Steric and electronic effects in metallophilic double ... - Semantic Scholar

PERSPECTIVE

www.rsc.org/dalton | Dalton Transactions

Steric and electronic effects in metallophilic double salts† Linda H. Doerrer* Received 5th October 2009, Accepted 17th December 2009 First published as an Advance Article on the web 28th January 2010 DOI: 10.1039/b920389c This Perspective highlights our efforts to assemble infinite chains of metal atoms in double salts of the form [M]+ [M]- under the complementary influences of metallophilic interactions and electrostatic attraction. Our design strategy necessarily incorporates the significant steric constraints of one-dimensional assemblies as well as the more subtle and challenging electronic tuning of weak dispersion forces.

One-dimensional metal chains Since harnessing electricity to do work during the Age of Enlightenment, methods and constructs have been sought for the evermore-efficient flow of electricity. Chemists have been intrigued by the idea of a one-atom wide conduit for decades.1 Such a wire made of metal atoms, naturally requires an insulating coat, which is provided by the ligands that fill out the metal coordination spheres perpendicular to the metal-metal vectors.2 Solution chemistry has taken a bottom-up approach to wire synthesis, often inspired by Magnus’ green salt, [Pt(NH3 )4 ][PtCl4 ].3,4 This simple but fascinating eponymous compound exhibits a green color and an infinite chain of Pt–Pt interactions in the solid state in which the cations and anions form a perfectly alternating stack.5 The related [Pt(NH2 R)4 ][PtCl4 ]6,7 and [Pt(NR3 )4 ][PtCl4 ]8 compounds have also been studied. The solid state physics of these materials continues to be an area of active research.9,10 A great deal of work has been done also with partially oxidized K2-x [Pt(CN)4 ] compounds1 and Boston University, Chemistry Department, 590 Commonwealth Ave, Boston, MA 02215, USA. E-mail: [email protected]; Fax: +1 (0)617-353-6466; Tel: +1 (0)617-358-4335 † In memoriam Professor Les Lessinger (1943–2009) Docendo discimus.

Linda H. Doerrer graduated from Cornell University in 1991 and received her PhD from MIT in 1997 under the guidance of Stephen J. Lippard. She subsequently held an NSF-NATO postdoctoral fellowship and a Junior Research Fellowship at St. John’s College, Oxford, with Malcolm L. H. Green, FRS at the University of Oxford. From 1999–2006 she served on the faculty at Barnard College, receiving the Emily Gregory award in 2002, Linda H. Doerrer and joined the faculty of Boston University in 2006. She has been recognized with a Camille and Henry Dreyfus Startup Grant, Cottrell College Science, Henry Dreyfus Teacher-Scholar Awards, and an NSF-CAREER award. This journal is © The Royal Society of Chemistry 2010

the related bis-oxalate derivatives,11,12 [Pt(ox)2 ]2-x . Both of these families of compounds form 1D chains of Pt atoms that are as long as the crystalline domains in which they are found. The stacking of two or more d 8 metal centers has also been achieved in bridged [Pt2 (P2 O5 H2 )4 ]4- complexes,13 [Rh2 (2,5-dimethyl-2,5-diisocyanohexane)4 ]2+ dimers14 and Ir2 (mpz)2 (CO)4 systems15 as well as non-bridged [Pt(diimine)X2 ]16-18 and [Rh(CO)2 (acac)],19 [Ir(CO)2 (acac)], [Ir(CO)3 Cl]20 and related systems.21 More recently various polychelating guanidinate ligand derivatives have been prepared that contain a chain fragment of less than ten first row transition metal ions, termed extended metal atom chains (EMACs).22,23 We have developed a new family of compounds somewhat like the Magnus’ green salt derivatives but with unitary charges on each component giving the general form [M]+ [M]- . This Perspective will focus primarily on work from our group but will also include other examples of so-called ‘double salts’ that contain metallophilic interactions with no bridging ligands, and form either pairwise combinations or an extended array of metallophilic contacts. This reduced scope necessarily excludes many fascinating compounds with bridging ligands for which the reader is referred to the literature with the Cambridge Structural Database as a guide.24

Metallophilicity Students beginning their studies of chemistry today are typically introduced to bonding partitioned between covalent and ionic extremes, supplemented by weaker intermolecular forces. Subsequently they will learn about metallic bonding, the “sea of electrons” present in delocalized systems and perhaps read a little about cluster bonding. Yet to be incorporated into standard textbooks is the phenomenon of metallophilicity, a name that indicates the affinity of two metal centers for one another. First described in detail based on the close Au ◊ ◊ ◊ Au contacts in several Au(I) derivatives,25 the phenomenon of two closedshell metal centers approaching closer than the sum of their van der Waals radii was termed aurophilicity,26 and now the word metallophilicity is used in a general sense for many metals. These interactions are understood to be a type of dispersion interaction between electron densities on larger and relatively reduced metal centers.27,28 The energy of such bonding is on the Dalton Trans., 2010, 39, 3543–3553 | 3543

order of hydrogen-bonding and can therefore exert a significant influence on solid-state structures.27 Scheme 1 shows an abbreviated periodic table of the elements in which blue indicates experimentally observed metallophilic interactions and red indicates only computational support. Absolutely required for metallophilic interactions is a closed-shell, or pseudoclosed shell valence electron configuration. Interactions between one or more metals with an open-shell have distinct and different bonding consequences that can include covalent metal-metal bond formation,29 as well as ferro- or antiferro-magnetic coupling.30,31 The d 10 configuration (Hg2+ , Au1+ , Ag1+ , Cu1+ , Pd0 , Pt0 ) is by far the most commonly observed, followed by d 8 (Au3+ , Pd2+ , Pt2+ , Ir1+ Rh1+ , Ru0 ), and s2 (Hg0 , Tl1+ , Pb2+ , Bi3+ ). The short-lived Rg1+ has been predicted to be a metallophilic species, but this behaviour has yet to be observed.32 Among all these metallic elements, the most examples have been observed with Au which is (not coincidentally) the most electronegative metal due to the greatest relativistic effects being experienced by its electrons.27,28,33,34 These aurophilic interactions have been regularly and comprehensively reviewed throughout their development.35-37 An example and reference for each metal are given in Table 1, which is intended to give the reader a group of references that will guide them into the riches of this bonding phenomenon and introduce the researchers involved, rather than be comprehensive.

Scheme 1 Metallophilicity in the periodic table.

