Skip to main content
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2002 Apr 16;99(8):4916–4921. doi: 10.1073/pnas.062643599

Cluster self-assembly of di[gold(I)]halonium cations

Hubert Schmidbaur 1,*, Armin Hamel 1, Norbert W Mitzel 1, Annette Schier 1, Stefan Nogai 1
PMCID: PMC122694  PMID: 11959944

Abstract

Treatment of gold(I) halide complexes of the type L-Au-X [where L = PPh3, PEt3 with X = Cl, Br, I, or L = 2,6-(MeO)2C6H3PPh2 with X = Cl] with AgSbF6 in the molar ratio 2:1 in dichloromethane/tetrahydrofuran at −78°C affords high yields of di[gold(I)]halonium salts of the formula {X[Au(PR3)]2}+ SbFInline graphic (2-8). A determination of the crystal structures of the four triarylphosphine complexes (2-4, 8) revealed the presence of novel tetranuclear dications with a highly symmetrical structure (point group S4) that arises from self-assembly of the dinuclear monocations through a set of four equivalent aurophilic Au–Au interactions. A comparison with two reference structures of corresponding chloronium perchlorate and bromonium tetrafluoroborate salts with monomeric, dinuclear cations shows that the geometry of the latter is greatly altered on dimerization to optimize the interactions between the closed-shell metal centers (Au: 5d10). Weak metallophilic bonding clearly becomes significant only in crystal lattices where anions with a larger ionic radius (SbFInline graphic vs. BFInline graphic, ClOInline graphic) reduce the otherwise dominant role of strong interionic Coulomb forces. The results indicate that aurophilic bonding is indeed an ubiquitous, quite dependable mode of intermetallic interactions provided that the right environment is chosen to allow the weak forces to become operative.


Halogen atoms X engaged in covalent bonding (R–X) are among the poorest donor centers, and this is particularly true for X = F and Cl. Exceedingly strong acceptor components R+ are required to employ the halogen function in dative bonding and to generate cationic or zwitterionic species in which the halogen atom represents a “halonium” center as in [R–X–R]+ cations (1).

Prototypes with R = H are the cations of the general formula [H2X]+, which are formed in the treatment of HF or HCl with superacids (2)—e.g., [H2Cl]+[SbF6]. The zig-zag chains of polymeric (HF)n also feature fluoronium centers bridging the hydrogen atoms via strong hydrogen bonds (3), and fragments of these chains are present in anions of the type [HnFn+1]. Structural data are typically available for [H2F3] or [H3F4], as shown in Scheme S1 (48).

Scheme 1.

Scheme 1

With R representing an alkyl group, the dialkylchloronium cations [R–Cl–R]+ can be recognized as adducts of alkyl chlorides R–Cl with carbocations R+, which play an important mechanistic role in many reactions under strongly acidic conditions (1, 9). These diorganochloronium salts are often rather labile and the structural characterization is an experimental challenge. Alkyl- or arylhalides are also generally very poor ligands for metal acceptors, and there is only a limited number of stable complexes with R–X molecules as ligands (10).

In fundamental discussions of the frontier orbital concept in the 1970s, gold(I) [Au]+ and its complexes of the simple formula [L–Au]+, with L representing a standard monodentate ligand, were recognized as “isolobal” with simple species such as the proton [H]+ and carbocations [R]+, respectively, which suggested many analogies in structure and bonding of seemingly unrelated compounds (1113). And in fact, these analogies were readily detected even when screening the most common binary compounds (14, 41). Thus the gold(I) halides AuCl (15), AuBr (16), and AuI (17) have zig-zag chain structures with “unsupported” halonium centers bridging the gold atoms, and gold(I) dihalides [AuX2] (18) resemble hydrogen dihalide anions [HX2] (Scheme S2). It should also be noted that gold(I) is one of the strongest acceptor cations in the family of metals (19). The extremely high electron affinity and electronegativity have their origin in relativistic effects, which reach a local maximum for the element gold in the periodic table (20, 21), and which are also obvious from the unusually small atomic radius of gold in both prominent oxidation states (22, 23).

Scheme 2.

