Evolution of linear absorption and nonlinear optical properties in v-shaped ruthenium(II)-based chromophores

Published on Web 01/15/2010

Evolution of Linear Absorption and Nonlinear Optical

Properties in V-Shaped Ruthenium(II)-Based Chromophores

Benjamin J. Coe,*^,†Simon P. Foxon,^†Elizabeth C. Harper,^†Madeleine Helliwell,^†

James Raftery,^†Catherine A. Swanson,^†Bruce S. Brunschwig,^‡Koen Clays,^§

Edith Franz,^§Javier Gar´ın,^|Jesu´s Orduna,^|Peter N. Horton,^⊥and

Michael B. Hursthouse^⊥

School of Chemistry, UniVersity of Manchester, Oxford Road, Manchester M13 9PL, U.K.,

Molecular Materials Research Center, Beckman Institute, MC 139-74, California Institute of

Technology, 1200 East California BouleVard, Pasadena, California 91125, Department of

Chemistry, UniVersity of LeuVen, Celestijnenlaan 200D, B-3001 LeuVen, Belgium, Departamento

de Qu´ımica Orga´nica, ICMA, UniVersidad de Zaragoza-CSIC, E-50009 Zaragoza, Spain, and

EPSRC National Crystallography SerVice, School of Chemistry, UniVersity of Southampton,

Highﬁeld, Southampton SO17 1BJ, U.K.

Received October 20, 2009; E-mail: b.coe@manchester.ac.uk.

Abstract: In this article, we describe a series of complexes with electron-rich cis-{Ru^II(NH₃)₄}²⁺centers

coordinated to two pyridyl ligands bearing N-methyl/arylpyridinium electron-acceptor groups. These V-shaped

dipolar species are new, extended members of a class of chromophores ﬁrst reported by us (Coe, B. J. et

al. J. Am. Chem. Soc. 2005, 127, 4845-4859). They have been isolated as their PF₆^-salts and characterized

by using various techniques including ¹H NMR and electronic absorption spectroscopies and cyclic

voltammetry. Reversible Ru^III/IIwaves show that the new complexes are potentially redox-switchable

chromophores. Single crystal X-ray structures have been obtained for four complex salts; three of these

crystallize noncentrosymmetrically, but with the individual molecular dipoles aligned largely antiparallel.

Very large molecular ﬁrst hyperpolarizabilities ꢀ have been determined by using hyper-Rayleigh scattering

(HRS) with an 800 nm laser and also via Stark (electroabsorption) spectroscopic studies on the intense,

visible d f π* metal-to-ligand charge-transfer (MLCT) and π f π* intraligand charge-transfer (ILCT) bands.

The latter measurements afford total nonresonant ꢀ₀responses as high as ca. 600 × 10^-30esu. These

pseudo-C_2vchromophores show two substantial components of the ꢀ tensor, ꢀ_zzzand ꢀ_zyy, although the

relative signiﬁcance of these varies with the physical method applied. According to HRS, ꢀ_zzzdominates in

all cases, whereas the Stark analyses indicate that ꢀ_zyyis dominant in the shorter chromophores, but ꢀ_zzz

and ꢀ_zyyare similar for the extended species. In contrast, ﬁnite ﬁeld calculations predict that ꢀ_zyyis always

the major component. Time-dependent density functional theory calculations predict increasing ILCT

character for the nominally MLCT transitions and accompanying blue-shifts of the visible absorptions, as

the ligand π-systems are extended. Such unusual behavior has also been observed with related 1D

complexes (Coe, B. J. et al. J. Am. Chem. Soc. 2004, 126, 3880-3891).

Introduction

commercialization of crystals of the organic salt E-4′-(dimethyl-

amino)-N-methyl-4-stilbazolium tosylate (DAST) that is used

Continued interest in molecular nonlinear optical (NLO)

materials derives from potential for various applications, includ-

ing optical data processing and biological imaging.¹The recent

for terahertz (THz) wave generation via nonlinear frequency

mixing²has provided renewed impetus to work within this broad

research ﬁeld. From a more fundamental perspective, many

studies have concerned organotransition metal complexes, which

offer great scope for imaginative molecular engineering with

the aim of realizing new, multifunctional NLO materials.³

Various demonstrations of the use of metal-based redox

chemistry to reversibly switch different types of NLO phenom-

ena illustrate the promise of metal complexes in this ﬁeld.⁴

Molecular quadratic (second-order) NLO effects derive from

the ﬁrst hyperpolarizability ꢀ, while cubic (third-order) phe-

^†University of Manchester.

^‡California Institute of Technology.

^§University of Leuven.

^|Universidad de Zaragoza.

^⊥University of Southampton.

(1) (a) Molecular Nonlinear Optics: Materials, Physics and DeVices; Zyss,

J.; Academic Press: Boston, 1994. (b) Organic Nonlinear Optical

Materials. Bosshard, Ch., Sutter, K., Preˆtre, Ph., Hulliger, J., Flo¨r-

sheimer, M., Kaatz, P., Gu¨nter, P. AdVances in Nonlinear Optics;

Gordon & Breach: Amsterdam, The Netherlands, 1995; Vol. 1. (c)

Nonlinear Optics of Organic Molecules and Polymers; Nalwa, H. S.,

Miyata, S., Eds.; CRC Press: Boca Raton, FL, 1997. (d) Marder, S. R.

Chem. Commun. 2006, 131–134. (e) Nonlinear Optical Properties of

Matter: From Molecules to Condensed Phases; Papadopoulos, M. G.,

Leszczynski, J., Sadlej, A. J., Eds.; Springer: Dordrecht, 2006.

(2) See for examples: (a) Taniuchi, T.; Ikeda, S.; Okada, S.; Nakanishi,

H. Jpn. J. Appl. Phys. 2005, 44, L652–L654. (b) Schneider, A.; Neis,

M.; Stillhart, M.; Ruiz, B.; Khan, R. U. A.; Gu¨nter, P. J. Opt. Soc.

Am. B 2006, 23, 1822–1835.

9

1706 J. AM. CHEM. SOC. 2010, 132, 1706–1723

10.1021/ja908667p  2010 American Chemical Society

V-Shaped Ruthenium(II)-Based Chromophores

A R T I C L E S

nomena depend on the second hyperpolarizability γ. These

tensor quantities relate to the essentially instantaneous response

of the molecular electronic charges to the applied oscillating

electric ﬁeld of a laser light beam, and the speed of response is

one signiﬁcant advantage of organic materials over the currently

used inorganic crystals such as lithium niobate (LiNbO₃) and

potassium titanyl phosphate (KTiOPO₄).¹The present study

focuses on quadratic properties, although the compounds

investigated are also likely to display substantial cubic effects.

Most known NLO chromophores, whether purely organic or

metal-containing, contain relatively large, polarizable π-conju-

gated frameworks with attached electron-donating (D) and

-accepting (A) substituents. For a simple dipolar molecule, the

quadratic optical nonlinearity is largely one-dimensional (1D)

in nature, i.e., dominated by a single ꢀ tensor component, but

one or more other components can become signiﬁcant when

multiple D and/or A groups are present. A wide range of

multidimensional NLO chromophores has been investigated over

recent years;⁵for quadratic applications, these offer several

potential advantages over more traditional 1D species such as

increased ꢀ responses without undesirable losses of transparency

in the visible region. Also, the lack of a permanent dipole in

octupolar systems may increase the likelihood of achieving

noncentrosymmetric bulk structures that are essential for nonzero

bulk quadratic NLO susceptibilities ꢁ⁽²⁾. Studies of multidimen-

sional NLO metallochromophores have been dominated by

octupolar D₃tris-chelates, typiﬁed by [Ru^II(2,2′-bpy)₃]²⁺(bpy

) bipyridyl) and its derivatives,^3n,6and neutral dipolar 2D

complexes that mostly contain Schiff base ligands.^3j,7

via ꢀ_zyycan be prevented because its polarization is perpendicular

to µ₁₂, and phase-matching between the fundamental and

harmonic waves may also be facilitated.^5b,gVarious neutral

dipolar 2D metal-based NLO chromophores have been studied,^3j,7

and we recently reported the ﬁrst such investigation with related

charged complexes; these have a cis-{Ru^II(NH₃)₄}²⁺D center

(4) (a) Coe, B. J.; Houbrechts, S.; Asselberghs, I.; Persoons, A. Angew. Chem.,

Int. Ed. 1999, 38, 366–369. (b) Weyland, T.; Ledoux, I.; Brasselet, S.;

Zyss, J.; Lapinte, C. Organometallics 2000, 19, 5235–5237. (c) Malaun,

M.; Reeves, Z. R.; Paul, R. L.; Jeffery, J. C.; McCleverty, J. A.; Ward,

M. D.; Asselberghs, I.; Clays, K.; Persoons, A. Chem. Commun. 2001,

49–50. (d) Malaun, M.; Kowallick, R.; McDonagh, A. M.; Marcaccio,

M.; Paul, R. L.; Asselberghs, I.; Clays, K.; Persoons, A.; Bildstein, B.;

Fiorini, C.; Nunzi, J.-M.; Ward, M. D.; McCleverty, J. A. J. Chem. Soc.,

Dalton Trans. 2001, 3025–3038. (e) Cifuentes, M. P.; Powell, C. E.;

Humphrey, M. G.; Heath, G. A.; Samoc, M.; Luther-Davies, B. J. Phys.

Chem. A 2001, 105, 9625–9627. (f) Paul, F.; Costuas, K.; Ledoux, I.;

Deveau, S.; Zyss, J.; Halet, J.-F.; Lapinte, C. Organometallics 2002, 21,

5229–5235. (g) Powell, C. E.; Cifuentes, M. P.; Morrall, J. P.; Stranger,

R.; Humphrey, M. G.; Samoc, M.; Luther-Davies, B.; Heath, G. A. J. Am.

Chem. Soc. 2003, 125, 602–610. (h) Asselberghs, I.; Clays, K.; Persoons,

A.; McDonagh, A. M.; Ward, M. D.; McCleverty, J. A. Chem. Phys.

Lett. 2003, 368, 408–411. (i) Powell, C. E.; Humphrey, M. G.; Cifuentes,

M. P.; Morrall, J. P.; Samoc, M.; Luther-Davies, B. J. Phys. Chem. A

2003, 107, 11264–11266. (j) Sporer, C.; Ratera, I.; Ruiz-Molina, D.;

Zhao, Y.-X.; Vidal-Gancedo, J.; Wurst, K.; Jaitner, P.; Clays, K.; Persoons,

A.; Rovira, C.; Veciana, J. Angew. Chem., Int. Ed. 2004, 43, 5266–5268.

(k) Cifuentes, M. P.; Powell, C. E.; Morrall, J. P.; McDonagh, A. M.;

Lucas, N. T.; Humphrey, M. G.; Samoc, M.; Houbrechts, S.; Asselberghs,

I.; Clays, K.; Persoons, A.; Isoshima, T. J. Am. Chem. Soc. 2006, 128,

10819–10832. (l) Samoc, M.; Gauthier, N.; Cifuentes, M. P.; Paul, F.;

Lapinte, C.; Humphrey, M. G. Angew. Chem., Int. Ed. 2006, 45, 7376–

7379. (m) Dalton, G. T.; Cifuentes, M. P.; Petrie, S.; Stranger, R.;

Humphrey, M. G.; Samoc, M. J. Am. Chem. Soc. 2007, 129, 11882–

11883. (n) Boubekeur-Lecaque, L.; Coe, B. J.; Clays, K.; Foerier, S.;

Verbiest, T.; Asselberghs, I. J. Am. Chem. Soc. 2008, 130, 3286–3287.

(o) Wahab, A.; Bhattacharya, M.; Ghosh, S.; Samuelson, A. G.; Das,

P. K. J. Phys. Chem. B 2008, 112, 2842–2847.

Dipolar 2D NLO chromophores consist primarily of V-shaped

(or Λ-shaped) molecules with either D-A-D or A-D-A

motifs.^{5b,e,g-k,m,r,x,y}Such C_2Vsymmetric species display elec-

tronic transitions for which the direction of the transition dipole

moment µ₁₂is perpendicular to the C₂(z) axis, and these are

associated with substantial off-diagonal tensor components ꢀ_zyy

(if the molecule occupies the yz plane). Materials comprising

such chromophores are attractive from several perspectives. In

the context of the quadratic NLO phenomenon second harmonic

generation (SHG), reabsorption of the harmonic light generated

(5) Examples (mostly purely organic): (a) Ledoux, I.; Zyss, J.; Siegel, J. S.;

Brienne, J.; Lehn, J.-M. Chem. Phys. Lett. 1990, 172, 440–444. (b)

Wortmann, R.; Kra¨mer, P.; Glania, C.; Lebus, S.; Detzer, N. Chem. Phys.

1993, 173, 99–108. (c) Zyss, J. J. Chem. Phys. 1993, 98, 6583–6599.

(d) Zyss, J.; Ledoux, I. Chem. ReV. 1994, 94, 77–105. (e) Moylan, C. R.;

Ermer, S.; Lovejoy, S. M.; McComb, I.-H.; Leung, D. S.; Wortmann,

R.; Krämer, P.; Twieg, R. J. J. Am. Chem. Soc. 1996, 118, 12950–12955.

(f) Wortmann, R.; Glania, C.; Kra¨mer, P.; Matschiner, R.; Wolff, J. J.;

Kraft, S.; Treptow, B.; Barbu, E.; La¨ngle, D.; Go¨rlitz, G. Chem.sEur. J.

1997, 3, 1765–1773. (g) Wolff, J. J.; La¨ngle, D.; Hillenbrand, D.;

Wortmann, R.; Matschiner, R.; Glania, C.; Kra¨mer, P. AdV. Mater. 1997,

9, 138–143. (h) Tomonari, M.; Ookubo, N.; Takada, T. Chem. Phys.

Lett. 1997, 266, 488–498. (i) Wolff, J. J.; Wortmann, R. AdV. Phys.

Org. Chem. 1999, 32, 121–217. (j) Liu, Y.-J.; Liu, Y.; Zhang, D.-J.;

Hu, H.-Q.; Liu, C.-B. J. Mol. Struct. 2001, 570, 43–51. (k) Ostroverkhov,

V.; Petschek, R. G.; Singer, K. D.; Twieg, R. J. Chem. Phys. Lett. 2001,

340, 109–115. (l) Cho, B. R.; Piao, M. J.; Son, K. H.; Lee, S. H.; Yoon,

S. J.; Jeon, S.-J.; Cho, M.-H. Chem.-Eur. J. 2002, 8, 3907–3916. (m)

Yang, M.-L.; Champagne, B. J. Phys. Chem. A 2003, 107, 3942–3951.

(n) Wortmann, R.; Lebus-Henn, S.; Reis, H.; Papadopoulos, M. G. J.

Mol. Struct. 2003, 633, 217–226. (o) Brunel, J.; Mongin, O.; Jutand,

A.; Ledoux, I.; Zyss, J.; Blanchard-Desce, M. Chem. Mater. 2003, 15,

4139–4148. (p) Ray, P. C.; Leszczynski, J. Chem. Phys. Lett. 2004, 399,

162–166. (q) Le Floc’h, V.; Brasselet, S.; Zyss, J.; Cho, B. R.; Lee, S. H.;

Jeon, S.-J.; Cho, M.-H.; Min, K. S.; Suh, M. P. AdV. Mater. 2005, 17,

196–200. (r) Cui, Y.-Z.; Fang, Q.; Huang, Z.-L.; Xue, G.; Yu, W.-T.;

Lei, H. Opt. Mater. 2005, 27, 1571–1575. (s) Hennrich, G.; Omenat,

A.; Asselberghs, I.; Foerier, S.; Clays, K.; Verbiest, T.; Serrano, J. L.

Angew. Chem., Int. Ed. 2006, 45, 4203–4206. (t) Kang, H.; Evmenenko,

G.; Dutta, P.; Clays, K.; Song, K.; Marks, T. J. J. Am. Chem. Soc. 2006,

128, 6194–6205. (u) Jeong, M.-Y.; Kim, H. M.; Jeon, S.-J.; Brasselet,

S.; Cho, B. R. AdV. Mater. 2007, 19, 2107–2111. (v) Li, H.-P.; Han,

K.; Tang, G.; Shen, X.-P.; Wang, H.-T.; Huang, Z.-M.; Zhang, Z.-H.;

Bai, L.; Wang, Z.-Y. Chem. Phys. Lett. 2007, 444, 80–84. (w) Moreno

Oliva, M.; Casado, J.; Lo´pez Navarrete, J. T.; Hennrich, G.; Ruiz Delgado,

M. C.; Orduna, J. J. Phys. Chem. C 2007, 111, 18778–18784. (x) Zrig,

S.; Koeckelberghs, G.; Verbiest, T.; Andrioletti, B.; Rose, E.; Persoons,

A.; Asselberghs, I.; Clays, K. J. Org. Chem. 2007, 72, 5855–5858. (y)

Liu, C.-G.; Qiu, Y.-Q.; Su, Z.-M.; Yang, G.-C.; Sun, S.-L. J. Phys. Chem.

C 2008, 112, 7021–7028. (z) Akdas-Kilig, H.; Roisnel, T.; Ledoux, I.;

Le Bozec, H. New J. Chem. 2009, 33, 1470–1473.

(3) (a) Kanis, D. R.; Ratner, M. A.; Marks, T. J. Chem. ReV. 1994, 94,

195–242. (b) Long, N. J. Angew. Chem., Int. Ed. Engl. 1995, 34, 21–

38. (c) Whittall, I. R.; McDonagh, A. M.; Humphrey, M. G.; Samoc,

M. AdV. Organomet. Chem. 1998, 42, 291–362. (d) Whittall, I. R.;

McDonagh, A. M.; Humphrey, M. G.; Samoc, M. AdV. Organomet.

Chem. 1998, 43, 349–405. (e) Heck, J.; Dabek, S.; Meyer-Friedrichsen,

T.; Wong, H. Coord. Chem. ReV. 1999, 190-192, 1217–1254. (f)

Gray, G. M.; Lawson, C. M. In Optoelectronic Properties of Inorganic

Compounds; Roundhill, D. M., Fackler, J. P., Jr., Eds.; Plenum: New

York, 1999; pp 1-27. (g) Shi, S. In Optoelectronic Properties of

Inorganic Compounds; Roundhill, D. M., Fackler, J. P., Jr., Eds.;

Plenum: New York, 1999; pp 55-105. (h) Le Bozec, H.; Renouard,

T. Eur. J. Inorg. Chem. 2000, 229–239. (i) Barlow, S.; Marder, S. R.

Chem. Commun. 2000, 1555–1562. (j) Lacroix, P. G. Eur. J. Inorg.