Because the most common technique for identifying metallophilic interactions is analysis of metal-metal distances obtained in crystallographic studies, some references values will be helpful. Collected in Table 2 are the pairwise sums of van der Waals radii38 for the metallophilic elements for which such radii are available. The reader’s attention is called to the very long distances between two metals with ns2 configurations. Highlighted in bold Table 1 Examples of metallophilic interactions Metal

Metallophilic species

Ru Rh Ir Pd Pt Cu Ag Au Hg Tl Pb Bi

[Ru(CO)4 ]• [Rh(CO)2 (NCCH3 )2 ]+ [Ir2 (m-OPy)2 (CO)4 ] [Pd2 Tl(P2 phen)3 ]+ [Pt(NR3 )4 ][PtCl4 ] {Cu(C6 F5 )}4 -C10 H8 (Ph3 PS)4 }{[CF3 C(O)OAg]6 [AuCH2 P(S)Ph]2 [Pd(salophen)]2 [Hg(C6 F5 )2 ] [AuTl(C6 F5 )2 (OPPh3 )2 ] • (Bu4 N)2 [Pb[Pt(C6 F5 )4 ]2 ] [Au(C6 F5 )2 ][Bi(C6 H4 CH2 NMe-2 )2 ]

3544 | Dalton Trans., 2010, 39, 3543–3553

Reference 39 40 41 42 6 43 44 45 46 47 48 49

are the three combinations most relevant to this Perspective. It is emphasized that van der Waals radii are difficult to determine precisely and therefore these values are not absolute maxima for the determination of metallophilic interactions in comparisons with experimental interatomic distances. In addition to single crystal diffraction studies, the presence of metallophilic interactions can be determined by more challenging but very valuable spectroscopic studies. For example, UV-vis50 and fluorescence50-53 studies, which reveal occupation of a s-type molecular orbital (Scheme 2, vide infra), or NMR studies42,54 which indicate changes in metal valence electron density are increasingly used. Raman studies50,55 are particularly valuable for directly measuring M-M interactions and conductance measurements56 can indicate solution-phase interactions as well. The earliest examples of two diamagnetic metal centers interacting strongly with each other were observed in the stacking of ~D4h Pt(II) moieties.57 The orbital interactions between squareplanar Pt(II) d 8 species were explored and understood via the intensive studies of [Pt(CN)4 ]2- ,58 [Pt(ox)2 ]2- ,59 and [Pt2 (POP)4 ]2pyrophosphite60 systems. Whereas the overlap of two or more d z2 orbitals in Pt chains had a clear steric rationale, the bonding of two d 10 centers was unexpected. Crystallographic studies clearly revealed distances closer than the sum of the van der Waals-radii in both intra- and intermolecular cases.45 Subsequent electronic structure analyses revealed a very similar type of orbital interaction to that observed in Pt(II) d8 pairs between Pd(0),61 Cu(I),62 and Au(I) d10 ions25 . Scheme 2 illustrates a generalized homodinuclear metallophilic interaction for two different electron configurations, d 8 and d 10 . In each case a filled orbital on one metal center, labeled F in Scheme 2, overlaps with an identical orbital on an adjacent molecule, forming filled bonding (FF) and antibonding (FF*) combinations. The metal centers involved are typically those with valence orbitals of higher prinicipal quantum numbers such that empty valence orbitals (E) of the same local symmetry are low enough in energy to mix with the filled orbitals (F) and create similar, but unfilled, bonding (EE) and antibonding combinations (EE*). It is the mixing of the filled and unfilled orbitals of like symmetry on each metal center that results in the stabilization of the FF and FF* orbitals, and results in a net lowering of energy in the metallophilic pair, relative to the individual precursors. In the absence of such mixing a repulsive interaction typical of twoorbitals and four-electrons would be expected.61-63 For both d 10 Au1+ , on the left of Scheme 2, and d 8 Pt2+ on the right, the empty 6pz orbitals are energetically low enough to mix with the 5d orbitals of like symmetry. This molecular-based picture was expanded to band structure descriptions of partially oxidized extended chains to explain the physical properties of those systems.57,64 It should be noted that the d x2 -y2 orbital is shown as the relevant orbital for Au(I) d 10 systems, because in a linear Au geometry the z-axis is conventionally taken along the Au-ligand vectors. In this case, the metallophilic orbital is in the xy-plane.25 As can be seen from Table 1, however, metallophilic interactions are observed not only between the same chemical moieties, but often between different ones, as long as the orbital overlaps have compatible local symmetry. In the latter cases, the above diagram must be modified somewhat. In the ionic compounds that are the focus of this Perspective, increasing or decreasing charge is clearly an important factor. This journal is © The Royal Society of Chemistry 2010

Table 2 Pairwise sums of metallophilic element van der Waals radii38

Pd Pt Cu Ag Au Hg Tl Pb

1.63 1.72 1.4 1.72 1.66 1.55 1.96 2.02

Pd

Pt

Cu

Ag

Au

Hg

Tl

Pb

1.63

1.72

1.4

1.72

1.66

1.55

1.96

2.02

3.26

3.35 3.44

3.03 3.12 2.8

3.35 3.44 3.12 3.44

3.29 3.38 3.06 3.38 3.32

3.18 3.27 2.95 3.27 3.21 3.1

3.59 3.68 3.36 3.68 3.62 3.51 3.92

3.65 3.74 3.42 3.74 3.68 3.57 3.98 4.04

Scheme 2 Generic molecular orbital scheme for metallophilic interactions between pairs of d 10 ions (left) and d 8 ions (right).

The equal energies of the homodinuclear case are perturbed when the two components have opposing charges, and the anion is then higher in energy than the cation. In this case, it is useful to consider the [M]- species a nucleophile and the [M]+ one an electrophile in the assembly of [M]+ [M]- salts, such that a close energetic match is desired between the [M]- HOMO and the [M]+ LUMO, as depicted in Scheme 3.

ligands are only present in the xy-plane and not along the z-axis, defined as the incipient M-M vector, as shown in Scheme 4.

Scheme 4

Scheme 3 Relative molecular orbital energies in [M]+ [M]- .

There are significant steric restrictions to the formation of metallophilic contacts as well as the electronic ones outlined above. If one desires not just a single metallophilic contact, but a series of them aligned in a 1D array, only certain circumstances are permitted. A chain of ligated metal ions with alternating charges requires building blocks whose steric properties permit stacking them together in extended arrays. Such geometries are limited to a ligand set that enforces a highly anisotropic environment in which This journal is © The Royal Society of Chemistry 2010

Metal ion geometries suited for stacking in 1D arrays.

One of the most interesting aspects of metallophilic bonding is that it can overcome electrostatic repulsion between cations or anions and bring together groups of like charge. Because electrostatic repulsions do not prevent metallophilic interactions, the use of cationic, anionic, and/or neutral building blocks gives rise to six possibilities from the pairwise combinations that are shown in Scheme 5. Among the hundreds of examples known, the [M]+ [M]- category is the focus of this Perspective. Many ions can form pairwise metallophilic interactions in [M]+ [M]- salts, but repeated assemblies of such pairs are much less common. As will be seen, there is a delicate balance among the attractive intermolecular forces and it is a challenge to orchestrate only one type of metallophilic interaction in a particular salt. We will first briefly define and review these double salts with metallophilic interactions from the literature and then describe in detail the systems under investigation in our laboratory.

Scheme 5

Metallophilic pairwise motifs.