Scheme 2

In light of this analogy, it was therefore not unexpected that in 1979 Uson et al. were able to show that gold(I) chloride complexes of the formula L–Au–Cl are prone to accept cations [L–Au]+ to form di(gold)chloronium salts with the novel cations [Cl(AuL)2]+. A stable salt was obtained and structurally characterized with L = PPh3 and the poorly coordinating perchlorate counterion (24, 25): {Cl[Au(PPh3)]2}+ ClOInline graphic (1). In subsequent studies in our own laboratories the phenomenon was confirmed with complexes bearing different ligands L, and the chemistry could be extended further to include gold(I) bromide and iodide (26). With tetrafluoroborate counterions, the complete set of halonium salts {X[Au(PR3)]2}+ BFInline graphic was isolated, which are among the most stable halonium species known in coordination chemistry. The strongly bent structural units Au–X–Au can be recognized as parts of the structural motif of the (AuX)n zig-zag chains (Scheme S2).

Our more recent work reported in this paper (and briefly in a preliminary communication; ref. 27) has shown that the di(gold)halonium salts have an extremely interesting supramolecular chemistry that only emerges as larger counterions are introduced. It appears that changes in the Coulomb energies of the ionic lattice allow weak forces based on closed-shell interactions Au–Au (d10–d10) between the metal centers to contribute significantly to the lattice energy (2831). In brief, the halonium monocations are found to undergo self-assembly against Coulomb repulsion to form dications based on tetranuclear gold(I) clusters.

Experimental Procedures

General Experimental Techniques.

All experiments were carried out under an atmosphere of pure and dry nitrogen either in Schlenk apparatus or in a glove box. Solvents were dried, degassed in a vacuum and saturated with nitrogen. Glassware was oven-dried and filled with nitrogen. Standard equipment was used throughout for all preparative, analytical, and spectroscopic work. The complexes Ph3PAuCl, Et3PAuCl, and [2,6-(MeO)2C6H3Ph2P]AuCl were prepared following literature methods and the former converted into the bromides or iodides in metathesis reactions with KBr/KI in a water/dichloromethane two-phase system (32). All other reagents were commercially available. graphic file with name pq06264350e1.gif

Preparative Procedure.

General.

Silver hexafluoroantimonate (AgSbF6, 1 equivalent) is suspended in dry tetrahydrofuran (20 ml) at −78°C under nitrogen and a solution of the (phosphine)gold(I) halide (LAuX, two equivalents) in dry dichloromethane (20 ml) is added with stirring. Stirring of the reaction mixture is continued at −78°C for 2 h. After warming to 20°C, the reaction mixture is filtered and the volume of the filtrate reduced to 5 ml in a vacuum. The product is precipitated by the addition of dry pentane, filtered, washed with pentane, and dried in a vacuum. Compounds 2-8 are colorless solids, soluble in chloroform, dichloromethane, and tetrahydrofuran, and insoluble in diethyl ether and pentane. The Et3P complexes are sensitive to light. Single crystals grown from the same solvent mixture (dichloromethane/tetrahydrofuran) by careful layering with pentane and cooling in the refrigerator may contain solvents as indicated in Table 1.

Table 1.

Crystal data, data collection, and structure refinement

[(Ph3PAu)4Cl2] [SbF6]22 [(Ph3PAu)4Br2] [SbF6]2 3 × 0.5 thf × 0.5 CH2Cl2 [(Ph3PAu)4I2] [SbF6]24 [(({MeO}2C6H3) Ph2 PAu)4Cl2] [SbF6]28
Crystal data
Formula C36H30Au2ClF6P2Sb* C38.5H35Au2BrClF6O0.5P2Sb C36H30Au2IF6P2Sb* C40H38Au2ClF6O4P2Sb
Mr 1189.68* 1312.65 1281.13* 1307.78
Crystal system Tetragonal Tetragonal Monoclinic Tetragonal
Space group I I C2 I4/m
a, Å 25.7515(2) 25.7428(2) 36.2971(4) 20.024(1)
b, Å 25.7515(2) 25.7428(2) 13.9107(3) 20.024(1)
c, Å 13.9439(2) 14.0698(2) 25.6609(5) 24.368(1)
α, ° 90 90 90 90
β, ° 90 90 134.417(1) 90
γ, ° 90 90 90 90
V, Å3 9,246.8(2) 9,324.0(2) 9,254.5(3) 9,771.07(7)
ρcalc, gcm−3 1.709* 1.870 1.839* 1.781
Z 8 8 8 8
F(000) 4,448* 4,920 4,763* 4,960
μ(Mo Kα), cm−1 70.82 78.86 76.86 67.17
Data collection
T (K) 163 143 143 143
Measured reflections 76,303 135,096 70,329 170,842
Unique reflections 10,584 [Rint = 0.066] 5,307 [Rint = 0.076] 20,417 [Rint = 0.0795] 5,362 [Rint = 0.0754]
Refinement
Refined parameters 435 462 868 269
Final R values [I ≥ 2σ(I)]
Flack Parameter racemic twin, BASF 0.232 racemic twin, BASF 0.496 racemic twin, BASF 0.332
R1 0.0384 0.0341 0.0543 0.0522
wR2 0.0919 0.0904 0.1206 0.1230
a/b 0.0492/35.09 0.0546/44.35 0.0468/58.30 0.0437/178.67
*

