Aggregation of [Au(CN)4]- anions: Examination by crystallography and 15N CP-MAS NMR and the structural factors influencing intermolecular Au...N interactions

Article abstract of DOI:10.1021/ic101782v

To investigate the factors influencing the formation of intermolecular Au...NC interactions between [Au(CN)₄]^- units, a series of [cation]ⁿ⁺[Au(CN)₄]_n double salts was synthesized, structurally characterized and probed by IR and ¹⁵N{¹H} CP-MAS NMR spectroscopy. Thus, [ ⁿBu₄N][Au(CN)₄], [AsPh₄][Au(CN) ₄], [N(PPh₃)₂][Au(CN)₄], [Co(1,10-phenanthroline)₃][Au(CN)₄]₂, and [Mn(2,2′;6′,2″-terpyridine)₂][Au(CN) ₄]₂ show [Au(CN)₄]^- anions that are well-separated from one another; no Au-Au or Au...NC interactions are present. trans-[Co(1,2-diaminoethane)₂Cl₂][Au(CN) ₄] forms a supramolecular structure, where trans-[Co(en) ₂Cl₂]⁺ and [Au(CN)₄]^- ions are found in separate layers connected by Au-CN...H-N hydrogen-bonding; weak Au...NC coordinate bonds complete octahedral Au(III) centers, and support a 2-D (4,4) network motif of [Au(CN)₄] ^--units. A similar structure-type is formed by [Co(NH ₃)₆][Au(CN)₄]₃· (H ₂O)₄. In [Ni(1,2-diaminoethane)₃][Au(CN) ₄]₂, intermolecular Au...NC interactions facilitate formation of 1-D chains of [Au(CN)₄]^- anions in the supramolecular structure, which are separated from one another by [Ni(en) ₃]²⁺ cations. In [1,4-diazabicyclo[2.2.2]octane-H][Au(CN) ₄], the monoprotonated amine cation forms a hydrogen-bond to the [Au(CN)₄]^- unit on one side, while coordinating to the axial sites of the gold(III) center through the unprotonated amine on the other, thereby generating a 2-D (4,4) net of cations and anions; an additional, uncoordinated [Au(CN)₄]^--unit lies in the central space of each grid. This body of structural data indicates that cations with hydrogen-bonding groups can induce intermolecular Au...NC interactions, while the cationic charge, shape, size, and aromaticity have little effect. While the μ_CN values are poor indicators of the presence or absence of N-cyano bridging between [Au(CN)₄]^--units (partly because of the very low intensity of the observed bands), ¹⁵N{¹H} CP-MAS NMR reveals well-defined, ordered cyanide groups in the six diamagnetic compounds with chemical shifts between 250 and 275 ppm; the resonances between 260 and 275 ppm can be assigned to C-bound terminal ligands, while those subject to CN...H-N bonding resonate lower, around 250-257 ppm. The ¹⁵N chemical shift also correlates with the intermolecular Au...N distances: the shortest Au-N distances also shift the ¹⁵N peak to lower frequency. This provides a real, spectroscopically measurable electronic effect associated with the crystallographic observation of intermolecular Au...NC interactions, thereby lending support for their viability.

Full text of DOI:10.1021/ic101782v

Inorg. Chem. 2011, 50, 1265–1274 1265
DOI: 10.1021/ic101782v
Aggregation of [Au(CN)4]- Anions: Examination by Crystallography and
15N CP-MAS NMR and the Structural Factors Influencing Intermolecular
Au N Interactions
3 3 3
Andrew R. Geisheimer,† John E. C. Wren,‡ Vladimir K. Michaelis,‡ Masayuki Kobayashi,†,§ Ken Sakai,§
Scott Kroeker,*,‡ and Daniel B. Leznoff*,†
†Department of Chemistry, Simon Fraser University, 8888 University Drive, Burnaby, British Columbia,
Canada V5A 1S6, ‡Department of Chemistry, University of Manitoba, Winnipeg, Manitoba, R3T 2N2 Canada,
§
and Department of Chemistry, Faculty of Science, Kyushu University, Hakozaki 6-10-1, Higashi-ku, Fukuoka,
812-8581 Japan
Received September 1, 2010
To investigate the factors influencing the formation of intermolecular Au NC interactions between [Au(CN)4]-
3 3 3
units, a series of [cation]nþ[Au(CN)4]n double salts was synthesized, structurally characterized and probed by IR and
15N{1H} CP-MAS NMR spectroscopy. Thus, [nBu4N][Au(CN)4], [AsPh4][Au(CN)4], [N(PPh3)2][Au(CN)4], [Co(1,10-
phenanthroline)3][Au(CN)4]2, and [Mn(2,20;60,200-terpyridine)2][Au(CN)4]2 show [Au(CN)4]- anions that are well-
separated from one another; no Au-Au or Au NC interactions are present. trans-[Co(1,2-diaminoethane)2Cl2]-
3 3 3
[Au(CN)4] forms a supramolecular structure, where trans-[Co(en)2Cl2]þ and [Au(CN)4]- ions are found in separate
layers connected by Au-CN H-N hydrogen-bonding; weak Au NC coordinate bonds complete octahedral
3 3 3
3 3 3
Au(III) centers, and support a 2-D (4,4) network motif of [Au(CN)4]--units. A similar structure-type is formed by
[Co(NH3)6][Au(CN)4]3 (H2O)4. In [Ni(1,2-diaminoethane)3][Au(CN)4]2, intermolecular Au NC interactions facil-
3
3 3 3
itate formation of 1-D chains of [Au(CN)4]- anions in the supramolecular structure, which are separated from one
another by [Ni(en)3]2þ cations. In [1,4-diazabicyclo[2.2.2]octane-H][Au(CN)4], the monoprotonated amine cation
forms a hydrogen-bond to the [Au(CN)4]- unit on one side, while coordinating to the axial sites of the gold(III) center
through the unprotonated amine on the other, thereby generating a 2-D (4,4) net of cations and anions; an additional,
uncoordinated [Au(CN)4]--unit lies in the central space of each grid. This body of structural data indicates that cations
with hydrogen-bonding groups can induce intermolecular Au NC interactions, while the cationic charge, shape,
3 3 3
size, and aromaticity have little effect. While the νCN values are poor indicators of the presence or absence of N-cyano
bridging between [Au(CN)4]--units (partly because of the very low intensity of the observed bands), 15N{1H} CP-MAS
NMR reveals well-defined, ordered cyanide groups in the six diamagnetic compounds with chemical shifts between
250 and 275 ppm; the resonances between 260 and 275 ppm can be assigned to C-bound terminal ligands, while
those subject to CN H-N bonding resonate lower, around 250-257 ppm. The 15N chemical shift also correlates
3 3 3
with the intermolecular Au N distances: the shortest Au-N distances also shift the 15N peak to lower frequency.
This provides a real, spectroscopically measurable electronic effect associated with the crystallographic observation of
intermolecular Au NC interactions, thereby lending support for their viability.
3 3 3
3 3 3
Introduction
properties observed for these materials, including mag-
netic,3-7 luminescent,8 gas storage,9,10 birefringent,11,12
The field of metal-organic coordination polymers has
received a great deal of focused research attention re-
cently.1,2 This is largely because of the wide range of
(4) Kurmoo, M. Chem. Soc. Rev. 2009, 38, 1353–1379.
(5) Lescouezec, R.; Toma, L. M.; Vaissermann, J.; Verdaguer, M.;
Delgado, F. S.; Ruiz-Perez, C.; Lloret, F.; Julve, M. Coord. Chem. Rev.
2005, 249, 2691–2729.
(6) Miller, J. S. MRS Bull. 2000, 25, 60–64.
(7) Shatruk, M.; Avendano, C.; Dunbar, K. R. Prog. Inorg. Chem. 2009,
56, 155–334.
*To whom correspondence should be addressed. E-mail: dleznoff@sfu.ca
(D.B.L.); scott_kroeker@umanitoba.ca (S.K.). Tel: 1-778-782-4887
(D.B.L.); 1-204-474-9335 (S.K.). Fax: 1-778-782-3765 (D.B.L.); 1-204-474-
7608 (S.K.).
(8) Allendorf, M. D.; Bauer, C. A.; Bhakta, R. K.; Houk, R. J. T. Chem.
Soc. Rev. 2009, 38, 1330–1352.
(1) Janiak, C. Dalton Trans. 2003, 2781–2804.