Chem. 2001, 339–348. (k) Di Bella, S. Chem. Soc. ReV. 2001, 30,

355–366. (l) Goovaerts, E.; Wenseleers, W. E.; Garcia, M. H.; Cross,

G. H. In Handbook of AdVanced Electronic and Photonic Materials

and DeVices; Nalwa, H. S., Ed.; Academic Press: San Diego, 2001;

Vol. 9, pp 127-191. (m) Coe, B. J. In ComprehensiVe Coordination

Chemistry II; McCleverty, J. A., Meyer, T. J., Eds.; Elsevier Pergamon:

Oxford, U.K, 2004; Vol. 9, pp 621-687. (n) Maury, O.; Le Bozec,

H. Acc. Chem. Res. 2005, 38, 691–704. (o) Cariati, E.; Pizzotti, M.;

Roberto, D.; Tessore, F.; Ugo, R. Coord. Chem. ReV. 2006, 250, 1210–

1233. (p) Coe, B. J. Acc. Chem. Res. 2006, 39, 383–393. (q) Zhang,

C.; Song, Y.-L.; Wang, X. Coord. Chem. ReV. 2007, 251, 111–141.

(r) Thompson, M. E.; Djurovich, P. E.; Barlow, S.; Marder, S. In

ComprehensiVe Organometallic Chemistry III; Crabtree, R. H., Min-

gos, D. M. P., Eds.; Elsevier: Oxford, U.K., 2006; Vol. 12, pp 101-

194. (s) Andraud, C.; Maury, O. Eur. J. Inorg. Chem. 2009, 4357–

4371.

9

J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1707

A R T I C L E S

Coe et al.

connected to two pyridinium (pyd) A groups.⁸Hyper-Rayleigh

scattering (HRS)⁹and electronic Stark effect (electroabsorption)

spectroscopic measurements¹⁰revealed large ꢀ_zyycomponents,

and time-dependent density functional theory (TD-DFT) cal-

culations proved very helpful in rationalizing the observed

properties.⁸In this article, we describe a series of new, extended

cis-{Ru^II(NH₃)₄}²⁺-based chromophores, with the focus being

on increasing the ꢀ responses and systematically probing how

the relative contributions of the two main tensor components

depend on the molecular structure.

4,4′-bipyridinium (MeQ⁺), N-phenyl-4,4′-bipyridinium (PhQ⁺), N-(4-

acetylphenyl)-4,4′-bipyridinium (4-AcPhQ⁺) or N-(2-pyrimidyl)-4,4′-

bipyridinium (2-PymQ⁺)],⁸4-[E-2-(4-pyridyl)vinyl]benzaldehyde,¹³

N-methylpicolinium iodide ([Mepic⁺]I),¹⁴E,E-1,4-bis(4-pyridyl)-1,3-

butadiene,¹⁵N-phenyl-4-[E-2-(4-pyridyl)vinyl]pyridinium chloride hy-

drate ([Phbpe⁺]Cl·2.75H₂O),¹⁶N-phenyl-4-[E-2-(4-pyridyl)vinyl]py-

ridinium hexaﬂuorophosphate ([Phbpe⁺]PF₆),¹⁷and N-methyl-4-[E,E,E-

6-(4-pyridyl)hexa-1,3,5-trienyl]pyridinium chloride hydrate ([Mebph⁺]-

Cl·2H₂O)¹⁸were synthesized by using previously published methods.

The compound N-methyl-4-[E-2-(4-pyridyl)vinyl]pyridinium iodide

([Mebpe⁺]I) has been reported previously,¹⁹but only limited synthetic

details and characterization data are available. All other reagents were

obtained commercially and used as supplied. Products were dried at

room temperature overnight in a vacuum desiccator (CaSO₄) prior to

characterization.

Experimental Section

Materials and Procedures. All reactions were performed under

an Ar atmosphere and in Ar-purged solvents. All reactions and

chromatographic puriﬁcations involving complex salts were per-

formed in the dark. The compounds cis-[Ru^IIICl₂(NH₃)₄]Cl,¹¹cis-

[Ru^II(NH₃)₄(H₂O)₂][PF₆]₂,¹²cis-[Ru^II(NH₃)₄(L^A)₂][PF₆]₄[L^A) N-methyl-

General Physical Measurements. ¹H NMR spectra were

recorded on a Varian Unity 400 or a Bruker UltraShield 500

spectrometer, and all shifts are quoted with respect to TMS. The

ﬁne splitting of pyridyl or phenyl ring AA′BB′ patterns is ignored,

and the signals are reported as simple doublets, with J values

referring to the two most intense peaks. Elemental analyses were

performed by the Microanalytical Laboratory, University of Manches-

ter, and UV-vis spectra were obtained by using either a Shimadzu

UV-2401 PC or a Thermo Helios beta spectrophotometer. Mass

spectra were recorded by using +electrospray on a Micromass

Platform II spectrometer. Cyclic voltammetric measurements were

performed by using an EG&G PAR model 283 potentiostat/

galvanostat. A single-compartment cell was used with a silver/silver

chloride reference electrode (3 M NaCl, saturated AgCl) separated

by a salt bridge from a Pt disk working electrode and Pt wire

auxiliary electrode. Acetonitrile was freshly distilled (from CaH₂),

and [N(C₄H₉-n)₄]PF₆, as supplied from Fluka, was used as the

supporting electrolyte. Solutions containing ca. 10^-3M analyte (0.1

M electrolyte) were deaerated by purging with N₂. All E_1/2values

(6) Examples: (a) Zyss, J.; Dhenaut, C.; Chauvan, T.; Ledoux, I. Chem.

Phys. Lett. 1993, 206, 409–414. (b) Dhenaut, C.; Ledoux, I.; Samuel,

I. D. W.; Zyss, J.; Bourgault, M.; Le Bozec, H. Nature 1995, 374,

339–342. (c) Vance, F. W.; Hupp, J. T. J. Am. Chem. Soc. 1999, 121,

4047–4053. (d) Le Bozec, H.; Renouard, T.; Bourgault, M.; Dhenaut,

C.; Brasselet, S.; Ledoux, I.; Zyss, J. Synth. Met. 2001, 124, 185–

189. (e) Le Bouder, T.; Maury, O.; Le Bozec, H.; Ledoux, I.; Zyss, J.

Chem. Commun. 2001, 2430–2431. (f) Le Bozec, H.; Le Bouder, T.;

Maury, O.; Bondon, A.; Ledoux, I.; Deveau, S.; Zyss, J. AdV. Mater.

2001, 13, 1677–1681. (g) Senechal, K.; Maury, O.; Le Bozec, H.;

Ledoux, I.; Zyss, J. J. Am. Chem. Soc. 2002, 124, 4560–4561. (h) Le

Bouder, T.; Maury, O.; Bondon, O.; Costuas, K.; Amouyal, E.; Ledoux,

I.; Zyss, J.; Le Bozec, H. J. Am. Chem. Soc. 2003, 125, 12284–12299.

(i) Viau, L.; Bidault, S.; Maury, O.; Brasselet, S.; Ledoux, I.; Zyss,

J.; Ishow, E.; Nakatani, K.; Le Bozec, H. J. Am. Chem. Soc. 2004,

126, 8386–8387. (j) Maury, O.; Viau, L.; Se´ne´chal, K.; Corre, B.;

Gue´gan, J.-P.; Renouard, T.; Ledoux, I.; Zyss, J.; Le Bozec, H.

Chem.sEur. J. 2004, 10, 4454–4466. (k) Bidault, S.; Brasselet, S.;

Zyss, J.; Maury, O.; Le Bozec, H. J. Chem. Phys. 2007, 126, 034312-

1-034312-13.

were calculated from (E_pa+ E_pc)/2 at a scan rate of 200 mV s^-1

.

Synthesis of N-Methyl-4-[E-2-(4-pyridyl)vinyl]pyridinium

Iodide, [Mebpe⁺]I. Iodomethane (0.06 mL, 0.963 mmol) was added

dropwise to a solution of E-1,4-bis(4-pyridyl)ethylene (182 mg,

0.999 mmol) in chloroform (10 mL), and the reaction was allowed

to stir at room temperature for 2 d. The yellow precipitate was

ﬁltered off, washed with chloroform, and dried: 319 mg, 96%; δ_H

(500 MHz, D₂O) 8.61 (2 H, d, J ) 6.9 Hz, C₅H₄N), 8.51 (2 H, d,

J ) 6.3 Hz, C₅H₄N), 8.05 (2 H, d, J ) 6.9 Hz, C₅H₄N), 7.67 (1 H,

d, J ) 16.4 Hz, CH), 7.61 (2 H, d, J ) 6.2 Hz, C₅H₄N), 7.50 (1 H,

d, J ) 16.2 Hz, CH), 4.26 (3 H, s, Me). Anal. Calcd for

C₁₃H₁₃IN₂·0.5H₂O: C, 46.87; H, 4.24; N, 8.41. Found: C, 46.65;

H, 4.04; N, 8.41. m/z ) 197 ([M - I]⁺).

(7) Examples: (a) Lacroix, P. G.; Di Bella, S.; Ledoux, I. Chem. Mater.

1996, 8, 541–545. (b) Di Bella, S.; Fragala`, I.; Ledoux, I.; Diaz-Garcia,

M. A.; Marks, T. J. J. Am. Chem. Soc. 1997, 119, 9550–9557. (c)

Averseng, F.; Lacroix, P. G.; Malfant, I.; Lenoble, G.; Cassoux, P.;

Nakatani, K.; Maltey-Fanton, I.; Delaire, J. A.; Aukauloo, A. Chem.

Mater. 1999, 11, 995–1002. (d) Briel, O.; Su¨nkel, K.; Krossing, I.;

No¨th, H.; Schma¨lzlin, E.; Meerholz, K.; Bra¨uchle, C.; Beck, W. Eur.

J. Inorg. Chem. 1999, 483–490. (e) Hilton, A.; Renouard, T.; Maury,

O.; Le Bozec, H.; Ledoux, I.; Zyss, J. Chem. Commun. 1999, 2521–

2522. (f) Di Bella, S.; Fragala`, I.; Ledoux, I.; Zyss, J. Chem.sEur. J.

2001, 7, 3738–3743. (g) Di Bella, S.; Fragala`, I. New J. Chem. 2002,

26, 285–290. (h) Di Bella, S.; Fragala`, I. Eur. J. Inorg. Chem. 2003,

2606–2611. (i) Maya, E. M.; Garc´ıa-Frutos, E. M.; Va´zquez, P.; Torres,

T.; Mart´ın, G.; Rojo, G.; Agullo´-Lo´pez, F.; Gonza´lez-Jonte, R. H.;

Ferro, V. R.; Garc´ıa de la Vega, J. M.; Ledoux, I.; Zyss, J. J. Phys.

Chem. A 2003, 107, 2110–2117. (j) Di Bella, S.; Fragala`, I.; Guerri,

A.; Dapporto, P.; Nakatani, K. Inorg. Chim. Acta 2004, 357, 1161–

1167. (k) Rigamonti, L.; Demartin, F.; Forni, A.; Righetto, S.; Pasini,

A. Inorg. Chem. 2006, 45, 10976–10989. (l) Tedim, J.; Patr´ıcio, S.;

Bessada, R.; Morais, R.; Sousa, C.; Marques, M. B.; Freire, C. Eur.

J. Inorg. Chem. 2006, 3425–3433.

Improved Synthesis of N-Methyl-4-[E,E-4-(4-pyridyl)buta-1,3-

dienyl]pyridinium Iodide, [Mebpb⁺]I. This compound was prepared

in a similar manner as [Mebpe⁺]I, using iodomethane (0.02 mL,

0.321 mmol) and E,E-1,4-bis(4-pyridyl)-1,3-butadiene (70 mg,

(13) Ichimura, K.; Watanabe, S. J. Polym. Sci., Polym. Chem. 1982, 20,

1419–1432.

(14) Murrill, P. J. Am. Chem. Soc. 1899, 21, 828–854.

(15) Coe, B. J.; Fitzgerald, E. C.; Helliwell, M.; Brunschwig, B. S.; Fitch,

A. G.; Harris, J. A.; Coles, S. J.; Horton, P. N.; Hursthouse, M. B.

Organometallics 2008, 27, 2730–2742.

(8) Coe, B. J.; Harris, J. A.; Jones, L. A.; Brunschwig, B. S.; Song, K.;

Clays, K.; Gar´ın, J.; Orduna, J.; Coles, S. J.; Hursthouse, M. B. J. Am.

Chem. Soc. 2005, 127, 4845–4859.

(9) (a) Terhune, R. W.; Maker, P. D.; Savage, C. M. Phys. ReV. Lett.

1965, 14, 681–684. (b) Clays, K.; Persoons, A. Phys. ReV. Lett. 1991,

66, 2980–2983. (c) Clays, K.; Persoons, A. ReV. Sci. Instrum. 1992,

63, 3285–3289. (d) Hendrickx, E.; Clays, K.; Persoons, A. Acc. Chem.

Res. 1998, 31, 675–683.

(16) Coe, B. J.; Harries, J. L.; Helliwell, M.; Jones, L. A.; Asselberghs, I.;

Clays, K.; Brunschwig, B. S.; Harris, J. A.; Gar´ın, J.; Orduna, J. J. Am.

Chem. Soc. 2006, 128, 12192–12204.

(17) Coe, B. J.; Harris, J. A.; Asselberghs, I.; Persoons, A.; Jeffery, J. C.;

Rees, L. H.; Gelbrich, T.; Hursthouse, M. B. J. Chem. Soc., Dalton

Trans. 1999, 3617–3625.

(10) (a) Liptay, W. In Excited States; Lim, E. C., Ed.; Academic Press:

New York, 1974; Vol. 1, pp 129-229. (b) Bublitz, G. U.; Boxer,

S. G. Annu. ReV. Phys. Chem. 1997, 48, 213–242. (c) Vance, F. W.;

Williams, R. D.; Hupp, J. T. Int. ReV. Phys. Chem. 1998, 17, 307–

329. (d) Brunschwig, B. S.; Creutz, C.; Sutin, N. Coord. Chem. ReV.

1998, 177, 61–79.

(18) Coe, B. J.; Jones, L. A.; Harris, J. A.; Brunschwig, B. S.; Asselberghs,

I.; Clays, K.; Persoons, A.; Gar´ın, J.; Orduna, J. J. Am. Chem. Soc.

2004, 126, 3880–3891.

(19) See for examples: (a) Bergmann, E. D.; Crane, F. E. Jr.; Fuoss, R. M.

J. Am. Chem. Soc. 1952, 74, 5979–5982. (b) Rembaum, A.; Hermann,

A. M.; Stewart, F. E.; Gutmann, F. J. Phys. Chem. 1969, 73, 513–

520. (c) Vansant, J.; Smets, G.; Declercq, J. P.; Germain, G.; Van

Meerssche, M. J. Org. Chem. 1980, 45, 1557–1565.

(11) Boggs, S. E.; Clarke, R. E.; Ford, P. C. Inorg. Chim. Acta 1996, 247,

129–130.

(12) Sugaya, T.; Sano, M. Inorg. Chem. 1993, 32, 5878–5879.

9

1708 J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010

V-Shaped Ruthenium(II)-Based Chromophores

A R T I C L E S

0.336 mmol) to give a yellow solid: 97 mg, 78%; δ_H(500 MHz,

D₂O) 8.50 (2 H, d, J ) 6.9 Hz, C₅H₄N), 8.44 (2 H, d, J ) 6.3 Hz,

C₅H₄N), 7.92 (2 H, d, J ) 6.9 Hz, C₅H₄N), 7.56 (1 H, dd, J )

11.0, 15.5 Hz, CH), 7.53 (2 H, d, J ) 6.6 Hz, C₅H₄N), 7.35 (1 H,

dd, J ) 10.7, 15.5 Hz, CH), 6.97 (1 H, d, J ) 15.8 Hz, CH), 6.91

(1 H, d, J ) 15.8 Hz, CH), 4.20 (3 H, s, Me). Anal. Calcd for

C₁₅H₁₅IN₂·H₂O: C, 48.93; H, 4.65; N, 7.61. Found: C, 49.22; H,

4.12; N, 7.52. m/z ) 223 ([M - I]⁺).

solution of [(2,4-DNPh)bpb⁺]Cl (370 mg) in water (50 mL), and

the mixture was heated at 80 °C for 1 h. The reaction mixture was

cooled to room temperature and decanted (to separate from the 2,4-

dinitroaniline byproduct), and the aqueous solution was concentrated

under reduced pressure. Further 2,4-dinitroaniline was ﬁltered off,

and the ﬁltrate was evaporated to dryness. The yellow-orange solid

was dissolved in ethanol (20 mL), and diethyl ether was added to

give a yellow precipitate that was ﬁltered off, washed with CH₂Cl₂

and then diethyl ether, and dried under vacuum: 172 mg, 30%

(based on 2,4-dinitrochlorobenzene); δ_H(400 MHz, CD₃SOCD₃)

9.27 (2 H, d, J ) 6.8 Hz, C₅H₄N), 8.74 (2 H, d, J ) 6.0 Hz, C₅H₄N),

8.43 (2 H, d, J ) 6.8 Hz, C₅H₄N), 8.08 (1 H, dd, J ) 15.2, 10.8

Hz, CH), 7.91-7.88 (4 H, Ph + C₅H₄N), 7.78-7.69 (4 H, Ph +

CH), 7.23 (1 H, d, J ) 16.0 Hz, CH), 7.19 (1 H, d, J ) 16.0 Hz,

CH). Anal. Calcd for C₂₀H₁₇ClN₂·4.5H₂O: C, 59.77; H, 6.52; N,

6.97. Found: C, 59.80; H, 5.97; N, 7.02. m/z: 285 ([M - Cl]⁺).

Synthesis of N-Phenyl-4-[E,E-4-(4-pyridyl)buta-1,3-dienyl]py-

ridinium Hexaﬂuorophosphate, [Phbpb⁺]PF₆. Aqueous NH₄PF₆

was added dropwise to a stirred solution of [Phbpb⁺]Cl·4.5H₂O

(151 mg, 0.376 mmol) in water (6 mL). The pale yellow precipitate

was ﬁltered off, washed with water, and dried: 164 mg, 98%; δ_H

(500 MHz, CD₃COCD₃) 9.26 (2 H, d, J ) 6.9 Hz, C₅H₄N), 8.64 (2

H, d, J ) 6.3 Hz, C₅H₄N), 8.48 (2 H, d, J ) 6.9 Hz, C₅H₄N), 8.05

(1 H, dd, J ) 10.7, 15.8 Hz, CH), 7.97-7.95 (2 H, Ph), 7.80-7.79

(3 H, Ph), 7.60 (1 H, dd, J ) 10.7, 15.5 Hz, CH), 7.59 (2 H, d, J

) 6.3 Hz, C₅H₄N), 7.27 (1 H, d, J ) 15.5 Hz, CH), 7.19 (1 H, d,

J ) 15.8 Hz, CH). Anal. Calcd for C₂₀H₁₇F₆N₂P·H₂O: C, 53.58;

H, 4.27; N, 6.25. Found: C, 53.49; H, 4.38; N, 6.02. m/z ) 285

([M - PF₆]⁺).

