Copper(I) cyanide networks: Synthesis, structure, and luminescence behavior. Part 2. Piperazine ligands and hexamethylenetetramine

Article abstract of DOI:10.1021/ic8005184

A variety of photoluminescent, and in some cases thermochromic, metal-organic networks of CuCN were self-assembled in aqueous reactions with amine ligands: (CuCN)₂(Pip) (1a), (CuCN)₂₀(Pip)7 (1b), (CuCN)₇(MePip)₂ (2), (CuCN)₂(Me₂PiP) (3a), (CuCN)₄(Me₂Pip) (3b), (CuCN)₇(EtPip) ₂ (4), (CuCN)₄(Et₂Pip) (5), (CuCN) ₃(BzPip)₂ (6a), (CuCN)₅(BzPiP)₂ (6b), (CuCN)₇(BzPip)₂ (6c), (CuCN)₄(BzPip) (6d), (CuCN)₂(Bz₂Pip) (7), (CuCN)(Ph₂CHPip) (8a), (CuCN)₂(Ph₂CHPiP) (8b), (CuCN)₃(HMTA) ₂ (9a), (CuCN)₅(HMTA)₂ (9b), and (CuCN) ₅(HMTA) (9c) (Pip = piperazine, MePip = N-methylpiperazine, Me ₂Pip = N,N′-dimethylpiperazine, EtPip = N-ethylpiperazine, Et₂PiP = N,N′-diethylpiperazine, BzPip = N-benzylpiperazine, Bz₂Pip = N,N′-dibenzylpiperazine, Ph₂CHPip = N-(diphenylmethyl)piperazine, and HMTA = hexamethylenetetramine). New X-ray structures are reported for 1b, 2, 3b, 4, 5, 6a, 6d, 7, 8b, 9b, and 9c. An important structural theme is the formation of (6,3) (CuCN) ₂(piperazine) sheets with or without threading of independent CuCN chains. Strong luminescence at ambient temperature is observed for all but complexes 6 and 7. All luminescent compounds show a broad emission band in the blue region at about 450 nm attributable to metal-to-ligand charge transfer behavior based on the large Stokes shift between excitation and emission maxima. 3,8, and 9 are thermochromic due to an additional lower energy emission band, which is absent at 77 K.

Full text of DOI:10.1021/ic8005184

Inorg. Chem. 2008, 47, 6931-6947
Copper(I) Cyanide Networks: Synthesis, Structure, and Luminescence
Behavior. Part 2. Piperazine Ligands and Hexamethylenetetramine1
Mi Jung Lim, Courtney A. Murray, Tristan A. Tronic, Kathryn E. deKrafft, Amanda N. Ley,
Jordan C. deButts, Robert D. Pike,* Haiyan Lu, and Howard H. Patterson*
Department of Chemistry, College of William and Mary, Williamsburg, Virginia 23187-8795, and
Department of Chemistry, UniVersity of Maine, Orono, Maine 04469-5706
Received March 21, 2008
A variety of photoluminescent, and in some cases thermochromic, metal-organic networks of CuCN were self-
assembled in aqueous reactions with amine ligands: (CuCN)2(Pip) (1a), (CuCN)20(Pip)7 (1b), (CuCN)7(MePip)2 (2),
(CuCN)2(Me2Pip) (3a), (CuCN)4(Me2Pip) (3b), (CuCN)7(EtPip)2 (4), (CuCN)4(Et2Pip) (5), (CuCN)3(BzPip)2 (6a),
(CuCN)5(BzPip)2 (6b), (CuCN)7(BzPip)2 (6c), (CuCN)4(BzPip) (6d), (CuCN)2(Bz2Pip) (7), (CuCN)(Ph2CHPip) (8a),
(CuCN)2(Ph2CHPip) (8b), (CuCN)3(HMTA)2 (9a), (CuCN)5(HMTA)2 (9b), and (CuCN)5(HMTA) (9c) (Pip ) piperazine,
MePip ) N-methylpiperazine, Me2Pip ) N,N′-dimethylpiperazine, EtPip ) N-ethylpiperazine, Et2Pip ) N,N′-
diethylpiperazine, BzPip ) N-benzylpiperazine, Bz2Pip ) N,N′-dibenzylpiperazine, Ph2CHPip ) N-(diphenylmeth-
yl)piperazine, and HMTA ) hexamethylenetetramine). New X-ray structures are reported for 1b, 2, 3b, 4, 5, 6a,
6d, 7, 8b, 9b, and 9c. An important structural theme is the formation of (6,3) (CuCN)2(piperazine) sheets with or
without threading of independent CuCN chains. Strong luminescence at ambient temperature is observed for all
but complexes 6 and 7. All luminescent compounds show a broad emission band in the blue region at about 450
nm attributable to metal-to-ligand charge transfer behavior based on the large Stokes shift between excitation and
emission maxima. 3, 8, and 9 are thermochromic due to an additional lower energy emission band, which is absent
at 77 K.
Chart 1
Introduction
We are currently studying the extensive metal-organic
network chemistry of copper(I) cyanide with bridging
nitrogen ligands (L).1–3 Because a single cyanide ligand can
bridge two, three, or even four metal centers, it serves as an
excellent network former. Porous networks have potential
applications in areas such as separations, gas storage, and
catalysis.4 In addition, highly luminescent metal-organic
materials might be used in sensing systems when molecular
absorption of small molecules leads to an alteration in
network luminescence behavior.5 Our recent work has shown
that diimine L networks of CuCN are at best weakly
luminescent.1 In the current study, we report the network
formation and luminescence behavior of CuCN with diamine
and tetramine ligands, in particular substituted piperazines
and hexamethylenetetramine (HMTA), as in Chart 1. Many
of the resulting networks are intensely luminescent.2 The
relationship between structure and luminescence is examined
herein. Some of this work has been previously communicated.2
In contrast to studies of diimine L with CuCN,1,3,6
relatively few diamine ligands have been used to extend the
CuCN network. Known networks are limited to the follow-
ing: (CuCN)2L (L ) Pip, Me2Pip, 1,4-butanediamine, and
N,N,N′,N′-tetramethylethylenediamine), and (CuCN)(N-phe-
nylpiperazine),7 as well as the tetramine complex
* To whom correspondence should be addressed. E-mail: rdpike@wm.edu,
Tel: 757-2212555, Fax: 757-2212715 (R.D.P.); E-mail: howardp@maine.edu,
Tel: 207-5811178, Fax: 207-5811191 (H.H.P.).
(1) Tronic, T. A.; deKrafft, K. E.; Lim, M. J.; Ley, A. N.; Pike, R. D.
Inorg. Chem. 2007, 46, 8897.
(2) Pike, R. D.; deKrafft, K. E.; Ley, A. N.; Tronic, T. A. Chem. Commun.
2007, 3732.
(3) Pagola, S.; Pike, R. D.; deKrafft, K. E.; Tronic, T. A. Acta Crystallogr.,
Sect. C 2008, 64, m132.
10.1021/ic8005184 CCC: $40.75  2008 American Chemical Society
Inorganic Chemistry, Vol. 47, No. 15, 2008 6931
Published on Web 07/02/2008