Dalton Trans., 2010, 39, 3543–3553 | 3545

Double salts The term “double salt” initially referred to (M)(M¢)(X)n species such as KAl(SO4 )2 ·12(H2 O) which are prepared from the K and Al sulfates and have distinct properties from the starting material.65 In our work, we use the term to indicate species with cations and anions that both contain metal ions, consistent with such use in the literature for platinum double salts66,67 of the type [L4 Pt]2+ [PtX4 ]2and other metals.68-70 Double salts consisting of [M]+ [M]- units that also exhibit metallophilic interactions are particularly common in [LAuX] systems in which L and X are relatively labile and often rearrange and stack as shown in Scheme 6. With ligands of up to moderate size, pairwise stacking is typically observed that exhibits a staggered conformation between the nearest-neighbor ions. These and many other aurophilic examples have been reviewed recently.35 The relative kinetic stability and electron-rich nature71 of Au1+ require that L groups have ideally soft s-donor and some p-acceptor character, such as imines,72 phosphines,73 thioethers,74 selenoethers,75 or isocyanides.76 A very few examples with secondary amines have been prepared,77 but these salts also have hydrogen bonding from the cation to the [AuX2 ]- anion which was recognized as a critical factor in stabilizing the [(R2 NH)2 Au(I)]+ coordination.

Scheme 6 Rearrangement of [LAuX] compounds into double salts.

The [AuX2 ]- anions in double salts are well-represented by the Au1+ halides for X = Cl, Br, I. Also very common is the [Au(CN)2 ]- anion which bonds to other transition metal centers with metallophilic bonding and can also bridge via Lewis base coordination through one or both nitrogen atom lone pairs.78,79 Less common are the organometallic anions with acetylide or aryl groups. Because it is so easily reduced, strong carbanions are only compatible with Au1+ if a means for delocalizing electron density onto the ligands exists. Therefore the groups found in [AuR2 ]species are the perhalogenated C6 F5 and C6 Cl5 aryl groups or acetylide groups with two empty p* orbitals. The [Cl3 Ge–AuGeCl3 ]- anion80 is highly unusual in having a group 14 element with sp3 hybridization bound to gold. Many examples exist of metallophilic anions that bond to ligand-free or only solvated Ag1+ or Tl1+ countercations, both of which can support a much less rigid coordination sphere than the typically linear Au1+ . Silver81 and thallium82,83 exhibit a promiscuous plasticity in their coordination which promotes extended arrays with various degrees of dimensionality. Therefore although compounds with these cations forming metallophilic contacts with anions could be considered double salts they are not usually referred to as such since the cations have such labile coordination spheres, in contrast to the other examples given here. Ligated silver cations that form argentophilic interactions are also known with strongly binding ligands including phosphines,84 N-heterocyclic carbenes,85,86 and amines.87 Silver anions are also consistently observed to form metallophilic interactions including 3546 | Dalton Trans., 2010, 39, 3543–3553

the [AgX2 ]- anions with halogens,86 pseudo halogens,79 and even carboxylates.87 Not observed in other metallophilic chemistry is the [Ag2 X4 ]2- anion85 with unusual trigonal planar metal coordination. Cuprophilic interactions are less common, but nevertheless welldocumented.88 One recent [Cu]+ [Cu]- example89 has a particularly ˚ between bis-pyrimidine short Cu ◊ ◊ ◊ Cu distance of 2.715 A [CuL2 ]+ derivative and [CuCl2 ]- . Occasionally, even when the steric requirements for extended array formation are met, infinite chains of metallophilic interactions are not observed. In a recent example,90 the d 8 [Au(bpy)Cl2 ]+ cation was crystallized with a mixture of the d 8 [AuBr2 (CN)2 ]and [AuBr2 Cl2 ]- , but no metallophilic interactions were observed. Thus when steric factors are not the reason, electronic factors must be playing a role, a point we discuss in more detail below.

New cations and old anions Our group has employed a metathetical route to double salts with the potential to form metallophilic interactions.91 We are certainly not the first to do so, for example, the platinum double salts [Pt(CNR)4 ]2+ [Pt(CN)4 ]2- precipitate in virtually quantitative yields and are used as vapochromic sensors.67,92 A notable difference between salts with all singly charged ions versus those with one or more doubly-charged ions is that the latter have substantially lower solubilities. Knowing that salts formed from cations and/or anions with charges greater than one had substantially diminished solubility, we chose to use only (cation)1+ and (anion)1- species so that the salts could be redissolved for purification and structural characterization. Judicious choice of counterions and solvent were used to effect virtually quantitative precipitation of the desired salt [M]+ [M]- from the still dissolved byproduct salt XZ, as shown in Scheme 7.

Scheme 7

Generic double salt synthesis.

We also made the important choice to use the polyimine ligands terpyridine (tpy) and bipyridine (bpy) with the d 8 centers Pt2+ and Au3+ , as shown in Scheme 8, in order to (i) minimize undesired ligand substitution reactions, (ii) enforce planarity of the ion and (iii) introduce the possibility of p-stacking as a complementary van der Waals force. The synthetic and spectroscopic details have been reviewed recently.93 Herein we focus discussion on the roles played by the omnipresent duo of steric and electronic effects. This synthetic approach led to a manifold of complexes that are summarized in Table 3. The cations (all with d 8 configurations) are the row-headers and the anions (d 8 or d 10 configurations) head the columns. Not all pairwise permutations were attempted, as indicated with no entry. In the cases where a compound was synthesized but no metallophilic interactions were observed, the result is so indicated. Some compounds were prepared but crystals suitable for single-crystal diffraction experiments could not be grown and these are indicated with “no X-ray”. Not all combinations of cations and anions that could form metallophilic contacts did so. Table 3 also shows the observed stacking patterns of cations and anions, whether pairwise or extended, and lists This journal is © The Royal Society of Chemistry 2010

No X-ray 13 No X-ray 20 None 4

An empty entry indicates that the compound has not been synthesized.

{(+)(-)}n 7 {(+)(+)} 11 {(+)(-)(+)}n 12 [Pt(tpy)Cl]+ [Pt(tpy)Br]+ [Pt(tpy)I]+ [Pt(tpy)(CN)]+ [Pt(t Bu3 tpy)Cl]+ [Au(bpy)Cl2 ]+ [Au(bpy)Br2 ]+

{(+)(-)(+)}n 5

[AuBr2 ]-

[AuI2 ]-

This journal is © The Royal Society of Chemistry 2010

[AuCl2 ]-

Addition of three t-butyl groups to the terpyridine ligand increases the solubility of its compounds, and was expected to have a steric effect as well. The structural study97 of [Pt(t Bu3 tpy)Cl]Cl, 3, revealed a similar head-to-tail pairing as seen in 1, but with one cation shifted significantly to the side of the other due to the bulky alkyl groups, as shown in Scheme 10. Some curvature of the two triimine ligands away from one another was also observed, and the ˚ . Even in the double salt closest Pt ◊ ◊ ◊ Pt contact was almost 5 A t [Pt( Bu3 tpy)Cl][AuCl2 ], 4, no Pt ◊ ◊ ◊ Pt or Pt ◊ ◊ ◊ Au contacts shorter ˚ were observed.97 The closest contact from the anion to than 4.6 A the Pt center is formed with one Cl atom from the anion at 3.830(1) ˚ . Interestingly, the palladium derivative98 [Pd(t Bu3 tpy)Cl]Cl does A ˚ . Examples of exhibit a metallophilic interaction at 3.4253(4) A metallophilic interactions among Pd compounds are far fewer