Improved Synthesis of N-Methyl-1,4-bis[E-2-(4-pyridyl)-

vinyl]benzene Iodide and Hexaﬂuorophosphate, [Mebpvb⁺]X

(X ) I^-or PF₆^-). Piperidine (3 drops) was added to a solution of

4-[E-2-(4-pyridyl)vinyl]benzaldehyde (383 mg, 1.83 mmol) and

[Mepic⁺]I (448 mg, 1.90 mmol) in absolute ethanol (5 mL). The

solution was heated under reﬂux for 2.5 h in the dark. The orange

precipitate was ﬁltered off, washed with a little ethanol and then

diethyl ether, and dried: 600 mg, 77%; δ_H(400 MHz, CD₃OD)

8.73 (2 H, d, J ) 6.9 Hz, C₅H₄N), 8.51 (2 H, d, J ) 6.3 Hz, C₅H₄N),

8.18 (2 H, d, J ) 6.9 Hz, C₅H₄N), 7.96 (1 H, d, J ) 16.3 Hz, CH),

7.80 (2 H, d, J ) 8.4 Hz, C₆H₄), 7.75 (2 H, d, J ) 8.4 Hz, C₆H₄),

7.62 (2 H, d, J ) 6.4 Hz, C₅H₄N), 7.60 (1 H, d, J ) 16.5 Hz, CH),

7.48 (1 H, d, J ) 16.2 Hz, CH), 7.31 (1 H, d, J ) 16.4 Hz, CH),

4.32 (3 H, s, Me). Anal. Calcd for C₂₁H₁₉IN₂: C, 59.17; H, 4.49;

N, 6.57. Found: C, 58.90; H, 4.49; N, 6.49. Aqueous NH₄PF₆was

added to a solution of [Mebpvb⁺]I (277 mg, 0.650 mmol) in

methanol/water (2:1, 75 mL). The yellow precipitate was ﬁltered

off, washed with water and dried. Puriﬁcation was effected by

reprecipitation from DMF/diethyl ether: 237 mg, 68%; δ_H(500

MHz, CD₃OCD₃) 8.96 (2 H, d, J ) 6.7 Hz, C₅H₄N), 8.64 (2 H, d,

J ) 6.2 Hz, C₅H₄N), 8.34 (2 H, d, J ) 6.8 Hz, C₅H₄N), 8.07 (1 H,

d, J ) 16.4 Hz, CH), 7.84 (2 H, d, J ) 8.5 Hz, C₆H₄), 7.82

(2 H, d, J ) 8.6 Hz, C₆H₄), 7.70-7.67 (3 H, CH + C₅H₄N), 7.64

(1 H, d, J ) 16.4 Hz, CH), 7.44 (1 H, d, J ) 16.5 Hz, CH), 4.54

(3 H, s, Me). Anal. Calcd for C₂₁H₁₉F₆N₂P·0.5HPF₆·0.5H₂O: C,

47.53; H, 3.57; N, 5.22. Found: C, 47.66; H, 3.56; N, 5.29. m/z )

299 ([M - PF₆]⁺).

Synthesis of N-(2-Pyrimidyl)-4-[E-2-(4-pyridyl)vinyl]pyridinium

Hexaﬂuorophosphate, [2-Pymbpe⁺]PF₆. A solution of 2-chloro-

pyrimidine (114 mg, 0.995 mmol) and E-1,2-bis(4-pyridyl)ethylene

(182 mg, 0.999 mmol) in methanol (5 mL) was stirred for 16 h

under reﬂux. After cooling to room temperature, the solution was

added to diethyl ether (200 mL), and the resulting precipitate was

ﬁltered off, washed with chloroform, and dried. The crude product

was dissolved in a minimum of water, and aqueous NH₄PF₆was

added. The precipitate was ﬁltered off, washed with water, a small

amount of ethanol, and chloroform, and dried. Puriﬁcation was

effected by reprecipitation from acetone/diethyl ether containing

one drop of triethylamine to afford a pale pink solid: 39 mg, 10%;

δ_H(500 MHz, CD₃COCD₃) 10.18 (2 H, d, J ) 7.3 Hz, C₅H₄N),

9.25 (2 H, d, J ) 4.7 Hz, C₄H₃N₂), 8.74 (2 H, d, J ) 6.0 Hz,

C₅H₄N), 8.66 (2 H, d, J ) 6.9 Hz, C₅H₄N), 8.26 (1 H, d, J ) 16.4

Hz, CH), 8.03 (1 H, d, J ) 16.4 Hz, CH), 8.00 (1 H, t, J ) 4.7 Hz,

C₄H₃N₂), 7.76 (2 H, d, J ) 6.0 Hz, C₅H₄N). Anal. Calcd for

C₁₆H₁₃F₆N₄P: C, 47.30; H, 3.23; N, 13.79. Found: C, 47.04; H,

3.15; N, 13.45. m/z ) 261 ([M - PF₆]⁺).

Synthesis of cis-[Ru^II(NH₃)₄(4-AcPhQ⁺)₂][BPh₄]₄(3B). This

compound was prepared exactly as reported previously as far as

the collection of the chloride salt.⁸A portion of this salt (173 mg,

0.201 mmol) was treated with aqueous NaBPh₄, and the precipitate

was ﬁltered off, washed with water and dried: 124 mg, 31%; δ_H

(500 MHz, CD₃COCD₃) 8.73 (4 H, d, J ) 6.9 Hz, C₅H₄N), 8.50 (4

H, d, J ) 7.3 Hz, C₅H₄N), 8.31 (4 H, d, J ) 9.1 Hz, C₆H₄), 7.94

(4 H, d, J ) 6.9 Hz, C₅H₄N), 7.76 (4 H, d, J ) 8.9 Hz, C₆H₄), 7.50

(4 H, d, J ) 7.0 Hz, C₅H₄N), 7.32-7.31 (32 H, BPh₄-H^3,5), 6.88

(32 H, t, J ) 7.4 Hz, BPh₄-H^2,6), 6.74 (16 H, t, J ) 7.1 Hz, BPh₄-

H⁴), 3.48 (6 H, s, NH₃), 2.96 (6 H, s, NH₃), 2.72 (6 H, s, Me).

Anal. Calcd for C₁₃₂H₁₂₂B₄N₈O₂Ru: C, 79.40; H, 6.16; N, 5.61.

Found: C, 79.63; H, 6.65; N, 5.34.

Synthesis of cis-[Ru^II(NH₃)₄(Mebpe⁺)₂][PF₆]₄(5). A solution

of cis-[Ru^IIICl₂(NH₃)₄]Cl (55 mg, 0.200 mmol) in degassed water

(8 mL) was acidiﬁed with triﬂuoroacetic acid (4 drops) and reduced

over Zn/Hg amalgam for 15 min with Ar agitation. The solution

was then ﬁltered under Ar into a ﬂask containing [Mebpe⁺]I·0.5H₂O

(129 mg, 0.387 mmol), and the reaction was stirred for 16 h at

room temperature. The resulting solution was loaded onto a

Sephadex C-25 column and eluted with increasing concentrations

of aqueous NaCl, ranging from 0.25 to 0.7 M. The main dark blue

fraction was collected, acetone (ca. 2000 mL) was added, and the

precipitate ﬁltered off, washed with acetone, and dried. This solid

was dissolved in water, and aqueous NH₄PF₆was added to afford

a dark blue precipitate that was ﬁltered off, washed with water,

and dried: 26 mg, 11%; δ_H(500 MHz, CD₃COCD₃) 9.00 (4 H, d,

J ) 6.9 Hz, C₅H₄N), 8.65 (4 H, d, J ) 6.9 Hz, C₅H₄N), 8.35 (4 H,

d, J ) 6.9 Hz, C₅H₄N), 8.03 (2 H, d, J ) 16.4 Hz, CH), 7.91 (2 H,

d, J ) 16.4 Hz, CH), 7.61 (4 H, d, J ) 6.9 Hz, C₅H₄N), 4.54 (6 H,

s_,Me), 3.31 (6 H, s, NH₃), 2.92 (6 H, s, NH₃). Anal. Calcd for

C₂₆H₃₈F₂₄N₈P₄Ru: C, 27.31; H, 3.35; N, 9.80. Found: C, 27.26; H,

3.22; N, 9.20.

Synthesis of N-Phenyl-4-[E,E-4-(4-pyridyl)buta-1,3-dienyl]py-

ridinium Hexaﬂuorophosphate, [Phbpb⁺]Cl. A solution of 2,4-

dinitrochlorobenzene (403 mg, 1.99 mmol) and E,E-1,4-bis(4-

pyridyl)-1,3-butadiene (416 mg, 2.00 mmol) in acetone (10 mL)

was stirred under reﬂux for 16 h. The mixture was cooled to room

temperature, and the bright yellow precipitate of crude [(2,4-

DNPh)bpb⁺]Cl was collected by ﬁltration, washed with acetone

and diethyl ether, and dried under vacuum: 520 mg; δ_H(400 MHz,

CD₃OD) 9.27 (1 H, d, J ) 2.4 Hz, C₆H₃), 9.06 (2 H, d, J ) 6.8

Hz, C₅H₄N), 8.91 (1 H, dd, J ) 8.8, 2.4 Hz, C₆H₃), 8.57 (2 H, d,

J ) 6.0 Hz, C₅H₄N), 8.39 (2 H, d, J ) 7.2 Hz, C₅H₄N), 8.29 (1 H,

d, J ) 8.8 Hz, C₆H₃), 8.02 (1 H, dd, J ) 15.5, 10.8 Hz, CH), 7.64

(2 H, d, J ) 6.0 Hz, C₅H₄N), 7.57 (1 H, dd, J ) 15.3, 10.8 Hz,

CH), 7.22 (1 H, d, J ) 15.6 Hz, CH), 7.21 (1 H, d, J ) 15.6 Hz,

CH). m/z: 375 ([M - Cl]⁺). Aniline (ca. 1 mL) was added to a

Synthesis of cis-[Ru^II(NH₃)₄(Phbpe⁺)₂][PF₆]₄(6). Method 1.

This compound was prepared and subjected to column chroma-

tography in a similar manner as 5 by using [Phbpe⁺]Cl·2.75H₂O

(118 mg, 0.343 mmol) instead of [Mebpe⁺]I ·0.5H₂O. After ﬁltration

of the acetone solution, the ﬁltrate was still colored and so was

concentrated under vacuum to remove all of the acetone. Solid

NH₄PF₆was added to the remaining aqueous solution, and the

9

J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1709

A R T I C L E S

Coe et al.

resulting precipitate ﬁltered off. This solid was combined with the

material isolated previously by the addition of acetone to the column

eluent and puriﬁed by reprecipitation from acetone with diethyl

ether. The resulting dark blue solid was ﬁltered off, washed with

diethyl ether, and dried: 32 mg, 14%; δ_H(500 MHz, CD₃COCD₃)

9.30 (4 H, d, J ) 6.6 Hz, C₅H₄N), 8.69 (4 H, d, J ) 6.3 Hz, C₅H₄N),

8.53 (4 H, d, J ) 6.9 Hz, C₅H₄N), 8.19 (2 H, d, J ) 16.7 Hz, CH),

8.04 (2 H, d, J ) 16.4 Hz, CH), 7.97-7.95 (4 H, Ph), 7.80-7.79

(6 H, Ph), 7.67 (4 H, d, J ) 6.6 Hz, C₅H₄N), 3.35 (6 H, s, NH₃),

2.95 (6 H, s, NH₃). Anal. Calcd for C₃₆H₄₂F₂₄N₈P₄Ru·2H₂O: C,

33.17; H, 3.56; N, 8.59. Found: C, 33.08; H, 3.10; N, 8.20. Method

2. cis-[Ru^II(NH₃)₄(OH₂)₂][PF₆]₂(49 mg, 0.099 mmol) was added

to a solution of [Phbpe⁺]PF₆(80 mg, 0.198 mmol) in acetone (10

mL), and the reaction was stirred for 5 h at room temperature. The

product was precipitated by the addition of a solution of LiCl in

acetone, ﬁltered off, washed with acetone, and dried. The solid was

dissolved in a minimum of water, loaded onto a Sephadex C-25

column, and puriﬁed exactly as for 5 to yield a dark blue solid: 20

mg, 16%; δ_H(500 MHz, CD₃COCD₃) 9.30 (4 H, d, J ) 6.6 Hz,

C₅H₄N), 8.69 (4 H, d, J ) 6.3 Hz, C₅H₄N), 8.53 (4 H, d, J ) 6.9

Hz, C₅H₄N), 8.19 (2 H, d, J ) 16.7 Hz, CH), 8.04 (2 H, d, J )

16.4 Hz, CH), 7.97-7.95 (4 H, Ph), 7.80-7.79 (6 H, Ph) 7.67 (4

H, d, J ) 6.6 Hz, C₅H₄N), 3.35 (6 H, s, NH₃), 2.95 (6 H, s, NH₃).

Anal. Calcd for C₃₆H₄₂F₂₄N₈P₄Ru·2H₂O: C, 33.17; H, 3.56; N, 8.59.

Found: C, 32.91; H, 3.18; N, 8.43.

δ_H(400 MHz, CD₃COCD₃) 8.86 (4 H, d, J ) 6.8 Hz, C₅H₄N),

8.48 (4 H, d, J ) 6.0 Hz, C₅H₄N), 8.19 (4 H, d, J ) 6.8 Hz, C₅H₄N),

7.78 (2 H, dd, J ) 10.4, 15.6 Hz, CH), 7.50 (2 H, dd, J ) 10.4,

16.0 Hz, CH), 7.42 (4 H, d, J ) 6.4 Hz, C₅H₄N), 7.02-6.87 (8 H,

m, CH), 4.48 (6 H, s, Me), 3.20 (6 H, s, NH₃), 2.83 (6 H, s, NH₃).

Anal. Calcd for C₃₄H₄₆F₂₄N₈P₄Ru: C, 32.73; H, 3.72; N, 8.98.

Found: C, 32.50; H, 3.44; N, 8.55.

Synthesis of cis-[Ru^II(NH₃)₄(Mebpvb⁺)₂][PF₆]₄(11). cis-

[Ru^II(NH₃)₄(OH₂)₂][PF₆]₂(106 mg, 0.214 mmol) was added to a

solution of [Mebpvb⁺]PF₆·0.5HPF₆·0.5H₂O (190 mg, 0.361 mmol)

in DMF (5 mL) and the reaction was stirred for 16 h at room

temperature. The addition of diethyl ether produced a precipitate,

which was dissolved in a minimum of acetone and loaded onto a

Sephadex C-25 column. The column was eluted with 5:3 aqueous

NaCl/acetone, using a steadily increasing NaCl concentration

(0.05-0.30 M). The main dark red fraction was collected, the

acetone was removed under vacuum, and aqueous NH₄PF₆was

added to give a dark red precipitate that was ﬁltered off, washed

with water, and dried: 51 mg, 20%; δ_H(400 MHz, CD₃COCD₃)

8.93 (4 H, d, J ) 6.8 Hz, C₅H₄N), 8.52 (4 H, d, J ) 6.4 Hz, C₅H₄N),

8.32 (4 H, d, J ) 7.2 Hz, C₅H₄N), 8.05 (2 H, d, J ) 16.4 Hz, CH),

7.83 (4 H, d, J ) 8.8 Hz, C₆H₄), 7.79 (4 H, d, J ) 8.8 Hz, C₆H₄),

7.70 (2 H, d, J ) 16.5 Hz, CH), 7.62 (2 H, d, J ) 16.0 Hz, CH),

7.54 (4 H, d, J ) 6.8 Hz, C₅H₄N), 7.43 (2 H, d, J ) 16.4 Hz, CH),

4.52 (6 H, s, Me), 3.18 (6 H, s, NH₃), 2.84 (6 H, s, NH₃). Anal.

Calcd for C₄₂H₅₀F₂₄N₈P₄Ru·2H₂O: C, 36.45; H, 3.93; N, 8.10.

Found: C, 36.65; H, 3.56; N, 7.46.

Synthesis of cis-[Ru^II(NH₃)₄(2-Pymbpe⁺)₂][PF₆]₄(7). This

compound was prepared and puriﬁed in a similar manner as 6

(method 2), by using cis-[Ru^II(NH₃)₄(OH₂)₂][PF₆]₂(53 mg, 0.107

mmol) and [Pymbpe⁺]PF₆(80 mg, 0.197 mmol) instead of

[Phbpe⁺]PF₆to afford a dark blue solid: 24 mg, 18%; δ_H(500 MHz,

CD₃COCD₃) 10.14 (4 H, d, J ) 7.3 Hz, C₅H₄N), 9.24 (4 H, d, J )

4.7 Hz, C₄H₃N₂), 8.73 (4 H, d, J ) 6.7 Hz, C₅H₄N), 8.59 (4 H, d,

J ) 7.3 Hz, C₅H₄N), 8.29 (2 H, d, J ) 16.4 Hz, CH), 8.11 (2 H,

d, J ) 16.1 Hz, CH), 7.99 (4 H, d, J ) 4.8 Hz, C₄H₃N₂), 7.70 (4

H, d, J ) 6.9 Hz, C₅H₄N), 3.39 (6 H, s, NH₃), 2.98 (6 H, s, NH₃).

Anal. Calcd for C₃₂H₃₈F₂₄N₁₂P₄Ru: C, 30.22; H, 3.01; N, 13.22.