(2) Czaja, A. U.; Trukhan, N.; Muller, U. Chem. Soc. Rev. 2009, 38, 1284–
1293.
(3) Batten, S. R.; Murray, K. S. Coord. Chem. Rev. 2003, 246, 103–130.
(9) James, S. L. Chem. Soc. Rev. 2003, 32, 276–288.
(10) Murray, L. J.; Dinca, M.; Long, J. R. Chem. Soc. Rev. 2009, 38, 1294–
1314.
r
2011 American Chemical Society
Published on Web 01/26/2011
pubs.acs.org/IC

1266 Inorganic Chemistry, Vol. 50, No. 4, 2011
Geisheimer et al.
and vapochromic13-15 properties. Many of these properties
depend on the formation of an ordered, highly dimensional
structure; as a result, an important goal in this field is to
gain an understanding of the factors that govern the supra-
molecular structure adopted for a particular coordination
polymer.16-19 Such control and knowledge would greatly
enhance the ability to design coordination polymers to
exhibit properties of interest.
metallic coordination polymers with Ni(II) and Cu(II) metal
centers as well as with hydroxide-bridged Cu(II) dimers.33-36
Instead, weakAu NC-Aucoordinatebondsfromterminal
3 3 3
N-cyano units to vacant axial sites of adjacent Au(III) atoms
were observed in several cases, resulting in five- or six-
coordinate Au(III) centers.33-36 However, a systematic study
has not been previously performed to investigate whether or
not certain conditions favor the formation of these weak
This goal is often challenging because of the multitude of
factors that control packing of a solid-state material during
crystallization. Weak forces such as hydrogen bonding, π-π
interactions, weak coordinate bonds, and metallophilic inter-
actions collectively influence the structure of a coordination
polymer.20,21 With the same order of magnitude of strength
as a hydrogen-bond, Au(I) and Ag(I) metallophilic interac-
tions have been studied widely.22-26 Heterobimetallic coor-
dination polymers based on [M(CN)2]- (M = Ag, Au) have
been able to increase structural dimensionality through such
metallophilic interactions, resulting in dramatically different
structures than would occur without the interactions.20 It
is known that aurophilic interactions in [Au(CN)2]- salts
are favored by cations able to participate in hydrogen
bonding, either as complex cations or protonated organic
heterocycles.27-29
Metallophilicity has also been observed in cyanometallates
containing d8 metals such as [Pt(CN)4]2-, where metallo-
philic interactions have strongly influenced some struc-
tures.14,15,20,30-32 However, for the isoelectronic d8 Au(III)
analogue, [Au(CN)4]-, noAu(III)-Au(III) interactionshave
been observed in previously reported [Au(CN)4]- complexes
and coordination polymers; these include several heterobi-
Au NC coordinate bonds. The knowledge of whether self-
3 3 3
aggregation is observed for [Au(CN)4]- and if so, how it is
induced, would be invaluable in the design of coordination
polymers incorporating [Au(CN)4]-.
In this light, we prepared a series of [cation]nþ[Au(CN)4]n
double salts, with the cations judiciously chosen to span a
range of cation charge, shape, aromaticity, and ability to
participate in hydrogen bonding; as a result, the effect of
these factors on the self-aggregation of [Au(CN)4]- anions
could be examined. In addition to the structural studies,
selected compounds were analyzed by solid-state 15N CP-
MAS NMR spectroscopy in order to probe if a shift in
the characteristic 15N resonance of the Au-CN moiety was
observed upon N-cyano coordination to Au(III) via the
aforementioned interanionic Au NC interactions. Over-
3 3 3
all, thisstudy providesspectroscopic support fortheexistence
of and insight into this weak interaction.
Experimental Section
General Experimental Procedures. All reactions were per-
formed in air using purified solvents. trans-[Co(en)2Cl2]Cl
(en = 1,2-diaminoethane) was prepared as previously re-
ported.37 [Hdabco]Cl (dabco = 1,4-diazabicyclo[2.2.2]octane)
was prepared by stoichiometric addition of HCl to dabco in
water, followed by solvent evaporation. All other reagents were
used as received from commercial sources. Infrared spectra
were measured as KBr pellets on a Thermo-Nicolet Nexus 670
FT-IR spectrometer with a resolution of 1 cm-1. Microanalyses
(C, H, N) were performed at Simon Fraser University by
Mr. Frank Haftbaradaran using a Carlo Erba (Model 1106)
CHN analyzer.
(11) Katz, M. J.; Kaluarachchi, H.; Batchelor, R. J.; Bokov, A. A.; Ye,
Z. G.; Leznoff, D. B. Angew. Chem., Int. Ed. 2007, 46, 8804–8807.
(12) Katz, M. J.; Leznoff, D. B. J. Am. Chem. Soc. 2009, 131, 18435–
18444.
(13) Lefebvre, J.; Batchelor, R. J.; Leznoff, D. B. J. Am. Chem. Soc. 2004,
126, 16117–16125.
(14) Buss, C. E.; Anderson, C. E.; Pomije, M. K.; Lutz, C. M.; Britton, D.;
Mann, K. R. J. Am. Chem. Soc. 1998, 120, 7783–7790.
(15) Drew, S. M.; Janzen, D. E.; Buss, C. E.; MacEwan, D. I.; Dublin,
K. M.; Mann, K. R. J. Am. Chem. Soc. 2001, 123, 8414–8415.
(16) Batten, S. R.; Murray, K. S. Aust. J. Chem. 2001, 54, 605–609.
(17) Batten, S. R.; Robson, R. Angew. Chem., Int. Ed. 1998, 37, 1460–
1494.
(18) Dunbar, K. R.; Heintz, R. A. Prog. Inorg. Chem. 1997, 45, 283–391.
(19) Kitagawa, S.; Kitaura, R.; Noro, S. Angew. Chem., Int. Ed. 2004, 43,
2334–2375.
Caution! Although difficulties have not been experienced, per-
chlorate salts are potentially explosive and thus should be handled
with care and used only in small quantities.
Synthesis of [nBu4N][Au(CN)4] (1). Dropwiseadditionof tetra-
butylammonium bromide, [nBu4N]Br, (0.039 g, 0.12 mmol) in a
2 mL aqueous solution to a 2 mL aqueous solution of potassium
tetracyanoaurate, KAu(CN)4, (0.037 g, 0.11 mmol) resulted in a
colorless solution. Slow evaporation of the solution yielded
X-ray quality, colorless crystals of [nBu4N][Au(CN)4] after
two weeks, which were vacuum-filtered and air-dried. Yield:
0.043 g, 72%. Anal. Calcd. for C20H36N5Au: C 44.20%, H
6.68%, N 12.89%. Found: C 44.31%, H 6.68%, N 13.00%. IR
(KBr): 2996(m) 2970(st), 2959(st), 2930(m), 2874(m), 2832(w),
2735(w), ν(CN): 2183(vw), 1488(st), 1467(m), 1382(m), 1186(w),
(20) Katz, M. J.; Sakai, K.; Leznoff, D. B. Chem. Soc. Rev. 2008, 37,
1884–1895.
(21) Roesky, H. W.; Andruh, M. Coord. Chem. Rev. 2003, 236, 91–119.
(22) Bardaji, M.; Laguna, A. J. Chem. Educ. 1999, 76, 201–203.
€
(23) Pyykko, P. Angew. Chem., Int. Ed. 2004, 43, 4412–4456.
€
(24) Pyykko, P. Chem. Soc. Rev. 2008, 37, 1967–1997.
(25) Schmidbaur, H. Gold Bull. 2000, 33, 3–10.
(26) Schmidbaur, H.; Schier, A. Chem. Soc. Rev. 2008, 37, 1931–1951.
(27) Pham, D. M.; Rios, D.; Olmstead, M. M.; Balch, A. L. Inorg. Chim.
Acta 2005, 358, 4261–4269.
1152(mw), 1106(mw), 1053(mw), 1026(m), 879(m), 747(m) cm-1
.
Polycrystalline [nBu4N][Au(CN)4] can also be prepared with
an identical IR spectrum and elemental analysis by mixing a
40 mL aqueous solution of [nBu4N]Br (1.072 g, 3.33 mmol) with
a 60 mL aqueous solution of KAu(CN)4 (1.000 g, 2.94 mmol).
A white precipitate of [nBu4N][Au(CN)4] forms immediately
and was isolated by vacuum filtration. Yield: 1.484 g, 92.9%.