Values are without contents of solvent voids. 

wR2 = {[Σw(FInline graphicFInline graphic)2]/Σ[w(FInline graphic)2]}1/2; w = 1/[σ2(FInline graphic) + (ap)2 + bp]; p = (FInline graphic+ 2FInline graphic)/3. 

Details.

2: AgSbF6 (83.3 mg, 0.243 mmol) (Ph3P)AuCl (240 mg, 0.485 mmol), yield 242 mg (84%), dec. temp. 143°C. 3: AgSbF6 (63.7 mg, 0.186 mmol) (Ph3P)AuBr (200 mg, 0.371 mmol), yield 186 mg (81%), dec. temp. 131°C. 4: AgSbF6 (88 mg, 0.256 mmol) (Ph3P)AuI (300 mg, 0.521 mmol), yield 253 mg (77%), dec. temp. 118°C. 5: AgSbF6 (49 mg, 0.143 mmol) (Et3P)AuCl (100 mg, 0.285 mmol), yield 112 mg (87%), dec. temp. 121°C. 6: AgSbF6 (74 mg, 0.215 mmol) (Et3P)AuBr (170 mg, 0.430 mmol), yield 161 mg (79%), dec. temp. 117°C. 7: AgSbF6 (78 mg, 0.226 mmol) (Et3P)AuI (200 mg, 0.452 mmol), yield 180 mg (80%), dec. temp. 114°C. 8: AgSbF6 (44.4 mg, 0.129 mmol), {[2,6-(MeO)2C6H3]Ph2P}AuCl (310 mg, 0.258 mmol), yield 144 mg (85%), dec. temp. 139°C.

Elemental analyses (%, calculated/found).

2, C72H60Au4Cl2F12P4Sb2 (2379.36) C 36.36/36.80, H 2.54/2.58. 3, C72H60Au4Br2F12P4Sb2 (2468.41) C 35.04/36.56, H 2.91/3.14. 4, C72H60Au4F12I2P4Sb2 (2562.26) C 33.75/34.04, H 2.36/2.80. 5, C24H60Au4Cl2F12P4Sb2 (1802.24) C 16.00/15.81, H 3.36/3.30. 6, C24H60Au4Br2F12P4Sb2 (1892.24) C 15.23/15.08, H 3.20/3.49. 7, C24H60Au4I2F12P4Sb2 (1984.04) C 14.51/14.60, H 3.04/3.21. 8, C80H76O8Au4Cl2F12P4Sb2 (2618.98) C 36.67/37.11, H 2.92/3.08.

MS [fast atom bombardment (FAB), 4-nitro-benzyl alcohol], parent ion M/2-SbF6 [m/z].

2, 954; 3, 998; 4, 1045; 5, 666; 6, 710; 7, 757; 8, 1074. 31P{1H} NMR (CDCl3, 25°C): 2, 31.60; 3, 32.86; 4, 33.52; 5, 47.46; 6, 46.34; 7, 45.56; 8, 10.74 (all singlet resonances). The 1H and 13C{1H} NMR spectra of the complexes 2-8 show the usual signal patterns for Et, Ph, and 2,6-(MeO)2C6H3 groups, respectively, with no anomalies. The data are virtually indistinguishable from those reported for the tetrafluoroborate salts (26).