Found: C, 30.16; H, 3.08; N, 12.72.

X-ray Crystallography. Crystals of the complex salts cis-

[Ru^II(NH₃)₄(MeQ⁺)₂][PF₆]₄·MeCN ·H₂O (1·MeCN ·H₂O) and cis-

[Ru^II(NH₃)₄(PhQ⁺)₂][PF₆]₄(2) were obtained by slow diffusion of

diethyl ether vapor into acetonitrile solutions at 4 °C. Data were

collected on a Nonius Kappa CCD area-detector X-ray diffracto-

meter controlled by the Collect software package.²⁰The data were

processed using Denzo²¹and semiempirical absorption corrections

were applied using SADABS.²²

Crystals of the complex salt cis-[Ru^II(NH₃)₄(4-AcPhQ⁺)₂][BPh₄]₄

(3B) were obtained by slow diffusion of diethyl ether vapor into

an acetonitrile solution at 4 °C, while those of the pro-ligand salts

[MeQ⁺]I, [Mebpe⁺]I·0.5H₂O and [Phbpe⁺]Cl·HCl·H₂O were

obtained via slow diffusion of diethyl ether vapor into methanol

solutions at room temperature (and containing a trace of HCl in

the latter case). Data were collected on a Bruker APEX CCD X-ray

diffractometer and processed using the Bruker SAINT²³software

package, using SADABS²²for absorption corrections.

Synthesis of cis-[Ru^II(NH₃)₄(Mebpb⁺)₂][PF₆]₄(8). This com-

pound was prepared and puriﬁed in a similar manner as 5

by using [Mebpb⁺]I·H₂O (174 mg, 0.473 mmol) instead of

[Mebpe⁺]I·0.5H₂O and 0.3-0.6 M aqueous NaCl for the column

chromatography. Further puriﬁcation was effected by reprecipitation

from acetone/diethyl ether to afford a dark blue solid: 57 mg, 24%;

δ_H(400 MHz, CD₃COCD₃) 8.90 (4 H, d, J ) 6.8 Hz, C₅H₄N),

8.54 (4 H, d, J ) 6.8 Hz, C₅H₄N), 8.23 (4 H, d, J ) 6.4 Hz, C₅H₄N),

7.83 (2 H, dd, J ) 10.8, 15.4 Hz, CH), 7.63 (2 H, dd, J ) 10.8,

15.4 Hz, CH), 7.48 (4 H, d, J ) 6.4 Hz, C₅H₄N), 7.14 (2 H, d, J

) 15.2 Hz, CH), 7.12 (2 H, d, J ) 15.2 Hz, CH), 4.50 (6 H, s,

Me), 3.23 (6 H, s, NH₃), 2.86 (6 H, s, NH₃). Anal. Calcd for

C₃₀H₄₂F₂₄N₈P₄Ru·H₂O: C, 29.69; H, 3.65; N, 9.23. Found: C, 29.44;

H, 3.38; N, 9.01.

The structures were solved by direct methods and reﬁned by

2

full-matrix least-squares on all F₀data using SHELXS-97 and

SHELXL-97, respectively.²⁴All non-hydrogen atoms were reﬁned

anisotropically. Hydrogen atoms were included in idealized posi-

tions using the riding model, with thermal parameters of 1.2 or 1.5

times those of the parent atoms. The asymmetric unit of

[Mebpe⁺]I·0.5H₂O contains two ion pairs and one water molecule.

The crystals of 2 gave diffuse diffraction and as such only some of

the high angle data were observed (still >90% completeness to

Synthesis of cis-[Ru^II(NH₃)₄(Phbpb⁺)₂][PF₆]₄(9). This com-

pound was prepared and puriﬁed in a similar manner as 6 (method

2) by using cis-[Ru^II(NH₃)₄(OH₂)₂][PF₆]₂(69 mg, 0.139 mmol) and

[Phbpb⁺]PF₆(121 mg, 0.281 mmol) instead of [Phbpe⁺]PF₆to

afford a dark blue solid: 25 mg, 14%; δ_H(400 MHz, CD₃COCD₃)

9.20 (4 H, d, J ) 6.8 Hz, C₅H₄N), 8.57 (4 H, d, J ) 6.0 Hz, C₅H₄N),

8.41 (4 H, d, J ) 6.8 Hz, C₅H₄N), 8.03-7.96 (6 H, Ph + CH),

7.79-7.78 (6 H, Ph), 7.68 (2 H, dd, J ) 10.4, 15.6 Hz, CH), 7.51

(4 H, d, J ) 6.4 Hz, C₅H₄N), 7.27 (2 H, d, J ) 15.6 Hz, CH), 7.20

(2 H, d, J ) 15.6 Hz, CH), 3.26 (6 H, s, NH₃), 2.88 (6 H, s, NH₃).

Anal. Calcd for C₄₀H₄₆F₂₄N₈P₄Ru: C, 36.40; H, 3.51; N, 8.49.

Found: C, 36.03; H, 3.46; N, 8.30.

-

27.5°). Also, three of the four PF₆anions in the asymmetric unit

are highly disordered and in each case were reﬁned using two major

components resulting in R-factors higher than standard. It was also

necessary to restrain all of the P-F bonds to be a similar length.

The crystal of 3B diffracted extremely weakly to 1.1 Å resolution

and the data were cut at this point. The phenyl rings were

constrained to be regular hexagons; the phenyl group C49-C54 is

disordered over 2 sites the occupancies of which were constrained

to sum to 1.0. The pyridyl rings were constrained as rigid groups,

Synthesis of cis-[Ru^II(NH₃)₄(Mebph⁺)₂][PF₆]₄(10). This

compound was prepared in a similar manner as 5 by us-

ing [Mebph⁺]Cl · 2H₂O (113 mg, 0.352 mmol) instead of

[Mebpe⁺]I·0.5H₂O and 0.35-0.6 M aqueous NaCl for the column

chromatography. Further puriﬁcation was effected by reprecipitation

from acetone/diethyl ether to afford a dark blue solid: 32 mg, 15%;

(20) Hooft, R. Collect, Data collection software; Nonius BV: Delft, The

Netherlands, 1998.

(21) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307–326.

(22) SADABS (Version 2.10); Bruker AXS Inc.: Madison, WI, 2003.

(23) SAINT (Version 6.45); Bruker AXS Inc.: Madison, WI, 2003.

(24) Sheldrick, G. M. SHELXL 97, Programs for Crystal Structure Analysis

(Release 97-2); University of Go¨ttingen: Go¨ttingen, Germany, 1997.

9

1710 J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010

V-Shaped Ruthenium(II)-Based Chromophores

A R T I C L E S

Table 1. Crystallographic Data and Reﬁnement Details for the Salts 1·MeCN·H₂O, 2, 3B, [MeQ⁺]I, [Mebpe⁺]I·0.5H₂O and

[Phbpe⁺]Cl·HCl·H₂O

1·MeCN·H₂O

2

3B

[MeQ⁺]I

[Mebpe⁺]I·0.5H₂O

[Phbpe⁺]Cl·HCl·H₂O

formula

M

crystal system

space group

a, Å

C₂₄H₃₉F₂₄N₉OP₄Ru

1150.59

monoclinic

P2₁

11.0075(11)

18.0498(19)

11.1488(11)

103.942(7)

2149.8(4)

2

C₃₂H₃₈F₂₄N₈P₄Ru

1215.65

monoclinic

P2₁/c

16.86(7)

18.60(3)

14.35(5)

95.0(3)

C_152.25H_165.88B₄N_11.38O_5.38Ru C₁₁H₁₁IN₂

C₁₃H₁₄IN₂O_0.5

333.16

monoclinic

P2₁/c

11.0700(9)

19.9870(15)

12.9010(10)

111.9900(10)

2646.8(4)

8

C₁₈H₁₈Cl₂N₂O

349.24

monoclinic

P2₁/c

16.0499(16)

6.9437(7)

16.4241(16)

115.326(2)

1654.5(3)

4

2385.40

orthorhombic

Fdd2

298.12

orthorhombic

Pbca

61.18(2)

28.994(10)

30.362(10)

90

16.4397(15)

7.3508(6)

18.6167(16)

90

b, Å

c, Å

ꢀ, deg

V, Å³

4484(26)

4

53858(31)

16

2249.7(3)

8

Z

T, K

120(2)

0.654

120(2)

0.630

100(2)

0.176

100(2)

2.810

100(2)

2.401

100(2)

0.398

µ, mm^-1

crystal size, mm³

crystal appearance dark red shard

reﬂections

collected

0.14 × 0.10 × 0.05 0.18 × 0.15 × 0.04 0.60 × 0.30 × 0.02

0.34 × 0.10 × 0.10 0.22 × 0.20 × 0.10 0.60 × 0.15 × 0.02

dark red slab

29890

black plate

46837

colorless needle

11936

black block

14755

colorless plate

9157

25951

independent

reﬂections (R_int

reﬂections with

I > 2σ(I)

9747 (0.0514)

8030

9482 (0.0807)

4930

10616 (0.2524)

4436

2307 (0.0338)

1901

5399 (0.0317)

4266

3381 (0.0285)

2739

)

goodness-of-

1.025

1.029

0.861

1.120

1.001

1.361

ﬁt on F²

ﬁnal R1, wR2

[I > 2σ(I)]

(all data)

0.0476, 0.0936

0.1418, 0.3328

0.0949, 0.2268

0.0355, 0.0797

0.0302, 0.0617

0.0476, 0.1372

a

0.0656, 0.1015

0.2326, 0.3951

0.1476, 0.2447

0.0452, 0.0838

0.0457, 0.0664

0.0614, 0.1433

peak and hole

0.667, -0.444

1.991, -1.269

0.560, -0.419

2.101, -0.595

0.812, -0.376

0.486, -0.303

(e Å^-3

)

a

2

The structures were reﬁned on F_ousing all data; the values of R1 are given for comparison with older reﬁnements based on F_owith a typical

threshold of F_o> 4σ(F_o).

using parameters from the much more precise and related structure

Na₂[Fe^II(CN)₅(MeQ⁺)]·9H₂O.¹⁶The crystal also contained unre-

ﬁnable solvent which was accounted for by using the SQUEEZE

procedure,²⁵leading to 3433 electrons per cell, which is equivalent

to about 54(MeCN + Et₂O) molecules per cell, occupying a void

volume of 15165 Å³. These additional solvent molecules were added

to the formula. All other calculations for 3B were carried out using

the SHELXTL package.²⁶Crystallographic data and reﬁnement

details are presented in Table 1.

preparation and measurements were carried out in the dark, and

the solvents were degassed before use in order to protect the

compounds from decomposition due to photoactivated ligand loss

and/or aerial oxidation. The reported ꢀ values are the averages taken

from measurements at different amplitude modulation frequencies.

HRS depolarization ratios²⁸were determined at 800 nm in

acetonitrile according to a published methodology.²⁹

Stark Spectroscopy. The Stark apparatus, experimental meth-

ods, and data collection procedure were as previously reported,³⁰

except that a Xe arc lamp was used as the light source in the

place of a W ﬁlament bulb. The Stark spectrum for each

compound was measured at least twice. To ﬁt the Stark data,

the absorption (ε/ν vs ν) spectrum was in most cases modeled

with a sum of Gaussian curves that reproduce the data and

separate the peaks. The ﬁrst and second derivatives of the

Gaussian curves were then used to ﬁt the Stark spectra with

Liptay’s equation.^10aThe dipole moment change ∆µ₁₂) µ_e-

µ_g(where µ_eand µ_gare the respective excited and ground state

dipole moments, associated with each of the optical transitions

considered in the ﬁt) was then calculated from the coefﬁcient

of the second derivative component. For all of the complex salts,

two or three Gaussian functions were necessary to ﬁt the metal-

to-ligand charge-transfer (MLCT) absorption spectrum, whereas

four curves were used to ﬁt the intraligand charge-transfer (ILCT)

bands of compounds 8-11. The latter absorptions were also

analyzed by using direct ﬁts to the experimental absorption

spectra. Butyronitrile was used as the glassing medium, for which

the local ﬁeld correction f_intis estimated as 1.33.³⁰A two-state

analysis of the intramolecular charge-transfer (ICT) transitions

gives

Hyper-Rayleigh Scattering. General details of the hyper-

Rayleigh scattering (HRS) experiment have been discussed

elsewhere,^9b-dand the experimental procedure and data analysis

protocol used for the fs measurements used in this study were as

previously described.²⁷Measurements were carried out in aceto-

nitrile, with crystal violet as an external reference (ꢀ_xxx,800) 338

× 10^-30esu in methanol; local ﬁeld factors at optical frequencies

were applied to account for the change in solvent).^27aAll measure-

ments were performed by using the 800 nm fundamental of a

regenerative mode-locked Ti³⁺:sapphire laser (Spectra Physics,

model Tsunami, 100 fs pulses, 1 W, 80 MHz). Dilute solutions

(10^-4-10^-6M) were used to ensure linear dependences of I_2ω/I_ω

2

on solute concentration, precluding the need for Lambert-Beer

correction factors. Samples were ﬁltered (Millipore, 0.45 µm), and

an absence of demodulation, i.e., constant values of ꢀ versus

frequency, showed that no ﬂuorescence contributions to the HRS

signals were present at 400 nm. This situation may indicate (i) a

lack of ﬂuorescence, (ii) spectral ﬁltering out of ﬂuorescence, or

(iii) a ﬂuorescence lifetime too short for its demodulation to be

observed within the bandwidth of the instrument. All of the sample

(25) van der Sluis, P.; Spek, A. L. Acta Crystallogr., Sect. A 1990, 46,

194.

∆µ²_ab) ∆µ²₁₂+ 4µ²₁₂

(1)

(26) SHELXTL (Version 6.10); Bruker AXS Inc.: Madison, WI, 2000.

(27) (a) Olbrechts, G.; Strobbe, R.; Clays, K.; Persoons, A. ReV. Sci.

Instrum. 1998, 69, 2233–22441. (b) Olbrechts, G.; Wostyn, K.; Clays,

K.; Persoons, A. Opt. Lett. 1999, 24, 403–405. (c) Clays, K.; Wostyn,

K.; Olbrechts, G.; Persoons, A.; Watanabe, A.; Nogi, K.; Duan, X.-

M.; Okada, S.; Oikawa, H.; Nakanishi, H.; Vogel, H.; Beljonne, D.;

Bre´das, J.-L. J. Opt. Soc. Am. B 2000, 17, 256–265. (d) Franz, E.;

Harper, E. C.; Coe, B. J.; Zahradnik, P.; Clays, K.; Asselberghs, I.

Proc. SPIE-Int. Soc. Opt. Eng. 2008, 6999, 699923-1–699923-11.

where ∆µ_abis the dipole moment change between the diabatic

states, and ∆µ₁₂is the observed (adiabatic) dipole moment

(28) Heesink, G. J. T.; Ruiter, A. G. T.; van Hulst, N. F.; Bo¨lger, B. Phys.

ReV. Lett. 1993, 71, 999–1002.

9

J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1711

A R T I C L E S

Coe et al.

Figure 1. Chemical structures of the complex salts investigated, showing the axis convention used in the theoretical calculations.

change. The value of µ₁₂can be determined from the oscillator

strength f_osof the transition by

hybrid functional B3P86³²and the LanL2DZ³³basis set. It should

be noted that quantum chemistry programs use the so-called

“Standard Orientation” that for a C₂group assigns the z axis to the

C₂axis of symmetry and the molecule is placed on the yz plane

with the x axis orthogonal to the molecule (as shown in Figure 1).

The same model chemistry was used for properties calculations.

Electronic transitions were calculated by means of the TD-DFT

method and the excited state dipole moments were calculated by

using the one particle RhoCI density. The default Gaussian 03

parameters were used in every case. Molecular orbital contours were

plotted by using Molekel 4.3.³⁴

1/2

f_os

|µ₁₂| )

(2)

-5

[

]

1.08 × 10

E

max

where E_maxis the energy of the ICT maximum (in wavenumbers)

and µ₁₂is in eÅ. The latter is converted into Debye units upon

multiplying by 4.803. The degree of delocalization c_band

2

electronic coupling matrix element H_abfor the diabatic states

are given by

Results and Discussion

∆µ²₁₂

1/2

1

2

c²_b) 1 -

(3)

(4)

2

(

)

Synthetic Studies. The new complex salts 5-11 were

prepared as extended homologues of the previously studied

series 1-4 (Figure 1),⁸with the primary aims of this study being

both to increase the ꢀ responses and to probe the effects of

ligand extension on the relative contributions of the different

tensor components. We are not aware of any previous studies

that systematically address such issues in 2D dipolar NLO

chromophores. Salts 5, 6, 8, 10, and 11 contain known pyridyl-

pyd ligands, while those in 7 and 9 are new. The compound

[2-Pymbpe⁺]PF₆was synthesized via a nucleophilic aromatic

substitution reaction between E-1,2-bis(4-pyridyl)ethylene (bpe)

and 2-chloropyrimidine (Scheme 1), in similar manner to the

reported preparation of [2-PymQ⁺]PF₆¹⁷that is used to form 4.

The extended cation in [Phbpb⁺]PF₆was prepared by treatment

of E,E-1,4-bis(4-pyridyl)-1,3-butadiene (bpb) with 2,4-dinitro-

chlorobenzene, followed by a Zincke reaction with aniline

[

]

∆µ₁₂+ 4µ₁₂

E_max(µ₁₂)

|H_ab| )

|

∆µ_ab

If the hyperpolarizability tensor ꢀ₀has only nonzero elements

along the ICT direction, then this quantity is given by

3∆µ₁₂(µ₁₂)²

ꢀ₀)

(5)

2

(E_max

)

A relative error of ( 20% is estimated for the ꢀ₀values derived

from the Stark data and using eq 5, while experimental errors of

(10% are estimated for µ₁₂, ∆µ₁₂, and ∆µ_ab, (15% for H_ab, and

2

(50% for c_b.

Computational Procedures. All theoretical calculations were

performed by using the Gaussian 03³¹program. The molecular

geometries were optimized assuming C₂symmetry by using the

(32) The B3P86 Functional consists of Becke’s three parameter hybrid

functional (Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652) with

the nonlocal correlation provided by the Perdew 86 expression: Perdew,

J. P. Phys. ReV. B 1986, 33, 8822–8824.

(29) (a) Hendrickx, E.; Boutton, C.; Clays, K.; Persoons, A.; van Es, S.;

Biemans, T.; Meijer, B. Chem. Phys. Lett. 1997, 270, 241–244. (b)

Boutton, C.; Clays, K.; Persoons, A.; Wada, T.; Sasabe, H. Chem.