Table 3 Double salt combinations and metallophilicity

Scheme 9 Metallophilic distances in [Pt(tpy)X]X cation pairing

{(+)(-)} 14 {(+)(-)} 15

No X-ray 17 No X-ray 16 {(+)(-)(+)}n 19

[AuCl(CN)]-

[Au(CN)2 ]-

[Au(ArF )2 ]-

[AuCl2 ]- /[AuCl4 ]-

A simple way to modulate the steric bulk around a metal is to use halides of increasing atomic number. The structures of the starting materials [Pt(tpy)X]+ X- in which X = Cl,94 1, and Br,95 2, Scheme 9, are instructive. When the crystals are grown from water, metallophilic contacts form pairs of cations that are ˚ , consistent oriented head-to-tail, with Pt ◊ ◊ ◊ Pt contacts ~ 3.40 A with the expected distances for metallophilic contacts (see Table 2). Notably, twice the difference in van der Waals radii38 between Cl ˚ ) and Br (1.85 A ˚ ) is significantly greater than the difference (1.75 A between the two Pt ◊ ◊ ◊ Pt contacts in compounds 1 and 2, 0.2 > ˚ , suggesting that the steric pressure of the greater bulk of Br 0.07 A is being opposed to some extent by some electronic influence, vide infra. At the time of writing, the crystal structure of [Pt(tpy)I]I is still unknown and therefore unavailable for comparison. Iodide is generally a less common ligand in metallophilic species except for the [AuI2 ]- anion74,76,96 which is often observed to stack with an [L2 M]+ unit as shown in Scheme 6. The iodide van der Waals ˚ is bigger than that of all the d-block ions in radius38 of 1.98 A Table 2 such that its presence could hinder close contact between potentially metallophilic units.

{(+)(-)(+)}n 18

Ligand steric influences

None 8 None 9 {(+)(+)} 10

[AuBr2 ]- /[AuBr4 ]-

“none” where none were observed. Each example will be discussed in detail below.

{(+)(-)}n 6

[AuBr4 ][AuCl4 ]-

Scheme 8 d 8 Cations chosen for double salts.

Dalton Trans., 2010, 39, 3543–3553 | 3547

Scheme 10

Effect of t-butyl substitution in [Pt(R3 tpy)Cl]+ .

than with Pt, and even when they are observed it is rare to see them only with the less electronegative metal. Thus it is likely that metallophilic interactions in compounds with the Pt-based cation of 3 are not impossible, because they have been seen with the isosteric Pd derivative, but have not yet been observed.

Ligand electronic influences Double salts that differ only in the substitution of various halogens and pseudo halogens are also informative with regard to electronic as well as steric effects. As described above, iodide is sufficiently larger than Pt(II) or Au(I) such that metallophilic contacts for metals bonded to iodide are primarily seen between linear [AuI2 ]and other linear [L2 M]+ species.74,76,96 Therefore we have chosen to focus on smaller pseudo-halogen ligands X in the cations from Scheme 8. To continue using the effective [Pt(tpy)X]+ building block and harness the proclivity of d 10 centers for metallophilic interactions, we targeted salts with the general form [Pt]+ [Au]- , and more specifically [Pt(tpy)X][AuY2 ], with X and Y not necessarily equivalent. The [Pt(tpy)Cl]+ cation was paired with [AuCl2 ]- in [Pt(tpy)Cl][AuCl2 ], 5,91 and exhibited the unusual cation-cation interaction shown at the left in Scheme 11. A single [AuCl2 ]anion is sandwiched between two [Pt(tpy)Cl]+ cations, forming a trimetallic cation. The remaining [AuCl2 ]- anion does not participate in metallophilic interactions at all, but rather the trimetallic cations stack with one another. In contrast, as shown in the middle of Scheme 11, in [Pt(tpy)Cl][Au(CN)2 ], 6, a perfectly alternating chain of the [Pt]+ and [Au]- ions, assembled in part by metallophilic interactions, was achieved.91 A higher temperature recrystallization of 6 gave another analogous chain but with the exchange of one CN and one Cl between the Pt and Au centers to form, [Pt(tpy)CN][AuCl(CN)], 7.91 Clearly the differences between Cl and CN affected the number of metallophilic interactions formed. It is notable that with neither [Au(CN)2 ]- nor [AuCl(CN)]- was any evidence obtained for bridging cyanide coordination, which is unusual as this anion often participates in both metallophilic interactions and secondary Lewis acid–base bonding.78,79 The compound [Au(bpy)Cl2 ][Au(CN)2 ], 20, has also been prepared,91 but no structural data are available. The Au–C≡N stretching frequencies are not reliable indicators of the presence or absence of cyanide bridging due to the unusual behaviour of gold, unfortunately.99 Because [AuCl2 ]- participates in metallophilic bonding in 5, there seemed to be no steric reason for the lack of a {(+)(-)}• stacking pattern in 5, and we undertook to understand the phenomenon from an electronic point of view. A DFT analysis95 of [AuCl2 ]- , [AuCl(CN)]- and [Au(CN)2 ]- showed signif3548 | Dalton Trans., 2010, 39, 3543–3553