Crystal structure determinations.

Specimens of suitable quality and size were mounted on the ends of quartz fibers in F06206R oil and used for intensity data collection on a Nonius DIP2020 diffractometer, employing graphite-monochromated Mo Kα radiation. The structures were solved by a combination of direct methods (SHELXS-97) and difference-Fourier syntheses and refined by full matrix least-squares calculations on F2 (SHELXL-97). The thermal motion of all non-hydrogen atoms was treated anisotropically, except for those of the thf solvent atoms in 3, which were refined isotropically. Because of the very large thermal parameters of the solvents CH2Cl2 and thf in the lattice of 3, their occupancy was lowered to 50%. The structures of 2 and 4 suffered from disordered and unidentified solvent in the lattice, which was not included in the refinement but was taken care of by the SQUEEZE-procedure (from PLATON). The volumes occupied by the solvent were 2,700 and 2,637 Å (3), respectively; the number of electrons per unit cell deduced by SQUEEZE were 257 and 167. The SbF6 anions in the crystal of 8 were disordered and refined in split positions and with reduced site occupation factors. All hydrogen atoms were placed in idealized calculated positions and allowed to ride on their corresponding carbon atoms with fixed isotropic contributions. Further information on crystal data, data collection, and structure refinement are summarized in Table 1. Selected interatomic distances and angles are shown in the corresponding figure legends and in Table 2. The data for 2 were included in a preliminary communication (27). Thermal parameters and tables of all interatomic distances and angles have been deposited with the Cambridge Crystallographic Data Centre (12 Union Road, Cambridge CB2 1EZ, U.K.). The data are available on request (CCDS-178462–178465).

Table 2.

Interatomic distances (Å) and angles (°) in the monomeric (1, 1a) (ref. 2426) and dimeric di(gold)halonium cations {[(R3P)Au]2X}+ Y (24, 8) (ref. 27)

Au…Au Au–X–Au P–Au–X
{[(Ph3P)Au]2Cl}+ (ClO4) (1) M: 3.085(2) 82.7(2) 177.1(2)
M: 3.035(2) 80.7(2) 177.6(2)
177.3(2)
176.1(3)
{[(Ph3P)Au]2Br}+ (BF4) (1a) M: 3.6477(1) 96.83(3) 177.87(3)
{[(Ph3P)Au]4Cl2}2+ 2(SbF6) (2) D: 3.0734(4) 101.4(1) 170.99(8)
D: 3.0788(4) 102.0(1) 171.12(7)
M: 3.645(1)
M: 3.679(1)
{[(Ph3P)Au]4Br2}2+ 2(SbF6) (3) D: 3.0960(6) 97.33(5) 169.19(7)
D: 3.0840(6) 96.57(5) 168.91(7)
M: 3.717(1)
M: 3.686(1)
{[(Ph3P)Au]4I2}2+ 2(SbF6) (4) D: 3.0684(7) 90.36(1) 165.44(9)
D: 3.1077(7) 88.47(4) 165.93(9)
D: 3.0974(7) 92.40(5) 166.56(9)
D: 3.1269(7) 91.22(5) 167.14(9)
M: 3.700(1)
M: 3.644(1)
M: 3.774(1)
M: 3.732(1)
{[(MeO)2C6H3Ph2PAu]4Cl2}2+ 2(SbF6) (8) D: 3.1282(4) M: 3.809(1) 106.2(1) 174.96(7)

If there is more than one entry for a given parameter, these data refer to different independent units in the crystal. (Au–Au)M refers to metal–metal contacts in a given monomer, and (Au–Au)D to metal–metal contacts between monomers in the aggregate (see Figs. 1 and 3 for examples). 

Results

Gold(I) halide complexes of tertiary phosphines (R3P)AuX are readily obtained in quantitative yield through standard methods (32). In the present study, a set of six compounds with R = Ph, Et and X = Cl, Br, I was prepared, complemented by one complex with a more bulky ligand, [2,6-(MeO)2C6H3Ph2P]AuCl. Treatment of a solution of any of these complexes in a mixed solvent of tetrahydrofuran and dichloromethane [1:1] with silver hexafluoroantimonate AgSbF6 in the molar ratio 2:1 at −78°C gave a virtually quantitative precipitate of silver chloride, which was readily separated by filtration. After partial evaporation of the solvent at ambient temperature, the products were precipitated by the addition of pentane as colorless solids in high yields and identified by microanalysis data and NMR and mass spectra (Experimental Procedures, Eq. 1).

