COMMUNICATION
Figure 1. Molecular structure of 1 in the crystal (displayed is a “dimeric”
unit). Ellipsoids are shown with 40% probability. Bond lengths (Å) and
angles (deg) of the anions: Au1-N1 2.015(6), Au1-N4 1.996(9), Au1-
N7 2.071(7), Au1-N10 2.04(1), Au2-N13 2.04(1), Au2-N16 1.95(1),
Au2-N19 2.02(1), Au2-N22 2.06(1), N1/4/7/10-N2/5/8/11 1.19(1)-1.21-
(1), N2/5/8/11-N3/6/9/12 1.15(1)-1.16(1), N13/16/19/22-N14/17/20/23
1.19(1)-1.24(1), N14/17/20/23-N15/18/21/24 1.13(1)-1.16(1), Au1‚‚‚Au2
3.507(3); N1-Au1-N4 99.7(3), N1-Au1-N7 173.3(3), N1-Au1-N10
88.6(3), N10-Au1-N4 171.6(4), Au1-N1/4/7/10-N2/5/8/11 116.6(7)-
124.3(7), N1/4/7/10-N2/5/8/11-N3/6/9/12 172(1)-176(1); corresponding
angles of the Au2 anion are similar.
for the deposition of gold on surfaces, which can serve as a
hydrogenation catalyst.12
The product from the decomposition is always colloidal
gold. The formation of dinitrogen is detected after short
periods by 14N NMR spectroscopy as evidence of slow
decomposition in solution (pressure formed) aside from gold
deposition. In addition to the friction and impact sensitivity
of these salts, light sensitivity is observed, similar to that of
the tetrachloroaurates(III).
Various attempts were required to determine the crystal
structure of 1 because of the additional sensitivity to X-ray
exposure [as observed already for tetrachloroaurate(III) salts],
shown in Figure 1.13
Figure 2. Packing diagram of 1 with a view along the a axis, showing
the stacking of the anions. Dashed lines represent alternating Au‚‚‚Au
contacts of 3.507(3) and 3.584(3) Å.
The gold atoms are surrounded by azide groups in the
expected square-planar fashion with bond lengths and angles
in the expected ranges, similar to those observed for [Ph4-
As][Au(N3)4].3 A striking difference in this structure is the
presence of polymeric stacking of the anion, consisting of
weak Au‚‚‚Au interactions of 3.507(3) and 3.584(3) Å.14
These distances are slightly below the sum of the van der
Waals radii (3.6 Å).15 There is clearly no inversion center
between the gold atoms, although it appears as such.
Figure 2 shows a plot of the packing diagram of 1,
featuring a stacking of monomeric anions in parallel layers
with the [Me4N]+ units located in between.
(10) Into a solution of 0.100 g (0.25 mmol) of H[AuCl4]‚3H2O in 10 mL
of water in a beaker is added 0.7 g (10.0 mmol) of NaN3. The color
changes from yellow to red. Into this mixture is added, with stirring,
a solution of 1.5 g (10.0 mmol) of [Me4N]Br in 10 mL of water/
ethanol. Upon cooling with ice, the salt [Me4N][Au(N3)4] (1)
precipitates as red-orange needles (0.065 g, 59%). 1H NMR (acetone-
d6): δ 3.43 (Me, s). 13C NMR (acetone-d6): δ 55.9 (Me, t, 1JC-14N
)
4.0 Hz). 14N NMR (acetone-d6): δ -132.8 (Nâ), -178 (Nγ), -279
(NR), -338.6 (Me4N). IR (KBr): 3034 w (νCH), 2044/2037/2013 vs
(νas,N ), 1482 m, 1447 w, 1414 w, 1380 w, 1264 m (νs,N ), 949 m, 581
To find out whether there exist intrinsic Au‚‚‚Au contacts
between individual [Au(N3)4]- anions, hybrid density func-
3
3
vw, 566 vw, 424 m cm-1. Raman (100 mW): 3036 (10)/2978 (11)/
2-
tional theory calculations on a possible [Au(N3)4]2 dimer
2953 (9)/2920 (10) (νCH), 2062 (33)/2035 (21)/2011 (15) (νas,N ), 1449
were performed starting with an Au‚‚‚Au distance of 2.0 Å.16
3
(18) (δCH), 1284 (14) (νs,N ), 1020 (12), 999 (13), 950 (12), 755 (14),
3
690 (13), 674 (13), 403 (100) (νAuN), 391 (32), 235 (33), 212 (30),
160 (15) cm-1. Anal. Calcd for C4H12N13Au: C, 10.9; H, 2.8; N, 41.5.
Found: C, 11.3; H, 2.9; N, 41.3.
(13) Triclinic, P1; a ) 9.973(2) Å, b ) 10.246(2) Å, c ) 13.625(3) Å; R
) 94.14(3)°, â ) 93.63(3)°, γ ) 110.42(3)°; V ) 1295.7(4) Å3; Z )
2; F ) 2.252 g/cm3; T ) 213 K; Rint ) 0.0253; R1 ) 0.0444, wR2 )
0.1116 (4σ data); R1 ) 0.0700, wR2 ) 0.1240 (all data); GOF )
0.954; largest diff peak/hole 2.582/-2.929 e/Å3.
(11) Into a solution of 0.2 mmol of [Me2NH2][AuCl4] or [NH4][AuCl4] in
5 mL of methanol in a beaker is added 1.7 mmol of AgN3. An
immediate color change from yellow to red is observed. The mixture
is stirred for 24 h and then separated carefully from insoluble silver
salts. Slow evaporation in the case of the dimethylammonium salt
yields pure [Me2NH2][Au(N3)4] (2) in almost quantitative yield. In
the case of [NH4][Au(N3)4] (3), attempts to isolate the pure salt mostly
result in explosions, probably during the crystallization process.
Spectroscopic data of 2. 1H NMR (acetone-d6): δ 3.78 (NH2, s), 3.01
(Me, s). 13C NMR (acetone-d6): δ 35.9 (Me, br). 14N NMR (acetone-
d6): δ -132.8 (Nâ), -178 (Nγ), -280 (NR), -358.0 (Me2NH2).
Spectroscopic data of 3. 1H NMR (CD3OD): δ 4.85 (NH4, s). 14N
NMR (CD3OD): δ -133.6 (Nâ), -179 (Nγ), -280 (NR), -368.8
(NH4). 14N NMR (H2O): δ -133.9 (Nâ), -179 (Nγ), -282 (NR),
-361.5 (NH4).
(14) Estimated standard deviation.
(15) (a) Schmidbaur, H. Gold: Progress in Chemistry, Biochemistry and
Technology; Wiley: New York, 1999. (b) Mingos, D. M. P. J. Chem.
Soc., Dalton Trans. 1996, 561. (c) Fackler, J. P., Jr. Inorg. Chem.
2002, 41, 6959. (d) Pyykko¨, P. Angew. Chem., Int. Ed. 2004, 43, 4412.
(e) Theoretical studies on AuIII-AuIII systems: Mendizabal, F.; Zapata-
Torres, G.; Olea-Azar, C. Chem. Phys. Lett. 2003, 382, 92. Mendizabal,
F.; Pyykko¨, P. Phys. Chem. Chem. Phys. 2004, 6, 900.
(16) The calculations were performed with the Gaussian03 program
package17a at the B3LYP level of theory using a cc-pVDZ basis set
for nitrogen and a small-core ECP-60-MDF relativistic pseudopotential
replacing 60 core electrons at the gold atoms.17b,c The remaining Au
valence electrons were treated with a basis set of the following
contraction: (12s12p9d3f2g)/[6s6p4d3f2g].
(12) Lopez-Sanchez, J. A.; Winfield, J. M.; Krumm, B.; Klapo¨tke, T. M.;
Lennon, D., manuscript in preparation.
9626 Inorganic Chemistry, Vol. 44, No. 26, 2005