Phys. Lett. 1998, 286, 101–106.

(30) (a) Shin, Y. K.; Brunschwig, B. S.; Creutz, C.; Sutin, N. J. Phys. Chem.

1996, 100, 8157–8169. (b) Coe, B. J.; Harris, J. A.; Brunschwig, B. S.

J. Phys. Chem. A 2002, 106, 897–905.

(31) Frisch, M. J. et al. Gaussian 03, ReVision B.05; Gaussian, Inc.:

Pittsburgh PA, 2003.

(33) D95 on ﬁrst row: Dunning, T. H.; Hay, P. J. In Modern Theoretical

Chemistry; Schaefer, H. F., III, Ed.; Plenum: New York, 1976; Vol.

3, p 1. Los Alamos ECP plus DZ on Na-Bi: (a) Hay, P. J.; Wadt,

W. R. J. Chem. Phys. 1985, 82, 270–283. (b) Wadt, W. R.; Hay, P. J.

J. Chem. Phys. 1985, 82, 284–298. (c) Hay, P. J.; Wadt, W. R.

J. Chem. Phys. 1985, 82, 299–310.

(34) Portmann, S.; Lu¨thi, H. P. Chimia 2000, 54, 766–770.

9

1712 J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010

V-Shaped Ruthenium(II)-Based Chromophores

A R T I C L E S

Scheme 1. Synthesis of New Pro-Ligand Salts [2-Pymbpe⁺]PF₆(a) and [Phbpb⁺]PF₆(b)

1

Table 2. Selected H NMR Data for Complex Salts

(Scheme 1), according to the method we have used to prepare

the PhQ⁺ligand present in 2.³⁵We have synthesized the salt

[Mebpb⁺]I previously by using a Knoevenagel-type condensa-

tion reaction between E-3-(4-pyridyl)propenal and N-methylpi-

colinium iodide, but the isolated yield was only 12%.¹⁸The

present approach involving methylation of bpb with methyl

iodide gives a much better yield, and both methods require the

preparation of two precursor compounds. In similar fashion, the

salts [Mebpvb⁺]I and [Mebpvb⁺]PF₆were synthesized previ-

ously by using the same reagents as reported herein,¹⁸but

[Mebpvb⁺]I was isolated in only 22% yield from a 24 h reaction

at room temperature. Here we report that the yield of this salt

is improved over 3-fold simply by heating under reﬂux for a

shorter time.

a

cis-[Ru^II(NH₃)₄(L^A)₂][PF₆]₄

salt (L^A)

pyridyl

R

NH₃

1 (MeQ⁺)^b

2 (PhQ⁺)^b

9.12 8.88

9.42 8.95

8.66 7.95 4.61

3.39 2.99

8.84 8.05 8.01-7.96, 3.45 3.04

7.85-7.79

3 (4-AcPhQ⁺)^b9.47 8.95

8.87 8.05 8.36, 8.15

8.11 9.27, 8.04

8.35 7.61 4.54

3.46 3.05

3.53 3.09

3.31 2.92

4 (2-PymQ⁺)^b10.26 9.01-8.92

5 (Mebpe⁺)

6 (Phbpe⁺)

9.00 8.65

9.30 8.69

8.53 7.67 7.97-7.95, 3.35 2.95

7.80-7.79

7 (2-Pymbpe⁺) 10.14 8.73

8.59 7.70 9.24, 7.99

8.23 7.48 4.50

3.39 2.98

3.23 2.86

8 (Mebpb⁺)

9 (Phbpb⁺)

8.90 8.54

9.20 8.57

8.41 7.51 7.79-7.78, 3.26 2.88

8.03-7.96

10 (Mebph⁺)

11 (Mebpvb⁺)

8.86 8.48

8.93 8.52

8.19 7.42 4.48

8.32 7.54 4.52

3.20 2.83

3.18 2.84

The new complex salts 5, 6, 8, and 10 were prepared

by following the procedure used previously for cis-

[Ru^II(NH₃)₄(L^A)₂][PF₆]₄[L^A) N-methyl-4,4′-bipyridinium

(MeQ⁺) 1, N-phenyl-4,4′-bipyridinium (PhQ⁺) 2, N-(4-acetylphe-

nyl)-4,4′-bipyridinium (4-AcPhQ⁺) 3, or N-(2-pyrimidyl)-4,4′-

bipyridinium (2-PymQ⁺) 4],⁸involving reactions of cis-

[Ru^II(NH₃)₄(H₂O)₂]²⁺generated in situ by reduction of cis-

[Ru^IIICl₂(NH₃)₄]⁺in aqueous solutions. However, column

chromatography on Sephadex was required to give pure

products, and the yields are in all cases relatively low (ca.

10-25%). In contrast, 1-4 were generally isolated in higher

yields of up to 50% with no need for chromatographic

puriﬁcation; the primary difference in the present work is that

smaller relative amounts of the pro-ligands were used, ca. 2

equiv as compared with ca. 4 equiv for 1-4. This change

obviously lowers the efﬁciency of coordination but was neces-

sary because all of the extended pro-ligand salts show decreased

solubilities when compared with their 4,4′-bipyridinium coun-

terparts, making the removal of unreacted excess pro-ligand

more challenging. Complex salts 6, 7, 9, and 11 were

synthesized by using the preisolated precursor salt cis-

[Ru^II(NH₃)₄(H₂O)₂][PF₆]₂,¹²which allows reactions in nonaque-

ous solvents such as acetone or DMF. However, the isolated

yields after column chromatography are not signiﬁcantly higher

than those obtained when using the in situ method, as conﬁrmed

by two different preparations of 6.

^aRecorded at 200, 400, or 500 MHz in acetone-d₆; all values are

given in ppm with respect to TMS. ^bData taken from ref 8 (but not

previously discussed).

1

Figure 2. Aromatic region of the H NMR spectrum of the complex salt

11 recorded at 400 MHz in acetone-d₆at 293 K.

presence of two singlet peaks in the regions of 3.53-3.18 and

3.09-2.83 ppm for the NH₃ligands conﬁrms the cis coordina-

tion geometry in all cases.

For the N-Mepyd series 1, 5, 8, and 10, extending the

conjugation causes all of the peaks to shift upﬁeld steadily, with

the smallest changes for the Me signal (Table 2). The relative

magnitudes of these shifts diminish with increasing chain length,

e.g., for the lowest ﬁeld pyridyl doublet (LFPD), the sequential

shifts are 0.12, 0.10, and 0.04 ppm on moving from 1 to 10.

This upﬁeld shifting indicates that a cumulative shielding

inﬂuence is exerted by the additional vinyl units, consistent with

their electron-rich nature. Altering the structure of the conjugated

chain by replacing a vinyl group with a phenylene ring in

1

¹H NMR Spectroscopy Studies. The H NMR spectra of all

the new Ru^IIcomplex salts are well-deﬁned when recorded in

deuterated acetone. Selected data are collected in Table 2, and

a representative spectrum of salt 11 is shown in Figure 2. The

(35) Coe, B. J.; Harris, J. A.; Harrington, L. J.; Jeffery, J. C.; Rees, L. H.;

Houbrechts, S.; Persoons, A. Inorg. Chem. 1998, 37, 3391–3399.

9

J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1713

A R T I C L E S

Coe et al.

Table 3. UV-vis Absorption and Electrochemical Data for Complex Salts cis-[Ru^II(NH₃)₄(L^A)₂][PF₆]₄in Acetonitrile

E_1/2, V vs Ag-AgCl (∆E_p, mV)^b

Ru^III/IIL^Areductions

salt (L^A)

λ

_max, nm^a(ε, 10³M^-1cm^-1

)

E_max(eV)

assignment

1 (MeQ⁺)^c

570 (17.5)

502 (15.1)

262 (33.3)

606 (20.4)

528 (16.5)

284 (31.2)

620 (22.6)

536 (17.5)

288 (41.8)

644 (21.5)

558 (17.2)

286 (44.6)

579 (20.8)

523 (19.2)

310 (49.8)

614 (22.4)

546 (19.0)

328 (56.1)

641 (21.8)

336 (58.9)

552 (21.4)

354 (67.1)

567 (22.9)

367 (73.8)

560 (24.3)

392 (84.5)

507 (24.4)

382 (100.1)

2.18

2.47

4.73

2.05

2.35

4.37

2.00

2.31

4.31

1.93

2.22

4.34

2.14

2.37

4.00

2.02

2.27

3.78

1.93

3.69

2.25

3.50

2.19

3.38

2.21

3.16

2.45

3.25

d f π*

π f π*

d f π*

π f π*

d f π*

π f π*

d f π*

π f π*

d f π*

π f π*

d f π*

π f π*

d f π*

π f π*

d f π*

π f π*

d f π*

π f π*

d f π*

π f π*

d f π*

π f π*

0.79 (85)

0.79 (70)

0.80 (70)

0.82 (100)

0.70 (95)

0.70 (115)

-0.81 (110)

-1.42 (65)

-1.55 (70)

-0.66 (110)

-1.26 (70)

-1.38 (65)

-0.55 (120)

-1.09 (290)^d

2 (PhQ⁺)^c

3 (4-AcPhQ⁺)^c

4 (2-PymQ⁺)^c

5 (Mebpe⁺)

6 (Phbpe⁺)

-0.42^e

-0.76^e

-0.62^e

7 (2-Pymbpe⁺)

8 (Mebpb⁺)

0.70 (100)

0.63 (100)

0.64 (95)

0.61 (95)

0.59 (95)

-0.40^e

-0.77^e

-0.63^e

-0.79^e

-0.92^e

9 (Phbpb⁺)

10 (Mebph⁺)

11 (Mebpvb⁺)

b

^aSolutions ca. 3-8 × 10^-5M. Measured in solutions ca. 10^-3M in analyte and 0.1 M in [N(C₄H₉-n)₄]PF₆at a platinum disk working electrode

e

with a scan rate of 200 mV s^-1. Ferrocene internal reference E_1/2) 0.44 V, ∆E_p) 70-90 mV. ^cData taken from ref 8. ^dShoulders evident. E_pcfor an

irreversible reduction process.

moving from 10 to 11 causes downﬁeld shifts in almost all of

the signals, showing that a relatively smaller degree of shielding

is caused by the aryl moiety. The N-Phpyd series 2, 6, and 9

also shows upﬁeld shifts upon extending the conjugation for

most of the proton signals. For a given increase in length, the

shift of the LFPD signal is completely independent of R, being

0.12 ppm moving between 2 and 6 and between 4 and 7.

Within the series 1-4, most of the signals show downﬁeld

shifts as the electron-withdrawing strength of the pyd unit

increases in the order R ) Me < Ph < 4-AcPh < 2-Pym (Table

2). These changes are logically smallest for the NH₃signals.

Because the LFPD signal shows the greatest dependence on R,

it is reasonable to assign this as being due to the protons adjacent

to the quaternized nitrogen atom. The position of this signal

with respect to those of the other three pyridyl doublets is also

consistent with the deshielding effect of the formal positive

charge. Note also that the shifts in the other pyridyl doublets

caused by extending the conjugated chain are generally larger

than those observed for the LFPD signal, in keeping with the

fact that two of these doublets are due to the four protons located

ortho to the vinyl unit(s). For the new complexes with n ) 1 or

2, replacing a N-Me with a N-Ph substituent also results in

downﬁeld shifts for all the proton signals, and the shift for the

LFPD signal is independent of the chain length at 0.30 ppm.

Signiﬁcant further downﬁeld shifts are also observed when

replacing a N-Ph group with a N-Pym substituent on moving

from 6 to 7, and the LFPD signal shows the same very large

shift (1.14 ppm) on moving from 1 to 4 or from 5 to 7.

Electronic Spectroscopy Studies. The UV-vis absorption

spectra of the new complex salts 5-11 were recorded in

acetonitrile, and the results are presented in Table 3, together

with data reported previously for 1-4⁸for comparison purposes.

Figure 3. UV-vis absorption spectra of 1 (pink), 5 (blue), 8 (green), and

10 (red) at 293 K in acetonitrile.

Representative spectra of the N-Mepyd series 1, 5, 8, and 10

are shown in Figure 3. All of the new complexes show intense

d(Ru^II) f π*(L^A) (L^A) pyridyl ligand) MLCT bands in the

visible region, together with UV absorptions due to intraligand

π f π* excitations. As with 1-4,⁸the new compounds 5 and

6 show two strongly overlapped MLCT bands. However, these

become completely merged in all of the other chromophores

so that no inﬂection points denoting two separate peaks are

evident for 7-11. For these compounds, the quoted λ_maxand

E_maxvalues should be treated with caution because the observed

bands are extremely broad (Figure 3).

As noted previously,⁸for each of 1-4, the lower energy

absorption band (MLCT-1) is more intense than the higher

energy one (MLCT-2), and the energy gap between these bands

is constant at ca. 0.3 eV. For the new complex salts 5 and 6,

the same pattern of intensity differences is observed, but the

9

1714 J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010

V-Shaped Ruthenium(II)-Based Chromophores

A R T I C L E S

estimated band maxima are closer together and separated by

ca. 0.25 eV. The merging of the MLCT bands for 7-11 means

that it is impossible to speculate further without recourse to

spectral deconvolution. The series 1-4 shows a familiar trend

of decreasing E_maxvalues for both MLCT bands as R changes

in the order Me > Ph > 4-AcPh > 2-Pym, due to the steadily

increasing acceptor strength of the pyd groups.^17,36The new

extended species 5-7 appear to show a similar pattern; although

7 shows only one broad band, its estimated E_maxvalue is lower

than both of those for 6. This general trend is also observed for

the apparently single bands for the n ) 2 chromophores in 8

and 9.

Previous studies with the pseudolinear complexes trans-

[Ru^II(NH₃)₄(L^D)(L^A)]³⁺(L^D) NH₃, pyridine or N-methylimi-

dazole; L^A) MeQ⁺, Mebpe⁺, Mebpb⁺, or Mebph⁺) revealed

an unexpected continual blue-shifting of the MLCT bands

beyond n ) 1.¹⁸In contrast, D-A polyene chromophores

normally display progressive red-shifts in their ICT bands with

the sequential addition of E-vinyl units. TD-DFT calculations

helped to rationalize this trend in the 1D Ru^IIammine complexes

as arising from an increase of ILCT character into the low

energy, nominally MLCT transition upon extension of the

π-conjugated system.¹⁸Unfortunately, the presence of two

strongly overlapped MLCT bands in the series 1, 5, 8, and 10

and 2, 6, and 9 means that it is difﬁcult to draw ﬁrm conclusions

regarding the effects of increasing n. However, it does appear

that moving from n ) 0 to 1 causes small red-shifts in both

MLCT bands, but then further extension leads to blue-shifting

of at least MLCT-1 which then merges into MLCT-2 (Figure

3). Replacing an E-vinyl with a 1,4-phenylene group in moving

from 10 to 11 produces a substantial blue shift for the combined

MLCT band of ca. 0.25 eV, and some overlap with the ILCT

band occurs in 11. Corresponding blue shifts of ca. 0.1-0.2

eV have been observed previously for the single MLCT bands

of pseudolinear Mebph⁺/Mebpvb⁺Ru^IIor Fe^IIcomplex

pairs,^18,37,38and these observations are consistent with results

obtained with related purely organic compounds.³⁹The ILCT

absorptions show the expected, progressive shifts to lower

energies accompanied by enhanced intensities as n increases,

as illustrated for the Mepyd series 1, 5, 8, and 10 in Figure 3.

These incremental red shifts decrease with each additional vinyl

unit, and the total difference in the ILCT E_maxbetween 1 and

10 is ca. 1.6 eV. The ILCT band of 11 is slightly blue-shifted

and considerably more intense when compared with that of 10.

Electrochemical Studies. The complex salts 5-11 were

studied by cyclic voltammetry in acetonitrile, and the results

are presented in Table 3, together with data reported previously

for 1-4⁸for comparison purposes. All of the new complexes

show reversible or quasi-reversible Ru^III/IIoxidation waves,

together with irreversible ligand-based reduction processes. In

contrast, 1-3 display two or more reversible or quasi-reversible

reduction waves, showing that the introduction of 4-vinyl

Figure 4. Representation of the molecular structure of the complex cation

in the salt 1·MeCN·H₂O, with the H atoms and solvents removed for clarity

(50% probability ellipsoids).

substituents on the coordinated pyridyl rings causes the reduced

radical species to become more chemically reactive. The

presence of reversible Ru^III/IIwaves shows that the new

complexes have potential to behave as redox-switchable NLO

chromophores.^4a,n

In the previously reported series 1-4, the Ru^III/IIE_1/2value

increases very slightly with the electron-accepting ability of R,

but no corresponding trend is observed in the new series 5-7,

indicating that the Ru-based HOMO is not signiﬁcantly coupled

to the pyd unit in these extended species. The sequential addition

of electron-rich vinyl units to L^Adestabilizes the HOMO and

makes the Ru^IIcenter progressively easier to oxidize; the

III/II

Ru

E

value decreases in increments of 90, 70, and 20 mV

1/2

on proceeding along the series 1, 5, 8, and 10. A similar trend

is observed for the N-aryl complexes, with a total difference of

-150 mV between 2 and 9 and -120 mV on moving from 4

III/II

to 7. The Ru

E

values for 10 and 11 are almost identical,

1/2

showing that the corresponding large blue-shift in the MLCT

band (see above) is mainly attributable to destabilization of the

L^A-based LUMO caused by replacing an E-vinyl with a 1,4-

phenylene unit. This change is a result of less effective π-orbital

overlap between the central aryl ring and the attached vinyl

groups and is evidenced by a substantial decrease in the potential

for (irreversible) L^A-based reduction on moving from 10 to 11.

Similar effects are also observed in the related 1D Ru^IIammine

complexes.¹⁸

Because 1-3 show several reversible reduction waves,

comparison of their ligand-based E_1/2values with the E_pcdata

for the new compounds is of limited validity. However, it is

worth noting that the latter values show little dependence on

the length of the π-conjugated system with a given A group.

The potentials for ligand-based reduction in 1-4 increase

with the electron-acceptor strength of the pyd unit, and a similar

behavior is observed for the E_pcvalues of the new series 5-7

and of the pair 8 and 9. Such shifts are consistent with

stabilization of the L^A-based LUMO upon increasing the

electron-withdrawing ability of R, an effect which is also

reﬂected in the red-shifting of the MLCT bands (see above).