icant differences in the relative energies of the filled s-orbitals that participate in metallophilic interactions. Shown in Scheme 12 are abbreviated energy level diagrams for the two homoleptic anions and their constituent fragments. The ligand field splitting for the 5d orbitals on linear Au1+ is shown in the center of the diagram. On the far left are the HOMO through HOMO-2 for the cyanide fragment and on the far right are the 3p orbitals of chloride. The second and fourth columns show the energy levels for the resultant [AuX2 ]- species. Both CN- and Cl- form s- and p-overlaps with Au d-orbitals to form s- and p-type molecular orbitals. In the case of the p-acceptor ligand on [Au(CN)2 ]- , the HOMO is of s-type which has the appropriate symmetry to interact with the filled d z2 orbital from the [Pt(tpy)Cl]+ . The HOMO of the [AuCl2 ]anion has p-symmetry and the HOMO-1 has s-symmetry. The [AuCl(CN)]- case95 (not shown) is intermediate between these two. Clearly the slight lowering of this s-type orbital in [AuCl2 ]- relative to the p orbital set, does not prevent metallophilic interactions but may account for the difference in stacking patterns between 5 and 6. Therefore metal-based species with fewer or no p-donor ligands are more energetically disposed to form metallophilic interactions. Based on this hypothesis we chose a different [AuX2 ]- anion with s-donating and p-accepting X groups, namely the robust [Au(C6 F5 )2 ]- anion.100,101 This anion is well-known in the aurophilic literature and has exhibited metallophilic interactions in dozens of compounds.102,103 Unusually for a gold organometallic compound it is resilient to light and moderate heat. The C6 F5 ligands do not participate in ligand exchange interactions, unlike Cl- and CN- above. This anion was attractive furthermore for enhancing solubility and bringing the potential for aromatic p–p stacking with polyimine ligands such as bipyridine and terpyridine. We were successful in preparing and crystallographically characterizing95 the three halide derivatives [Pt(tpy)X][Au(C6 F5 )2 ] for X = Cl, 8, Br, 9, and I, 10. Indeed the distinct properties of the [Au(C6 F5 )2 ]- anion did affect the cation-anion interactions, but not in the way we had proposed. The anion steric bulk was comparable to the cation bulk such that in none of these three cases was solvent captured in the lattice. Compounds 8 and 9 have virtually identical structures in which the closest contact between the cation and anion in each case is between the Pt atom and an ipso-carbon atom from a C6 F5 ring, as shown in Scheme 13. Thus with the less-electron rich metal centers in [Pt(tpy)Cl]+ and [Pt(tpy)Br]+ , any Pt ◊ ◊ ◊ Pt or Pt ◊ ◊ ◊ Au metallophilic interaction loses out to other van der Waals forces. The precise tipping point among these different, weak intermolecular forces is clearly difficult to predict. Only in the more electron-rich iodide case, 10, were there signs of metallophilic interactions, and then only between the [Pt(tpy)I]+ ions as observed in 11, vide infra. As shown in Scheme 14, there are three crystallographically unique Pt ◊ ◊ ◊ Pt contacts in the crystal ˚ , is slightly longer structure. The shortest contact, 3.6376(7) A than most commonly accepted distances for Pt ◊ ◊ ◊ Pt contacts (see Table 2). This distance is bracketed by and between two ˚ . Between each pair of those slightly longer distances of 3.6817(8) A ˚ that medium length distances is a Pt ◊ ◊ ◊ Pt distance of 4.3046(9) A is significantly longer than any known metallophilic contact. If ˚ contacts have some metallophilic one accepts that the 3.6–3.7 A character, then there is a tetrameric stack exhibiting metallophilic interactions within the unit of four cations, but no significant interaction between the tetramers. This journal is © The Royal Society of Chemistry 2010

Scheme 11 Stacking in [Pt(tpy)X][AuY2 ] species (left to right, 5, 6, 7). [L3 PtCl] = [Pt(tpy)Cl]+ .

Scheme 12 Orbital interaction diagram for [AuX2 ]- .

Scheme 13 Sideways (left) and top-down (right) views of ion stacking in [Pt(tpy)X][Au(C6 F5 )2 ], X = Cl, 8, Br, 9.

In another family of double salts where structures of three halide analogs are available, metallophilic interactions were seen in all three cases. These salts have the stoichiometry [Pt(tpy)X][AuX2 ], for X = Cl, 5,91 Br, 12,104 I, 11.105 As discussed earlier and shown in This journal is © The Royal Society of Chemistry 2010

Scheme 11, in 5 all the [Pt(tpy)Cl]+ cations are involved in a stack united by an infinite chain of metallophilic interactions.91 Each pair of cations sandwiches one [AuCl2 ]- anion and the cationic trimers {[Pt(tpy)X]2 [AuX2 ]}+ , stack on top of one Dalton Trans., 2010, 39, 3543–3553 | 3549

Scheme 14 Cation packing in [Pt(tpy)I][Au(C6 F5 )2 ], 10.

Scheme 15 Metallophilic distances in {[Pt(tpy)X]2 [AuX2 ]}+ stacks.

another. The cations have an alternating head-to-tail orientation as shown in Scheme 15. The remaining [AuCl2 ]- ions occupy the channels between the chains as well as one CH3 CN solvent molecule per anion. The bromide derivative 12 has an analogous structure except that the lattice solvent is DMSO. The Pt ◊ ◊ ◊ Pt and Pt ◊ ◊ ◊ Au distances for both compounds are indicated in Scheme 15 as well. In contrast to the infinite stacking of the {[M]+ [M]- [M]+ }+ motifs, the iodide complex, 11, exhibits only pairwise stacking ˚ .105 Fig. 1 shows between the [Pt(tpy)I]+ cations at 3.5278(3) A how the M-M and M-X distances vary as a function of X in all three compounds. As expected from typical size-distance considerations, as the halogens increase in size, the M-X distances (blue squares and green triangles) increase linearly. The effect on the M-M contacts is less pronounced and less straightforward to interpret. The Au ◊ ◊ ◊ Pt contact (red diamonds) lengthens slightly upon changing from Cl to Br, and there is no such contact in the iodide derivative 11. There are metallophilic Pt ◊ ◊ ◊ Pt interactions in each case (black squares), but the shortest distance is observed in bromide 12 rather than chloride 5, as indicated in Scheme 15 and Fig. 1. This result suggests that the greater electron density of the Pt-containing unit, and perhaps indirectly the Au-containing unit, exerts a dominant influence over the increased steric bulk of the halogen. These three compounds clearly highlight the balance between steric and electronic influences on metallophilic contact formation. With decreasing halogen electronegativity, we now expect that metallophilic interactions are more favored. With increasing halogen size, however, a limit is reached. The Br case, 12, shows 3550 | Dalton Trans., 2010, 39, 3543–3553

Fig. 1 Interatomic distances in [Pt(tpy)X][AuX2 ] systems X = Cl, 5, Br, 12, I, 11.

the optimum balance as determined by the shortest intercation Pt ◊ ◊ ◊ Pt distances. These data therefore suggest that increasing the electron density on the metal center, by whatever means, would favor metallophilic interactions. Another means would be to add more electron donating substituents on the polyimine ligands and thereby increase the electron density at the metal center. Such a change was observed in the building blocks [Au(bpy)Cl2 ]Cl and [Au(Me2 bpy)Cl2 ]Cl.93 The cations of the former alternate with chloride anions in a chain, but the dialkylated ligand engenders a pairwise stacking of the cations with Au ◊ ◊ ◊ Au distance of ˚ . The trialkylated t Bu3 tpy ligand was also expected 3.5875(6) A to increase electron density at the metal center, but the increased steric bulk has an opposing influence, vide supra, in compounds 1 and 3.

Metal steric and electronic influences The radial extent of metal-based orbitals is naturally an important factor in the interaction of the filled orbitals that is the basis for metallophilicity. As discussed earlier, metallophilicity is most This journal is © The Royal Society of Chemistry 2010