The NMR spectra of the compounds 2-5 (in chloroform-d1 solution at 25°C) were found to be similar to those reported for the corresponding tetrafluoroborates [X(AuPR3)2]+BFInline graphic reported previously (26). This finding suggests that the new hexafluoroantimonates and the tetrafluoroborates have the same cations in solution. The mass spectra (FAB, from p-nitrobenzyl alcohol matrix) also show similarities between the corresponding SbFInline graphic and BFInline graphic salts.

Single crystals suitable for x-ray diffraction studies could be grown for all four triarylphosphine complexes (2–4, 8), whereas the triethylphosphine complexes (5–7) proved unstable in the crystal growth procedures or gave poor crystal quality. Preliminary work with trimethylphosphine ligands also lead to premature decomposition of the products.

The crystals of the chloronium and bromonium compounds 2 and 3 are isomorphous, tetragonal of space group I4̄, with the same number of formula units (Z = 8) in unit cells of very similar dimensions, whereas crystals of the iodonium compound 4 are monoclinic, space group C2 (Z = 4). Crystals of compound 8 are again tetragonal, of space group I4/m (Z = 4). In all four cases the cations appear as large tetranuclear dimers (dications), which are well separated from the SbF6 anions. Voids between these ions are filled with solvent molecules that were found disordered in 2 and 4, but could be resolved for compound 3. In the structure of 8, there are two different sites for the anions that are not completely filled, but this deficit could be accounted for by reduced populations in both positions (with conservation of electroneutrality). The anions do not show any structural irregularities in any of the complexes. As a representative example, Fig. 1 shows the contents of the unit cell of compound 3, and Table 1 gives crystal data and structure solution details for compounds 24 and 8.

Figure 1.

Figure 1

Unit cell of the crystalline phase 3 × 0.5 C4H8O × 0.5 CH2Cl2. The asymmetric unit comprises one quarter of each dication and one complete anion. See the supporting information, which is published on the PNAS web site, www.pnas.org.

The crystals of the two isomorphous compounds 2 and 3 contain two independent tetranuclear units with very similar dimensions and with crystallographically imposed S4 symmetry (Fig. 2). Crystals of complex 4 also show two independent but similar tetranuclear units, albeit with reduced crystallographic symmetry (point group C2, but still close to S4; Fig. 3). The asymmetric unit of 8 contains only one-quarter of a tetranuclear unit from which the complete dication can be generated through the symmetry operations of point group S4 (Fig. 4).

Figure 2.

Figure 2

Structure of one of the two dications of compound 3 with atomic numbering (ORTEP drawing with 50% probability ellipsoids; H-atoms omitted for clarity). Selected bond lengths (Å) and angles for both cations Au1/Au2 (°): Au1–P1 2.259 (2), Au1–Br1 2 475 (1), Au1–Au1A 3.0960 (6); Au2–P2 2.259 (2), Au2–Br2 2.469 (1), Au2–Au2A 3.0840 (6). P1–Au1–Br1 169.19 (7), Au1–Br1–Au1C 97.33 (5); P2–Au2–Br2 168.91 (7), Au2–Br2–Au2C 96.57 (5).

Figure 3.

Figure 3

Structure of the dications of compound 4 with atomic numbering (ORTEP drawing with 50% probability ellipsoids; H-atoms omitted for clarity). Selected bond lengths (Å) and angles (°): Au1–P1 2.271 (3), Au1–I1 2.608 (1), Au1–Au2 3.0684 (7), Au1–Au2a 3.1077 (7), Au2–P2 2.263 (3), Au2–I2 2.612 (1), Au3–P3 2.269 (3), Au3–I3 2.615 (1), Au3–Au4 3.0974 (7), Au3–Au4A 3.1269 (7), Au4–P4 2.276 (3), Au4–I4 2.612 (1); P1–Au1–I1 165.93 (9), P2–Au2–I2 165.44 (9), P3–Au3–I3 167.14 (9), P4–Au4–I4 166.56 (9), Au1–I1–Au1A 90.36 (1), Au2–I2–Au2a 88.47 (4), Au3–I3–Au3A 92.40 (5), Au4–I4–Au4A 91.22 (5).