(36) Coe, B. J.; Jones, L. A.; Harris, J. A.; Sanderson, E. E.; Brunschwig,

B. S.; Asselberghs, I.; Clays, K.; Persoons, A. Dalton Trans. 2003,

2335–2341.

(37) Coe, B. J.; Harries, J. L.; Harris, J. A.; Brunschwig, B. S.; Horton,

P. N.; Hursthouse, M. B. Inorg. Chem. 2006, 45, 11019–11029.

(38) Coe, B. J.; Foxon, S. P.; Harper, E. C.; Raftery, J.; Shaw, R.; Swanson,

C. A.; Asselberghs, I.; Clays, K.; Brunschwig, B. S.; Fitch, A. G.

Inorg. Chem. 2009, 48, 1370–1379.

Crystallographic Studies. Single crystal X-ray structures have

been obtained for the complex salts cis-[Ru^II(NH₃)₄(MeQ⁺)₂]-

(PF₆)₄·MeCN·H₂O (1·MeCN·H₂O), cis-[Ru^II(NH₃)₄(PhQ⁺)₂]-

(PF₆)₄(2), and cis-[Ru^II(NH₃)₄(4-AcPhQ⁺)₂](BPh₄)₄(3B) and also

for the pro-ligand salts [MeQ⁺]I, [Mebpe⁺]I·0.5H₂O and

[Phbpe⁺]Cl·HCl·H₂O. Representations of the molecular structures

are shown in Figures 4-6 and S1-S3 (Supporting Information),

(39) (a) Alain, V.; Re´doglia, S.; Blanchard-Desce, M.; Lebus, S.; Lukaszuk,

K.; Wortmann, R.; Gubler, U.; Bosshard, C.; Gu¨nter, P. Chem. Phys.

1999, 245, 51–71. (b) Luo, J.-D.; Hua, J.-L.; Qin, J.-G.; Cheng, J.-Q.;

Shen, Y.-C.; Lu, Z.-H.; Wang, P.; Ye, C. Chem. Commun. 2001, 171–

172.

9

J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1715

A R T I C L E S

Coe et al.

Table 4. Selected Interatomic Distances (Å) and Angles (deg) for

the Complex Salts 1·MeCN·H₂O, 2, and 3B^a

1·MeCN·H₂O

2

3B

Ru-N(L^A)

2.048(4)

2.041(4)

2.168(4)

2.162(4)

2.138(3)

2.146(3)

2.034(16)

2.056(16)

2.163(18)

2.167(16)

2.184(17)

2.140(15)

89.3(6)

91.4(7)

179.1(6)

90.9(6)

90.7(7)

89.9(7)

90.2(6)

179.8(7)

92.0(7)

87.9(7)

90.1(7)

177.2(6)

89.6(6)

90.1(7)

87.9(7)

25.6, 33.0

38.4, 49.4

2.031(7)

2.053(7)

2.160(11)

2.191(13)

2.141(13)

2.091(12)

91.2(4)

90.9(5)

179.2(5)

90.2(5)

93.0(5)

92.6(4)

88.8(5)

178.2(5)

89.8(4)

86.3(5)

88.3(6)

175.3(5)

89.8(5)

91.2(5)

87.8(6)

24.7, 11.1

40.3, 44.9

Ru-N(L^A)

Ru-NH₃(trans-L^A)

Ru-NH₃(trans-NH₃)

N(L^A)-Ru-N(L^A)

89.75(12)

89.84(15)

178.82(17)

90.09(17)

91.42(15)

90.29(16)

91.16(17)

178.54(18)

90.07(15)

90.89(16)

88.26(16)

178.69(15)

89.01(13)

87.84(16)

91.39(15)

52.9, 25.8

N(L^A)-Ru-N(ax-NH₃)

N(L^A)-Ru-N(trans-NH₃)

N(L^A)-Ru-N(eq-NH₃)

N(L^A)-Ru-N(ax-NH₃)

N(L^A)-Ru-N(eq-NH₃)

N(L^A)-Ru-N(trans-NH₃)

N(L^A)-Ru-N(ax-NH₃)

N(ax-NH₃)-Ru-N(eq-NH₃)

N(ax-NH₃)-Ru-N(ax-NH₃)

N(eq-NH₃)-Ru-N(eq-NH₃)

N(eq-NH₃)-Ru-N(ax-NH₃)

dihedral angles 1^b

Figure 5. Representation of the molecular structure of the complex cation

in the salt 2, with the H atoms removed for clarity (50% probability

ellipsoids).

dihedral angles 2^c

^aax ) axial; eq ) equatorial. ^bBetween the planes of the two

pyridyl rings within the two ligands. ^cBetween the planes of the

quaternized pyridyl and N-aryl substituent within the two ligands.

is clearly attributable to steric interactions between the ortho

hydrogen atoms. Extensive twisting is also found within the

L^Aligands, with dihedral angles between the aryl rings of up

to ca. 53° (Table 4). In the best quality structure, 1·MeCN·H₂O,

the Ru-pyridyl bond distances (Ru1-N1 and Ru1-N21) are

similar to but ca. 0.1 Å shorter than the average of the Ru-NH₃

distances. This observation is largely attributable to π-back-

bonding to the pyridyl ligands. The Ru-NH₃bonds located trans

to the MeQ⁺ligands are longer by ca. 0.02 Å when compared

to the mutually trans ones, showing that MeQ⁺exerts a modest

structural trans effect.⁴²The previously reported structure of

cis-[Ru^II(NH₃)₄(isn)₂](ClO₄)₂shows very similar behavior.⁴⁰

Comparable differences in bond distances are also generally

observed in the structures of 2 and 3B, although the uncertainties

on these data are quite large.

For our purposes, the most notable feature of these structures

is that 1·MeCN·H₂O and 3B adopt noncentrosymmetric space

groups that might potentially be suitable for bulk quadratic NLO

effects. However, inspection of the packing arrangements (e.g.,

Figure S4 in Supporting Information) unfortunately reveals that

in both compounds the individual molecular dipoles are arranged

in an antiparallel fashion so that these materials are actually

nonpolar and therefore not capable of showing NLO activity.

It is nevertheless quite possible that anion metathesis may

produce materials in which the orientation of the complex

chromophores becomes favorable for such effects.

Figure 6. Representation of the molecular structure of the complex cation

in the salt 3B, with the H atoms and disordered solvents removed for clarity

(50% probability ellipsoids).

and selected interatomic distances and angles are presented in

Tables 4 and S1 (Supporting Information).

The structures of the three pro-ligand salts show normal

geometric parameters. The dihedral angles between the two

pyridyl rings are 24.2°, 8.8/2.8°, and 15.7° for [MeQ⁺]I,

[Mebpe⁺]I·0.5H₂O, and [Phbpe⁺]Cl·HCl·H₂O, respectively,

revealing increased planarity for the bpe-based cations. The

reported structure of the related N,N′-dimethylated compound

[Me₂bpe²⁺]I₂shows a dihedral angle of ca. 8°.^19cThe dihedral

angle between the quaternized pyridyl and attached phenyl rings

in [Phbpe⁺]Cl·HCl·H₂O is 53.4°, attributable in part to steric

interactions between the ortho hydrogen atoms of the two rings.

The structures of the complex salts 1·MeCN·H₂O, 2, and

3B conﬁrm that the pyridyl ligands are coordinated in a cis

fashion. Very few closely related structures have been published;

Richardson et al. have described cis-[Ru^II(NH₃)₄(isn)₂](ClO₄)₂

(isn

)

isonicotinamide) and cis-[Ru^III(NH₃)₄(isn)₂]-

(ClO₄)₃·H₂O,⁴⁰while Clarke et al. reported the structure of cis-

[Ru^III(NH₃)₄(im)₂]Br₃(im ) imidazole).⁴¹In our three com-

plexes, the coordinated pyridyl rings display a concerted canting,

forming angles of ca. 40-50° with respect to the yz plane. A

similar effect is observed in cis-[Ru^II(NH₃)₄(isn)₂](ClO₄)₂⁴⁰and

We have also obtained an X-ray crystal structure of the

complex salt 9, from crystals grown by diffusion of diethyl ether

vapor into a nitromethane solution. Although the result is of a

rather low quality, it does conﬁrm the molecular structure of

the complex and also reﬁnes best in the noncentrosymmetric

space group Cc, albeit with the molecular dipoles essentially

canceled as in 1 · MeCN · H₂O and 3B (see above). The data

collection, structure solution, and reﬁnement details for 9 are

(40) Richardson, D. E.; Walker, D. D.; Sutton, J. E.; Hodgson, K. O.; Taube,

H. Inorg. Chem. 1979, 18, 2216–2221.

(41) Clarke, M. J.; Bailey, V. M.; Doan, P. E.; Hiller, C. D.; LaChance-

Galang, K. J.; Daghlian, H.; Mandal, S.; Bastos, C. M.; Lang, D. Inorg.

Chem. 1996, 35, 4896–4903.

(42) Coe, B. J.; Glenwright, S. J. Coord. Chem. ReV. 2000, 203, 5–80.

9

1716 J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010

V-Shaped Ruthenium(II)-Based Chromophores

A R T I C L E S

Table 5. HRS Data and Depolarization Ratios for Complex Salts 1-11

〈ꢀ²

(10^-30esu)^a

〉

ꢀ

HRS

b

c

F

d

salt (L^A)

ꢀ₈₀₀(10^-30esu)

k

ꢀ_zzz(10^-30esu)

ꢀ_zyy(10^-30esu)

1 (MeQ⁺)^e

58 ( 9

53 ( 8

139 ( 21

127 ( 19

142 ( 21

143 ( 21

292 ( 16

221 ( 22

387 ( 22

444 ( 78

904 ( 65

395 ( 22

1063 ( 112

2.32 ( 0.05

2.07 ( 0.06

2.09 ( 0.05

2.11 ( 0.06

2.24 ( 0.05

2.16 ( 0.05

2.07 ( 0.08

1.94 ( 0.06

2.30 ( 0.04

2.14 ( 0.05

2.37 ( 0.05

-0.36 ( 0.01

-0.43 ( 0.01

-0.42 ( 0.01

-0.38 ( 0.01

-0.41 ( 0.01

-0.43 ( 0.02

-0.48 ( 0.01

-0.37 ( 0.01

-0.41 ( 0.01

-0.35 ( 0.01

142 ( 22

126 ( 20

141 ( 22

143 ( 22

297 ( 18

222 (23

385 ( 26

431 ( 77

923 ( 68

397 ( 24

1092 ( 117

-51 ( 8

-54 ( 8

2 (PhQ⁺)^e

3 (4-AcPhQ⁺)^e

4 (2-PymQ⁺)^e

5 (Mebpe⁺)

6 (Phbpe⁺)

59 ( 9

-61 ( 9

59 ( 9

121 ( 7

92 ( 9

160 (9

184 ( 32

374 ( 27

164 ( 9

440 ( 46

-60 ( 9

-113 ( 7

-91 ( 9

-166 ( 11

-207 ( 37

-341 ( 25

-163 ( 10

-382 ( 41

7 (2-Pymbpe⁺)

8 (Mebpb⁺)

9 (Phbpb⁺)

10 (Mebph⁺)

11 (Mebpvb⁺)

^aThe total molecular HRS response without any assumption of symmetry or contributing tensor elements, measured in acetonitrile by using an 800

nm Ti³⁺:sapphire laser. The quoted cgs units (esu) can be converted into SI units (C³m³J^-2) by dividing by a factor of 2.693 × 10²⁰. ^bFirst

hyperpolarizability derived by assuming a single major tensor component. ^cDepolarization ratio. ^dHyperpolarizability tensor components derived from

the total molecular HRS response and depolarization ratio measurements by using eqs 7-9. ^eData taken from ref 8.

largely the same as those for 3B, and further details are included

in the Supporting Information (Figure S5 and Table S2).

Hyper-Rayleigh Scattering Studies. The ꢀ values of 5-11

were measured in acetonitrile solutions by using the HRS

technique^9,27with an 800 nm Ti³⁺:sapphire laser, and the results

are collected in Table 5, together with the previously published

data for 1-4.⁸This laser was chosen because most of the

complexes do not absorb strongly at either 800 nm or at the

second harmonic (SH) of 400 nm, their MLCT bands falling

conveniently in between these wavelengths (Figure 3). However,

the extended chromophores in 10 and 11 do show intense ILCT

bands near to 400 nm (Table 3), and the results obtained for

these salts are therefore strongly enhanced by resonance. Other

available laser fundamentals, e.g., 1064 or 1300 nm, would be

unsuitable for most of the samples due to very strong absorption

at their SH wavelengths. The ꢀ₈₀₀data shown are based on the

assumption of a single ꢀ component, ꢀ_zzz, and are derived from

the total HRS intensity 〈ꢀ²_HRS〉 by using eq 6.

with a more strongly accepting N-Phpyd unit is observed only

in the case of 8 and 9, between which the ca. 2-fold increase is

similar to that observed for the ꢀ₀values derived from 1064

nm HRS for the 1D Ru^IIammine complexes of MeQ⁺/PhQ⁺

and Mebpe⁺/Phbpe⁺.^17,35,36The complex of the new 2-Pymbpe⁺

ligand (in 7) has a ꢀ₈₀₀value somewhat larger than those of its

Mebpe⁺or Phbpe⁺counterparts, consistent with the greater

electron accepting ability of the pyd units. The Mebpvb⁺complex

in 11 has the largest ꢀ₈₀₀value among the compounds studied, but

this response is inevitably substantially resonance-enhanced due

to the intense ILCT band at λ_max) 384 nm (Table 3).

Although the ꢀ₈₀₀data shown in Table 5 are based on the

assumption of a single ꢀ component, ꢀ_zzz, the electronic

structures and hence hyperpolarizabilities of these complexes

actually display substantial 2D character. For a chromophore

with C_2Vsymmetry, there are ﬁve nonzero components of the ꢀ

tensor, ꢀ_zzz, ꢀ_zyy, ꢀ_zxx, ꢀ_yyz, and ꢀ_xxz. Assuming Kleinman

symmetry, ꢀ_zyy) ꢀ_yyzand ꢀ_zxx) ꢀ_xxz. Furthermore, if we assume

an essentially 2D structure, then ꢀ_zxx) ꢀ_xxz) 0, so only the

components ꢀ_zzzand ꢀ_zyyare signiﬁcant. In order to obtain further

information about the importance of “off-diagonal” tensor

components, we have measured HRS depolarization ratios F

for 5-11, and these are included in Table 5, together with the

corresponding previously published data for 1-4. The quantity

F is the ratio of the intensities of the scattered SH light polarized

parallel and perpendicular to the polarization direction of the

fundamental beam.²⁸A F value of 5 is the upper limit for purely

dipolar symmetry, corresponding with a single dipolar ꢀ tensor

component or a C_2Vsymmetric system with ꢀ_zzz) ꢀ_zyy, and in

the limit of ideal experimental conditions, and F ) 1.5 is the

lower limit for complete octupolar symmetry. The value of 3.4

determined for the reference compound Disperse Red 1 indicates

a single major tensor component and an appropriate experi-

mental setup. The low F values of ca. 1.9-2.4 obtained for 1-11

conﬁrm that the hyperpolarizabilities of these chromophores are

substantially 2D in nature.

1

7

1

35

〈ꢀ²_HRS〉 )

+

ꢀ²₈₀₀

(6)

(

)

Our previously reported ꢀ₈₀₀values obtained for the 4,4′-

bipyridinium series 1-4 are surprising in that they do not show

any signiﬁcant increase in the NLO response as the electron

acceptor strength of the pyd unit increases.⁸This observation

contrasts with the results of earlier 1064 nm HRS studies with

related 1D dipolar Ru^IIammine complex chromophores of the

same N-R-4,4′-bipyridinium ligands, for which substantial

increases in ꢀ₀(the estimated static ﬁrst hyperpolarizability) are

observed on changing R from Me to an aryl substituent.^17,35,36

It is therefore noteworthy that the HRS measurements with the

new compounds 5-11 do show some substantial variations in

ꢀ as the ligand structure changes.

Within the N-Mepyd series 1, 5, 8, and 10, ꢀ₈₀₀varies in a

manner that mirrors the trend in ꢀ₀observed with the corre-

sponding 1D complexes, i.e., steady and substantial increases

occur as n changes from 0 to 2, but then a small decrease is

observed for the n ) 3 (Mebph⁺) complex. This unusual

behavior has been rationalized with the aid of TD-DFT

calculations¹⁸and arises from the increasing ILCT contribution

to the nominally MLCT transition. A similar relationship

between ꢀ₈₀₀and n is evident with the N-Ph series 2, 6, and 9

and also for the pair of N-(2-Pym) species 4 and 7. However,

the previously noted increase in ꢀ when replacing a N-Mepyd

The values of ꢀ_zzzand ꢀ_zyycan be determined from 〈ꢀ_H²_RS

and F as follows:

〉

〈ꢀ²_HRS〉 ) 〈ꢀ_Z²_ZZ〉 + 〈ꢀ_Y²_ZZ

〉

〈ꢀ²_ZZZ

〈ꢀ²_YZZ

〉

(7)

F )

{

〉

The HRS intensities with parallel polarization for funda-

mental and SH wavelengths, 〈ꢀ²_ZZZ〉, and for perpendicular

9

J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1717

A R T I C L E S

Coe et al.

Table 6. Electronic Absorption and Stark Spectroscopic Data for Complex Salts 1-11^a

b

c

d

e

f

g

h

i

k

l

salt (L^A)

λ_max(nm) E_max(eV) f_os

µ₁₂(D) ∆µ₁₂(D) ∆µ_ab(D) r₁₂(Å) r_ab(Å) c_b²

H_ab(10³cm^-1

)

ꢀ₀^j(10^-30esu) Σ[ꢀ₀]_MLCT(10^-30esu) Σ[ꢀ₀]_ICT(10^-30esu)

1 (MeQ⁺)^m

628

518

632

512

684

558

731

589

719

667

577

763

704

599

822

748

658

718

662

578

369

755

705

602

385

717

667

583

407

652

608

545

405

1.98

2.39

1.96

2.42

1.81

2.22

1.97

2.10

1.72

1.86

2.15

1.62

1.76

2.07

1.51

1.66

1.88

1.73

1.87

2.15

3.36

1.64

1.76

2.06

3.22

1.73

1.86

2.13

3.05

1.90

2.04

2.28

3.06

0.34

0.24

0.41

0.20

0.40

0.26

0.40

0.27

0.04

0.15

0.31

0.05

0.16

0.34

0.07

0.31

0.06

0.13

0.31

0.95

0.03

0.19

0.30

1.02

0.04

0.17

0.27

6.7

5.2

7.4

4.6

7.6

5.6

7.9

5.9

2.4

4.5

6.0

2.7

4.9

6.6

3.6

3.4

6.6

3.0

4.3

6.2

8.7

2.3

5.3

6.2

9.2

2.6

4.8

5.8

10.1

11.3

13.8

9.9

16.8

15.3

20.3

13.5

19.5

17.1

19.7

16.9

17.8

16.5

21.0

19.5

17.8

22.2

19.9

17.2

24.8

20.4

20.0

24.3

20.7

22.0

23.0

25.4

21.9

25.2

24.7

25.7

24.1

25.7

26.2

28.8

25.0

2.1

2.4

2.9

2.1

2.5

2.7

2.4

2.6

3.6

2.9

3.5

3.9

3.1

3.7

3.9

3.3

4.4

4.1

3.8

4.3

2.3

4.5

4.3

4.6

2.5

5.1

4.7

4.8

2.3

5.2

5.6

2.6

3.5 0.20

3.2 0.13

4.2 0.16

2.8 0.13

4.1 0.19

3.6 0.12

4.1 0.20

3.5 0.14

3.7 0.02

3.4 0.09

4.4 0.10

4.1 0.02

3.7 0.08

4.6 0.09

4.1 0.02

3.6 0.08

5.2 0.10

4.2 0.02

4.2 0.05

5.0 0.07

4.3 0.77

4.6 0.01

4.8 0.06

5.3 0.06

4.5 0.77

5.2 0.01

5.1 0.04

5.3 0.05

5.0 0.73

5.4 0.01

5.4 0.02

6.0 0.03

5.2 0.76

6.4

6.5

5.8

6.7

5.7

5.8

5.5

5.9

1.9

4.2

5.2

1.9

4.0

5.0

2.2

2.7

4.0

2.0

3.3

4.4

11.4

1.3

3.3

4.0

10.9

1.4

2.9

3.8

11.0

1.6

2.5

3.3

10.6

137

62

231

42

248

95

298

110

37

94

155

62

134

213

123

80

302

66

199

2 (PhQ⁺)^m

273

3 (4-AcPhQ⁺)^m

4 (2-PymQ⁺)^m

5 (Mebpe⁺)

12.1

13.0

11.7

12.3

17.2

13.8

17.0

18.7

14.9

17.9

18.5

15.8

21.0

19.5

18.0

20.9

11.3

21.5

20.5

22.2

11.9

24.7

22.7

23.0

11.0

25.1

24.9

26.9

12.7

343

408

286 (131)

6 (Phbpe⁺)

409 (196)

505 (203)

381 (179)

7 (2-Pymbpe⁺)

8 (Mebpb⁺)

469

611

589

524

113

202

88

9 (Phbpb⁺)

48

499 (266)

431 (238)

340 (174)

218

233

112

61

177

193

158

62

10 (Mebph⁺)

11 (Mebpvb⁺)

1.33 10.7

0.06

0.13

0.24

2.8

4.0

5.2

112

166

184

1.34 10.8

^aMeasured in butyronitrile glasses at 77 K. ^bFor the two or three ﬁtted Gaussian components of the MLCT bands; f_osvalues were obtained from

(4.60 × 10^-9M cm²)ε_max× fw_1/2where ε_maxis the maximal molar extinction coefﬁcient and fw_1/2is the full width at half height (in wavenumbers). The

experimentally observed data are given for the ILCT bands of 8-11; f_osvalues were obtained from (4.32 × 10^-9M cm²)A where A is the numerically

integrated area under the absorption peak. ^cCalculated from eq 2. ^dCalculated from f_int∆µ₁₂using f_int) 1.33. ^eCalculated from eq 1. ^fDelocalized

electron-transfer distance calculated from ∆µ₁₂/e. ^gEffective (localized) electron-transfer distance calculated from ∆µ_ab/e. ^hCalculated from eq 3.

ⁱCalculated from eq 4. ^jCalculated from eq 5. ^kThe sum of the individual ꢀ₀components for the MLCT bands; the values in parentheses derive from

only the two lowest energy MLCT transitions for 5-11 (taken as corresponding with ꢀ_zyy). ^lThe sum of the individual ꢀ₀components for both types of

ICT bands (i.e., MLCT and ILCT). ^mData taken in part from ref 8.

polarization, 〈ꢀ²_YZZ〉, are given in terms of the molecular tensor

components ꢀ_zzzand ꢀ_zyyaccording to

Malachite Green have concluded that F measurements can give

a misleading indication of the NLO character of such chro-

mophores, due to Kleinman symmetry breaking.^5k,43This effect

becomes important when using a SH wavelength that lies on

the high energy side of the ICT bands, and the importance of

considering Kleinman-disallowed contributions has also been

noted in studies with 2D Ni^IISchiff base complexes^7fand a

Zn^IIphthalocyanine derivative.⁷ⁱBecause it is likely that a

comparable situation may pertain to 1-11, the HRS-derived ꢀ

tensor components (Table 5) should therefore be treated with

caution.

1

〈ꢀ²_ZZZ〉 ) ꢀ²_zzz

7

6

35

9

35

+

ꢀ_zzzꢀ_zyy

+

ꢀ²_zyy

(8)

1

2

11

105

〈ꢀ²_YZZ〉 ) ꢀ²_zzz

-

ꢀ_zzzꢀ_zyy

+

ꢀ²_zyy

{

35

105

and F can be expressed in terms of the parameter k ) ꢀ_zyy

ꢀ_zzzby

/

15 + 18k + 27k²

F )

(9)

3 - 2k + 11k²

Stark Spectroscopic Studies. Complex salts 5-11 have been

studied by Stark spectroscopy¹⁰in butyronitrile glasses at 77

K, and the results are presented in Table 6. As for the previously

reported 1-4, Gaussian-ﬁtting of the MLCT absorption spectra

was necessary to successfully model the Stark data, using two

or three curves in each case. For 1-4, two of these curves

contributed most of the Stark signal, so the data yielded by the

insigniﬁcant curves have been neglected.⁸For 5-11, data are

included for all three curves; the lowest energy one is always

of relatively low intensity but nevertheless signiﬁcant. The ILCT

bands for 8-11 were analyzed both by direct ﬁtting to the

experimental absorption proﬁles and by using Gaussian-ﬁtting

with four curves. The ILCT absorptions of 1-7 could not be

analyzed because of the operational cutoff of our Stark

Application of eqs 7-9 to our experimental data affords the

ꢀ_zzzand ꢀ_zyyvalues shown in Table 5. For the previously

described series 1-4, the derived ꢀ_zzzvalues are 2-3 times

larger than the corresponding ꢀ_zyyvalues, and the same applies

to all of the new salts 5-11. No consistent trend is observed

on changing n; the N-Phpyd series 2, 6, and 9 appears to show

an increasing dominance of the ꢀ_zzzcomponent, but an almost

completely opposite trend is shown by the N-Mepyd series 1,

5, 8, and 10 and the N-(2-Pym)pyd pair 4 and 7.

It is worth noting that previous polarized HRS studies

involving the C_2Vtriarylmethane dyes Brilliant Green and

(43) Kaatz, P.; Shelton, D. P. J. Chem. Phys. 1996, 105, 3918–3929.

9

1718 J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010

V-Shaped Ruthenium(II)-Based Chromophores

A R T I C L E S

Figure 7. Stark spectra and calculated ﬁts for the MLCT bands of 5, 6, and 11 in an external electric ﬁeld of 3.67 × 10⁷V m^-1. (Top panel) Absorption

spectrum illustrating Gaussian curves used in data ﬁtting. (Bottom panel) Electroabsorption spectrum, experimental (blue) and ﬁts (green) according to the

Liptay equation.^10a

Table 7. Calculated Electronic Transitions (TD-B3P86/LanL2DZ) and Finite Field Hyperpolarizabilities for the Complex Cations in Salts

1-12

c

d

]

cation structure

E_max(eV)

∆µ₁₂(D)

µ₁₂(D)

f_os

symmetry

major contributions

ꢀ_zyy(10^-30esu)

ꢀ_zzz(10^-30esu)

Σ[ꢀ_FF

(10^-30esu)

1

2.62

3.00

2.33

2.42

2.13

2.17

2.62

2.99

2.38

2.52

2.87

2.27

2.37

2.42

2.39

2.52

2.86

2.39

2.56

2.85

2.32

2.48

2.50

2.31

2.51

2.77

2.88

2.43

2.63

2.79

13.6

14.1

9.8

6.4^a

3.9^b

9.0^a

4.1^b

7.5^a

3.1^b

7.3^a

4.6^b

4.4^a

7.3^a

5.1^b

10.3^a

4.7^b

3.3^b

6.0^a

7.9^a

5.9^b

7.6^a

7.9^a

6.6^b

12.1^a

5.8^a

6.8^b

10.7^a

10.9^a

9.6^b

6.7^b

13.7^a

8.8^a

12.3^b

0.40

0.17

0.71

0.15

0.45

0.08

0.52

0.24

0.17

0.50

0.28

0.91

0.20

0.10

0.33

0.60

0.37

0.53

0.61

0.47

1.28

0.32

0.43

0.99

1.12

0.96

0.49

1.74

0.78

1.60

B

A

B

A

B

A

B

A

B

A

B

A

B

A

B

A

B

A

B

A

B

A

HOMO-1 f LUMO+1

HOMO f LUMO+1

HOMO-2 f LUMO

HOMO-3 f LUMO

HOMO-2 f LUMO

HOMO-3 f LUMO

HOMO-3 f LUMO+1

HOMO-2 f LUMO+1

HOMO f LUMO

HOMO-1 f LUMO+1

HOMO f LUMO+1

HOMO f LUMO

HOMO-1 f LUMO

HOMO f LUMO+1

HOMO-3 f LUMO

HOMO f LUMO+1

HOMO f LUMO

HOMO-1 f LUMO+1

HOMO f LUMO+1

HOMO f LUMO

HOMO-1 f LUMO+1

HOMO f LUMO+1

HOMO f LUMO

HOMO-1 f LUMO+1

HOMO f LUMO+1

HOMO-2 f LUMO

HOMO f LUMO

71

76

18

42

89

2

3

4

5

13.1

17.1

18.6

13.7

14.4

16.5

16.2

17.0

8.7

118

996

41

142

2

1138

43

115

29

54

144

136

92

6

82

81

15.8

2.8

7

14.7

15.9

16.7

14.5

17.0

18.5

1.0

0.1

17.5

9.1

10.7

15.0

6.8

5.3

5.2

11

32

45

27

8

131

63

163

108

9

10

131

158

81

11

67

HOMO-1 f LUMO+1

HOMO-1 f LUMO

4.1

14

^aTransition dipole moment along the y axis. ^bTransition dipole moment along the z axis. ^cObtained from FF calculations. ^dSum of the two

calculated ꢀ tensor components.

spectrometer (ca. 350 nm), but these high energy transitions

are not in any case expected to contribute substantially to the

NLO responses. Representative absorption spectra and electro-

absorption spectra in the MLCT region for the salts 5, 6, and

11 are shown in Figure 7.

pyridyl rings (moving from 1 to 5, from 2 to 6, and from 4 to

7), further extension of the conjugated systems leads to little or

no additional changes in the maxima of the ﬁtted Gaussian

curves.

When analyzing the band intensity data in cases involving

spectral deconvolution, it is appropriate to consider the total f_os

and µ₁₂values for the various components, since the superposi-

tion of these gives the overall absorption proﬁle. In contrast,

As observed previously in 1-4 and related compounds,^8,18,30b

the MLCT bands of 5-11 display large red-shifts on moving

from acetonitrile solution to butyronitrile glass (Tables 3 and

6). It is noteworthy that while the two E_maxvalues decrease

substantially on inserting an E-vinyl bridge between the two

2

for the parameters ∆µ₁₂, ∆µ_ab, r₁₂, r_ab, c_b, and H_ab, it is necessary

to compare averaged values because the ﬁtted Gaussian

9

J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1719

A R T I C L E S

Coe et al.

components represent hypothetical ICT transitions that are (at

least approximately) codirectional, not additive.

bands, tolerable ﬁts were obtained simply by ﬁtting to the

experimental ILCT absorption proﬁles and the data thus derived

are included in Table 6. Notably, rather better ﬁts are obtained

when using deconvolution with four Gaussian curves (see

Supporting Information, Table S3 and Figures S6-S9), but the

total [ꢀ₀]_ILCTvalues that arise from such treatment are very

similar to those obtained without deconvolution. The results of

these analyses are reminiscent of our previous studies with 1D

Ru^IIammine complexes of the same series of pyridyl polyenyl

Within the N-Mepyd series 1, 5, 8, 10, and 11, little variation

is observed in the total f_osor µ₁₂values, but ∆µ₁₂and ∆µ_ab

both increase in a steady and predictable fashion as the

conjugated system is extended, with the latter always being the

larger of the two dipole moment changes. The delocalized and

localized electron-transfer distances r₁₂and r_abnecessarily show

accompanying increasing trends. The magnitudes of c_b²and H_ab

also diminish with increasing molecular length, with the

averaged values for 10 and 11 being respectively about 15%

and 40% of those of 1, logically reﬂecting the smaller extent

of D-A π-electronic coupling in the extended systems. Similar

ligands.¹⁸The ILCT bands have larger f_os, c_b, and H_abvalues

2

but smaller values of ∆µ₁₂and r₁₂when compared with the

accompanying MLCT bands. These observations are consistent

with stronger π-orbital overlap occurring over shorter distances

for the ligand-based transitions. The [ꢀ₀]_ILCTvalues are sub-

stantial and somewhat larger than those of the related 1D

chromophores,¹⁸due to the presence of two ligands in 8-11.

[ꢀ₀]_ILCTincreases in the order 8 < 9 < 10 < 11; although the

differences between each pair may not be signiﬁcant, the overall

trend of increasing with extension of the π-system is clear, and

the total enhancement is ca. 2-fold on moving from 8 to 11.

Given that the general direction of the ICT excitations is the

same, it is reasonable to consider the total NLO responses as

being the sum of those associated with the MLCT and ILCT

transitions, Σ[ꢀ₀]_ICT. The derived values (Table 6) are very large

and show the expected increases on extending the π-conjugated

pathlengths (moving from 8 to 10) or increasing the electron-

accepting strength of the pyd unit (moving from 8 to 9). As for

the Σ[ꢀ₀]_MLCTvalues, the Σ[ꢀ₀]_ICTvalue for 11 is possibly a

little smaller than that of 10. The extent of contribution of the

ILCT transitions to the total ꢀ₀response increases with the

length of the π-conjugated systems, ranging from ca. 20% for

8 and 9 to as much as ca. 35% for 11.

Regarding the trend of the overall NLO response increasing

as the π-conjugated systems extend, the results of these Stark

analyses generally agree well with the HRS data for 1-10

(Table 5). Because the ꢀ₈₀₀value of 11 is strongly resonance-

enhanced (see above), the Stark approach is likely to be more

reliable in this instance. The only minor discrepancy is that the

HRS results show the Mebph⁺-containing complex in 10 to have

a smaller ꢀ₈₀₀value when compared with its Mebpb⁺analogue

in 8, while the Stark-derived Σ[ꢀ₀]_ICTdata indicate the opposite.

However, for both types of measurement, the differences

between the total ꢀ values determined for these two compounds

are probably not signiﬁcant. The effects of increasing the

electron accepting strength of the pyd unit are represented

consistently by the Stark-based data but less so by the HRS

results (especially for the pairs 1/2 and 5/6). This reasonably

good qualitative level of agreement is satisfying, especially given

that the two sets of data are derived in quite different ways.

While the Stark analysis indirectly affords estimated ꢀ responses

associated with the (ﬁtted) ICT transitions, the HRS measure-

ments are direct but subject to the variable effects of resonance.

Furthermore, the HRS and Stark experiments are carried out

under different physical conditions of medium and temperature,

which affect the molecular optical properties. Although in two

cases (5 and 8), the ꢀ₈₀₀and Σ[ꢀ₀]_ICTvalues are very similar,

such a good degree of quantitative agreement is undoubtedly

only coincidental because ꢀ₈₀₀is enhanced due to resonance.

2

dependences of ∆µ₁₂, ∆µ_ab, r₁₂, r_ab, c_b, and H_abon the

conjugation length are evident with the N-Phpyd series 2, 6,

and 9 and also for the N-(2-Pym)pyd pair 4 and 7.

As in our previous report describing 1-4,⁸application of the

two-state model (eq 5)⁴⁴to the electronic transitions provides

estimates of the corresponding ꢀ₀components (Table 6).

Theoretical analyses predict that such complex chromophores

display two dominant MLCT transitions;^8,45the lowest energy

transition is polarized along the y axis and associated with the

off-diagonal ꢀ_zyyresponse, while the other transition is polarized

along the z axis and associated with ꢀ_zzz. Further theoretical

studies with the extended chromophores 5-11 indicate that the

two lowest energy transitions are associated with ꢀ_zyywhile the

next one(s) are associated with ꢀ_zzz(see below). Therefore, it

appears reasonable to consider the sums of the Stark-derived

ꢀ₀values for the two lowest energy Gaussian components as

corresponding with ꢀ_zyy(Table 6), while the data for the highest

energy curve lead to estimated ꢀ_zzzvalues. The main point of

note for 1-4 is the observation that the ꢀ_zyyresponse dominates

over ꢀ_zzzin all cases.⁸In contrast, this situation appears to be

reversed in the new extended chromophores in 5-7, although

the large error limits on the ꢀ₀values mean that the difference

is probably only signiﬁcant for 7. These observations indicate

that the ꢀ_zzzcomponent becomes dominant on inserting E-vinyl

linkages into the ligands, and such an effect is logical because

the overall elongation is more pronounced in the z direction

when compared with the y direction. However, the data for the

series 1, 5, 8, and 10 do not show a steadily increasing

dominance of ꢀ_zzzwith elongation, rather the ratio ꢀ_zzz/ꢀ_zyy

changes from ca. 0.5 to 1.2, then to 1.1, and ﬁnally 0.8. The

corresponding N-Phpyd series 2, 6, and 9 shows the same

general trend. In 11, ꢀ_zyyapparently still dominates over ꢀ_zzz,

but the difference between these two components is probably

not signiﬁcant.