common in the heaviest metallic elements including the late transition metals and heavier, chalcophilic p-block metals. The heaviest transition metals are not necessarily the largest, however, as a result of the lanthanide contraction and relativistic effects. Thus the silver radius is larger than that of gold in some measures (atomic, covalent, vdW65 ) but gold makes more metallophilic compounds.28 In developing our intuition about these systems, we describe metallophilic interactions as favored by relatively higher electron densities at the metal centers, as well as higher electronegativities. Another subtle difference that influences metallophilic stacking and exemplifies the idea of metal electron density influence is the metal oxidation state. Au(I) dominates the known metallophilic examples, whereas Au(III) examples were only experimentally documented recently.91,106 Unlike the case of d 8 Pt2+ , metallophilic stacking between d 8 Au3+ -containing units is extremely rare. The first published106 example contains infinite chains of the kinetically stable [Au(N3 )4 ]- anions with Me4 N+ cations. In our double salt systems we prepared the d 8 -d 8 double salts [Au(bpy)Cl2 ][AuCl4 ], 13, [Au(bpy)Cl2 ][AuBr4 ], 14, and [Au(bpy)Br2 ][AuBr4 ], 15. A preliminary report was made107 for 13 and 15 but no structural data were provided. We structurally characterized 14 and 15, both of which revealed only pairwise stacking of the [L2 AuX2 ]+ cations and [AuX4 ]- anions91 as shown in Scheme 16. Both distances are relatively long for aurophilic interactions and interestingly the allbromide derivative does not have a shorter Au ◊ ◊ ◊ Au distance. Contracted orbitals accompany an increase in oxidation state, which then limits the availability of the d z2 orbital for bonding. Therefore it is not surprising that only pairwise contacts were observed in the [Au(III)]+ [Au(III)]- structures, but an infinite chain was seen with the more electron rich [Au(N3 )4 ]- anions. The pairwise stacking motif in 14 and 15 can be rationalized as insufficient nucleophilicity, or insufficient radial extent of the d z2 orbital, or both, on the part of the [AuX4 ]- anions

Scheme 16 Pairwise stacking in [Au(bpy)X2 ][AuBr4 ], X=Cl, 14, X=Br, 15.

We also synthesized [Pt(II)]+ [Au(III)]- double salts of the form [Pt(tpy)X][AuX4 ], namely [Pt(tpy)Cl][AuCl4 ], 16, and [Pt(tpy)Br][AuBr4 ], 17. Analytically pure materials were readily obtained91 but to obtain X-ray diffraction quality crystals was more demanding. Gentle heating of the compounds to obtain super-saturated solutions lead to partial reduction of Au(III) to Au(I) in both cases. The new mixed-valent compounds {[Pt(tpy)Cl]2 [AuCl2 ]}[AuCl4 ], 18, and {[Pt(tpy)Br]2 [AuBr2 ]}[AuBr4 ], 19 were obtained and studied crystallographically.91 As indicated by the formulae, cations identical to those in 5 and 12 were obtained with the [AuX4 ]- anions, and have the same cation stacking patterns shown in Scheme 15. It is not at all surprising This journal is © The Royal Society of Chemistry 2010

that the [AuX4 ]- ions lie outside the metallophilic chains because the highly charged Au(III) ion is one of the least metallophilic that we have discussed. It is expected that the [Pt(tpy)X]+ moieties stack preferentially either with themselves or with the [AuX2 ]- anions.

Crystallization solvent influences Before closing it is appropriate and important to mention the role of lattice solvent. The efforts of many researchers in the more than 300 existing papers describing metallophilic interactions clearly show that solvent can influence the formation of metallophilic contacts. Either the solvent is present in the lattice itself and directly influences the packing via hydrogen bonding or steric pressure, for example, or different structures result from different recrystallization solvents even when the solvent does not form part of the lattice. For double salts in particular, if there is a great size mismatch between cations and anions, lattice solvent makes efficient crystal packing possible. Energetically the solvent influence is not surprising since the energy of metallophilic interactions has been determined to be on the order of hydrogen bonding ones.28 Therefore, the absence of metallophilic contacts in a particular structure does not necessarily mean than alternative solvent conditions could not be found under which the interactions would form.

Summary and future directions Our work in double salt complexes with the potential for metallophilic contact formation was based on the known importance of metal ion size and electron configuration. We have shown that even with favorable steric circumstances, the additional driving force of electrostatic attraction is not always enough to force metallophilic contacts. Ions of like charge may be preferentially found to stack with one another rather than with oppositely charged anions. The cation-anion interactions may also be aligned without direct metal-metal contact if some other atomic pair has a preferred electrostatic combination. Because metallophilicity results from interatomic overlap of electron density, any factor that influences a metal center’s electron density, or another metal’s access to it, can impact the presence, absence, or extent of metallophilic interactions. As in most chemical studies, steric and electronic perturbations are connected and changing one often affects the other. Our endeavors to date suggest the following guidelines for the formation of metallophilic contacts with double salts. A cation’s inherent positive charge somewhat diminishes its favorablity for metallophilic interactions, therefore the other electronic factors in the cation should be maximized: a d 10 rather than d 8 electron configuration and pacceptor ligands rather than p-donor ones. The greater radial extent of metal orbitals in anions favors metallophilic formation and therefore if higher oxidation states or pseudo-closed shells rather than closed shells are required, they are better placed in the anions. The inclusion of cooperative p-stacking effects between cations and anions remains challenging because the particular points of p–p attraction may not generate M-M vector alignment. There is substantial congruence between the patterns observed in these metallophilic double salt compounds and the requirements for one-dimensional conductors as described by Miller Dalton Trans., 2010, 39, 3543–3553 | 3551

and colleagues.57,58,108 This should come as no surprise given the central role played by square planar Pt2+ species in both families of compounds. Future generations of our double salt compounds will build on the electronic lessons learned in order to develop new 1D conductors that retain the solution processibility and robust synthetic conditions exploited to date, but incorporate new features of technological interest such as light harvesting and longlived excitonic states.

Acknowledgements I am grateful to all my research group colleagues for their efforts and contributions. This work was supported by Boston University and the National Science Foundation (NSF-CCF 829890 to LHD, NSF-REU 0649114 for W. A. Barksdale, W. M. Kochemba, and purchase of NMR spectrometer NSF-CHE 0619339).