Lim et al.
Instruments Q500 in the dynamic (variable temperature) mode with
a maximum heating rate of 50 °C/min to 900 °C under 60 mL/min
N2 flow.
(CuCN)3(HMTA)2, in which HMTA behaves in a bidentate
fashion.8 In addition to the relative paucity of reported
diamine-CuCN networks is the fact that only 1:1, 3:2, and
2:1 CuCN/L stoichiometries have been encountered thus far.
Each of the known (CuCN)2L materials exhibits a simple
network of CuCN chains cross-linked by L. This produces
hexagonally tiled 2D sheets with 3-coordinate copper centers,
best described as a (6,3) network.9 Given the diverse array
of copper-rich networks identified for diimine L having
CuCN/L ratios as high as 4:1,1,6 it seemed likely that a variety
of copper-rich products would also be found using pipera-
zines and HMTA.
Ligand Syntheses. BzPip.10 A solution of benzyl chloride (10.0
g, 79.0 mmol) and Pip (27.2 g, 316 mmol) in 250 mL toluene was
heated to 85 °C for 2 h. The resulting yellow solution was filtered
and evaporated. The oily residue was extracted with 2 M HCl and
CH2Cl2 (50 mL each). The aqueous fraction was then basified to
ca. pH 14 with solid NaOH. It was then re-extracted with CH2Cl2
(2 × 100 mL). The organic layer was washed with brine and water
and then dried over anhydrous Na2SO4. Evaporation left a colorless
oil which was dried in vacuo (8.97 g, 50.9 mmol, 64.4%). 1H NMR
(400 MHz, CDCl3) δ 7.30 (m, 5H, Ph), 3.49 (s, 2H, CH2Ph), 2.88
(m, 2H CH2), 2.41 (br s, 2H, CH2), 1.61 (br s, 1H, NH). 13C{1H}
NMR (100 MHz, CDCl3) δ 138.22, 129.31, 128.31, 127.12, 63.91,
54.73, 46.34.
Experimental Section
Materials and Methods. All reagents were purchased from
Aldrich or Acros and used without purification, except for BzPip
and Bz2Pip, which were prepared as described below. All water
used was ultrafiltered deionized quality and was thoroughly
degassed with argon. Analyses for carbon, hydrogen, and nitrogen
were carried out by Atlantic Microlabs, Norcross, GA. Ambient
temperature-luminescence measurements were carried out on well-
ground powders using a PerkinElmer LS 55 spectrofluorimeter.
Steady-state photoluminescence at ambient and reduced temperature
were collected using a Photon Technology International Model
Quanta-Master-1046 spectrophotometer equipped with a 75 W
xenon lamp. Wavelengths were selected with two excitation
monochromators and a single emission monochromator. A model
LT-3-110-Heli-Trans cryogenic liquid transfer system equipped
with a temperature controller was used to record luminescence
spectra as a function of temperature. IR measurements were made
on KBr pellets using a Digilab FTS 7000 FTIR spectrophotometer.
Thermogravimetric analyses (TGA) were conducted using a TA
Bz2Pip.11 Solutions of Pip (0.315 g, 3.66 mmol) and freshly
distilled benzaldehyde (64.6 g, 6.09 mmol) dissolved in separate
10 mL portions of CH2Cl2 were combined, forming a clear, colorless
solution. Solid NaBH(OAc)3 (1.27 g, 5.97 mmol) was added, and
the resulting suspension was stirred for 4 h under N2. The mixture
was washed with saturated aq NaHCO3 (20 mL), and the aqueous
layer was re-extracted three times with 25 mL CH2Cl2. The organic
layers were combined and dried over anhydrous Na2SO4. Evapora-
tion left a white solid that was dried in vacuo (0.398 g, 49.1%). 1H
NMR (400 MHz, CDCl3) δ 7.30 (m, 10H, Ph), 3.51 (s, 4H, CH2Ph),
2.48 (s, 8H, CH2CH2). 13C{1H} NMR (100 MHz, CDCl3) δ 138.21,
129.56, 128.48, 127.33, 63.39, 53.35.
Network Syntheses. (CuCN)2(Pip), 1a. Copper(I) cyanide
(0.268 g, 3.00 mmol) and KCN (0.130 g, 2.00 mmol) were
suspended in 20 mL H2O. The mixture was heated under N2 and
Pip (0.0861 g, 1.00 mmol) was added. The suspension was
vigorously stirred at about 80 °C for 1 h. The suspended solid was
collected by means of filtration, washed with H2O, ethanol, and
diethyl ether, and then dried under vacuum. A pale-yellow powder
(4) (a) Muller, U.; Schubert, M.; Teich, F.; Puetter, H.; Schierle-Arndt,
K.; Pastre´, J. J. Mater. Chem. 2006, 16, 626. (b) Dincaj, M.; Yu, A. F.;
Long, J. R. J. Am. Chem. Soc. 2006, 128, 8904. (c) Rowsell, J. L. C.;
Yaghi, O. M. J. Am. Chem. Soc. 2006, 128, 1304. (d) Forster, P. M.;
Cheetham, A. K. Top. Catal. 2003, 24, 79. (e) Kesanli, B.; Lin, W.
Coord. Chem. ReV. 2003, 246, 305.
was isolated (1.37 g, 5.16 mmol, 51.6%). IR (KBr pellet, cm-1
)
3244 (w), 3153 (w), 3101 (m), 3038 (m), 2104 (s), 1575 (m), 1439
(m), 1405 (m), 1382 (m), 1103 (m), 990 (w), 968 (m), 867 (m).
Anal. Calcd for C6H10N4Cu2: C, 27.17; H, 3.80; N, 21.12. Found:
C, 27.48; H, 3.85; N, 21.73. TGA Calcd for CuCN: 67.5. Found:
68.0 (175-245 °C).
(5) (a) Exstrom, C. L.; Sowa, J. L., Jr.; Daws, C. A.; Janzen, D.; Mann,
K. R.; Moore, G. A.; Stewart, F. F. Chem. Mater. 1995, 7, 15. (b)
Daws, C. A.; Exstrom, C. L.; Sowa, J. L., Jr.; Mann, K. R. Chem.
Mater. 1997, 9, 363. (c) Vickery, J. C.; Olmstead, M. M.; Fung, E. Y.;
Balch, A. L. Angew. Chem, Int. Ed. 1997, 36, 1179. (d) Cariati, E.;
Bu, X.; Ford, P. C. Chem. Mater. 2000, 12, 3385. (e) Drew, S. M.;
Janzen, D. E.; Buss, C. E.; MacEwen, D. I.; Dublin, K. I.; Mann,
K. R. J. Am. Chem. Soc. 2001, 123, 8414. (f) Buss, C. E.; Mann,
K. R. J. Am. Chem. Soc. 2002, 124, 1031. (g) Ferna´ndez, E. J.; Lo´pez-
de-Luzuriaga, J. M.; Monge, M.; Olmos, M. E.; Pe´rez, J.; Laguna,
A.; Mohamed, A. A.; Fackler, J. P., Jr. J. Am. Chem. Soc. 2003, 125,
2022. (h) Lu, W.; Chan, M. C. W.; Zhu, N.; Che, C.-M.; He, Z.; Wong,
K.-Y. Chem.sEur. J. 2003, 9, 6155.
(CuCN)20(Pip)7, 1b.2 The procedure was identical to that used
for 1a, using 4.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
Pip. A cream powder was isolated (85.1%). IR (KBr pellet, cm-1
)
3246 (m), 2990 (w), 2911 (m), 2881 (m), 2857 (m), 2124 (s), 1449
(m), 1311 (s), 1277 (w), 1115 (m), 1099 (s), 990 (w), 870 (s), 624
(w). Anal. Calcd for C48H70N34Cu20: C, 24.08; H, 2.95; N 19.89.
Found: C, 24.27; H, 2.97; N, 20.05. TGA Calcd for CuCN: 74.8.
Found: 74.9 (185-240 °C).
(6) (a) Cromer, D. T.; Larson, A. C. Acta Crystallogr., Sect. B 1972, 28,
1052. (b) Chesnut, D. J.; Kusnetzow, A.; Zubieta, J. J. Chem. Soc.,
Dalton Trans. 1998, 4081. (c) Teichert, O.; Sheldrick, W. S. Z. Anorg.
Allg. Chem. 1999, 625, 1860. (d) Kuhlman, R.; Schimek, G. L.; Kolis,
J. W. Polyhedron 1999, 18, 1379. (e) Teichert, O.; Sheldrick, W. S.
Z. Anorg. Allg. Chem. 2000, 626, 1509. (f) Chesnut, D. J.; Plewak,
D.; Zubieta, J. J. Chem. Soc., Dalton Trans. 2001, 2567. (g) Hanika-
Heidl, H.; Etaiw, S. E. H.; Ibrahim, M. S.; El-din, A. S. B.; Fischer,
R. D. J. Organomet. Chem. 2003, 684, 329. (h) Greve, J.; Na¨ther, C.
Z. Naturforsch. 2004, 59b, 1325. (i) Hibble, S. J.; Chippindale, A. M.
Z. Anorg. Allg. Chem. 2005, 631, 542. (j) Jess, I.; Na¨ther, C. Acta
Crystallogr., Sect. E 2006, 62, m721. (k) Lin, J.-D.; Li, Z.-H.; Li, T.;
Li, J.-R.; Du, S.-W. Inorg. Chem. Commun. 2006, 6, 675.
(CuCN)7(MePip)2, 2. The procedure was identical to that used
for 1a, using 4.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
MePip. A white powder was isolated (63.0%). IR (KBr pellet, cm-1
)
3296 (m), 3233 (w), 2951 (w), 2866 (w), 2837 (m), 2819 (w), 2156
(m), 2145 (m), 2122 (s), 1450 (m), 1311 (m), 1283 (s), 1261 (w),
1125 (m), 1055 (w), 1032 (w), 1003 (m), 973 (w), 944 (w) 860
(s), 779 (w), 625 (w). Anal. Calcd for C17H24N11Cu7: C, 24.68; H,
2.92; N, 18.62. Found: C, 24.48; H, 2.88; N, 18.70. TGA Calcd
for CuCN: 75.9 Found 76.6 (175-230 °C).
(7) Stocker, F. B.; Staeva, T. P.; Rienstra, C. M.; Britton, D. Inorg. Chem.
1999, 38, 984.
(10) Capuano, B.; Crosby, I. T.; Lloyd, E. J.; Taylor, D. A. Aust. J. Chem.
2002, 55, 565.
(11) Denmark, S. E.; Fu, J. Org. Lett. 2002, 4, 1951.
(8) Stocker, F. B. Inorg. Chem. 1991, 30, 1472.
(9) Batten, S. R.; Robson, R. Angew. Chem., Int. Ed. 1998, 37, 1460.
6932 Inorganic Chemistry, Vol. 47, No. 15, 2008