Figure 4.

Figure 4

Structure of the dication in compound 8.

Inspection of the details of the structures of the four dications reveals that {X[Au(PR3)]2}+ cations have become intimately associated into clusters through intercationic gold–gold contacts. The four gold atoms are arranged at the corners of an equilateral bisphenoid that can be described as a tetrahedron with two opposite edges lengthened considerably as compared with the remaining four edges. Taken together for all four compounds, these four short edges are in the surprisingly narrow range from 3.0684(7) to 3.1282(4) Å regardless of the nature of the halonium atoms (Cl, Br, I) and of the ligands PR3 (2 as compared with 8). These short contacts clearly qualify for aurophilic interactions between the metal atoms (19, 2831). By contrast, the two opposite long edges are in the range 3.645(1) to 3.809(1) Å (for all four complexes) and are thus well beyond the limits where aurophilic bonding should be considered (33).

The surprising agreement of the Au–Au contacts in the tetranuclear dicationic units regardless of the nature of the halogen atoms is the result of very pronounced changes in the bond angles at the halonium centers: Whereas Au–Cl–Au angles are 101.4(1)° and 102.0(1)° for 2, the Au–Br–Au angles in 3 are 97.33(5)° and 96.57(5)°, and the Au–I–Au angles in 4 only 88.47(4), 90.36(4), 91.22(5), and 92.40(5)°. These differences show that the dimerization of the cations—where necessary—is associated primarily with readjustments of the Au–X–Au bond angles in a direction that close contacts of the metal atoms of the approaching monomers are optimized.

This idea is supported by the crystal structure of the tetrafluoroborate/perchlorate salts {X[Au(PPh3)]2}+BFInline graphic/ClO4 (X = Cl, Br) in which the cations are dinuclear monomers. In the chloronium perchlorate 1 the Au–Cl–Au angle (24, 25) is as small as 82.7(2)° as compared with 101.7(1)° (average) in 2, a widening of no less than 19°. For the bromonium tetrafluoroborate 1a the Au–Br–Au angle (26) is 96.83(3)°, almost the same as in the tetranuclear units of 3 [96.95(5)° average]. Clearly for the Br-bridged units no angle change is necessary because of a perfect match of the monomers on dimerization. For iodonium monomers no data are available (Table 2).

There is also a geometrical response to dimerization to be detected in the P–Au–X angles. Note that these angles are expected to be linear for any L–Au–X compound (32). For 2, the angles P–Au–Cl are found at 170.99(8)° and 171.12(7)°, but for 3 the corresponding values are further down at 169.19(7)° and 168.91(7)°, and for 4 finally at 165.44(9)°, 165.93(9)°, 166.56(9)°, and 166.56(9)°. These deviations from linearity of 9° (2), 11° (3), and 14° (4) mean a bending of the P–Au–X axes that brings the gold atoms closer together.

In the classical zig-zag structures of the binary gold(I) halides AuX (ref. 32; Scheme S2), the Au–X distances are 2.36 Å (Cl), 2.40 Å (Br), and 2.60 Å (I). The Au–X bond lengths determined for 2-4 are similar with average values of 2.31 Å (2, Cl), 2.48 Å (3, Br), and 2.61 Å (4, I). However, this comparison is of limited value because of the different effects of the close packing of the chains in the binary phases (with different modifications) on one hand, and the efficient shielding of the Au–X–Au units by bulky ligands in 2-4 on the other. There is also an influence of variations of the bulkiness of the tertiary phosphine as exemplified by the data for 2 and 8: For the latter there is a lengthening of the Au–Cl bonds [to 2.382(2) Å] and a widening of the Au–Cl–Au angle [to 106.2(1)°] and the P–Au–Cl angle [to 174.96(7)°] as compared with 2, but as yet the four short Au–Au contacts remain almost constant [at 3.1282(4) Å].

Discussion

The structural studies have shown that the generation of di[gold(I)]halonium hexafluoroantimonates (2–8) leads to salts with tetranuclear dications through a self-assembly of two monocations as drawn in Scheme S3.

Scheme 3.