Notes and references 1 J. S. Miller, ed., Extended Linear Chain Compounds, Plenum Press, New York, 1982, Three Vols. 2 J. K. Bera and K. R. Dunbar, Angew. Chem., Int. Ed., 2002, 41, 4453– 4457. 3 G. Magnus, Ann. Chim. Phys. S´er. 2, 1829, 40, 110. 4 J. R. Miller, J. Chem. Soc., 1965, 713–720. 5 M. Atoji, J. W. Richardson and R. E. Rundle, J. Am. Chem. Soc., 1957, 79, 3017–3020. 6 W. Caseri, Platinum Met. Rev., 2004, 48, 91–100. 7 M. Fontana, W. Caseri and P. Smith, Platinum Met. Rev., 2006, 50, 112–117. 8 E. M. Cradwick, D. Hall and R. K. Phillips, Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 1971, 27, 480–484. 9 E.-G. Kim, K. Schmidt, W. R. Caseri, T. Kreouzis, N. StingelinStutzmann and J.-L. Bredas, Adv. Mater., 2006, 18, 2039–2043. 10 M. G. Debije, M. P. De Haas, J. M. Warman, M. Fontana, N. Stutzmann, M. Kristiansen, W. R. Caseri, P. Smith, S. Hoffmann and T. I. Solling, Adv. Funct. Mater., 2004, 14, 323–328. 11 R. Mattes and K. Krogmann, Z. Anorg. Allg. Chem., 1964, 332, 247– 256. 12 K. Krogmann, Angew. Chem., Int. Ed. Engl., 1969, 8, 35–42. 13 C. M. Che, L. G. Butler, H. B. Gray, R. M. Crooks and W. H. Woodruff, J. Am. Chem. Soc., 1983, 105, 5492–5494. 14 J. R. Winkler, J. L. Marshall, T. L. Netzel and H. B. Gray, J. Am. Chem. Soc., 1986, 108, 2263–2266. 15 J. L. Marshall, M. D. Hopkins, V. M. Miskowski and H. B. Gray, Inorg. Chem., 1992, 31, 5034–5040. 16 V. M. Miskowski and V. H. Houlding, Inorg. Chem., 1991, 30, 4446– 4452. 17 V. M. Miskowski and V. H. Houlding, Inorg. Chem., 1989, 28, 1529– 1533. 18 W. B. Connick, V. M. Miskowski, V. H. Houlding and H. B. Gray, Inorg. Chem., 2000, 39, 2585–2592. 19 F. Huq and A. C. Skapski, J. Cryst. Mol. Struct., 1974, 4, 411. 20 A. H. Reis, Jr., V. S. Hagley and S. W. Peterson, J. Am. Chem. Soc., 1977, 99, 4184–4186. 21 P. W. DeHaven and V. L. Goedken, Inorg. Chem., 1979, 18, 827–831. 22 F. A. Cotton and C. A. Murillo, Eur. J. Inorg. Chem., 2006, 4209–4218. 23 J. F. Berry, F. A. Cotton, P. Lei, T. Lu and C. A. Murillo, Inorg. Chem., 2003, 42, 3534–3539. 24 F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 380–388. 25 Y. Jiang, S. Alvarez and R. Hoffmann, Inorg. Chem., 1985, 24, 749– 757. 26 F. Scherbaum, A. Grohmann, B. Huber, C. Krueger and H. Schmidbaur, Angew. Chem., Int. Ed. Engl., 1988, 27, 1544–15446. ¨ Angew. Chem., Int. Ed., 2004, 43, 4412–4456. 27 P. Pyykko, ¨ Chem. Rev., 1997, 97, 597–636. 28 P. Pyykko, 29 F. A. Cotton, C. A. Murillo and R. A. Walton, Multiple Bonds Between Metal Atoms, Springer Science, New York, 2005, pp. 818. 30 F. A. Cotton, G. Wilkinson, C. A. Murillo and M. Bochmann, Advanced Inorganic Chemistry, John Wiley & Sons, New York, 1999, pp. 1355.

3552 | Dalton Trans., 2010, 39, 3543–3553

31 A. R. West, Basic Solid State Chemistry, John Wiley & Sons, New York, 1999, pp. 480. 32 E. O’Grady and N. Kaltsoyannis, Phys. Chem. Chem. Phys., 2004, 6, 680–687. 33 H. Schmidbaur, S. Cronje, B. Djordjevic and O. Schuster, Chem. Phys., 2005, 311, 151–161. 34 P. Schwerdtfeger and M. Lein, in Gold Chemistry, ed. F. Mohr, WileyVWH, New York, 2009, pp. 183–248. 35 H. Schmidbaur and A. Schier, Chem. Soc. Rev., 2008, 37, 1931–1951. 36 H. Schmidbaur, Gold Bull., 2000, 33, 3–10. 37 H. Schmidbaur, Gold Bull., 1990, 23, 11–21. 38 A. Bondi, J. Phys. Chem., 1964, 68, 441–451. 39 N. Masciocchi, M. Moret, P. Cairati, F. Ragaini and A. Sironi, J. Chem. Soc., Dalton Trans., 1993, 471–475. 40 M. E. Prater, L. E. Pence, R. Clerac, G. M. Finniss, C. Campana, P. Auban-Senzier, D. Jerome, E. Canadell and K. R. Dunbar, J. Am. Chem. Soc., 1999, 121, 8005–8016. 41 B. E. Villarroya, C. Tejel, M.-M. Rohmer, L. A. Oro, M. A. Ciriano and M. Benard, Inorg. Chem., 2005, 44, 6536–6544. 42 V. J. Catalano, B. L. Bennett, R. L. Yson and B. C. Noll, J. Am. Chem. Soc., 2000, 122, 10056–10062. 43 A. Doshi, K. Venkatasubbaiah, A. L. Rheingold and F. Jakle, Chem. Commun., 2008, 4264–4266. 44 B. Djordjevic, O. Schuster and H. Schmidbaur, Inorg. Chem., 2005, 44, 673–676. 45 A. M. Mazany and J. P. Fackler, Jr., J. Am. Chem. Soc., 1984, 106, 801–802. 46 M. Kim, T. J. Taylor and F. P. Gabbai, J. Am. Chem. Soc., 2008, 130, 6332–6333. 47 O. Crespo, A. Laguna, E. J. Fernandez, J. M. Lopez-de-Luzuriaga, A. Mendia, M. Monge, E. Olmos and P. G. Jones, Chem. Commun., 1998, 2233–2234. 48 R. Uson, J. Fornies, L. R. Falvello, M. A. Uson and I. Uson, Inorg. Chem., 1992, 31, 3697–3698. 49 E. J. Fernandez, A. Laguna, J. M. Lopez-de-Luzuriaga, M. Monge, M. Nema, M. E. Olmos, J. Perez and C. Silvestru, Chem. Commun., 2007, 571–573. 50 C.-M. Che and S.-W. Lai, Coord. Chem. Rev., 2005, 249, 1296–1309. 51 V. W.-W. Yam and E. C.-C. Cheng, Chem. Soc. Rev., 2008, 37, 1806– 1813. 52 M. A. Rawashdeh-Omary, M. A. Omary, H. H. Patterson and J. P. Fackler, Jr., J. Am. Chem. Soc., 2001, 123, 11237–11247. 53 E. J. Fernandez, A. Laguna and J. M. Lopez-de-Luzuriaga, Dalton Trans., 2007, 1969–1981. 54 E. J. Fernandez, C. Hardacre, A. Laguna, M. C. Lagunas, J. M. Lopezde-Luzuriaga, M. Monge, M. Montiel, M. E. Olmos, R. C. Puelles and E. Sanchez-Forcada, Chem.–Eur. J., 2009, 15, 6222–6233. 55 F. Jalilehvand, M. Maliarik, M. Sandstroem, J. Mink, I. Persson, P. Persson, I. Toth and J. Glaser, Inorg. Chem., 2001, 40, 3889–3899. 56 J. Fornies, A. Garcia, E. Lalinde and M. T. Moreno, Inorg. Chem., 2008, 47, 3651–3660. 57 J. S. Miller and A. J. Epstein, Prog. Inorg. Chem., 1976, 20, 1–151. 58 J. S. Miller, Inorg. Chem., 1976, 15, 2357–2360. 59 A. E. Underhill, D. M. Watkins, J. M. Williams and K. Carneiro, in Extended Linear Chain Compounds, ed. J. S. Miller, Plenum Press, New York, 1982, vol. 1. 60 D. M. Roundhill, H. B. Gray and C. M. Che, Acc. Chem. Res., 1989, 22, 55–61. 61 A. Dedieu and R. Hoffmann, J. Am. Chem. Soc., 1978, 100, 2074– 2079. 62 P. K. Mehrotra and R. Hoffmann, Inorg. Chem., 1978, 17, 2187–2189. 63 K. M. Merz, Jr. and R. Hoffmann, Inorg. Chem., 1988, 27, 2120–2127. 64 M.-H. Whangbo, in Extended Linear Chain Compounds, ed. J. S. Miller, Plenum Press, New York, 1982, vol. 2. 65 A. F. Holleman and E. Wiberg, Inorganic Chemistry, Academic Press, San Diego, CA, 2001, pp. 1884. 66 V. H. Houlding and A. J. Frank, Inorg. Chem., 1985, 24, 3664–3668. 67 S. M. Drew, D. E. Janzen, C. E. Buss, D. I. MacEwan, K. M. Dublin and K. R. Mann, J. Am. Chem. Soc., 2001, 123, 8414–8415. 68 V. Urban, T. Pretsch and H. Hartl, Angew. Chem., Int. Ed., 2005, 44, 2794–2797. 69 N. D. Draper, R. J. Batchelor, P. M. Aguiar, S. Kroeker and D. B. Leznoff, Inorg. Chem., 2004, 43, 6557–6567. 70 Q.-M. Wang and T. C. W. Mak, J. Am. Chem. Soc., 2001, 123, 7594– 7600.