(CuCN)2(Me2Pip), 3a. The procedure was identical to that used
for 1a, using 3.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
Me2Pip. A white powder was isolated (52.7%). IR (KBr pellet,
cm-1) 2968 (m), 2826 (m), 2816 (m), 2135 (s), 1456 (s), 1290
(m), 1165 (m), 1123 (w), 1095 (w), 1049 (w), 1011 (s), 907 (w),
814 (w). Anal. Calcd for C8H14N4Cu2: C, 32.76; H, 4.81; N, 19.10.
Found: C, 32.94; H, 4.91; N, 19.19. TGA Calcd for
(CuCN)4(Me2Pip): 80.5 (135-175 °C). Found: 80.7 (175-215 °C).
Calcd for CuCN: 61.1. Found: 62.0 (180-215 °C).
(CuCN)4(Me2Pip), 3b. The procedure was identical to that used
for 1a, using 4.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
Me2Pip. A white powder was isolated (59.1%). IR (KBr pellet,
cm-1) 2974 (m), 2947 (w), 2832 (m), 2820 (m), 2149 (m), 2127
(s), 1453 (s), 1367 (w), 1284 (s), 1151 (m), 1119 (m), 1089 (m),
1037 (m), 1005 (s), 910 (w), 817 (s), 633 (w). Anal. Calcd for
C10H14N6Cu4: C, 25.42; H, 2.99; N, 17.79. Found: C, 25.65; H,
3.01; N, 17.94. TGA Calcd for CuCN: 75.8. Found: 76.1 (180-220
°C).
(CuCN)4(BzPip), 6d. The procedure was identical to that used
for 1a, using 5.00 mmol CuCN, 3.00 mmol KCN, and 1.00 mmol
BzPip. An off-white powder was isolated (70.5%). IR (KBr pellet,
cm-1) 3026 (w), 2971 (w), 2930 (w), 2836 (m), 2162 (s), 2119 (s),
1446 (m), 1167 (m), 1096 (m), 961 (s), 857 (m), 768 (m), 745
(m), 702 (s). Anal. Calcd for C15H16N6Cu4: C, 33.71; H, 3.02; N,
15.72. Found: C, 33.22; H, 2.98; N, 15.64. TGA Calcd for CuCN:
67.0. Found: 67.8 (145-215 °C).
(CuCN)2(Bz2Pip), 7. The procedure was identical to that used
for 1a, using 2.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
Bz2Pip. A light-tan powder was isolated (33.0%). IR (KBr pellet,
cm-1) 3061 (w), 3026 (w), 2967 (w), 2874 (w), 2832 (m), 2805
(m) 2126 (s), 1492 (w), 1451 (s), 1399 (m), 1345 (m), 1265 (m),
1130 (m), 1114 (m), 997 (m), 937 (m), 833 (m), 754 (s), 702 (s).
Anal. Calcd for C20H22N4Cu2: C, 53.92; H, 4.98; N, 12.58. Found:
C, 53.87; H, 4.98; N, 12.50. TGA Calcd for CuCN: 40.2. Found:
40.6 (180-215 °C).
(CuCN)(Ph2CHPip), 8a. The procedure was identical to that
used for 1a, using 2.00 mmol CuCN, 2.00 mmol KCN, and 1.00
mmol Ph2CHPip. A white powder was isolated (36.4%). IR (KBr
pellet, cm-1) 3059 (w), 3026 (w), 2960 (m), 2850 (w), 2802 (w),
2134 (s), 2130 (s), 1490 (m), 1450 (m), 1318 (w), 1123 (m), 1072
(w), 1006 (w), 920 (w), 869 (s), 751 (s), 702 (s). Anal. Calcd for
C18H20N3Cu: C, 63.23; H, 5.90; N, 12.29. Found: C, 62.96; H, 5.83;
N, 12.43. TGA Calcd for (CuCN)2(Ph2CHPip): 63.1. Found: 63.3
(150-185 °C). Calcd for CuCN: 26.2. Found: 26.6 (185-215 °C).
(CuCN)2(Ph2CHPip), 8b. The procedure was identical to that
used for 1a, using 3.00 mmol CuCN, 2.00 mmol KCN, and 1.00
mmol Ph2CHPip. A white powder was isolated (67.4%). IR (KBr
pellet, cm-1) 3265 (w), 3026 (w), 2972 (w), 2808 (w), 2768 (w),
2738 (w), 2127 (s), 1490 (m), 1449 (s), 1342 (w), 1310 (m), 1129
(m), 1072 (m), 1007 (m), 851 (m), 752 (m), 706 (s). Anal. Calcd
for C19H20N4Cu2: C, 52.89; H, 4.67; N, 12.98. Found: C, 53.57; H,
4.66; N, 12.65. TGA Calcd for CuCN: 41.5. Found: 41.6 (145-240
°C).
(CuCN)7(EtPip)2, 4. The procedure was identical to that used
for 1a, using 4.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
EtPip. A white powder was isolated (65.7%). IR (KBr pellet, cm-1
)
3286 (w), 3236 (w), 2974 (w), 2836 (m), 2145 (m), 2120 (s), 1444
(w), 1315 (w), 1225 (m), 1094 (m), 1028 (w), 930 (m), 856 (m),
964 (w), 771 (w), 625 (w). Anal. Calcd for C19H28N11Cu7: C, 26.68;
H, 3.30; N, 18.01. Found: C, 27.38; H, 3.32; N, 17.72. TGA Calcd
for CuCN: 73.3. Found: 72.0 (180-230 °C).
(CuCN)4(Et2Pip), 5. The procedure was identical to that used
for 1a, using 5.00 mmol CuCN, 3.00 mmol KCN, and 1.00 mmol
Et2Pip. A white powder was isolated (89.0%). IR (KBr pellet, cm-1
)
2970 (m), 2928 (w), 2900 (w), 2879 (w), 2839 (m), 2152 (m), 2119
(s), 1462 (w), 1445 (w), 1380 (w), 1316 (w), 1269 (w), 1142 (m),
1103 (m), 1018 (m), 941 (m), 831 (w), 784 (w). Anal. Calcd for
C12H18N6Cu4: C, 28.80; H, 3.62; N, 16.79. Found: C, 28.76; H,
3.63; N, 16.81. TGA Calcd for CuCN: 73.3. Found: 72.1 (205-245
°C).
(CuCN)3(BzPip)2, 6a. The procedure was identical to that used
for 1a, using 2.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
BzPip. A pale yellow powder was isolated (40.3%). IR (KBr pellet,
cm-1) 3285 (m), 3326 (w), 3027 (w), 2957 (m), 2853 (m), 2809
(m), 2124 (s), 2111 (s), 1491 (w), 1452 (m) 1339 (w), 1316 (m),
1119 (m), 1071 (w), 989 (w), 862 (m), 746 (m), 701 (m). Anal.
Calcd for C25H32N7Cu3: C, 48.34; H, 5.19; N, 15.78. Found: C,
48.36; H, 5.13; N, 15.79. TGA Calcd for (CuCN)3(BzPip): 71.6.
Found: 72.4 (125-155 °C). Calcd for CuCN: 43.3. Found: 43.5
(155-220 °C).
(CuCN)3(HMTA)2, 9a. The procedure was identical to that used
for 1a, using 1.00 mmol CuCN, 1.00 mmol KCN and 1.00 mmol
HMTA. A white powder was isolated (30.0%). IR (KBr pellet,
cm-1) 2956 (w), 2945 (w), 2920 (w), 2875 (w), 2132 (m), 2114
(m), 1460 (m), 1368 (w), 1243 (m), 1231 (m), 1222 (m),
1048 (m), 1020 (m), 990 (s), 829 (m), 820 (m), 802 (m), 783 (w),
708 (m), 683 (m), 662 (m). Anal. Calcd for C15H24N11Cu3: C, 32.81;
H, 4.41; N, 28.06. Found: C, 32.87; H, 4.39; N, 28.06. TGA Calcd
for (CuCN)5(HMTA): 64.3. Found: 64.5 (215-270 °C). Calcd for
CuCN: 48.9. Found: 49.6 (290-335 °C).
(CuCN)5(HMTA)2, 9b. The procedure was identical to that used
for 1a, using 2.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
HMTA and heating for 5 h. A white powder was isolated (39.7%).
IR (KBr pellet, cm-1) 2995 (w), 2990 (w), 2956 (w), 2930 (w),
2881 (w), 2108 (m), 2098 (m), 2088 (m), 2072 (m), 1454 (m),
1366 (w), 1238 (s), 1223 (s), 1062 (m), 1026 (s), 1006 (s), 989 (s),
928 (w), 841 (m), 808 (m), 701 (m), 670 (m). Anal. Calcd for
C17H24N13Cu5: C, 28.04; H, 3.32; N, 25.01. Found: C, 28.32; H,
3.41; N, 25.19. TGA Calcd for (CuCN)5(HMTA): 80.7. Found: 81.8
(240-270 °C). Calcd for CuCN: 61.5. Found: 63.7 (270-330 °C).
(CuCN)5(HMTA), 9c. The procedure was identical to that used
for 1a, using 4.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
HMTA. A white powder was isolated (59.1%). IR (KBr pellet,
cm-1) 2950 (w), 2158 (w), 2148 (w), 2106 (w), 2080 (m), 1462
(w), 1236 (m), 1003 (s), 827 (m), 810 (m), 707 (w), 700 (m), 689
(w). Anal. Calcd for C11H12N9Cu5: C, 22.47; H, 2.06; N, 21.44.
Found: C, 22.65; H, 2.07; N, 21.47. TGA Calcd for CuCN: 76.2.
Found: 77.8 (290-330 °C).
(CuCN)5(BzPip)2, 6b. The procedure was identical to that used
for 1a, using 3.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
BzPip. A cream powder was isolated (58.6%). IR (KBr pellet, cm-1
)
3282 (w), 3250 (w), 3026 (w), 2925 (w), 2893 (w), 2841 (m), 2161
(m), 2127 (s), 1491 (w), 1451 (m), 1315 (w), 1159 (w), 1101 (m),
987 (w), 965 (m), 862 (m), 745 (m), 702 (m). Anal. Calcd for
C27H32N9Cu5: C, 40.52; H, 4.03; N, 15.75. Found: C, 41.04; H,
4.16; N, 15.77. TGA Calcd for (CuCN)4(BzPip): 83.5. Found: 83.7
(125-145 °C). Calcd for CuCN: 56.0. Found: 55.9 (145-200 °C).
(CuCN)7(BzPip)2, 6c. The procedure was identical to that used
for 1a, using 4.00 mmol CuCN, 2.00 mmol KCN, and 1.00 mmol
BzPip. An off-white powder was isolated (72.8%). IR (KBr pellet,
cm-1) 3246 (m), 3024 (w), 2926 (m), 2835 (m), 2811 (m), 2162
(m), 2128 (s), 2123 (s), 2090 (s) 1490 (w), 1439 (m), 1161 (w),
1095 (m), 1022 (w), 984 (m), 962 (m), 742 (s), 700 (s). Anal. Calcd
for C29H32N11Cu7: C, 35.56; H, 3.29; N, 15.73. Found: C, 35.87;
H, 3.36; N, 15.53. TGA Calcd for CuCN: 64.0. Found: 63.4
(140-220 °C).
Inorganic Chemistry, Vol. 47, No. 15, 2008 6933

Lim et al.
Table 1. Synthetic Results for Aqueous Reflux Reactions
ligand
General Method for Hydrothermal (HT) Reactions. Copper(I)
cyanide and KCN were suspended in 5.0 mL H2O in a 23 mL
Teflon-lined Parr acid digestion vessel. The L ligand was added,
and the resulting suspension was stirred briefly. The vessel was
sealed and heated at 175 °C for 4 days. After overnight cooling to
room temperature, the suspended solid was collected by means of
filtration, washed with H2O, ethanol, and diethyl ether, and then
dried under vacuum. Crystals of sufficient quality for X-ray analysis
were then selected. The following mmol ratios of CuCN/KCN/L
were used to produce X-ray crystals: 1b (3:1:1), 2 (2:1:1), 3b
(4:1:1), 4 (2:1:1), 5 (2:1:1), 6a (1:1:1), 6d (2:1:1), 7 (2:1:1), 8b
(4:1:1), 9b (4:1:1), and 9c (5:2:1).
1:1:1a
2:2:1a
3:2:1a
4:2:1a
5:3:1a
Pip
1a (2:1)
2 (7:2)
3a (2:1)
4 (7:2)
1a
2
3a
4
1b (20:7)
2
3b (4:1)
4
1b
2
3b
4
MePip
Me2Pip
EtPip
Et2Pip
5 (4:1)
5
5
5
BzPip
6a (3:2)
7 (2:1)
8a (1:1)
9b (5:2)
6b (5:2)
7
8b (2:1)
9c (5:1)
6c (7:2)
7
8b
9c
6d (4:1)
7
8b
9c
Bz2Pip
Ph2CHPip
HMTA
9a (3:2)
a Mixing ratio: mmol CuCN/mmol KCN/mmol L.
the preparations of a diverse group of highly luminescent
materials can be effected by a simple organic solvent-free
reaction carried out on a hotplate. The reactions are easily
scaled up to multigram levels. The commercial low temper-
ature polymorph of CuCN was used.18 As indicated in Table
1, four different reactant ratios were used, leading to as few
as one and as many as four different products. In those cases
where multiple network products were found, increasing
CuCN/L ratios led to more copper-rich products. Synthesis
yields were fair to good, generally being higher for the more
copper-rich products. All of the products were white, or
nearly white, powders in visible light, however, and as will
be described below, their luminescence responses varied
widely.
The hotplate syntheses produced powders which were
analytically pure and of good crystallinity (as indicated by
sharp X-ray powder peaks), but sizable crystals were not
present. Therefore, HT syntheses at 175 °C in Teflon-lined
bombs were used to produce X-ray quality crystals. Where
possible, X-ray diffraction patterns of the hotplate products
were compared to calculated powder patterns from X-ray
structures to confirm phase identities. As shown in Table 1,
17 distinct products were realized from bulk synthesis.
Eleven of these compounds (1b, 2, 3b, 4, 5, 6a, 6d, 7, 8b,
9b, and 9c) were characterized as new X-ray structures from
HT crystals and three (1a, 3a, and 9a) matched existing
literature structures.7,8 In all cases, powder-diffraction pat-
terns from the bulk products sufficiently matched those
calculated from the corresponding single crystals. This close
correspondence between bulk powders and HT crystals for
the diamine and tetramine ligands in the current study stands
in contrast to our results with diimine L, which showed few
such matches.1
X-ray Analysis. Single-crystal determinations were carried out
using a Bruker SMART Apex II diffractometer using graphite-
monochromated Cu KR radiation.12 The data were corrected for
Lorentz and polarization13 effects and absorption using SADABS.14
The structures were solved by use of direct methods or Patterson
map. Least squares refinement on F2 was used for all reflections.
Structure solution, refinement, and the calculation of derived results
were performed using the SHELXTL15 software package. The non-
hydrogen atoms were refined anisotropically with one exception.
The (ordered) cyano atom C5 in 9c could not successfully be made
anisotropic despite attempts to model it in multiple positions and
despite removal of several offending reflections. For 2 and 3b,
hydrogen atoms were located by standard difference Fourier
techniques and were refined with isotropic thermal parameters. In
all other cases, hydrogen atoms were located and then placed in
theoretical positions. In most, but not all cases, the cyano carbon
and nitrogen atoms were disordered. When required by crystal-
lographic symmetry, occupancies of 50% carbon and 50% nitrogen
were used; otherwise relative carbon and nitrogen occupancies were
refined. The former and latter cases are referred to in the text as
symmetrically and non-symmetrically disordered, respectively. All
ethyl groups in 4 and 5 were disordered across mirror planes. This
disorder was successfully modeled.
Powder-diffraction analysis was carried out on the instrument
described above. Samples were ground and prepared as mulls using
Paratone-N oil. Four 180 s frames were collected, covering 8-100°
2θ. Frames were merged using the SMART Apex II software12 and
were further processed using DIFFRAC-Plus and EVA software.16
Simulated powder patterns from single-crystal determinations were
generated using the Mercury17 program. Experimental and calcu-
lated powder diffraction results are provided in the Supporting
Information.
Results and Discussion
Synthesis. In our previous contributions, we have shown
that copper(I) cyanide networks can be produced via
overnight reflux reactions with KCN and bridging ligands
(L).1–3 We have since refined and simplified our strategy,
determining that one hour at about 80 °C is usually sufficient
to afford complete conversions to crystalline products. Thus,
Reactions of CuCN with Pip produced two networks, the
known 2:1 product 1a and the remarkable 20:7 product 1b.
Use of MePip or EtPip yielded networks having 7:2 ratio (2
and 4), irrespective of the reaction stoichiometry employed.
Reactions of Me2Pip produced the known 2:1 network 3a
and the new 4:1 3b. In contrast, only the 4:1 product 5 was
found for Et2Pip. The BzPip ligand gave rise to four distinct
networks having 3:2 (6a), 5:2 (6b), 7:2 (6c), and 4:1 (6d)
ratios. The bulkier Bz2Pip and Ph2CHPip were the only
ligands that failed to give rise to copper-rich products having
ratios >2:1. For Bz2Pip only the 2:1 7 was found and for
Ph2CHPip 1:1 8a and 2:1 8b formed. Finally, the tetradentate
(12) SMART Apex II, Data Collection Software, Version 2.1; Bruker AXS
Inc.: Madison, WI, 2005.
(13) SAINT Plus, Data Reduction Software, Version 7.34a; Bruker AXS
Inc.: Madison, WI, 2005.
(14) Sheldrick, G. M. SADABS; UniVersity of Go¨ttingen: Go¨ttingen,
Germany, 2005.
(15) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112.
(16) DIFFRAC Plus, Version 10.0 and EVA, release 2004; Bruker AXS
Inc.: Madison, WI, 2005.
(17) Mercury, Version 1.5; Cambridge Crystallographic Data Centre:
Cambridge, U.K., 2006.
(18) Hibble, S. J.; Eversfield, S. G.; Cowley, A. R.; Chippendale, A. M.
Angew. Chem., Int. Ed. 2004, 43, 628.
6934 Inorganic Chemistry, Vol. 47, No. 15, 2008