Scheme 3

The structures of the products are particularly illustrative examples for the significance of weak “aurophilic” (“metallophilic”) bonding for the building of supramolecular structures (2831, 33, 34).

Three observations are important: (i) The self-assembly of two monocations to give a dication takes place against Coulomb repulsion. (ii) The intimate aggregation is based solely on interactions between closed shell (d10) metal atoms [Au(I)]. These interactions are known—from experimental and theoretical data—to be associated with bond energies in the order of 7–12 kcal/mol and thus comparable to those of standard hydrogen bonds (2831). Although naturally less ubiquitous than hydrogen bonding, metallophilic bonding appears to be an equally dependable contributor to supramolecular architecture whenever low-coordinate, heavy late transition elements are present in the building blocks. (iii) The association in the solid state depends on the nature of the counterion. Because it takes place against electrostatic forces, the Coulomb energy arising from nearest neighbor cation–anion interactions is of prime importance. Larger anions (SbFInline graphic as compared with BFInline graphic) therefore enhance the chances of aurophilic interactions to be realized in a given system. It should also be noted that even solvent molecules in the crystal may influence the lattice energy quite considerably (2831). Separation of the monomers also occurs by complete solvation upon dissolution in suitable solvents.

The halonium salts are not the first class where association between polyaurated onium salts is ligand and anion dependent: Pyramidal tetra(gold)arsonium cations and tri(gold)oxonium and -sulfonium cations {[(R3P)Au]nE}+X (E = As/n = 4; E = O,S/n = 3) were found as monomers and as dimers in lattices with different components (R3P and X), and the dimers may have different structures (29, 3537). Likewise, methanium complexes {[(R3P)Au]4CH}+BFInline graphic appear to be associated in the crystal (38), whereas analogous C-substituted cations are monomeric (39). Quantum-chemical and force-field calculations have shown for simple model compounds that there is a delicate energy balance between the structural alternatives (40).

Supplementary Material

Supporting Information

Acknowledgments

This work was generously supported by Deutsche Forschungsgemeinschaft, Fonds der Chemischen Industrie, Volkswagenstiftung, Alexander von Humboldt Stiftung, Deutscher Akademischer Austauschdienst, Degussa AG, and Heraeus GmbH.

Footnotes

This paper was submitted directly (Track II) to the PNAS office.