The effects of conjugation extension are also manifested in

the estimated total MLCT-associated ꢀ responses, Σ[ꢀ₀]_MLCT

.

For the N-Mepyd series 1, 5, 8, and 10, a steady increase is

observed, with a total change of ca. 120%. A similar trend is

found for the series 2, 6, and 9 and also for the pair 4 and 7. In

all cases, increasing the electron acceptor strength of the pyd

unit produces the expected enhancements in the NLO response;

for example, the Σ[ꢀ₀]_MLCTvalue for 7 is ca. 75% larger than

that of 5. For 11, the value of Σ[ꢀ₀]_MLCTis possibly a little

smaller than that of 10.

Their relatively low energies also allow Stark analysis of the

Regarding the ꢀ components for 1-4, there is a clear

inconsistency between the HRS and Stark results.⁸The HRS

data for the new extended chromophores in 5-11 show ꢀ_zzzto

be dominant in all cases, while the Stark results generally

indicate that both tensor components are of similar magnitude.

ILCT bands of complex salts 8-11. In contrast to the MLCT

(44) (a) Oudar, J. L.; Chemla, D. S. J. Chem. Phys. 1977, 66, 2664–2668.

(b) Oudar, J. L. J. Chem. Phys. 1977, 67, 446–457.

(45) Zwickel, A. M.; Creutz, C. Inorg. Chem. 1971, 10, 2395–2399.

9

1720 J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010

V-Shaped Ruthenium(II)-Based Chromophores

A R T I C L E S

Theoretical Studies. The electronic structures of the com-

plexes in 2-11 have been studied by density functional theory

(DFT) methods. The parameters involved in the two-state model

(eq 5) were determined by time-dependent DFT (TD-DFT)

calculations, and the ꢀ values were calculated from the

numerical second derivative of the dipole moment with respect

to the applied electric ﬁeld by using the ﬁnite ﬁeld (FF)

approach. The results of these theoretical calculations, together

with those reported previously for the complex in 1, are gathered

in Table 7. A recent comparative study involving several

quantum mechanical methods for predicting ꢀ₀values (including

FF-DFT) concluded that such approaches are relatively reliable

when considering essentially linear dipolar species,⁴⁶but it is

worth noting that the FF-DFT method has been applied only

rarely to 2D chromophores.^5y,8,47Although in previous studies

with related 1D Ru^IIammine chromophores we did also calculate

high energy intraligand transitions,¹⁸the present 2D species are

inherently more complex, leading to greater numbers of excited

states. In order to limit the computational costs, we have

therefore for 1-11 calculated only the transitions below 3 eV

(above 413 nm), and Table 7 includes those with f_osabove 0.1.

Even so, this work has involved the calculation of a relatively

high number of transitions.

The overall picture derived from the theoretical calculations

on the complexes in 2-11 is similar to that described previously

for the complex in 1.⁸Thus, all of the complex cations display

C₂symmetry, and therefore their molecular orbitals may be

symmetric (a) or antisymmetric (b) with respect to the 2-fold

axis (z in the standard orientation). The electronic transitions

between orbitals with the same symmetry (A transitions) are

polarized along the z axis and contribute to the ꢀ_zzzcomponent

of the hyperpolarizability tensor, while the transitions between

orbitals with different symmetries (B transitions) are polarized

perpendicular to the z axis and therefore contribute to ꢀ_zyy.

Figures 8 and 9 show the topologies of the molecular orbitals

involved in the lowest energy transitions of the MeQ⁺complex

in 1 and of the Mebph⁺complex in 10. The corresponding

depictions for 2-9 and 11 are shown in Figures S10-S18 in

Supporting Information. Compared to the chromophore in 1,

the presence of the extended ligand π-systems produces a

somewhat more complicated scheme in 10 since the highest

occupied molecular orbitals (HOMO and HOMO-1) are no

longer of purely metal d-orbital character but also include

important contributions from ligand-based π orbitals located

primarily on the electron-rich hexatrienyl chains. The extent of

ligand contributions to the HOMO and HOMO-1 increases

steadily moving along the series 1, 5, 8 (Figures S13 and S16

in Supporting Information) and 10. Therefore, as the molecular

size increases, it becomes difﬁcult to differentiate MLCT from

ILCT transitions.

Figure 8. Illustration of the 0.04 contour surface diagrams of the molecular

orbitals of 1 involved in the two lowest energy transitions.

Figure 9. Illustration of the 0.04 contour surface diagrams of the molecular

orbitals of 10 involved in the four lowest energy transitions. The axis

convention is as for Figure 8.

As also observed in previous related studies,^8,18the TD-DFT-

derived excitation energies are substantially overestimated when

compared with the corresponding experimental data (Tables 3

and 6), although the second lowest energy transition for the

complex in 3 and the lowest energy transition for the complex

in 11 provide notable exceptions to this general trend. TD-DFT

calculations do predict the experimentally observed decrease

in visible excitation energies on moving along the series Me >

Ph > 4-AcPh (1 f 3), but completely fail to predict the further

red-shift observed for the 2-Pym derivative in 4. Indeed, the

predicted E_maxvalues for 1 and 4 are essentially identical. The

same disagreement between theory and experiment applies to

the results for the n ) 1 chromophores, with indistinguishable

energies for 5 and 7. Thus, according to these calculations the

electronic inﬂuence of a 2-Pym substituent is the same as that

of a Me group in these complex chromophores. This result is

an interesting anomaly, given the well established electron-

withdrawing nature of this heterocyclic group. There are clear

differences in the calculated donor MOs of the 2-Pym deriva-

(46) Isborn, C. M.; Leclercq, A.; Vila, F. D.; Dalton, L. R.; Bre´das, J. L.;

Eichinger, B. E.; Robinson, B. H. J. Phys. Chem. A 2007, 111, 1319–

1327.

(47) Coe, B. J.; Harris, J. A.; Brunschwig, B. S.; Gar´ın, J.; Orduna, J. J. Am.

Chem. Soc. 2005, 127, 3284–3285.

9

J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1721

A R T I C L E S

Coe et al.

tions. For both N-Mepyd and N-Phpyd complexes, small

increases in E_maxare mostly predicted on moving to n ) 2, but

then small decreases occur for n ) 3 in the N-Mepyd series

(8 f 10). Because of the presence of strongly overlapped bands,

it is not possible to compare these results with the spectral data

recorded at 293 K in acetonitrile (Table 3), and the deconvoluted

spectra recorded at 77 K in butyronitrile (Table 6) show no

trends (see above). Our previous TD-DFT studies performed

on 1D {Ru^II(NH₃)₅}²⁺complexes of the same series of pyridyl

polyenyl ligands revealed a clear agreement with experimental

observations that show the MLCT bands to undergo blue-shifts

on moving from n ) 1 to 3.¹⁸This behavior was attributed to

the increased ILCT character of the lowest energy transition

of the extended systems, and this phenomenon is also observed

in the present 2D complexes (see above). The replacement of

an E-vinyl unit with a 1,4-phenylene ring on passing from 10

to 11 is predicted to cause blue-shifts in the two lowest energy

transitions, but the magnitude of these changes (ca. 0.1 eV) is

underestimated when compared with the data recorded in

acetonitrile solution or butyronitrile glass (ca. 0.2 eV).

Almost half of the predicted µ₁₂values for the complexes in

1-5 agree closely with those measured at 77 K in butyronitrile

(Table 6); for example, the two values for the lowest energy

transitions of the complexes in 3 and 5 are essentially identical.

The other µ₁₂values show larger discrepancies between experi-

ment and theory that follow no clear pattern. The predicted ∆µ₁₂

values are generally larger than those observed for the complexes

in 1-4. For the other compounds, which require three signiﬁcant

Gaussian components to ﬁt the Stark spectra, the level of

agreement with TD-DFT is worse and the predicted ∆µ₁₂values

are mostly smaller than those measured. The observed trend of

∆µ₁₂increasing with extension of the conjugated systems is only

partially reproduced by the calculations. Thus, the TD-DFT-

derived ∆µ₁₂values do generally increase on moving from n )

0 to 1, but then the expected pattern is largely lost as the more

extended chromophores mostly yield relatively small ∆µ₁₂

values. It is hence apparent that the accuracy of the TD-DFT

approach is greater with the smaller complexes. This overall

moderate level of quantitative agreement between the predicted

and experimental parameters is partly a consequence of inherent

limitations of the TD-DFT approach, but also arises from the

fact that the calculations were performed on isolated molecules,

while the experimental measurements involved solutions or solid

matrices. It is therefore of little value to use the predicted

parameters in the two-state model (eq 5) to derive ꢀ₀values, so

the latter are better modeled by using the FF approach. However,

DFT methods do provide reasonable predictions of at least some

of the experimentally observed trends and yield precise repre-

sentations of the electronic structures of the complexes studied.

In agreement with the Stark spectroscopic studies, TD-DFT

predicts that for the complex cations in 1-4 there are only two

possible transitions below 3.00 eV (Table 7). The lowest energy

transition is polarized along the y axis, displays a higher

oscillator strength, and contributes to the ꢀ_zyytensor component

that therefore dominates over ꢀ_zzz. This latter prediction is

consistent with the FF calculations but, as noted previously,⁸is

the opposite of the pattern observed in the HRS data (Table 5).

The fact that the Stark analyses and both types of theoretical

studies are in qualitative agreement for these four compounds

lends credence to our original suggestion that resonance effects

are probably responsible for the discrepancy with HRS.⁸

Figure 10. Illustration of the 0.04 contour surface diagrams of the donor

molecular orbitals of 2 (top) and 4 (bottom). The axis convention is as for

Figure 8.

tives when compared to their Ph/4-AcPh analogues. In the latter

(in 2, 3, 6 and 9), these MOs include important contributions

from the outer aryl rings, while the electron-deﬁcient 2-Pym

rings in 4 and 7 do not contribute to these MOs which are

formed almost exclusively from Ru d-orbitals (see for example

Figure 10). These results are consistent with the higher TD-

DFT ∆µ₁₂values for the complexes in 4 and 7 when compared

with 2 and 6, respectively.

Surprisingly, the prediction of some donor ability of the

phenyl rings still leads to calculated red-shifts of the MLCT

transitions for the Ph/4-AcPh derivatives when compared to their

methyl analogues. The contributions of the aromatic groups

increase the energies of the HOMO and HOMO-1, resulting

in decreased energy gaps for the MLCT transitions. The 2-Pym

groups in 4 and 7 show no contributions to any of the relevant

MOs and therefore their MLCT energy gaps are unaltered with

respect to the methyl derivatives 1 and 5.

It is noteworthy that previous TD-DFT studies with purely

organic stilbazolium chromophores with 4-(dimethylamino)phe-

nyl electron donors do more accurately model the observed

differences between these pyd acceptors, predicting a 0.2 eV

decrease in the ICT energy on replacing a N-Mepyd with a N-(2-

Pym)pyd group.⁴⁸In that case, the latter contributes to the

LUMO, resulting in a lower orbital energy and a decreased

HOMO-LUMO gap. Also in contrast with the present studies,

these earlier calculations predicted no contribution from the Ph

substituent to the HOMO in the N-Phpyd chromophore.⁴⁸

Comparison of the TD-DFT results for the pairs 1/5, 2/6, and

4/7 reveals decreases in the E_maxvalues on addition of a vinyl

bridge in each case, in keeping with the experimental observa-

When compared with the complex cations in 1-4, those in

5-11 display a more complicated electronic excitation behavior,

(48) Coe, B. J.; Beljonne, D.; Vogel, H.; Gar´ın, J.; Orduna, J. J. Phys.

Chem. A 2005, 109, 10052–10057.

9

1722 J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010

V-Shaped Ruthenium(II)-Based Chromophores

A R T I C L E S

with three possible transitions below 3.00 eV for all except 10,

which shows four (Table 7). These results are in general

consistent with the need for three Gaussian curves that all

contribute signiﬁcantly when ﬁtting the Stark spectra for these

extended chromophores (Table 6). However, the FF calculations

predict in every case a continued dominance of ꢀ_zyyover ꢀ_zzz,

and TD-DFT indicates that this is due to the higher oscillator

strengths associated with the B transitions. This latter result does

not agree with either the Stark or HRS results, but it is unclear

at present which approach can be expected to afford the most

reliable results. While HRS studies are inevitably inﬂuenced

by resonance factors, the Stark-based method becomes less

reliable as the required extent of spectral deconvolution

increases. On the other hand, the quantitative accuracy of TD-

DFT predictions is certainly limited and all other theoretical

modeling approaches have their limitations. It is noteworthy

that previous FF-DFT calculations on a diquat derivative predict

a substantially greater dominance of ꢀ_zyyover ꢀ_zzz(with a ratio

of 4.8) when compared with the results derived from the coupled

perturbed Hartree-Fock method (ꢀ_zyy/ꢀ_zzzratio ) 2.1).⁴⁷An

important aspect when considering the new extended chro-

mophores is that Stark studies conﬁrm that their ꢀ responses

are associated substantially with the high energy ILCT transi-

tions, as well as the visible MLCT ones. Given that FF-DFT

calculations predict the total hyperpolarizability of a molecule,

it is therefore quite reasonable to expect some considerable

deviation from the Stark measurements which have allowed us

to estimate the different tensor components for only the MLCT

bands (see above). In addition, it should be noted that calcula-

tions carried out at zero frequency where Kleinman symmetry

is exact may well afford results that differ somewhat from those

of HRS measurements made under conditions where Kleinman

symmetry breaking may occur (see above).

Disregarding the issues of tensor components and absolute

magnitudes, the FF-calculated overall trends in the total ꢀ

responses show only partial agreement with the results of the

Stark and HRS measurements. Within the N-Mepyd series,

Σ[ꢀ_FF] increases steadily with the conjugation length up to n )

2 (1 f 5 f 8), but then there is little change or even a small

decrease on passing to n ) 3 (in 10). This pattern is the same

as that observed in the ꢀ₈₀₀values for these compounds and

also when applying 1064 nm HRS or Stark measurements to

related 1D complexes.¹⁸For the N-Phpyd and N-(2-Pym)pyd

chromophores, Σ[ꢀ_FF] also increases on moving from n ) 0 to

1 but is decreased in 9 with respect to 6. The Σ[ꢀ_FF] value also

increases with increasing the acceptor strength for n ) 0 (1 <

2 < 3), but these calculations mirror the TD-DFT results in

failing to describe the relatively enhanced electron-withdrawing

effect of the N-(2-Pym)pyd group in 4. Furthermore, for the n

) 1 or 2 chromophores, the FF results are completely opposed

to most of the experimental observations in indicating that the

total ꢀ response decreases on replacing R ) Me with Ph or

2-Pym. The FF method gives an unrealistically small total ꢀ

response for the complex in 11 that is about half that of the

complex in 10 and even smaller than that predicted for its

counterpart in 1. These and previous¹⁸Stark studies show that

replacing Mebph⁺with Mebpvb⁺causes ꢀ₀to decrease by ca.

5-25%. The results of these calculations therefore appear to

be more reliable with the shorter chromophores, as also found

with the TD-DFT approach.

Conclusion

We have synthesized and characterized a number of new

additions to the ﬁrst family of charged 2D dipolar NLO

chromophores based on reversibly redox-switchable transition

metal centers. Three new materials with noncentrosymmetric

crystal structures have been identiﬁed, but unfortunately these

are effectively nonpolar and so not expected to display bulk

quadratic NLO effects. Both fs HRS measurements at 800 nm

and Stark spectroscopic studies show that these cis-{Ru^II-

(NH₃)₄}²⁺species possess very large NLO responses with

substantial 2D character. According to the Stark analyses,

increased total ꢀ₀responses as high as ca. 600 × 10^-30esu have

been achieved; these are unusually large and stand out especially

among 2D chromophores. We have for the ﬁrst time applied

the TD-DFT method to probe the effects of varying the electron

acceptor strength of pyd units in metal complexes. Surprisingly

and in contrast with previous related studies with purely organic

stilbazolium derivatives, the results of these calculations do not

reproduce the especially high electron-withdrawing ability of

the N-(2-Pym)pyd group. However, TD-DFT does predict an

increased contribution of ILCT character to the nominally

MLCT transitions and accompanying blue-shifts of the visible

absorption bands as the π-conjugated systems of the ligands

are extended. These results are reminiscent of our previous work

with related 1D complexes of the same pyd-substituted ligands.

In terms of ꢀ components, the HRS data for the new extended

chromophores indicate that ꢀ_zzzis always dominant, while the

Stark results generally indicate that ꢀ_zzzand ꢀ_zyyare of similar

magnitude. In contrast, FF calculations predict signiﬁcant

dominance of ꢀ_zyyover ꢀ_zzzfor all of the complexes. The

discrepancies between these various physical methods are not

overly surprising given the importance of resonance and

Kleinman symmetry breaking effects in HRS measurements,

the indirectness of the Stark approach, and the limited quantita-

tive accuracy of current theoretical methods. Nevertheless, these

results are very promising in the context of practical applications

that exploit large ꢀ_zyyresponses, so future studies will focus on

the incorporation of our new chromophores into polar crystalline,

thin ﬁlm, or other materials.

Acknowledgment. We thank the EPSRC for support (grant EP/

D070732/1) and also the Fund for Scientiﬁc Research-Flanders

(FWO-V, G.0312.08), the University of Leuven (GOA/2006/3),

MCyT-FEDER (CTQ2005-01368, CTQ2008-02942), the NSF

(grant CHE-0802907, “Powering the Planet: an NSF Center for

Chemical Innovation”) and Gobierno de Aragon-Fondo Social

Europeo (E39). We are grateful to Dr. Inge Asselberghs for

technical assistance with the analysis of the HRS data.

Supporting Information Available: Crystallographic informa-

tion in CIF format; Figures S1-S18, Tables S1-S3; Cartesian

coordinates of theoretically optimized geometries; complete ref

31. This material is available free of charge via the Internet at

http://pubs.acs.org.

JA908667P

9

J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1723

Article Doi

Evolution of linear absorption and nonlinear optical properties in v-shaped ruthenium(II)-based chromophores

DOI: 10.1021/ja908667p

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja908667p

Full text of DOI:10.1021/ja908667p

Products guided by the article

R&D Labs maybe for 1228272-41-0

Relevant to this article

Hot Product