HMTA yielded three networks: the known complex 3:2
complex (9a) and new 5:2 (9b) and 5:1 (9c) species. The
previously reported synthesis of 9a involved addition of
HMTA to a hot aqueous solution of CuCN and thiosulfate
or the reduction of Cu(OAc)2 in the presence of water,
ammonia, and formalin.7,8 Thus, the new procedure was more
convenient, albeit with equally low-product yields in the 30%
range.
X-ray Structures. Eleven new X-ray structures emerged
from the current study. Refinement details for all structures
are summarized in Table 2 and selected bond lengths and
angles are given in Table 3. A common theme in CuCN
networks is the (6,3) lattice, which consists of 3-coordinate
copper centers linked by cyano and L ligands to form 2D
hexagonal sheets and results in (CuCN)2L stoichiometry.
Lattices or sublattices with (6,3) tiling are encountered herein
for networks 1b, 2, 3a,7 4, 5, 6d, and 7. Of these, 7 is the
most straightforward example because the (6,3) is the only
lattice present. Interestingly, the 2:1 network 1a does not
show the (6,3) motif but rather forms a 3D network with
four metal centers bridged by a single CN and associated
Cu · · · Cu interactions.7
distance (Cu · · ·Cu ) 2.8156(10)-3.0378(11) Å). All of these
weak interactions are between the two subnetworks and are
unsupported by any ligand bridging.
The MePip and EtPip 2 and 4 are isomorphic and
isostructural to one another (Figure 2). Each has the formula
(CuCN)7L2 and crystallizes in the monoclinic C2/m space
group. Each shows seven independent copper atoms, all of
which lie on a single mirror plane. All cyanides are
disordered; X1/X1 and X2/X2 are symmetrically disordered
(X ) disordered C/N). As was the case with 1b, two
sublattice types are present. In direct analogy to 1b, sublattice
A contains 3-coordinate metal atoms and forms a (6,3)
(CuCN)2L stepped sheet running in the a,b plane. There are
two independent A sublattices (Cu4/Cu5 and Cu6/Cu7),
which are related by 2-fold rotation and stack in XYYX...
sequence. Cyano-Cu-cyano angles for the 3-coordinate A
centers are again large: 141.59(13)-145.85(14)° in 2 and
141.42(18)-145.77(17)° in 4. The B sublattices in 2 and 4
are simple CuCN chains, lacking any L cross-links. The
chains, which propagate parallel to the c axis and thread
through the (6,3) sheets, undulate, owing to the <180°
cyano-Cu-cyano bond angles: 152.71(15)-172.05(13)° in
2 and 146.3(2)-171.21(17)° in 4. As is the case for 1b, the
sublattices are weakly linked by near-van der Waals interac-
tions of 2.6087(7), 2.7369(7) and 2.8824(7) Å in 2 and
2.7076(11), 2.7824(9), and 2.9639(10) Å in 4. The A and B
sublattices occur in 3:2 ratio: 2[(CuCN)2L] ·3CuCN. Copper
atoms are coordinated in axial positions to the piperazine
chairs in both 2 and 4.
A straightforward (6,3) network is recognized for
(CuCN)2(Me2Pip) (3a),7 but no such formulation appears to
exist for L ) Et2Pip. The (CuCN)4L 3b and 5 are similar
but are not isostructural (see Figures 3 and 4). Complex 3b
crystallizes in the orthorhombic Cccm space group and 5
crystallizes in C2/m. Both structures are composed of two
half-independent copper atoms and one-half- and two-
quarter-independent cyano groups, all of which are sym-
metrically disordered. In these structures a single (6,3) A
stepped sheet sublattice (Cu1) is penetrated by a CuCN chain
B sublattice (Cu2). Although all of the B chains in 5
propagate in a parallel direction, alternate B chains in 3b
run in orthogonal directions. However, in both 3b and 5 (as
with 2 and 4) all CuCN chains run through the hexagonal
voids in the A sublattice. In these networks, the sublattice
ratio is [(CuCN)2L] · 2CuCN. Copper atoms are coordinated
in axial positions to the piperazine chairs in both 3b and 5.
The cyano-Cu-cyano angles in 3b and 5 are 142.99(10) and
143.97(13)° for Cu1 and 148.28(13) and 158.37(11)° for
Cu2. Near-van der Waals Cu · · · Cu interactions between the
sublattices measuring 2.5893(7) and 2.6715(6) Å in 3b and
5, respectively, are present.
The crystal structure of 1b has been described in a
preliminary communication.2 This compound crystallizes in
j
the triclinic space group P1. There are 20 crystallographically
independent copper atoms and 20 independent cyano groups,
all of which are non-symmetrically disordered. The X-ray
structure of 1b is depicted in Figure 1. Two nonidentical
sublattices are present. Sublattice A (containing copper atoms
Cu9-Cu20) is a (6,3) lattice lying in the a,b plane and
consisting of hexagonal Cu6(CN)4(Pip)2 units. There are three
independent, but essentially identical, A sublattices (Cu9/
Cu10/Cu15/Cu16, Cu11/Cu14/Cu17/Cu20, and Cu12/Cu13/
Cu18/Cu19). The three A layers stack in an XYZZYX...
pattern. Despite the 3-coordination of the copper atoms, the
cyano-Cu-cyano angles are relatively obtuse, measuring
141.58(19)-143.6(2)°. These large angles suggest a pre-
dominance of CuCN chain bonding over Pip coordination.
For 1b and most of the other (6,3) sublattices, metal-atom
attachment to the piperazines is axial. As a result, the ligand
is oriented perpendicular to the A sheet plane and creates a
series of zigzag steps in the sheets.
Sublattice B (containing Cu1-Cu8) consists mostly of
roughly linear 2-coordinate copper atoms, forming CuCN
chains, which are cross-linked at every eighth copper by a
Pip unit. This arrangement in the B sublattice produces
rippled sheets of large tiled Cu18(CN)16(Pip)2 rings, which
are threaded through the hexagonal voids in the A sublattices.
Sublattices A and B have stoichiometries (CuCN)2(Pip) and
(CuCN)8(Pip), respectively. The 6:1 A/B ratio gives rise to
a 6[(CuCN)2(Pip)] · [(CuCN)8(Pip)] formulation, or more
simply, (CuCN)20(Pip)7. The Pip centroids in the two sublat-
tices lie roughly in line with one another, but the rings are
rotated by about 90° with respect to one another and are canted
at different angles as well. Interestingly, although copper is
coordinated in axial positions to Pip in the A sublattice, it is in
equatorial position in the B sublattice. There are a total of 10
Cu · · ·Cu interactions that are close to the 2.8 Å van der Waals
Network 6a, (CuCN)3(BzPip)2, crystallizes in the chiral
monoclinic space group P21. Six independent copper atoms
and six cyano groups are present. Five of the latter are non-
symmetrically disordered and one (C2/N2) is fully ordered.
The resulting network is composed of planar rows of
undulating CuCN chains running roughly parallel to the a
axis (Figure 5). The chains are cross-linked by bridging L
Inorganic Chemistry, Vol. 47, No. 15, 2008 6935