References

  • 1.Olah G A. Halonium Ions. New York: Wiley-Interscience; 1975. [Google Scholar]
  • 2.Olah G A, Prakash G K S, Sommer J. Superacids. New York: Wiley; 1985. [DOI] [PubMed] [Google Scholar]
  • 3.Atoji M, Lipscomb W N. Acta Crystallogr. 1954;7:173–175. [Google Scholar]
  • 4.Forrester J D, Senko M E, Zalkin A, Templeton D H. Acta Crystallogr. 1963;16:58–65. [Google Scholar]
  • 5.Coyle B A, Schroeder L W, Ibers J A. J Solid State Chem. 1970;1:386–391. [Google Scholar]
  • 6.Ibers J A. J Chem Phys. 1964;40:402–409. [Google Scholar]
  • 7.Emsley J. Chem Soc Rev. 1980;9:91–104. [Google Scholar]
  • 8.Mootz D, Poll W. Z Anorg Allg Chem. 1982;484:158–164. [Google Scholar]
  • 9.Olah G A, DeMember J R. J Am Chem Soc. 1969;91:2113–2115. [Google Scholar]
  • 10.Tellers D M, Bergman R G. J Am Chem Soc. 2001;123:11508–11509. doi: 10.1021/ja0164982. [DOI] [PubMed] [Google Scholar]
  • 11.Hoffmann R. Angew Chem Int Ed Engl. 1982;21:711–725. [Google Scholar]
  • 12. Halpern, J. (1968) Discuss. Faraday Soc. 467–476.
  • 13.Stone F G A. Acc Chem Res. 1981;14:318–325. [Google Scholar]
  • 14.Hönes G, Huisl W, Keim R, Schwager B, editors. Gmelin Handbook of Inorganic Chemistry Gold. Berlin: Springer; 1992. , Suppl. Vol. B1. [Google Scholar]
  • 15.Strähle J, Lörcher K P. Z Naturforsch. 1974;29b:266–267. [Google Scholar]
  • 16.Janssen E M W, Wiegers G A. J Less-Common Met. 1978;57:P47–P57. [Google Scholar]
  • 17.Weiss A, Weiss A. Z Naturforsch. 1956;11b:604–605. [Google Scholar]
  • 18.Braunstein P, Müller A, Bögge H. Inorg Chem. 1986;25:2104–2106. [Google Scholar]
  • 19.Schmidbaur H. Interdisc Sci Rev. 1992;17:213–220. [Google Scholar]
  • 20.Pyykkö P, Desclaux J P. Acc Chem Res. 1979;12:276–281. [Google Scholar]
  • 21.Hrusak J, Hertwig R H, Schröder D, Schwerdtfeger P, Koch W, Schwarz H. Organometallics. 1995;14:1284–1291. [Google Scholar]
  • 22.Bayler A, Schier A, Bowmaker G A, Schmidbaur H. J Am Chem Soc. 1996;118:7006–7008. [Google Scholar]
  • 23. Tripathi, U. M., Bauer, A. & Schmidbaur, H. (1997) J. Chem. Soc. Dalton Trans. 2865–2868.
  • 24.Uson R, Laguna A, Castrillo M V. Synth React Inorg Metalorg Chem. 1979;9:317–324. [Google Scholar]
  • 25.Jones P G, Sheldrick G M, Uson R, Laguna A. Acta Crystallogr B. 1980;36:1486–1489. [Google Scholar]
  • 26.Bayler A, Bauer A, Schmidbaur H. Chem Ber Recueil. 1997;130:115–118. [Google Scholar]
  • 27.Hamel A, Mitzel N W, Schmidbaur H. J Am Chem Soc. 2001;123:5106–5107. doi: 10.1021/ja010398e. [DOI] [PubMed] [Google Scholar]
  • 28.Schmidbaur H. Gold Bull. 2000;33:3–9. [Google Scholar]
  • 29.Schmidbaur H. Chem Soc Rev. 1995;24:391–400. [Google Scholar]
  • 30.Schmidbaur H. Gold Bull. 1990;23:11–21. [Google Scholar]
  • 31.Scherbaum F, Grohmann F, Huber B, Krüger C, Schmidbaur H. Angew Chem Int Ed Engl. 1988;27:1544–1546. [Google Scholar]
  • 32.Laguna A. In: Gold: Progress in Chemistry, Biochemistry and Technology. Schmidbaur H, editor. Chichester, U.K.: Wiley; 1999. pp. 349–428. [Google Scholar]
  • 33.Pyykkö P. Chem Rev. 1997;97:597–636. doi: 10.1021/cr940396v. [DOI] [PubMed] [Google Scholar]
  • 34.Braga D, Grepioni F, Desiraju G R. Chem Rev. 1998;98:1375–1405. doi: 10.1021/cr960091b. [DOI] [PubMed] [Google Scholar]
  • 35.Angermaier K, Schmidbaur H. Inorg Chem. 1994;33:2069–2070. [Google Scholar]
  • 36.Zeller E, Beruda H, Kolb A, Bissinger P, Riede J, Schmidbaur H. Nature (London) 1991;352:141–143. [Google Scholar]
  • 37.Schmidbaur H, Hofreiter S, Paul M. Nature (London) 1995;377:503–504. [Google Scholar]
  • 38.Schmidbaur H, Gabbai F P, Schier A, Riede J. Organometallics. 1995;14:4969–4971. [Google Scholar]
  • 39.Steigelmann O, Bissinger P, Schmidbaur H. Z Naturforsch. 1993;48b:72–78. [Google Scholar]
  • 40.Chung S-C, Krüger S, Schmidbaur H, Rösch N. Inorg Chem. 1996;35:5387–5395. [Google Scholar]
  • 41.Hönes G, Huisl W, Keim R, Schwager B, editors. Gmelin Handbook of Inorganic Chemistry Gold. Berlin: Springer; 1994. , Suppl. Vol. B2. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
pnas_99_8_4916__1.pdf (15.8KB, pdf)
pnas_99_8_4916__2.pdf (22.6KB, pdf)
pnas_99_8_4916__3.pdf (19.6KB, pdf)
pnas_99_8_4916__4.pdf (15.9KB, pdf)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES