cifuentes, I. R. Whittall,
M. G. Humphrey, M. Samoc, B. Luther-Davies and D. C. R.
Hockless, J. Organomet. Chem., 1996, 526, 99; (g) I. R. Whittall,
M. G. Humphrey, S. Houbrechts, A. Persoons and D. C. R.
Hockless, Organometallics, 1996, 15, 5738; (h) I. R. Whittall,
M. G. Humphrey, M. Samoc and B. Luther-Davies, Angew. Chem.,
1997, 109, 386; Angew. Chem., Int. Ed. Engl., 1997, 36, 370; (i) I. R.
Whittall, M. P. Cifuentes, M. G. Humphrey, B. Luther-Davies,
M. Samoc, S. Houbrechts, A. Persoons, G. A. Heath and D. C. R.
Hockless, J. Organomet. Chem., in the press; (j) I. R. Whittall,
M. P. Cifuentes, M. G. Humphrey, B. Luther-Davies, M. Samoc,
S. Houbrechts, A. Persoons, G. A. Heath, D. Bogsanyi and D. C. R.
Hockless, Organometallics, 1997, 16, 2631; (k) S. Houbrechts,
K. Clays, A. Persoons, V. Cadierno, M. P. Gamasa, J. Gimeno,
I. R. Whittall and M. G. Humphrey, Proc. SPIE-Int. Soc. Opt. Eng.,
1996, 2852, 98; (l) M. G. Humphrey, Chem. Aust., 1996, 63, 442; (m)
S. Houbrechts, K. Clays, A. Persoons, V. Cadierno, M. P. Gamasa
and J. Gimeno, Organometallics, 1996, 15, 5266.
4 C. W. Dirk, H. E. Katz, M. L. Schilling and L. A. King, Chem.
Mater., 1990, 2, 700.
5 L.-T. Cheng, W. Tam, S. R. Marder, A. E. Steigman, G. Rikken and
C. W. Spangler, J. Phys. Chem., 1991, 95, 10 643.
6 A. K.-Y. Jen, V. P. Rao, K. Y. Wong and K. J. Drost, J. Chem. Soc.,
Chem. Commun., 1993, 90.
7 D. F. Shriver and M. A. Drezdzon, The Manipulation of Air
Sensitive Compounds, Wiley, New York, 1986.
8 T. Sakamoto, M. Shiraiwa, Y. Kondo and H. Yamanka, Synthesis,
1983, 312.
9 K. W. Barnett, J. Chem. Educ., 1974, 51, 422.
10 M. I. Bruce, C. Hameister, A. G. Swincer and R. C. Wallis, Inorg.
Synth., 1982, 21, 78.
11 R. F. Heck, Palladium Reagents in Organic Syntheses, Academic
Press, London, 1985, p. 18.
Conclusion
The first structural characterizations of N-heterocyclic metal
acetylide complexes are not consistent with an appreciable
vinylidene contribution to the ground-state geometry. One
᎐
example, [Ni{2-(C᎐C)C H NNO -5}(PPh )(η-C H )] 3, packs
᎐
5
3
2
3
5
5
non-centrosymmetrically in the crystal lattice, satisfying an
important requirement for observable bulk second-order non-
linearity. As the alignment of the molecules in the unit cell for
this complex is not the most favourable, larger χ(2) should be
possible with appropriate crystal engineering to enforce opti-
mum molecular alignment in the lattice. Cyclic voltammetry
reveals that the NiII/III couples for the pyridylacetylide com-
plexes are not significantly different from those of the phenyl-
acetylide complexes, but that the ruthenium pyridylacetylide
complexes are >0.1 V harder to oxidize than their phen-
II/III
Ϫ
ylacetylide analogues. The parameter EЊM Ϫ EЊNO /NO
correctly predicts the relative quadratic NLO merits of2 th2e
ruthenium and nickel complexes, despite the less-than-ideal
reversible response for the NiII/III couple. The HRS results
confirm our earlier ordering of ligated metal centres as
donor groups, namely Ru(PPh3)2(η-C5H5) > Ni(PPh3)(η-C5-
H5) > Au(PPh3), and are consistent with an increase in non-
linearity upon replacement of trimethylphosphine with tri-
phenylphosphine. The present study permits
a cautious
assessment of the effect upon the molecular quadratic optical
non-linearities of metal acetylides of the replacement of a
phenylacetylide ligand by a pyridylacetylide group. The optical
non-linearity of the nickel nitropyridylacetylide complex is
comparable to that of the nickel nitrophenylacetylide com-
pound. However, an increase in non-linearity is observed for
the ruthenium and gold complexes when nitropyridylacetylide
replaces nitrophenylacetylide. Further studies of the optical
non-linearities of organometallic complexes are currently
underway.
12 M. I. Bruce, B. K. Nicholson and O. B. Shawkataly, Inorg. Synth.,
1989, 26, 324.
13 M. I. Bruce, E. Horn, J. G. Matisons and M. R. Snow, Aust.
J. Chem., 1984, 37, 1163.
14 Single Crystal Structue Analysis Software, Version 1.6c, Molecular
Structure Corporation, The Woodlands, TX, 1993.
Acknowledgements
15 M. Stähelin, D. M. Burland and J. E. Rice, Chem. Phys. Lett., 1992,
191, 245.
16 K. Clays and A. Persoons, Rev. Sci. Instrum., 1992, 63, 3285;
E. Hendrickx, C. Dehu, K. Clays, J. L. Brédas and A. Persoons,
ACS Symp. Ser., 1995, 601, 82; S. Houbrechts, K. Clays,
A. Persoons, Z. Pikramenou and J.-M. Lehn, Chem. Phys. Lett.,
1996, 258, 485.
17 C. K. Johnson, ORTEP, Report ORNL-5138, Oak Ridge National
Laboratory, Oak Ridge, TN, 1976.
18 T. L. Gilchrist, Heterocyclic Chemistry, Pitman, London, 1985.
We thank the Australian Research Council (M. G. H.), the
Belgian Government (Grant No. IUAP-P4-11) (A. P.), the Fund
for Scientific Research-Flanders (Grant No. G.0308.96) (A. P.),
the Institute of Advanced Studies (G. A. H.), the University
of Leuven (Grant No. GOA 95/1) (A. P.), and the Flemish
Institute of Advancement of the Scientific and Technological
Research in Industry (C. B.) for support of this work, and
Johnson-Matthey Technology Centre (M. G. H.) for the loan
of ruthenium salts. M. P. C. holds an ARC Australian Post-
doctoral Research Fellowship, M. G. H. an ARC Australian
Research Fellowship, and S. H. is a Research Assistant of the
Fund for Scientific Research-Flanders .
Received 2nd April 1997; Paper 7/02249B
4174
J. Chem. Soc., Dalton Trans., 1997, Pages 4167–4174