Lim et al.
Table 2. Crystal and Structure Refinement Data
1b
2
3b
CCDC deposit no.
color and habit
size, mm3
formula
fw
space group
a, Å
b, Å
647 769
678 228
678 225
colorless prism
0.26 × 0.13 × 0.05
C5H7Cu2N3
236.22
Cccm (No. 66)
12.10260(10)
14.16340(10)
9.37560(10)
90
colorless prism
0.19 × 0.13 × 0.06
C48H70Cu20N34
2394.18
colorless blade
0.42 × 0.23 × 0.05
C17H24Cu7N11
827.25
C2/m (No. 12)
13.9138(2)
9.41970(10)
21.1231(3)
90
105.9470(10)
90
2661.93(6)
4
2.064
1624
6.272
Cu KR (1.54178)
100
0.0255; 0.0672
1.149
j
P1 (No. 2)
8.4553(2)
15.7200(3)
27.6308(6)
91.9900(10)
93.9660(10)
97.2830(10)
3530.75(14)
2
2.190
2352
6.591
Cu KR (1.54178)
296
0.0506; 0.1529
1.054
c, Å
R, deg
ꢀ, deg
90
90
γ, deg
V, Å3
1607.11(2)
8
1.953
928
5.930
Cu KR (1.54178)
100
0.0229; 0.0541
1.081
Z
F
F000
calcd, g cm-3
µ(Cu KR), mm-1
radiation (λ, Å)
T, K
residuals:a R; Rw
GOF
4
5
6a
CCDC deposit no.
color and habit
size, mm3
formula
fw
678 229
678 230
678 234
colorless blade
0.25 × 0.09 × 0.03
C19H28Cu7N11
855.30
colorless blade
0.31 × 0.05 × 0.02
C6H9Cu2N3
250.24
colorless plate
0.30 × 0.23 × 0.02
C50H64Cu6N14
1242.39
space group
a, Å
b, Å
C2/m (No. 12)
13.8422(2)
9.41320(10)
21.6258(2)
90
101.9110(10)
90
2757.16(6)
4
2.060
1688
6.082
Cu KR (1.54178)
296
0.0298; 0.0716
1.080
C2/m (No. 12)
13.1088(3)
9.3972(2)
9.5547(4)
90
133.1290(10)
90
859.00(4)
4
1.935
496
5.590
Cu KR (1.54178)
296
0.0216; 0.0570
1.043
P21 (No. 4)
19.3691(4)
6.82790(10)
21.6030(4)
90
108.0350(10)
90
2716.63(9)
2
1.519
1272
5.995
Cu KR (1.54178)
200
0.0273; 0.0692
0.994
c, Å
R, deg
ꢀ, deg
γ, deg
V, Å3
Z
F
F000
calcd, g cm-3
µ(Cu KR), mm-1
radiation (λ, Å)
T, K
residuals:a R; Rw
GOF
6d
7
8b
CCDC deposit no.
color and habit
size, mm3
formula
fw
678 233
678 231
678 232
colorless blade
0.24 × 0.11 × 0.02
C15H16Cu4N6
534.50
colorless plate
0.17 × 0.13 × 0.04
C10H11Cu1N2
222.75
colorless prism
0.31 × 0.14 × 0.06
C19H21Cu2N4
432.48
space group
a, Å
b, Å
Pnma (No. 62)
18.8238(5)
9.2760(2)
10.9423(3)
90
P21/c (No. 14)
9.37890(10)
13.1246(2)
9.49180(10)
90
118.1060(10)
90
1030.61(2)
4
1.436
456
2.609
Cu KR (1.54178)
296
0.0262; 0.0731
1.068
Pbca (No. 61)
11.58880(10)
8.73360(10)
35.7412(3)
90
c, Å
R, deg
ꢀ, deg
90
90
90
90
γ, deg
V, Å3
1910.63(8)
4
1.858
1056
5.995
Cu KR (1.54178)
100
0.0201; 0.0523
1.068
3617.44(6)
8
1.588
1768
1.588
Cu KR (1.54178)
100
0.0265; 0.0717
1.038 m
Z
F
F000
calcd, g cm-3
µ(Cu KR), mm-1
radiation (λ, Å)
T, K
residuals:a R; Rw
GOF
9b
9c
CCDC deposit no.
color and habit
size, mm3
678 226
678 227
colorless block
0.16 × 0.15 × 0.10
colorless needle
0.15 × 0.03 × 0.02
6936 Inorganic Chemistry, Vol. 47, No. 15, 2008

Table 2. Continued
9b
9c
formula
fw
space group
a, Å
b, Å
c, Å
R, deg
ꢀ, deg
γ, deg
V, Å3
Z
C34H48Cu10N26
1456.38
C2/c (No. 15)
21.7147(9)
6.9228(3)
15.6273(7)
90
111.294(2)
90
2188.82(16)
2
2.210
1448
5.691
Cu KR (1.54178)
173
0.0250; 0.0636
1.107
C44H48Cu20N36
2351.98
P21/c (No. 14)
7.0745(2)
29.9449(7)
14.7420(3)
90
92.7030(10)
90
3119.55(13)
2
2.504
2288
7.669
Cu KR (1.54178)
173
0.0331; 0.0747
1.029
F
calcd, g cm-3
F000
µ(Cu KR), mm-1
radiation (λ, Å)
T, K
residuals:a R; Rw
GOF
a R ) R1 ) ∑||Fo| - |Fc||/∑|Fo| for observed data only. Rw ) wR2 ) {∑[w(Fo - Fc )2]/∑[w(Fo )2]}1/2 for all data.
2
2
2
in the c direction, forming 2D double sheets, which are
capped by nonbridging L. All copper atoms are 3-coordinate,
having cyano-Cu-cyano bond angles of 128.53(12)-
140.28(12)°. Each metal atom coordinates two cyanides and
a single BzPip. The four BzPip ligands in the structure show
interesting contrasts. Two are bridging, connecting Cu1 &
Cu2 and Cu4 & Cu5, linking pairs of CuCN chains. These
BzPip groups bond copper in equatorial positions at the
H-substituted end and in axial positions at the benzyl-
substituted end. The remaining two BzPip ligands are
terminal. These BzPip ligands bond Cu3 and Cu6 at the
H-substituted end in axial positions. The benzyl-substituted
ends are not coordinated. Among the six Cu-N(BzPip)
bonds, the two Cu-NBz bonds are longer than the Cu-NH
bonds, by 0.02-0.05 Å. Two Cu · · · Cu interactions are
present: Cu3 · · ·Cu4 ) 2.6650(6) and C1 · · ·Cu6 ) 2.9644(6)
Å. Cyanide X4/X4, which bridges Cu4 and Cu5, shows a weak
bridging interaction to Cu3. This is indicated by the slight
bending of Cu4-X4-X4 angle (156.1(3)°) toward Cu3.
As shown in Figure 6, 6d (which crystallizes in the
orthorhombic group Pnma) is structurally related to the other
(CuCN)4L 3b and 5. A highly rippled (6,3) sheet (sublattice
A) is created by 3-coordinate atoms Cu3 and Cu4. Sublattice
B CuCN chains are formed by Cu1 and Cu2, running parallel
to the a axis and passing through the A sheets. The
cyano-Cu-cyano angles are 157.84(10) and 162.49(10)°
for 2-coordinate centers and 138.59(10) and 140.67(10)° for
the 3-coordinate centers. All four copper atoms are half-
independent. Two of three cyano groups are also centered
about special positions; all three cyano groups are non-
symmetrically disordered. The half-independent BzPip co-
ordinates copper equatorially on the NH side and axially on
the benzyl end; the Cu-NBz bond is ca. 0.04 Å longer than
the Cu-NH. As with 3b and 5, metal-metal interactions
(2.6171(5) and 2.6387(5) Å) are present, linking the two
sublattices.
attachments to the disubstituted piperazine are axial, resulting
in a series of zigzag steps in the sheet. The Cu-NBz value of
2.2149(14) is the longest Cu-N(L) distance in the current study.
The cyano-Cu-cyano angle is 147.82(8)°. Predictably, there
are no Cu · · ·Cu interactions. Of the (6,3) networks in this study,
only 7 does not appear to allow penetration by a second
sublattice. For all of the smaller piperazines studied (Pip, MePip,
Me2Pip, EtPip, Et2Pip, and BzPip), threading of CuCN through
the honeycomb (6,3) lattice results in copper-rich (i.e., >2:1)
phases. The likely explanation is that benzyl groups in 7 block
access to the channels.
The very bulky substitution on one end of Ph2CHPip
precludes metal coordination, leading to the unique structure
of 8b, which is shown in Figure 8. The space group is
orthorhombic Pbca. A zigzag chain running parallel to the
b axis is formed by 3-coordinate Cu1 and cyano X1/X1.
Connected to Cu1 and decorating the chain along the a axis
are X2-X2-Cu2-L units. Because the metal is coordinated
axially to the terminal Ph2CHPip units, the ligands are oriented
away from the plane of the chain, in roughly the c direction.
Ligand-free sides of adjacent CuCN chains face one another,
with the nearest approach of Cu1 · · ·Cu2 ) 2.8976(4) Å. Both
copper atoms show coordination environments that are unique
in the current study. Three-coordinate Cu1 lies in Cu(CN)3
coordination. The angles around Cu1 deviate by <5° from 120°.
The Cu(CN)3 coordination sphere is very unusual in CuCN-L
networks, having been seen previously only for the (CuCN)3(2,4-
diaminopyrimidine) and (CuCN)5(pyridazine)2.1 Two-coordinate
Cu2 is in cyano-Cu-L coordination and shows a 163.78(7)°
bond angle. In previously characterized monodentate amine
complexes of CuCN, decoration of the chains consists merely
of pendant amine ligands (L), rather than cyano-Cu-L side
chains.7,19,20 The only precedent for this 2-coordinate arrange-
ment is the cyano-Cu-OH2 unit seen in (CuCN)3(H2O).21
Two new CuCN-HMTA networks (9b and 9c) were
elucidated during the current study. The literature structure
The structure of 7, in space group P21/c, is that of a simple
(6,3) sheet running parallel to the b,c plane (Figure 7). In other
words, 7 consists of an A sublattice only. A single 3-coordinate
copper atom is bridged by a non-symmetrically disordered
cyanide and a half-independent Bz2Pip unit. Both metal atom
(19) Dyason, J. C.; Healy, P. C.; Engelhardt, L. M.; Pakawatchai, C.;
Patrick, V. A.; White, A. H. J. Chem. Soc., Dalton Trans. 1985, 839.
(20) Bowmaker, G. A.; Lim, K. C.; Skelton, B. W.; White, A. H. Z.
Naturforsch. 2004, 59b, 1264.
(21) Kildea, J. D.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1985, 38,
1329.
Inorganic Chemistry, Vol. 47, No. 15, 2008 6937