This journal is © The Royal Society of Chemistry 2010

71 F. Mohr, ed., Gold Chemistry, Wiley-VCH, New York, 2009. 72 W. Conzelmann, W. Hiller, J. Straehle and G. M. Sheldrick, Z. Anorg. Allg. Chem., 1984, 512, 169–176. 73 O. Schuster, U. Monkowius, H. Schmidbaur, R. S. Ray, S. Krueger and N. Roesch, Organometallics, 2006, 25, 1004–1011. 74 S. Ahrland, B. Noren and A. Oskarsson, Inorg. Chem., 1985, 24, 1330– 1333. 75 S. Ahrland, K. Dreisch, B. Noren and A. Oskarsson, Mater. Chem. Phys., 1993, 35, 281–289. 76 W. C. Kaska, H. A. Mayer, M. R. J. Elsegood, P. N. Horton, M. B. Hursthouse, C. Redshaw and S. M. Humphrey, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2004, 60, m563–m565. 77 B. Ahrens, S. Friedrichs, R. Herbst-Irmer and P. G. Jones, Eur. J. Inorg. Chem., 2000, 2017–2029. 78 M. J. Katz, K. Sakai and D. B. Leznoff, Chem. Soc. Rev., 2008, 37, 1884–1895. 79 J. R. Stork, D. Rios, D. Pham, V. Bicocca, M. M. Olmstead and A. L. Balch, Inorg. Chem., 2005, 44, 3466–3472. 80 A. Bauer and H. Schmidbaur, J. Am. Chem. Soc., 1996, 118, 5324– 5325. 81 E. J. Fernandez, J. M. Lopez-de-Luzuriaga, M. Monge, M. E. Olmos, R. C. Puelles, A. Laguna, A. A. Mohamed and J. P. Fackler, Jr., Inorg. Chem., 2008, 47, 8069–8076. 82 E. J. Fernandez, A. Laguna, J. M. Lopez-de-Luzuriaga, M. Montiel, M. E. Olmos and J. Perez, Organometallics, 2005, 24, 1631– 1637. 83 E. J. Fernandez, J. M. Lopez-de-Luzuriaga, M. E. Olmos, J. Perez, A. Laguna and M. C. Lagunas, Inorg. Chem., 2005, 44, 6012–6018. 84 P. W. R. Corfield and H. M. M. Shearer, Acta Crystallogr., 1966, 20, 502–508. 85 M. V. Baker, D. H. Brown, R. A. Haque, B. W. Skelton and A. H. White, Dalton Trans., 2004, 3756–3764. 86 S. K. Schneider, W. A. Herrmann and E. Herdtweck, Z. Anorg. Allg. Chem., 2003, 629, 2363–2370. 87 Z. L. You, Z. M. Wang, Y. Zou and H. L. Zhu, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2004, 61, 6–m8. 88 M. A. Carvajal, S. Alvarez and J. J. Novoa, Chem.–Eur. J., 2004, 10, 2117–2132.

This journal is © The Royal Society of Chemistry 2010

89 S. H. Oakley, M. P. Coles and P. B. Hitchcock, Inorg. Chem., 2004, 43, 5168–5172. 90 B. Pitteri, M. Bortoluzzi and V. Bertolasi, Transition Met. Chem., 2008, 33, 649–654. 91 R. Hayoun, D. K. Zhong, A. L. Rheingold and L. H. Doerrer, Inorg. Chem., 2006, 45, 6120–6122. 92 S. M. Drew, J. E. Mann, B. J. Marquardt and K. R. Mann, Sens. Actuators, B, 2004, 97, 307–312. 93 L. H. Doerrer, Comments Inorg. Chem., 2008, 29, 93–127. 94 C. S. Angle, A. G. DiPasquale, A. L. Rheingold and L. H. Doerrer, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2006, 62, m340– m342. 95 V. Phillips, J. A. Golen, A. L. Rheingold and L. H. Doerrer, manus. in prep., 2009. 96 D. Schneider, O. Schuster and H. Schmidbaur, Organometallics, 2005, 24, 3547–3551. 97 M. I. Kahn, J. A. Golen, A. L. Rheingold and L. H. Doerrer, manus. in prep., 2009. 98 E. J. MacLean, R. I. Robinson, S. J. Teat, C. Wilson and S. Woodward, J. Chem. Soc., Dalton Trans., 2002, 3518–3524. 99 H. Schmidbaur, Gold: Progress in Chemistry, Biochemistry, and Technology, John Wiley & Sons, New York, 1999, pp. 894. 100 R. Uson, A. Laguna and J. Vicente, J. Chem. Soc., Chem. Commun., 1976, 353–354. 101 R. Uson and A. Laguna, Coord. Chem. Rev., 1986, 70, 1–50. 102 E. J. Fernandez, A. Laguna and M. E. Olmos, Coord. Chem. Rev., 2008, 252, 1630–1667. 103 A. Luquin, E. Cerrada and M. Laguna, in Gold Chemistry, ed. F. Mohr, Wiley-VWH, New York, 2009, pp. 93-182. 104 M. I. Kahn, J. A. Golen, A. L. Rheingold and L. H. Doerrer, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2009, 65, m1135. 105 C. S. Angle, K. J. Woolard, M. I. Kahn, J. A. Golen, A. L. Rheingold and L. H. Doerrer, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2007, 63, m231–m234. ¨ 106 T. M. Klapotke, B. Krumm, J.-C. Galvez-Ruiz and H. Noeth, Inorg. Chem., 2005, 44, 9625–9627. 107 C. M. Harris and T. N. Lockyer, J. Chem. Soc., 1959, 3083–3085. 108 J. S. Miller, Adv. Chem., 1976, 150, 18–30.

Dalton Trans., 2010, 39, 3543–3553 | 3553