Synthesis of [AsPh4][Au(CN)4] (2). A 4 mL 50% v/v methanol/
water solution of tetraphenylarsonium chloride, [AsPh4]Cl,
(28) Stender, M.; Olmstead, M. M.; Balch, A. L.; Rios, D.; Attar, S.
Dalton Trans. 2003, 4282–4287.
(29) Stender, M.; White-Morris, R. L.; Olmstead, M. M.; Balch, A. L.
Inorg. Chem. 2003, 42, 4504–4506.
€
(30) Pyykko, P. Chem. Rev. 1997, 97, 597–636.
(31) Doerrer, L. H. Comm. Inorg. Chem. 2008, 29, 93–127.
(32) Angle, C. S.; Woolard, K. J.; Kahn, M. I.; Golen, J. A.; Rheingold,
A. L.; Doerrer, L. H. Acta Crystallogr. C 2007, 63, M231–M234.
(33) Katz, M. J.; Kaluarachchi, H.; Batchelor, R. J.; Schatte, G.; Leznoff,
D. B. Cryst. Growth Des. 2007, 7, 1946–1948.
(34) Katz, M. J.; Aguiar, P. M.; Batchelor, R. J.; Bokov, A. A.; Ye, Z.-G.;
Kroeker, S.; Leznoff, D. B. J. Am. Chem. Soc. 2006, 126, 3669–3676.
(35) Shorrock, C. J.; Jong, H.; Batchelor, R. J.; Leznoff, D. B. Inorg.
Chem. 2003, 42, 3917–3924.
(36) Vitoria, P.; Muga, I.; Gutierrez-Zorrilla, J. M.; Luque, A.; Roman,
P.; Lezama, L.; Zuniga, F. J.; Beitia, J. I. Inorg. Chem. 2003, 42, 960–969.
(37) Bailar, J. C. Inorg. Synth. 1946, 2, 222–225.

Article
Inorganic Chemistry, Vol. 50, No. 4, 2011 1267
(0.055 g, 0.12 mmol) was added to a 2 mL aqueous solution of
KAu(CN)4 (0.036 g, 0.11 mmol). Slow evaporation of the
solution yielded colorless, X-ray quality crystals of [AsPh4]-
[Au(CN)4] after 10 days, which were vacuum-filtered and air-
dried. Yield: 0.044 g, 92%. Anal. Calcd. for C28H20N4AsAu:
C 49.14%, H 2.95%, N 8.19%. Found: C 49.13%, H 2.88%, N
8.29%. IR (ATR): 3160(w), 3085(w), 3060(w), 3026(w), 2658(w),
2998(w), ν(CN): 2192(vw), 2180(vw), 1558(w), 1484(m), 1439(st),
1341(mw), 1310(mw), 1189(mw), 1162(w), 1081(st), 1023(mw),
Synthesis of [Ni(en)3][Au(CN)4]2 (7). To an 2 mL aqueous
solution of nickel(II) perchlorate, Ni(ClO4)2 6H2O, (0.035 g,
3
0.096 mmol) 3 mL of an aqueous ethylenediamine (en) stock
solution (0.1 M, 0.3 mmol) was added, resulting in a purple
solution. A 2 mL aqueous solution of KAu(CN)4 (0.066 g,
0.19 mmol) was added to this solution, which did not cause a
change in color or formation of a precipitate; the resultant
solution was covered with parafilm. After four days, large,
purple, X-ray quality crystals of [Ni(en)3][Au(CN)4]2 were re-
covered by vacuum-filtration and subsequently air-dried. Yield:
0.047 g, 58%. Anal. Calcd. for C14H24N14Au2Ni: C 19.99%,
H 2.88%, N 23.32%. Found: C 20.30%, H 2.69%, N 23.57%.
IR (ATR): 3341(m), 3291(m), 2960(w), 2895(w) ν(CN): 2185(w),
2176(w), 1604(m), 1579(m), 1467(w), 1327(w), 1286(w), 1098(m),
997(st), 922(w), 843(w) cm-1
.
Synthesis of [N(PPh3)2][Au(CN)4] (3). Dropwise addition
of bis(triphenylphosphine)iminium chloride, [N(PPh3)2]Cl,
(0.063 g, 0.11 mmol) in a 6 mL 67% v/v methanol/water solution
to a 2 mL aqueous solution of KAu(CN)4 (0.036 g, 0.11 mmol)
resulted in a white precipitate, which was dissolved upon further
addition of 20 mL methanol. X-ray quality colorless crystals of
[N(PPh3)2][Au(CN)4] formed during slow evaporation of the
solution (for about two weeks); they were then vacuum-filtered
and air-dried. Yield: 0.034 g, 54%. Anal. Calcd. for C40H30-
N5AuP2: C 57.22%, H 3.60%, N 8.34%. Found: C 57.22%, H
3.46%, N 8.20%. IR (ATR): 3060(w), 3017(w), 2995(w), ν(CN):
2179(vw), 1588(mw), 1480(mw), 1438(m), 1334(m), 1302(m),
1181(mw), 1159(w), 1114(st), 998(m), 968(w), 928(w), 854(w),
1023(st), 967(st) cm-1
.
Synthesis of [Co(NH3)6][Au(CN)4]3 4H2O (8). A 3 mL aque-
3
ous solution of hexaamminecobalt(III) chloride, [Co(NH3)6]Cl3,
(0.026 g, 0.14 mmol) was added to a 4 mL aqueous solution of
KAu(CN)4 (0.102 g, 0.300 mmol), which did not alter the color of
the initial orange solution. The solution was covered with para-
film and yielded orange, X-ray quality crystals of [Co(NH3)6]-
[Au(CN)4]3 4H2O after five weeks, which were then vacuum-
3
filtered and air-dried. Yield: 0.065 g, 57%. Anal. Calcd. for
C12H26N18Au3CoO4: C 12.68%, H 2.31%, N 22.19%. Found:
C 12.87%, H 2.06%, N 21.91%. IR (KBr): 3653(m), 3644(m),
3229(s,br), 3174(s,br), ν(CN): 2191(m), 2189(m), 1600(st), 1349(m),
753(m), 721(st), 691(st) cm-1
.
Synthesis of [Co(phen)3][Au(CN)4]2 (4). A 2 mL 50% v/v
methanol/water solution of 1,10-phenanthroline (phen) (0.036
g, 0.18 mmol) was added to a 3 mL 50% v/v methanol/water
1325(st), 1259(w), 987(w), 822(st), 473(m), 458(m) cm-1
.
Synthesis of [Hdabco][Au(CN)4] (9). Adding a 3 mL aqueous
solution of [Hdabco]Cl (0.032 g, 0.22 mmol) to a 5 mL aqueous
solution KAu(CN)4 (0.068 g, 0.20 mmol) resulted in a colorless
solution. The solution was covered with parafilm and colorless
X-ray quality crystals of [Hdabco][Au(CN)4] were deposited,
which were filtered and air-dried. Retention of the filtrate
and further slow evaporation yielded additional crystals of
[Hdabco][Au(CN)4]. Yield: 0.049 g, 59%. Anal. Calcd. for
C10H13N6Au: C 29.00%, H 3.16%, N 20.29%. Found: C
29.14%, H 3.00%, N 19.83%. IR (KBr): 3054(w), 2958(w),
2890(w), 2678(w), ν(CN): 2196(w), 2184(w), 2175(w), 1461(m),
1326(st), 1175(st), 1088(st), 1057(st), 1014(st), 941(m), 837(m),
solution of cobalt(II) perchlorate, Co(ClO4)2 6H2O, (0.023 g,
3
0.063 mmol) resulting in an orange solution. A 10 mL 50% v/v
methanol/water solution of KAu(CN)4 (0.068 g, 0.20 mmol)
was then added to the solution, which resulted in forma-
tion of orange crystals of [Co(phen)3][Au(CN)4]2 after 1 day,
which were then vacuum-filtered and air-dried. Yield:
0.055 g, 73%. Anal. Calcd. for C34H24N14Au2Co: C 43.98%,
H 2.01%, N 16.32%. Found: C 43.88%, H 1.78%, N 15.96%.
IR (KBr): 3436(m), 3128(w), 3067(m), 3020(w), 2943(w),
2836(w), ν(CN): 2199(w), 2190(w), 2185(w), 2180(w), 1626(m),
1587(m), 1519(st), 1426(st), 1343(m), 1225(m), 1144(m), 1104(m),
1024(w), 847(st), 786(m), 725(st), 644(m), 473(m), 458(st),
798(st), 772(m), 626(st), 602(st), 458(m) cm-1
.
419(st) cm-1
.
X-ray Crystallographic Analysis. Crystallographic data for all
compounds are collected in Table 1. X-ray crystallographic
analysis was performed by mounting crystals onto glass fibers
using epoxy adhesive and then placing them into a cold nitrogen
stream. The diffraction data was collected using a Bruker
SMART APEX II CCD area detector diffractometer positioned
6.0 cm from the crystal. The X-ray source was a monochro-
Synthesis of [Mn(terpy)2][Au(CN)4]2 (5). A 3 mL 50% v/v
methanol/water solution of 2,20;60,200-terpyridine (terpy) (0.024
g, 0.10 mmol) was added to a 2 mL 50% v/v methanol/water
solution of manganese(II) chloride, MnCl2 6H2O, (0.019 g,
3
0.052 mmol) resulting in a yellow solution. A 2 mL aqueous
solution of KAu(CN)4 (0.069 g, 0.20 mmol) was then added to
the solution without any color change. The yellow solution was
allowed to slowly evaporate and yellow, X-ray quality crystals
of [Mn(terpy)2][Au(CN)4]2 were deposited after six days, which
were then vacuum-filtered and air-dried. Yield: 0.024 g, 43%.
Anal. Calcd. for C38H22N14Au2Mn: C 40.62%, H 1.97%, N
17.45%. Found: C 40.62%, H 1.86%, N 17.07%. IR (KBr):
3113(w), 3062(w), 3054(w), 3031(w), 3021(w), ν(CN): 2188(w),
1572(m), 1595(st), 1475(m), 1450(st), 1318(m), 1248(m), 1162(m),
1115(m), 1108(m), 1099(st), 1086(st), 1014(m), 775(st), 651(m),
˚
mated Mo KR radiation (λ = 0.71073 A) from a rotating anode
with a mirror focusing apparatus operated at 1.2 kW (50 kV,
24 mA). The frames were collected with a scan width of 0.5° in ω
and were integrated with the Bruker SAINT software. Correc-
tions for absorption were made by SADABS.
The structure of each compound was solved using the Sir 92
routine and expanded using Fourier techniques within the
CRYSTALS38 software package. Hydrogen atoms were placed
geometrically, oriented with probable hydrogen bonds, and
refined with a riding model to link the shifts of the respective
carbon and nitrogen atoms. The crystal of 6 was of poor quality
and twinned and was therefore analyzed using twinning decon-
volution software. Diagrams were made using Ortep-3 (version
2.02)39 and POV-Ray (version 3.6.0).40
624(m), 458(st) cm-1
.
Synthesis of trans-[Co(en)2Cl2][Au(CN)4] (6). A 5 mL 80% v/v
methanol/water solution of trans-(bis-ethylenediamine) cobalt-
(III) chloride, trans-[Co(en)2Cl2]Cl, (0.030 g, 0.11 mmol) was
added to a 5 mL 80% v/v methanol/water solution of KAu-
(CN)4 (0.035 g, 0.10 mmol). The solution was covered with
parafilm. After four days, X-ray quality, green crystals of
trans-[Co(en)2Cl2][Au(CN)4] had formed, and were vacuum-
filtered and air-dried. Yield: 0.030 g, 53%. Anal. Calcd. for
C8H16N8AuCl2Co: C 17.44%, H 2.93%, N 20.33%. Found: C
17.65%, H 2.58%, N 20.17%. IR (KBr): 3443(m), 3301(st),
3281(st), 3237(st), 3135(m), 3126(m), 3001(w), 2986(w), 2959-
(mw), ν(CN): 2186(mw), 1584(st), 1450(w), 1282(m), 1218(st),
Nitrogen-15 CP MAS NMR. 15N {1H} cross-polarization
(CP) magic-angle spinning (MAS) NMR was acquired on a
(38) Betteridge, P. W.; Carruthers, J. R.; Cooper, R. I.; Prout, K.;
Watkin, D. J. J. Appl. Crystallogr. 2003, 36, 1487–1487.
(39) Farrugia, L. J. J. Appl. Crystallogr. 1997, 5, 565–565.
(40) Fenn, T. D.; Ringe, D.; Petsko, G. A. J. Appl. Crystallogr. 2003, 36,
944–947.
1124(st), 1054(st), 1003(m), 794(w), 589(m) cm-1
.

1268 Inorganic Chemistry, Vol. 50, No. 4, 2011
Geisheimer et al.
Table 1. Crystallographic Data for Complexes 1-5
[nBu4N][Au(CN)4] (1) [AsPh4][Au(CN)4] (2) [N(PPh3)2][Au(CN)4] (3) [Co(phen)3][Au(CN)4]2 (4) [Mn(terpy)2][Au(CN)4]2 (5)
empirical formula
formula weight
crystal system
space group
C20H36N5Au
507.47
monoclinic
P21/n
14.6965(6)
8.4684(3)
19.6174(8)
90
105.485(2)
90
2352.87(16)
4
293
C28H20N4AsAu
684.38
triclinic
P1
10.5193(9)
10.5990(9)
12.7155(11)
96.057(4)
103.596(4)
106.681(4)
1297.0(2)
2
293
1.752
6.960
0.0232
0.0296
C40H30N5AuP2
839.63
monoclinic
P21/c
16.7592(5)
14.6440(4)
15.6117(4)
90
103.1830(10)
90
3730.48(18)
4
293
1.495
4.063
C44H24N14Au2Co
1201.64
triclinic
P1
11.2733(3)
11.3956(3)
16.4575(4)
78.6460(10)
79.2060(10)
83.8300(10)
2030.70(9)
2
293
1.965
7.663
0.0331
0.0405
C38H22N14Au2Mn
1123.55
triclinic
P1
12.925(3)
13.158(4)
13.497(4)
76.64(3)
62.548(18)
84.31(3)
1981.7(10)
2
293
1.883
7.745
0.0442
0.0622
˚
a (A)
˚
b (A)
˚
c (A)
R (deg)
β (deg)
γ (deg)
3
˚
V (A )
Z
T (K)
F
calcd (g/cm3)
1.534
6.265
0.0293
0.0368
μ (mm-1
)
R [Io g 2.5σ(Io)]
Rw [Io g 2.5σ(Io)]
0.0232
0.0274
Table 2. Summary of Crystallographic Data for Compounds 6-9
trans-[CoCl2(en)2][Au(CN)4] (6)
[Ni(en)3][Au(CN)4]2 (7)
[Co(NH3)6][Au(CN)4]3 4H2O (8)
3
[Hdabco][Au(CN)4] (9)
empirical formula
formula weight
crystal system
space group
C8H16N8AuCl2Co
551.07
monoclinic
P21/a
8.7523(5)
8.0191(5)
11.7635(8)
90
94.729(3)
90
822.82(9)
2
293
2.224
10.239
C14H24N14Au2Ni
841.08
monoclinic
P21/c
13.865(2)
12.345(2)
16.213(3)
90
112.002(2)
90
2573.0(8)
4
293
2.171
12.129
C12H25N18Au3CoO4
1135.28
orthorhombic
P212121
7.40440(10)
18.9411(3)
21.4134(3)
90
C20H26N12Au2
828.44
monoclinic
P21/c
7.1769(3)
15.5957(5)
12.3167(4)
90
104.269(2)
90
1336.06(8)
2
293
2.059
11.000
˚
a (A)
˚
b (A)
˚
c (A)
R (deg)
β (deg)
γ (deg)
90
90
3
˚
V (A )
Z
T (K)
3003.18(8)
4
293
2.511
15.203
F
calcd (g/cm3)
μ (mm-1
)
R [Io g 2.5σ(Io)]
Rw [Io g 2.5σ(Io)]
0.1011
0.1072
0.0246
0.0266
0.0220
0.0290
0.0267
0.0314
Varian UNITYInova 600 (B0 = 14.1T) at 60.81 MHz, using a
3.2 mm double-resonance MAS probe (Varian-Chemagnetics)
with a spinning frequency of 10 000 ( 3 Hz and 22 μL (or 36 μL,
thin walled) ZrO2 rotors. CP experiments were acquired on
natural abundance samples with 7 to 10 ms contact times, a
aqueous solution containing two equivalents of KAu-
(CN)4. For compounds 1-5, once the solutions were
mixed, the solvents were allowed to slowly evaporate until
crystals suitable for X-ray analysis were deposited, typi-
cally taking 1-14 days.
1
4.2 μs excitation pulse on H (optimized for the Hartmann-
Thecrystal structuresof compounds1-4 show standard,
unremarkable bond lengths and angles. The [Au(CN)4]-
anions are well-separated from one another in these
compounds; no Au-Au or Au-N interactions are pre-
sent since the shortest Au-Au distances range from 4.323
Hahn match), optimized recycle delays between 5 to 40 s, and
8k to 50k coadded transients. Chemical shifts are reported
with respect to saturated ammonium nitrate (NH4þ, 0.0 ppm),
using 15N-enriched solid glycine (11.59 ppm) as a secondary
reference.41,42
˚
to 8.249 A (well outside the range of Au-Au metallo-
Results
philic interactions) while the shortest Au N distances
˚
range from 3.266(2) (in 2) to 6.137(4) A (Table 3). These
3 3 3
Synthesis and Structural Characterization. [nBu4N]-
[Au(CN)4] (1), [AsPh4][Au(CN)4] (2), [N(PPh3)2][Au(CN)4]
(3), [Co(phen)3][Au(CN)4]2 (4), [Mn(terpy)2][Au(CN)4]2
(5). To prepare crystals of 1, 2, and 3, aqueous solutions
of [nBu4N]Br, [AsPh4]Cl, or [N(PPh3)2]Cl, respectively,
were added to an aqueous solution of K[Au(CN)4].
Crystals of 4 were prepared by mixing 50% MeOH/
Au N distances are at the limit of or are longer than the
sum of the van der Waals radii of Au and N: 3.12-
3 3 3
43
3.27 A. Only for 5 is weak aggregation of [Au(CN)4]-
˚
anions through very long Au N interactions ob-
served: weak trans-axial coordination of Au(2) through
3 3 3
Au-N interactions results in [Au(CN)4]- trimers
H2O solutions of phen and Co(ClO4)2 6H2O, followed
3
˚
(Au(2)-N(11) = 3.228(17) A; Table S1) and an octahe-
dral Au(III) center (Supporting Information Figure S1).
by addition of two equivalents of KAu(CN)4. Crystals
of 5 were prepared by mixing 50% MeOH/H2O solutions
This Au N interaction is also longer than for most
3 3 3
of terpy and MnCl2 6H2O, followed by addition of an
3
previously reported compounds, which have been in the
35
˚
range of 2.963(13)-3.3035(8) A.
(41) Hayashi, S.; Hayamizu, K. Bull. Chem. Soc. Jpn. 1991, 64, 688–690.
(42) Shoji, A.; Ozaki, T.; Fujito, T.; Deguchi, K.; Ando, S.; Ando, I.
J. Am. Chem. Soc. 1990, 112, 4693–4697.
(43) Bondi, A. J. Phys. Chem. 1964, 68, 441–451.

Article
Inorganic Chemistry, Vol. 50, No. 4, 2011 1269
Table 3. Comparison of Compounds 1-9, Their Minimum Au-Au Distances, Au-N Distances, and Observed νCN IR Bands
˚
˚
compound
minimum Au-Au distances (A)
minimum Au-N distances (A)
6.137(4)
3.266(2)
5.665(5)
3.357(6)
3.228(17)
3.072(13)
observed νCN bands in IR (cm-1)
[nBu4N][Au(CN)4] (1)
[AsPh4][Au(CN)4] (2)
[N(PPh3)2][Au(CN)4] (3)
[Co(phen)3][Au(CN)4]2 (4)
[Mn(terpy)2][Au(CN)4]2 (5)
trans-[Co(en)2Cl2][Au(CN)4] (6)
[Ni(en)3][Au(CN)4]2 (7)
8.118
4.342
8.249
4.815
4.963
5.935
4.401
4.975, 5.099
6.158
2183
2192, 2181, 2179
2179
2199, 2190, 2185, 2180
2188
2186
2185, 2176
2191, 2189
2196, 2184, 2175
3.056(5), 3.082(5)
3.004(6), 3.010(6), 3.022(6), 3.052(7)
2.958(6)
[Co(NH3)6][Au(CN)4]3 (H2O)4 (8)
[Hdabco][Au(CN)4] (9)
3
Figure 1. (a) Asymmetric unit of 6 showing the chain formed by hydrogen bonding between N(11) H(21)*;N(2)* and (b) the view of the solid state
3 3 3
crystal viewed down the a-axis showing layering that occurs between [trans-Co(en)2Cl2]þ and [Au(CN)4]- ions. Hydrogen atoms are removed for clarity.
[trans-Co(en)2Cl2][Au(CN)4] (6). A solution of metha-
nolic trans-[Co(en)2Cl2]Cl mixed with methanolic KAu-
(CN)4 deposited green crystals of 6 after four days. The
IR spectrum showed a νCN peak at 2186 cm-1, slightly
shifted from the 2183 cm-1 of [nBu4N][Au(CN)4] (1).
The crystal structure of 6 contains a cobalt(III) cation
octahedrally coordinated by chloride and en ligands in a
trans fashion (Figure 1a). Selected bond lengths and
angles are listed in Supporting Information Table S2.
Hydrogen bonding occurs between the [Co(en)2Cl2]þ and
[Au(CN)4]- ions, such that the N-cyano end of CN(11)
acts as a hydrogen bond acceptor and N(2) acts as
a hydrogen bond donor (N(11) H(21)*-N(2)* =
3 3 3
Figure 2. View of the (4,4) [Au(CN)4]- net formed by crystallographi-
˚
2.997(17) A; C(11)-N(11)-N(2)* = 119.6(9)°). This
˚
cally equivalent Au N interactions of 3.072(13) A in 6. trans-[Co-
length is consistent with hydrogen bonds previously ob-
served in a similar structural study investigating hydrogen
bonding between [M(NH3)6][Au(CN)2]x salts, which were
3 3 3
(en)2Cl2]þ cations have been removed for clarity.
gold-gold interactions are observed in this structure;
hydrogen bonding between the cations and nitriles of the
27
˚
typically slightly greater than 3.0 A. Each en ligand is
involved in formation of these hydrogen bonds, which are
propagated infinitely throughout the crystal and aid the
formation of a supramolecular structure where trans-
[Co(en)2Cl2]þ and [Au(CN)4]- ions form separate layers
(Figure 1b). This type of structure encourages self-aggre-
gation of [Au(CN)4]- anions because they are not physi-
cally isolated from one another, as found in 1-5.
anions, and Au NC interactions contribute to cation/
anion and anion/anion aggregation, respectively.
3 3 3
Some five- and six-coordinate Au(III) complexes are
known, although Au(III) is normally square planar.44
Five-coordinate Au(III) complexes45,46 typically incorpo-
rate a chelating ligand exhibiting some rigidity, such as
2,20-bipyridine, phenanthroline, or 2,20:60:200-terpyridine,
which allows one pyridyl group to coordinate to a basal
site on the Au(III) center while the other pyridyl group
Indeed, weak Au NC coordinate bonds are observed
3 3 3
between N(12) and Au(1)‡ such that the Au atom is octa-
hedral, but exhibiting a great deal of axial distortion
‡
‡
˚
(Au(1) -N(12) = 3.072(13) A; Au(1) -N(12)-C(12) =
145.4(13)°). These N-cyano-gold interactions are similar
in length to previously described compounds.35 The
N-cyano ends of C(12)†-N(12)† coordinate to adjacent
gold atoms, forming a 2-D (4,4) net motif (Figure 2).17 No
(44) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; 3rd ed.;
Interscience: New York, 1972.
(45) Kilpin, K. J.; Henderson, W.; Nicholson, B. K. Inorg. Chim. Acta
2009, 362, 5080–5084.
(46) Sommerer, S. O.; MacBeth, C. E.; Jircitano, A. J.; Abboud, K. A.
Acta Crystallogr. C 1997, 53, 1551–1553.

1270 Inorganic Chemistry, Vol. 50, No. 4, 2011
Geisheimer et al.
Figure 3. (a) Asymmetric unit of 7 and (b) the 1-D chain formed by intermolecular Au N interactions. Hydrogen atoms are omitted for clarity.
3 3 3
coordinates to the apical site.47-51 A six-coordinate Au-
(III) complex based on a macrocyclic ligand with four
equatorial nitrogen donors and axial chloride ligands has
been reported; a great disparity in bond lengths is ob-
served between the equatorial and axial ligands (also
The asymmetric unit of 8 consists of an octahedral hexa-
amminecobalt(III) cation, three [Au(CN)4]- units, and
four cocrystallized H2O molecules (Figure 4a). The bond
lengths and angles for the [Au(CN)4]- anions are typical
(Supporting Information Table S4).
˚
seen here in 6), which average 2.049(8) and 3.097(4) A,
Viewing the supramolecular structure along the a-axis
reveals alternating layers of [Co(NH3)6]3þ and [Au-
(CN)4]- ions with the four water molecules lying within
the cationic layer (Figure 4b), similar to that observed in
respectively.52
[Ni(en)3][Au(CN)4]2 (7). Mixing an aqueous solution of
[Ni(en)3]2þ with [Au(CN)4]- resulted in the deposition of
large purple crystals of 7. The IR spectrum of 7 shows νCN
6. Within the [Au(CN)4]- layer in 8, Au N interactions
3 3 3
peaks at 2185 and 2176 cm-1
.
are observed between the adjacent [Au(CN)4]- anions.
There are four crystallographically unique, relatively
short Au N interactions, generating octahedral coor-
The X-ray crystal structure of 7 shows an octahedrally
coordinated Ni(II) cation, bound by three bidentate en
ligands with a Λ stereochemistry and [Au(CN)4]- anions
(Figure 3a). The Ni(1)-N, Au-C, and C-N bond
lengths are unremarkable (Supporting Information Table
S3). In the supramolecular structure there is an absence
of hydrogen bonding despite the ability of the en amino
groups to act as hydrogen bond donors. However, inter-
3 3 3
dination of Au(2), while Au(1) and Au(3) each display
square pyramidal coordination (Au(1)-N(24) = 3.052(7)
0
˚
˚
A; Au(2)-N(13) = 3.010(6) A, Au(2)-N(31) = 3.022(6);
‡
˚
Au(3)-N(34) = 3.004(6) A). This weak axial coordina-
tion has little structural impact on the C-bound CN-
ligands in the equatorial plane. Figure 5 shows the network
of aggregated [Au(CN)4]- anions.
molecular Au N interactions are observed between
Au(2) and N(11)* as well as between Au(1) and N(21)0
3 3 3
Hydrogen bonding plays a large role in the packing of
8; terminal N-cyano units act as hydrogen bond acceptors
to both coordinated amine ligands and interstitial water
molecules. Hydrogen bonding also facilitates the layering
between cations and anions; the terminal N-cyano units
0
˚
˚
(Au(2)-N(11)*=3.056(5) A; Au(1)-N(21) =3.082(5) A).
These Au N interactions cause both Au(1) and Au(2)
3 3 3
to be five-coordinate, adopting a square pyramidal geo-
metry. However, the apical coordination of both Au(1)
and Au(2) only causes slight distortion of the basal CN-
ligands away from planarity for Au(1) and almost no
distortion about Au(2) (C(1)-Au(1)-C(3) = 175.39(19)°,
C(2)-Au(1)-C(4) = 177.7(2)°; C(5)-Au(2)-C(7) =
all participate in either hydrogen bonds, short Au
interactions, or weak Au N contacts slightly longer
N
3 3 3
3 3 3
˚
than 3.27 A.
[Hdabco][Au(CN)4] (9). Mixing aqueous solutions of
[Hdabco]Cl and K[Au(CN)4] resulted in crystals of 9. The
IR spectrum showed the presence of C-H bonds attri-
butable to the dabco ligand and νCN modes at 2196 and
2184 cm-1. The crystal structure of 9 consists of the
[Hdabco]þ cation and two [Au(CN)4]- anions (Figure 6a).
A hydrogen bond is present between the protonated
amine of the [Hdabco]þ cation, N(2), and the N(11)-
179.1(2)°, C(6)-Au(2)-C(8)=179.7(3)°). These Au
N
3 3 3
interactions facilitate formation of [Au(CN)4]- 1-D
chains in the supramolecular structure, as seen in
Figure 3b. The 1-D chains of [Au(CN)4]- anions are
separated from one another by [Ni(en)3]2þ cations.
[Co(NH3)6][Au(CN)4]3 4H2O (8). The reaction of aqu-
3
eous [Co(NH3)6]3þ and [Au(CN)4]- resulted in an orange
solution that yielded orange crystals of 8. The IR spec-
cyano unit of Au(1) (Figure 6, N(11) H(23)-N(2) =
trum of 8 has two νCN bands at 2189 and 2191 cm-1
.
3 3 3
˚
2.869(5) A, C(11)-N(11) H(23)-N(2) = 162.3(3)°).
3 3 3
This hydrogen bond is substantially shorter than those
˚
seen in 6 (2.997(17) A). The crystallographically equiva-
(47) Cinellu, M. A.; Zucca, A.; Stoccoro, S.; Minghetti, G.; Manassero,
M.; Sansoni, M. J. Chem. Soc., Dalton Trans. 1996, 4217–4225.
(48) Ferretti, V.; Gilli, P.; Bertolasi, V.; Marangoni, G.; Pitteri, B.;
Chessa, G. Acta Crystallogr. C 1992, 48, 814–817.
(49) Marangoni, G.; Pitteri, B.; Bertolasi, V.; Gilli, G.; Ferretti, V.
J. Chem. Soc., Dalton Trans. 1986, 1941–1944.
(50) O’Connor, C. J.; Sinn, E. Inorg. Chem. 1978, 17, 2067–2071.
(51) Vicente, J.; Chicote, M. T.; Bermudez, M. D.; Jones, P. G.; Fittschen,
C.; Sheldrick, G. M. J. Chem. Soc., Dalton Trans. 1986, 2361–2366.
(52) Suh, M. P.; Kim, I. S.; Shim, B. Y.; Hong, D.; Yoon, T. S. Inorg.
Chem. 1996, 35, 3595–3598.
lent trans N(11)-C(11) unit also participates in an equi-
valent hydrogen bond, such that each Au(1) [Au(CN)4]-
anion bridges two Hdabcoþ cations through hydrogen
bonding. Selected bond lengths and angles are collected in
Supporting Information Table S5.
In addition to the hydrogen bonding, two [Hdabco]þ
cations coordinate to the vacant axial sites of Au(1)

Article
Inorganic Chemistry, Vol. 50, No. 4, 2011 1271
Figure 4. (a) Asymmetric unit of [Co(NH3)6][Au(CN)4]3 4H2O (8) and (b) a view of the packing in the crystal structure, while looking down the a-axis,
3
showing layers of [Au(CN)4]- anions alternating with layers of [Co(NH3)6]3þ cations and water molecules. Hydrogen atoms in b are removed for clarity.
Figure 5. Network of [Au(CN)4]- anions formed by Au N interactions in 8.
3 3 3
through an unprotonated amine, N(1). Thus, Au(1) is for-
mally octahedral, with a large axial elongation (Au(1)-
sum of the van der Waals radii) in the X-ray crystal
structure, then the interaction is considered to be signifi-
cant. Given the limitations of such an analysis, especially
00
˚
N(1) = 2.932(2) A). It is worth noting that this Au
N
3 3 3
interaction is shorter than the Au NC-Au interac-
when the interaction in question (Au NC) has rela-
3 3 3
3 3 3
tions observed in 6-8, suggesting that the alkyl amine
group of the [Hdabco]þ ligand is more Lewis-basic than
the terminal N-cyano groups of [Au(CN)4]-. The hydro-
gen bonds and long gold-amine coordinate bonds pro-
pagate chains of [Hdabco]þ and [Au(CN)4]- in two
dimensions, thereby forming a (4,4) net (Figure 6b).17 The
anion containing Au(2) lies in the open spaces formed by
this net and does not participate in hydrogen bonding
or Au-N interactions. Adjacent nets pack with weak
van der Waals forces; there do not appear to be any
tively few crystallographic examples overall, independent
(nonstructural) spectroscopic evidence for the signifi-
cance of such an interaction is invaluable. With this in
mind, Table 3 compares 1-9 by their shortest Au-Au
distances, Au N distances (or length of Au N inter-
3 3 3
3 3 3
actions, where appropriate), and observed νCN bands in
the infrared spectra. IR spectroscopy is normally very
useful in identifying bridging of cyanometallate-contain-
ing compounds based on red- or blue-shifting of the νCN
mode.18 KAu(CN)4 has a νCN band at 2189 cm-1, while 1
and 3 have weak νCN resonances at 2183 and 2179 cm-1
respectively; these latter compounds have only well-iso-
lated N-cyano units (crystallographically), suggesting
that any substantial shifting from these values could
provide a good spectroscopic signature for bridging,
gold-gold or Au N interactions supporting packing
between layers.
3 3 3
Correlation of Structural and IR Spectral Features. The
identification and characterization of a weak intermolec-
ular bonding interaction in coordination polymers often
relies predominantly on a distance argument, that is, if the
atoms in question lie within a certain threshold (e.g., the
intermolecular Au NC units. Curiously, 2 and 4 also
3 3 3
both show νCN modes at roughly 2180 cm-1, along with

1272 Inorganic Chemistry, Vol. 50, No. 4, 2011
Geisheimer et al.
Figure 6. (a) The asymmetric unit of 9. In (b), [Au(CN)4]- and Hdabcoþ ions form an extended network through hydrogen bonding and Au
interactions.
N
3 3 3
additional blue-shifted peaks that suggest that some
N-cyano units are bridging rather than terminal, despite
X-ray evidence to the contrary. Furthermore, the IR
spectrum of 5 has a single νCN mode at 2188 cm-1, the
blue-shifting of which normally suggests that the N-cyano
units in this structure may be exclusively bridging to
another metal. Even though it should be recognized that
the νCN values for compounds 1-5 are all between the
fairly small range of 2179-2188 cm-1 (Table 3), overall,
there appears to be no trend in the νCN values that
correlates readily with Au NC distances.
3 3 3
The apparent contradictions between the structural
features of 1-5 and their IR spectra show that it is dif-
ficult to make structural assertions based on the IR
spectra for [Au(CN)4]--containing compounds. Thus,
Figure 7. Approximate 15N NMR chemical shift regions for cyanide
the νCN values appear to be a poor indicator of the
presence or absence of N-cyano bridging between metals;
this may be partly because of the very low intensity of
the observed bands. It is worth noting that the weak
peaks inherent to [Au(CN)4]-, as previously noted by
others,35,36 confound IR analysis such that identification
of peaks is much more difficult than in cyanometallates
with strongly absorbing νCN bands such as [Au(CN)2]- or
groups from this and previous studies.34,54-58
close proximity to an electron-donating atom (often other
metals, e.g., Zn, Hg, Au), occur in an intermediate 15N
chemical shift range of 225-255 ppm. N-bound terminal
cyanides have the lowest chemical shifts. For example,
AuNC has a 15N chemical shift of ∼200 ppm.54 Figure 7
illustrates that these regions can be used as a rough
“fingerprint” for determining cyanide speciation.34,54-58
15N{1H} CP-MAS NMR of the natural abundance
samples reveals well-defined, ordered cyanide groups in
the six diamagnetic compounds (1-3, 6, 8, 9, see Table 4)
with chemical shifts in the range expected for C-bound
ligands (320-260 ppm).55,61 Compounds 1, 2, and 3 each
[Pt(CN)4]2- 18,53 In this light, we turned to solid-state 15
N
.
MAS NMR as a potential tool to shed spectroscopic
insight on the Au NC interactions that were observed
crystallographically.
3 3 3
Nitrogen-15 CP MAS NMR. The nitrogen-15 isotropic
chemical shift has been shown to be sensitive to three
types of cyanide interactions observed in metal-cyanide
complexes: C-bound terminal, bridging, and N-bound
terminal cyanide ligands.54,55 Terminal C-bound cyanide
ligands are observed in the high 15N chemical shift range
from 260 to 320 ppm.55 Bridging cyanides, where the
nitrogen on the cyanide ligand (i.e., the N-cyano unit) is in
(56) Kroeker, S.; Wasylishen, R. E.; Hanna, J. V. J. Am. Chem. Soc. 1999,
121, 1582–1590.
(57) Bryce, D. L.; Wasylishen, R. E. Inorg. Chem. 2002, 41, 4131–4138.
(58) Al-Maythalony, B. A.; Fettouhi, M.; Wazeer, M. I. M.; Isab, A. A.
Inorg. Chem. Commun. 2009, 12, 540–543.
(59) Juranic, N.; Lichter, R. L. J. Am. Chem. Soc. 1983, 105, 406–410.
(60) Gobetto, R.; Nervi, C.; Valfre, E.; Chierotti, M. R.; Braga, D.;
Maini, L.; Grepioni, F.; Harris, R. K.; Ghi, P. Y. Chem. Mater. 2005, 17,
1457–1466.
(61) Pines, A.; Gibby, M. G.; Waugh, J. S. J. Chem. Phys. 1973, 59, 569–
590.
(53) Nakamoto, K. Infrared and Raman Spectra of Inorganic and Co-
ordination Compounds; 5th ed.; John Wiley & Sons: New York, 1997.
(54) Harris, K. J.; Wasylishen, R. E. Inorg. Chem. 2009, 48, 2316–2332.
(55) Schwarz, P.; Siebel, E.; Fischer, R. D.; Davies, N. A.; Apperley,
D. C.; Harris, R. K. Chem.;Eur. J. 1998, 4, 919–926.

Article
Inorganic Chemistry, Vol. 50, No. 4, 2011 1273
Table 4. 15N NMR Isotropic Chemical Shifts for Au(CN)4- Salts
15N chemical shifts
((2 ppm, unless otherwise indicated)
nitrogen site
(number of crystallographically inequivalent sites)
cmpd
name
1
2
3
6
8
9
[nBu4N][Au(CN)4]
[AsPh4][Au(CN)4]
[N(PPh3)2][Au(CN)4]
[Co(en)2Cl2][Au(CN)4]
[Co(NH3)6][Au(CN)4]3 (H2O)4
275
270
272
264.3(5)
258
270(1)
46.4a
CN(4)
CN(4)
CN(4)
CN(2)
CN(12)
CN(4)
Bu4N(1)
PPN(1)
en(2)
NH3(6)
Hdabco(2)
257.2(5)
250(1)
-30b
3
[Hdabco][Au(CN)4]
19c
-12.7c
a nBu4N.55 b en.59 c Hdabco.60
the chemical shifts are at or above 270 ppm. In 2, there is a
˚
range of Au-N distances from the relatively short 3.27 A,
through the intermediate distance of 4.12 A, to two dis-
˚
˚
tances greater than 9 A. The corresponding peak is broad,
with greatest intensity around 270 ppm and significant
intensity as low as 260 ppm, implying that the shortest
Au-N distances are shifted to lower frequency. The non-
H-bonded peak in 6hasa veryshort Au-N distance of 3.07
˚
A and its shift of 264 ppm is quite low relative to the others.
Likewise, 9 shows evidence of two distinct resonances
around 270 ppm: the peak with the strongest intensity
appears at the highest frequency and likely corresponds to
the two cyanides with the longest Au-N distances (4.93
˚
and 5.25 A), while the lower frequency shoulder with half
the intensity can be assigned to the cyanide with a Au-N
distance of 3.51 A. Compound 8 has 12 inequivalent
˚
cyanides with a wide range of Au-N distances, but its
shift(s) is dominated by extensive hydrogen-bonding.
Hence, while the primary influence on 15N chemical shifts
in this system is the well-known effect of hydrogen-bonding,
it appears that closer proximity to Au shifts the 15N peaks to
low frequency, a correlation which can be used for peak
assignments and possibly for the determination of unknown
structures. In a general sense, it is clear that, for the first
time, there is a real, spectroscopically measurable electronic
effect associated with the crystallographic observation of
Figure 8. 15N chemical shifts of the cyanide region for diamagnetic
[Au(CN)4]- salts.
feature a single broad peak between 260 and 275 ppm,
representing the overlap of unresolved crystallographi-
cally inequivalent cyanides in structurally similar term-
inal environments (Figure 8). Compound 9 yields a
narrower composite peak at ca. 270 ppm constituting
75% of the signal, which can be assigned to the three
terminal cyanides. Likewise, 6 possesses two cyanide
species, one of which appears at 264 ppm and represents
the terminal cyanide. Compounds 6, 8, and 9 contain
H-bond donating ligands and crystallographic evidence
intermolecular Au NC interactions, thereby lending sup-
3 3 3
port for their chemical significance.
Discussion
Structural patterns in 1-9 agree with previously reported
[Au(CN)4]--based compounds,33-35 which indicate self-ag-
gregation of [Au(CN)4]- in the solid state appears to rely on
Au N rather than aurophilic interactions. This preference
of CN H-N bonding. Each of these spectra exhibits
3 3 3
3 3 3
may be due to the Lewis acidity of Au(III) relative to the
isoelectronic[Pt(CN)4]2- (for whichmetallophilic interactions
are common). In turn, the Lewis basic N-cyano groups may
bind to the Au(III) centers by coordination to vacant axial
sites. Another contributing factor is likely the fact that
(because of the higher oxidation state), the Au-based orbitals
that would be needed for Au(III)-Au(III) bonding are
contracted relative to the Pt(II) analogues. Indeed, only
very weak-field ligands are found to support Au(III)-based
metallophilicity: such Au(III)-Au(III) aurophilic interac-
tions have been observed only in [N(CH3)4][Au(N3)4] and
[Au(bipy)Cl2][AuBr4] (bipy = 2,20-bipyridine), with (long)
peaks at lower chemical shifts (250-257 ppm), consistent
with previous work indicating that hydrogen bonding
induces a shift to lower 15N frequencies.55,62 For example,
CN H-O bonding in a series cobalt cyanide compounds
3 3 3
reduces the shift range by about 15-20 ppm with respect to
terminal cyanides.55 Compounds 6 and 8 also contain
terminal cyanides without H-bonding which appear around
264 and 270 ppm, respectively. Hence, gold-bound cyanides
have a 15N chemical shift region between 260 and 275 ppm
that can be assigned to C-bound terminal ligands, while
those subject to N H-N bonding resonate in a lower
3 3 3
frequency chemical shift region of 250-257 ppm.
There also appears to be a relationship between the 15N
chemical shift and Au-N distances. In 1 and 3, where the
˚
Au-Au bond lengths of 3.507(3), 3.508(3), and 3.584(3) A
63,64
˚
for the former compound and 3.518(1) A for the latter;
˚
nearest Au-N distances are all very long (5.66-11.12 A),
(63) Hayoun, R.; Zhong, D. K.; Rheingold, A. L.; Doerrer, L. H. Inorg.
Chem. 2006, 45, 6120–6122.
€
€
(62) Lorente, P.; Shenderovich, I. G.; Golubev, N. S.; Denisov, G. S.;
(64) Klapotke, T. M.; Krumm, B.; Galvez-Ruiz, J. C.; Noth, H. Inorg.
Bunthowsky, G.; Limbach, H.-H. Magn. Reson. Chem. 2001, 39, S18–S29.
Chem. 2005, 44, 9625–9627.

1274 Inorganic Chemistry, Vol. 50, No. 4, 2011
Geisheimer et al.
both sets of distances are still relatively long compared to
those observed for Au(I), which are frequently shorter than
3.1 A.
This interpretation of the (minimal) effect of cation charge
on Au N interaction formation is also consistent with
3 3 3
20
˚
the series of structures that contain hydrogen bonding
In this study, the cations were judiciously chosen to
provide comparisons between [Au(CN)4]--containing com-
pounds based on the variations in shape, charge (þ1 or þ2)
and the presence or absence of groups with available
π-systems, as well as their abilities to participate in H-bonding.
Among all of these factors, from examining the summary of
cations (6-9) in that there is no trend in the Au
N
3 3 3
distances that would suggest that the presence or strength
of the observed Au N interactions depend on the rela-
3 3 3
tive number of [Au(CN)4]- anions.
Similarly, the shape and aromaticity of the cations appears
to play a minimal, if any, role in self-aggregation of [Au(CN)4]-
anions. Compounds 1-3 differ by the shape of, and ability to
π-π stack their aromatic groups in, their respective cations,
but nonetheless the [Au(CN)4]- anions are isolated from one
another in all three structures.
Au N distances in Table 3 and utilizing the sum of the van
3 3 3
˚
der Waals radii limit for Au and N (3.27 A) as a cutoff, it is
clear that Au NC interactions are much shorter when the
3 3 3
countercation is capable of participating in hydrogen bond-
ing with terminal N-cyano units of the [Au(CN)4]- anion.
Thus, compounds 6-9, where the cation is a hydrogen
Conclusions
The series of solid-state structures of [Au(CN)4]- contain-
ing salts illustrate that when N-cyano units bridge the central
Au(III) atom and another group (be it a metal cation or
H-bond), the Lewis acidity of the central Au(III) appears to
increase, which induces self-aggregation of the [Au(CN)4]-
bond donor, all exhibit significant Au N interactions in
3 3 3
˚
the solid state, ranging from 2.958(6) to 3.082(5) A (Table 3).
Most of these compounds (except [Ni(en)3][Au(CN)4]2) show
terminal N-cyano groups participating in hydrogen bonding
to the cation, which acts to withdraw electron density from
the [Au(CN)4]- molecule to some degree. Formation of this
hydrogen bonding may increase the Lewis acidity of the
anions by intermolecular Au NC interactions; no aurophi-
3 3 3
lic interactions are observed. The distances of the N-cyano
ligands to the vacant Au(III)-axial sites are long, but within
the sum of the van der Waals radii. In addition to the
crystallographic evidence, 15N CP MAS NMR data provide
Au(III) center, making intermolecular Au N interactions
3 3 3
more favorable.
In 9, hydrogen bonds are formed between the protonated
aliphatic amine of [Hdabco]þ and the terminal N-cyano
groups. As a result, long axial binding to the central Au(III)
atom is observed through the unprotonated aliphatic amine
from [Hdabco]þ. This appears to be a more favorable
solid corroboration in support of the existence of Au NC
3 3 3
interactions. The combination of the 15N chemical shifts
for CN- units presented herein and those of previously
published works reveal distinct ranges which may serve
to distinguish among various metal-cyanide interactions in
unknown structures. Thus, the designed incorporation of
hydrogen bond donor groups into the cation is a viable route
to increasing the dimensionality of [Au(CN)4]- based materi-
Au N interaction that would be formed by coordination
3 3 3
of a terminal N-cyano group, as the free [Au(CN)4]- mole-
cule remains in the interstitial spaces of the network. This
preference suggests that, unsurprisingly, the dabco amine
group is more basic than the N-cyano units of [Au(CN)4]-.
Previously synthesized coordination polymers employing
[Au(CN)4]- have also shown a number of Au-N interac-
als via Au N interactions, while the cationic charge, shape,
3 3 3
size, and aromaticity of the cation do not appear to have an
appreciable effect on the ability of [Au(CN)4]- molecules to
˚
tions ranging in length from 2.963(13) to 3.052(9) A, slightly
form Au N interactions.
shorter than in 6-9.35 Their presence can be rationalized in a
similar fashion: binding of the N-cyano group to a metal
cation redistributes the electron density from Au(III) to the
metal cation, thereby increasing the Lewis acidity of the
3 3 3
Acknowledgment. Weare grateful toNSERC of Canada
(D.B.L., S.K.), the World Gold Council GROW program
(D.B.L.), the Natural Resources Canada Internship pro-
gram (A.R.G.), and a Grant-in-Aid for Scientific Re-
search (A) (No. 17205008) from the Ministry of Educa-
tion, Culture, Sports, Science, and Technology of Japan
(K.S.) for financial support. V.K.M. is grateful to NSERC
for a postgraduate scholarship. Infrastructure support
from the Canada Foundation for Innovation and the
Manitoba Research Innovation Fund (S.K.) is gratefully
acknowledged. D.B.L. was generously supported by a
JSPS Long-term Invitation Fellowship for 10 months
while on sabbatical with his gracious host K.S.
Au(III) center and promoting intermolecular Au N inter-
3 3 3
actions. Thus, in addition to hydrogen bonding, it appears
that the generation of Au(III)-CN-M bridges in coordina-
tion polymers is also favorable to formation of intermolecu-
lar Au N interactions.
3 3 3
On the other hand, cationic charge does not appear to play
a role in inducing the formation of Au N interactions. This
3 3 3
assessment is based on comparing structures with steric,
noncoordinating, monovalent cations in 2 and 3 versus
bulky, divalent cations in 4 and 5. Au N interactions
3 3 3
are absent in all of these structures (with the exception of
the marginal ones in 5). This suggests that self-aggregation
is only very slightly influenced by cationic charge for
cations with a relatively uniform electronic topology,
such as those with large aromatic or aliphatic groups.
Supporting Information Available: Crystallographic data in
cif format for compounds 1-9, additional figures (5), and tables
of bond lengths/angles for 5-9. This material is available free of
charge via the Internet at http://pubs.acs.org.