Lim et al.
Table 3. Selected Bond Lengths and Angles for All Complexesa
1b, (CuCN)20(Pip)7
Cu-X
X-X
Cu-N
Cu-Cu
1.843(5)-1.900(4)
1.128(8)-1.168(6)
2.049(7)-2.144(4)
2.8156(10)-3.0378(11)
Cu-X-X
160.9(6)-178.8(4)
134.5(3)-143.6(2)
165.2(3)-179.76(19)
103.7(2)-114.8(2)
X-Cub-X
X-Cuc-X
X-Cu-N
2, (CuCN)7(MePip)2
Cu-X
X-X
Cu-N
Cu-Cu
1.851(3)-1.896(2)
1.149(7)-1.169(7)
2.081(3)-2.143(3)
2.6087(7)-2.8824(7)
Cu-X-X
X-Cub-X
X-Cuc-X
X-Cu-N
172.2(3)-179.9(3)
141.59(13)-145.85(14)
152.71(15)-172.05(13)
106.70(7)-109.16(7)
3b, (CuCN)4(Me2Pip)
Cu-X
1.860(3)-1.884(2)
1.147(4)-1.149(6)
2.112(2)
Cu-X-X
173.8(4)-177.0(2)
143.99(13)
148.28(13)
X-X
X1-Cu1b-X1
X2-Cu2c-X3
X1-Cu1-N4
Cu1-N4
Cu1-Cu2
2.5893(7)
107.74(7)
4, (CuCN)7(EtPip)2
Cu-X
X-X
Cu-N
Cu-Cu
1.858(4)-1.900(3)
1.144(4)-1.153(5)
2.100(3)-2.158(3)
2.6539(9)-2.9639(10)
Cu-X-X
X-Cub-X
X-Cuc-X
X-Cu-N
169.9(4)-179.6(4)
141.21(16)-145.77(17)
146.3(2)-171.21(17)
106.43(8)-109.38(8)
5, (CuCN)4(Et2Pip)
Cu-X
1.863(3)-1.8927(16)
1.141(5)-1.149(3)
2.137(2)
Cu-X-X
176.0(2)-179.2(3)
158.37(12)
108.50(5)
X-X
X2-Cu2c-X3
X1-Cu1-N4
X1-Cu1b-X1′
Cu1-N4
Cu1-Cu2
2.6715(6)
142.99(10)
6a, (CuCN)3(BzPip)2
Cu-X
X-X
Cu-N
Cu-Cu
1.872(3)-1.953(3)
1.144(4)-1.160(4)
2.100(2)-2.124(2)
2.6650(6), 2.9644(6)
Cu-X-X
X-Cub-X
X-Cu-N
156.1(3)-176.5(3)
128.53(12)-140.28(12)
100.09(15)-124.97(14)
6d, (CuCN)4(BzPip)
Cu-X
X-X
Cu-N
Cu-Cu
1.852(2)-1.8966(17)
1.162(3)-1.165(3)
2.141(2), 2.100(2)
2.6171(5), 2.6387(5)
Cu-X-X
X-Cub-X
X-Cuc-X
X-Cu-N
172.8(2)-179.8(2)
138.59(10), 140.67(10)
157.84(10), 162.49(10)
109.64(5), 110.16(5)
7, (CuCN)2(Bz2Pip)
Cu1-X1
X1-X1′
Cu1-N2
1.873(2), 1.9024(17)
1.144(3)
2.2149(14)
Cu1-X1-X1′
X1-Cu1b-X1
X1-Cu1-N2
175.66(17), 176.25(18)
147.82(8)
98.63(6), 113.53(7)
8b, (CuCN)2(Ph2CHPip)
Cu1-X1
X-X
Cu2-N3
Cu1-Cu2
1.8963(19)-1.9417(18)
1.157(3), 1.162(3)
1.9349(15)
Cu-X-X
169.96(16)-176.50(16)
114.65(7)-124.43(8)
163.78(7)
X1-Cu1b-X2
X2-Cu2c-N3
2.8976(4)
9b, (CuCN)5(HMTA)2
Cu-X
Cu-Xd
X-X
Cu-N
Cu-Cu
1.926(2)-1.940(2)
1.993(2)-2.380(2)
1.156(3)-1.163(5)
2.1664(19)-2.1913(18)
2.4381(6), 2.6682(7)
Cu-X-X
Cu-Xd-X
X-Cu-X
Cu-Xd-Cu
X-Cu-N
N-Cu-N
164.77(19)-173.1(3)
119.68(18)-165.4(2)
103.90(8)-121.48(9)
74.61(7), 71.20(8)
92.34(7)-127.04(8)
107.03(10)
9c, (CuCN)5(HMTA)
Cu-X
Cu-Xd
X-X
Cu-N
Cu-Cu
1.830(4)-2.028(4)
1.920(5)-2.444(4)
1.147(6)-1.167(6)
2.103(4)-2.217(4)
2.4098(12)-2.5577(13)
Cu-X-X
Cu-Xd-X
X-Cub-X
X-Cuc-X
Cu-Xd-Cu
X-Cu-N
166.7(4)-178.8(4)
109.4(3)-176.9(4)
94.44(14)-127.50(16)
166.66(17), 168.67(17)
68.39(13)-75.22(15)
92.90(14)-123.21(15)
a X indicates cyanide C/N, N indicates L ligand. b 3-Coordinate copper atom. c 2-Coordinate copper atom. d X atom bridges two copper atoms.
of 9a reveals only a slight alteration to the familiar (6,3)
motif, with all HMTA ligands bidentate and alternating 3-
and 4-coordinate copper atoms.8 The resulting sheets are tiled
from alternating Cu4(CN)2(HMTA)2 and Cu6(CN)4(HMTA)2
rings. In contrast, 9b and 9c (Figures 9 and 10) contain 3-
and 4-coordinate HMTA units, respectively, and are far more
structurally complex 3D materials. There is also a distinct
tendency toward cyano-bridged dimer formation. This is
probably the result of the forced close proximity of copper
centers in 9b and 9c.
In 9b (monoclinic C2/c), each of the two fully independent
copper atoms (Cu1 and Cu2) forms part of a cyano-bridged
Cu2(CN)2 dimer with short Cu · · ·Cu interactions (2.4381(6),
2.6682(7) Å). Each dimer links to four cyanides (including
the two that bridge the metal atoms) and two HMTA units.
The half-independent Cu3 is 4-coordinate, binding two
6938 Inorganic Chemistry, Vol. 47, No. 15, 2008

Figure 1. X-ray crystal structure of (CuCN)20(Pip)7, 1b. Hydrogen atoms and Cu · · ·Cu interactions omitted for clarity. (a) Thermal ellipsoid view (50%).
(b) Projection down crystallographic a axis. Atom identities for all ball and stick projections: black circles ) Cu atoms; gray circles ) cyano C/N atoms;
wireframe ) L ligand.
Figure 2. X-ray crystal structure of (CuCN)7(MePip)2, 2. Hydrogen atoms omitted for clarity (a) Thermal ellipsoid view (50%). All Cu· · ·Cu interactions
are shown with dashed bonds. (b) Projection down crystallographic b-axis. Cu · · ·Cu interactions are not shown.
cyanides and two HMTA units. Copper atoms are coordi-
nated to three of the four HMTA nitrogen atoms. The 3D
structure of 9c (P21/c) reveals 10 fully independent copper
atoms and 2 independent HMTA units, both of which
coordinate copper at all four nitrogen atoms. Four Cu2(CN)2
dimers like those in 9b are present; each shows a short
Cu· · ·Cu interaction (2.4098(12)-2.5577(13) Å). However,
unlike 9b two 2-coordinate copper atoms (Cu7 and Cu10)
are also present. Each is coordinated in nearly linear fashion
to two cyanides. These atoms are reminiscent of the
2-coordinate metal centers in 1b, 2, 3b, 4, 5, and 6d;
however, in the present case they are part of the main
network, rather than independent CuCN chains. These
2-coordinate centers appear to help create some extra space
in this very dense (ca. 2.50 g/cm3) network. The potentially
tetradentate HMTA ligand has now been shown to form bi-,
tri-, and tetradentate coordination in CuCN networks. 9c
represents one of relatively few occurrences of tetradentate
HMTA in network materials; most of the recognized
examples are based on Ag(I).22
Survey of CuCN-L Networks. There are now some 55
known CuCN-L (L ) bridging bidentate diamine or diimine
ligand) networks, 32 of which are associated with this or
our previous study.1–3,6–8 The products break down by
stoichiometry as follows: 6 (CuCN)L, 4 (CuCN)3L2, 21
(CuCN)2L, 4 (CuCN)5L2, 8 (CuCN)3L (including the 20:7
Inorganic Chemistry, Vol. 47, No. 15, 2008 6939

Lim et al.
Figure 3. X-ray crystal structure of (CuCN)4(Me2Pip), 3b. Hydrogen atoms omitted for clarity (a) Thermal ellipsoid view (50%). All Cu · · ·Cu interactions
are shown with dashed bonds. (b) Projection down crystallographic c axis. Cu · · ·Cu interactions are not shown.
Figure 4. X-ray crystal structure of (CuCN)4(Et2Pip), 5. Hydrogen atoms omitted for clarity. (a) Thermal ellipsoid view (50%). All Cu · · ·Cu interactions
are shown with dashed bonds. (b) Projection down crystallographic b axis. Cu· · ·Cu interactions are not shown.
product 1b), 6 (CuCN)7L2, and 6 (CuCN)4L. All 1:1
CuCN-L compounds exhibit 4-coordinate copper. The 3:2
compounds display either 2- and 4- or 3- and 4-coordination.
The 2:1 networks all show only 3-coordinate copper. Finally,
the 5:2, 3:1, 7:2, and 4:1 species have both 2- and
3-coordination. It will be noted that even number ratios are
more common than odd ratios by more than two-to-one
margin, suggesting a preference for the relative structural
simplicity associated with these ratios.
The most common network type is (CuCN)2L. This fact
suggests a preference for 3-coordinate copper. One common
arrangement for such complexes is the (6,3) network,
consisting of cross-linked CuCN chains forming tiled
Cu6(CN)4L2 rings. Such is the case for 3a7 and 7 in the
current study, and also for L ) 2-aminopyrimidine, qui-
noxaline, and phenazine.1,6f Copper-copper bridging is
sometimes found in 2:1 complexes, producing a 3D honey-
comb arrangement, as occurs for 1a (Cu · · · Cu ) 2.581(1)
Å)7 and for several other 2:1 compounds.1 However, in the
present case, the simple (6,3) sheet arrangement is of greater
interest, because it lies at the heart of many of the most
luminescent networks (see below). The 3a and 7 networks
are very similar, having obtuse cyanide-Cu-cyanide angles
and large hexagonal channels.
The hexagonal channels in the (6,3) networks allow for
penetration by a second sublattice, which usually takes the
form of CuCN chains. This is not so for 7, in which benzyl
groups block access to the channels. Nevertheless, for L )
6940 Inorganic Chemistry, Vol. 47, No. 15, 2008

Figure 5. X-ray crystal structure of (CuCN)3(BzPip)2, 6a. Hydrogen atoms omitted for clarity. (a) Thermal ellipsoid view (50%). All Cu · · ·Cu interactions
are shown with dashed bonds. (b) Projection down crystallographic b axis. Cu · · ·Cu interactions are not shown.
Figure 6. X-ray crystal structure of (CuCN)4(BzPip), 6d. Hydrogen atoms omitted for clarity. (a) Thermal ellipsoid view (50%). All Cu · · ·Cu interactions
are shown with dashed bonds. (b) Projection down crystallographic b axis.
Figure 7. X-ray crystal structure of (CuCN)2(Bz2Pip), 7. Hydrogen atoms omitted for clarity. (a) Thermal ellipsoid view (50%). (b) Projection down
crystallographic a axis.
Pip, MePip, Me2Pip, EtPip, Et2Pip, and BzPip, Cu-rich (i.e.,
3b, 4, 5, and 6d) have CuCN threading and all but 6d are
>2:1) phases have been found. All of these products (1b, 2,
luminescent. The CuCN chains run roughly perpendicular
Inorganic Chemistry, Vol. 47, No. 15, 2008 6941

Lim et al.
Figure 8. X-ray crystal structure of (CuCN)2(Ph2CHPip), 8b. Hydrogen atoms omitted for clarity. (a) Thermal ellipsoid view (50%). (b) Projection down
the crystallographic c axis. Cu · · ·Cu interactions are not shown. (c) Projection down crystallographic b axis. Cu · · ·Cu interactions are not shown.
Figure 9. X-ray crystal structure of (CuCN)5(HMTA)2, 9b. Hydrogen atoms omitted for clarity. (a) Thermal ellipsoid view (50%). All Cu· · ·Cu interactions
are shown with dashed bonds. (b) Projection down crystallographic b axis. Cu · · ·Cu interactions are not shown.
Thermal Analysis. As in our previous study, thermo-
gravimetric analysis (TGA) of CuCN-L networks yielded
smooth loss of L, thus providing sufficiently reliable data
for rapid determination of product stoichiometry.1 Mass
losses, temperature ranges, and interpretations for the bulk
powders in the current study provided in the Experimental
to the (6,3) sheets, passing through the channels, which are
roughly 10 × 8 Å in all cases. Similar behavior has been
found for (CuCN)3(pyrazine) and (CuCN)n(4,4′-bipyridine)
(n ) 3.5, 4).6e,f,i In several cases, including (CuCN)7-
(pyrimidine)2, (CuCN)4(quinazoline), (CuCN)7(phthalazine)2,
both 2- and 3-coordinate copper centers are actually incor-
porated into a single L-cross-linked lattice.1 The unusual
(CuCN)4(1,4-bis(imidazole-1-ylmethyl)benzene) contains 2-co-
ordinate copper atoms both within the cross-linked sublattice
and as independent chains.6k Uniquely, in 1b the CuCN
chains coordinate a Pip ligand at every eighth copper center,
producing a sublattice of long interlocking Cu18(CN)16(Pip)2
rings.2 In all of these compounds, the 3-coordinate cyanide-
Cu-cyanide angles are very open (134.5(3)-157.1(3)°),
suggesting significant 2-coordinate (CuCN) character. In
contrast, it will be noted that the 3-coordinate Cu(CN)3 angles
in 8b are very close to 120°.
(22) (a) Bertelli, M.; Carlucci, L.; Ciani, G.; Proserpio, D. M.; Sironi, A.
J. Mater. Chem. 1997, 7, 127. (b) Carlucci, L.; Ciani, G.; von
Gudenberg, D. W.; Proserpio, D. M.; Sironi, A. Chem. Commun. 1997,
631. (c) Carlucci, L.; Ciani, G.; Proserpio, D. M.; Rizzato, S. J. Solid
State Chem. 2000, 152, 211. (d) Tong, M.-L.; Zheng, S.-L.; Chen,
X.-M. Chem.sEur. J. 2000, 6, 3729. (e) Bu, W.-M.; Ye, L.; Fan,
Y.-G. Chem. Lett. 2000, 152. (f) Moulton, B.; Lu, J.; Zaworotko, M. J.
J. Am. Chem. Soc. 2001, 123, 9224. (g) Zheng, S.-L.; Tong, M.-L.;
Tan, S.-D.; Wang, Y.; Shi, J.-X.; Tong, Y.-X.; Lee, H. K.; Chen, X.-
M. Organometallics 2001, 20, 5319. (h) Zheng, S.-L.; Zhang, J.-P.;
Chen, X.-M.; Ng, S.-W. J. Solid State Chem. 2003, 172, 45. (i) Konar,
S.; Mukherjee, P. S.; Drew, M. G. B.; Ribas, J.; Chaudhuri, N. R.
Inorg. Chem. 2003, 42, 2545. (j) Fang, Q.; Zhu, G.; Xue, M.; Sun, J.;
Wei, Y.; Qiu, S.; Xu, R. Angew. Chem., Int. Ed. 2005, 44, 3845.
6942 Inorganic Chemistry, Vol. 47, No. 15, 2008

Figure 10. X-ray crystal structure of (CuCN)5(HMTA), 9c. Hydrogen atoms omitted for clarity. (a) Thermal ellipsoid view (50%). All Cu · · ·Cu interactions
are shown with dashed bonds. (b) Projection down crystallographic a axis. Cu · · ·Cu interactions are not shown.
Section; the TGA traces are included in the Supporting
Information. Decomposition temperatures for the piperazine
complexes are generally modest, typically commencing
around 150-180 °C. However, the HMTA materials exhibit
high thermal stability. It is noteworthy that increasing the
number of metal centers bound to HMTA produces denser,
more highly cross-linked networks, which show increasing
decomposition temperature in TGA results. Thus, 9a (bi-
dentate), 9b (tridentate), and 9c (tetradentate) begin to lose
HMTA at 215, 240, and 290 °C, respectively. In all cases,
the multistep decomposition of CuCN began at about 400
°C. Relatively copper-deficient networks of BzPip and
Ph2CHPip and HMTA showed plateaus prior to the ultimate
formation of CuCN. Based upon TGA mass percent values,
transformation sequences (1) through (5) are strongly
indicated:
(CuCN)2L species, 1a, 3a, 7, and 8b were found to show
only a single Ct N band. This fact is consistent with the
relatively simple structures of these networks. However, the
other network that showed a single cyanide band, 1b, had
the most complex structure, with 20 separate cyanide groups.
It is generally recognized that an inverse correlation exists
between Ct N stretching frequency and Cu-X (X ) C/N)
bond length.20 Thus in our previous study, it was noted that
in each case where weak cyano bridging produced a long
Cu-X interaction of >2.0 Å, a band of <2100 cm-1was
present in the IR spectrum.1 In the present study, only
6c, 9b and 9c exhibited Ct N stretches of <2100 cm-1
.
The structure of 6c is as of yet unknown. However, both
9b and 9c have long Cu-X distances as part of triply
bridging cyanide arrangements. Although 1a shows cya-
nide bridging of four metal centers, it lacks a low
frequency Ct N stretch because all Cu-X bond lengths
in 1a are relatively short.7
(CuCN)3(BzPip)2(6a) f (CuCN)3(BzPip) f CuCN (1)
(CuCN)5(BzPip)2(6b) f (CuCN)4(BzPip)(6d) f CuCN
Luminescence Spectroscopy. Solid-state luminescence
measurements were carried out at ambient temperature using
powder samples. Selected samples were examined for
luminescence at 77 and 4 K. The results are summarized in
Table 4 and several traces are presented in Figures 11–13.
Additional spectra are included in the Supporting Informa-
tion. As shown in part a of Figure 11, copper(I) cyanide itself
shows solid-state luminescence behavior, having a broad
excitation feature with peaks at about 288, 307, and 345 nm.
A fairly sharp emission peak centered at 392 nm moves to
412 nm at 77 K. Because this peak lies at the UV-vis border,
CuCN shows a faint purple emission under a black-light.
Our previous studies of CuCN networks with diimine L
ligands indicated substantial to complete quenching of this
luminescence upon network formation. In stark contrast to
these previous results, the networks based on the diamine
ligands showed a variety of luminescence intensities, ranging
from almost no response above baseline to intensity com-
parable to that of CuCN itself.
(2)
(CuCN)(Ph2CHPip)(8a) f (CuCN)2(Ph2CHPip)(8b) f
CuCN (3)
(CuCN)3(HMTA)2(9a) f (CuCN)5(HMTA)(9c) f CuCN
(4)
(CuCN)5(HMTA)2(9b) f (CuCN)5(HMTA)(9c) f CuCN
(5)
The most interesting decomposition patterns arise in the
diverse CuCN-BzPip system. All suggested intermediates
in the CuCN-BzPip sequence, except (CuCN)3(BzPip),
represent known compositions. However, it is intriguing that
6a and 6b appear to progress through different intermediate
stoichiometries during their decomposition reactions.
Infrared Spectroscopy. Infrared analysis of the powder
samples revealed the expected intense cyanide stretching
bands, occurring in the 2075-2162 cm-1 range. The
Inorganic Chemistry, Vol. 47, No. 15, 2008 6943

Lim et al.
Table 4. Luminescence results
b
complexa
T, K
excitation λmax, nm
emission λmax, nm (color)
Stokes shift, cm-1
CuCN
298
77
298
77
298
77
298
77
298
77
298
77
298
77
4
298
77
298
77
298
77
4
285, 309, 350
288, 307, 345
282, 308
391 (purple)
412 (purple)
458 (blue)
(blue)
443 (blue)
443 (blue)
445 (blue)
445 (blue)
485 (yellow-green)
(blue)
487 (green)
412, 460 (blue)
445 (blue)
449 (blue)
450
436 (purple)
415 (purple)
484 (yellow-green)
(blue)
443, 575 (orange)
442, 592 (purple)
377, 444
2996
10 634
6413
6940
9501
9954
7379
(CuCN)2(Pip), 1a
(CuCN)20(Pip)7, 1b
(CuCN)7(MePip)2, 2
(CuCN)2(Me2Pip), 3a
(CuCN)4(Me2Pip), 3b
(CuCN)7(EtPip)2, 4
286, 322, 345
345
284, 340
340
281, 332
284, 328
325
284, 335
310, 335
312, 341
284, 315
307
(CuCN)4(Et2Pip), 5
8810
(CuCN)(Ph2CHPip), 8a
(CuCN)2(Ph2CHPip), 8b
282, 317
10 885
8580
281, 321
323
325
(CuCN)3(HMTA)2, 9a
(CuCN)5(HMTA)2, 9b
(CuCN)5(HMTA), 9c
298
77
298
77
298
77
4
281, 304
390, 475, 505 (blue-green)
(purple)
470, 510 (yellow)
(blue)
417, 522 (yellow-green)
417, 522 (blue)
416, 451
11 842
11836
8914
285, 302
282, 304
304
295
a Other complexes had no luminescence response. b Calculated between longest excitation λmax and shortest emission λmax
.
All of the network complexes, except 6a-d and 7, were
luminescent at room temperature. In all cases broad excitation
features, not significantly different from those of CuCN, were
seen in the 250-350 nm range. As many as two emission
bands were observed. A high energy (HE) band in the
435-450 nm range was noted for all luminescent products.
This band is probably analogous to the CuCN 392 nm
emission band. For some compounds a second low energy
(LE) band was also noted in the 470-575 nm range.
Complexes lacking the LE band (1a, 1b, 2, 4, and 5) showed
intense blue luminescence, whereas those having both HE
and LE bands (3a, 3b, 8a, 8b, 9a, 9b, and 9c) displayed
luminescence of varying intensity ranging from green to
orange at ambient temperature. For all species except 8b and
9a-c, the LE band overlapped the HE band, resulting in a
single broad emission feature.
Low-temperature luminescence studies (Table 4, Figures
11–13) revealed significant thermochromism in some cases.
As noted above, the sharp CuCN emission band moves from
391 nm at ambient temperature to 412 nm at 77 K. For the
network complexes, cooling to 77 K usually had no effect
on excitation or HE emission wavelength. Several com-
pounds lacking LE bands (1b, 2, 4, and 5) were examined
at reduced temperature. They showed a simple increase in
intensity and no change in λmax. The behavior of 2 is
representative and is shown in part b of Figure 11. The only
exception was 5, for which HE λmax fell slightly, from 436
to 415 nm. In some cases, such as 4 (part c of Figure 11),
careful examination of the luminescence behavior at 77 K
revealed that the broad excitation band is actually composed
of three peaks (280, 320, and 354 nm) and the emission band
is composed of two peaks (412 and 449 nm). The strongest
excitation and emission peaks for 4 at 320 and 449 nm are
coupled, as are the somewhat weaker intensity peaks at 354
and 412 nm. Similar behavior was noted for 3b (Supporting
Information). Because 3b and 4 do not show separate LE
features, it is not clear whether these closely spaced emission
peaks are two components of the HE band or whether they
represent the HE and LE bands.
Thermochromism. The LE band showed thermochromic
behavior, most strikingly so for 8b and 9c which have widely
separated HE and LE bands (443 and 575 nm for 8b and
417 and 522 nm for 9c). Variable-temperature spectra and
photographs of these compounds are shown Figures
12and 13. 3a, 3b, 8a, 9a, and 9b are also thermochromic,
shifting from green or yellow-green to blue emission upon
cooling to 77 K (Supporting Information for additional
photos). The marked color change in 8b and 9c at low
temperature is due to significant increase in intensity of the
HE band and at least partial disappearance of the LE feature.
In the case of 9c, a new shoulder at 451 nm becomes
apparent at 4 K. It is suggested that this represents splitting
of the HE band (as is noted for 3b and 4, e.g.), rather than
persistence of the LE band.
Luminescence Interpretation. Luminescence behavior in
Cu(I) complexes (other than Cu4X4 and related clusters) is
commonly attributed to a Cu d10 f d9s1 or d9p1 metal-
centered (MC) transition;23 however, when unsaturated
ligands are present metal-to-ligand charge transfer (MLCT)
can also be invoked.6k,24 Copper-copper interactions of <3
Å are absent in both the low- and high-temperature forms
of CuCN, ruling out metal-to-metal charge transfer as a likely
source of luminescence.18 Preliminary data from a DFT study
in our laboratory using Gaussian98 suggests that the CuCN
6944 Inorganic Chemistry, Vol. 47, No. 15, 2008

Figure 12. Luminescence of 8b, (CuCN)2(Ph2CHPip): a) spectra at 298,
77 and 4 K. b) photos (left to right) room light/298 K, UV light/298 K, UV
light/77 K.
Figure 11. Luminescence spectra: (a) CuCN at 298 and 77 K; (b) 2,
(CuCN)7(MePip)2 at 298 and 77 K; (c) 4, (CuCN)7(EtPip)2 at 77 K.
HOMO is of metal 3dz2/4s character, whereas the LUMO
combines metal 4p and π*CN contributions. This would
indicate that a combination of MC and MLCT underlies the
CuCN HE band. Turning attention to CuCN-L complexes,
we have previously noted that the excitation spectra of 1b
appears very similar to that of CuCN.2 This now appears to
be the case for all of the luminescent complexes reported
herein. When excitation fine structure is observable, as is
the case for 4 (part c of Figure 11), the observed peaks (280,
320, and 354 nm) are very similar in position and relative
intensity to those of CuCN (288, 307, and 345 nm).
Therefore, it seems safe to propose that the photophysics of
CuCN are not significantly altered upon network expansion
with piperazines. Moving from CuCN to the CuCN-
piperazine network complexes, there is an apparent emission
red shift in this HE band of about 50 nm. However, careful
examination of the spectra reveals that in at least some of
the networks the HE emission band is composed of a pair
of features that become more readily discernible at low
temperature (part c of Figure 11). One of these emission
Figure 13. Luminescence of 9c, (CuCN)5(HMTA):)a) spectra at 298, 77
and 4 K; (b) photos (left to right) room light/298 K, UV light/298 K, UV
light/77 K.
features is identical to that of CuCN (λmax ) 412 nm), and
the second shows a red shift to about 440-460 nm. The
412 nm band is more prominent when excitation is carried
out at longer wavelengths. We suggest that this 412 nm band
is the result of the 2-coordinate copper centers that form the
Inorganic Chemistry, Vol. 47, No. 15, 2008 6945

Lim et al.
CuCN chains (the B sublattice), which are present in all of
the networks having >2:1 CuCN/L stoichiometry. If this
explanation is correct, then the 440-460 nm emission would
reasonably be associated with the 3-coordinate copper centers
in the A sublattice. Whereas the 440-460 nm peak is
observable in all the luminescent complexes reported herein,
the 412 peak is discernible only in a few cases, such as 3b
and 4, both of which have been confirmed by X-ray as
containing 2-coordinate copper centers.
Some of the complexes described herein show a LE feature
in addition to the pervasive HE band. The two bands show
markedly different temperature behavior. Whereas the HE
band consistently increases in relative intensity when the
temperature is lowered to 77 K, the LE band is diminished
in intensity under these conditions. 8b and 9c in particular
have relatively long wavelength LE bands that are strongly
temperature dependent. Low-temperature luminescence spec-
troscopy results have shown that the 8b and 9c LE bands
disappear completely at reduced temperature (parts d and e
of Figure 11). This change in the relative intensity of the
HE and LE bands with temperature is due to energy transfer
between the high-energy sublattices and low-energy sublat-
tices. At 77 K, the barrier to energy transfer is larger than
kT and very little energy transfer occurs, but at 298 K energy
transfer occurs easily and the LE band is observed. The
results reported herein are similar to those we have reported
for metal dicyanides such as K[AuxAg1-x(CN)2], as well as
for K[Au(CN)2] and K[Ag(CN)2] both pure and doped as
impurities in KCl.25
It is remarkable that the complexes of Me2Pip (3a and
3b) show both HE and LE bands, giving rise to yellow to
green luminescence, whereas those of Pip, MePip, EtPip,
and Et2Pip produce blue luminescent materials (HE band
only). This is the case in spite of the fact that 3b (Me2Pip)
and 5 (Et2Pip) are structurally similar to one another.
In Table 4, we tabulate the various Stokes shift values
by taking the difference between the longest wavelength
excitation band and the HE emission at room temperature.
When the Stokes shift is of the order of 8000 cm-1 or
greater, the HE band is most likely a MLCT transition
because there is a bond order difference between the
HOMO and LUMO states. However, when the Stokes shift
is much smaller, then the HE emission band is most likely
a MC transition. This is because the bond order difference
of the HOMO and the LUMO states should be small
because both the HOMO and LUMO states are of the same
metal origin.
The absence of measurable luminescence behavior in the
BzPip and Bz2Pip species is most readily ascribed to
electronic coupling in the excited state to π* orbitals in the
benzyl groups, facilitating nonradiative relaxation. However,
it should be noted that the mere presence of aromatic features
is not enough to ensure quenching of the copper-based
luminescence in these networks. Results for 8a and 8b and
preliminary results with N-phenylpiperazine show lumines-
cent products. A word is also in order with regard to Cu · · ·Cu
interactions and luminescence. As stated above, CuCN itself
is luminescent and yet lacks such interactions. Although such
close contacts are seen in many of networks reported herein,
they are probably not connected with the luminescence seen.
This assertion is supported by the fact that CuCN and its
amine networks show related photophysical behavior and also
by the nearly identical behavior of 3a and 3b, the former of
which lacks CuCN threading and the associated Cu · · ·Cu
interactions.
Conclusion
In summary, we have described a large number of very
stable metal-organic networks that combine CuCN and
L ) various substituted piperazines or HTMA. The
products are formed in simple aqueous reactions. A very
common structural theme in these networks is the hex-
agonally tiled (6,3) sheet composed of CuCN chains cross-
linked by L giving rise to 2:1 CuCN/L stoichiometry.
Additional CuCN chains can be incorporated by threading
through the hexagonal channels in the (6,3) sheets,
resulting in 7:2 or 4:1 stoichiometry. Bond angles for both
2- and 3-coordinate copper are highly variable. The
orientation of the piperazine ligands tends to favor axial
coordination of copper, but exceptions do occur. The bulky
Ph2CHPip ligand remains monodentate, producing a very
unusual structure in which a CuCN chain is decorated with
CNCuL units. The HMTA unit can be bi, tri, or tetraden-
tate in CuCN complexes, leading to extensive Cu2(CN)2
dimer formation in the latter two cases.
Networks lacking benzyl groups show moderate to strong
photoluminescence, which is probably associated with MLCT
transitions. Networks containing Pip, MePip, EtPip, and
Et2Pip show strong blue luminescence, which results from
an apparent combination of a sharp CuCN-like emission
feature and a broader emission band that is probably
associated with 3-coordinate copper centers. This HE band
intensifies at lower temperature. In complexes containing
Me2Pip, Ph2CHPip, and HMTA, a LE band has also been
identified. Although its origin is unclear, its intensity is
sharply diminished at reduced temperature, resulting in
interesting thermochromic behavior.
(23) (a) Ford, P. C.; Cariati, E.; Bourassa, J. Chem. ReV. 1999, 99, 3625,
and references cited therein. (b) Vega, A.; Saillard, J.-Y. Inorg. Chem.
2004, 43, 4012. (c) Lin, Y.-Y.; Li, S.-W.; Che, C.-M.; Fu, W.-F.; Zhou,
Z.-Y.; Zhu, N. Inorg. Chem. 2005, 44, 1511. (d) Bi, M.; Li, G.; Zou,
Y.; Shi, Z.; Feng, S. Inorg. Chem. 2007, 46, 604.
(24) (a) Na¨ther, C.; Greve, J.; Jess, I.; Wickleder, C. Solid State Sci. 2003,
5, 1167. (b) Liu, X.; Guo, G.-C.; Wu, A.-Q.; Cai, L.-Z.; Huang, J.-S.
Inorg. Chem. 2005, 44, 4282. (c) Araki, H.; Tsuge, K.; Sasaki, Y.;
Ishizaka, S.; Kitamura, N. Inorg. Chem. 2005, 44, 9667. (d) Araki,
H.; Tsuge, K.; Sasaki, Y.; Ishizaka, S.; Kitamura, N. Inorg. Chem.
2007, 46, 10032.
Acknowledgment. Grateful acknowledgement is made to
the donors of the American Chemical Society Petroleum
Research Fund (44891-B3). We also acknowledge a Howard
Hughes Medical Institute grant through the Undergraduate
Biological Sciences Education Program to the College of
(25) (a) Omary, M. A.; Hall, D. R.; Shankle, G. E.; Siemiarczuk, A.;
Patterson, H. H. J. Phys. Chem. B 1999, 103, 3845. (b) Rawashdeh-
Omary, M. A.; Omary, M. A.; Shankle, G. E.; Patterson, H. H. J.
Phys. Chem. B 2000, 104, 6143. (c) Hettiarachchi, S. R.; Schaefer,
R. L.; Yson, R. L.; Staples, R. J.; Herbst-Irmer, R.; Patterson, H. H.
Inorg. Chem. 2007, 46, 6997.
6946 Inorganic Chemistry, Vol. 47, No. 15, 2008

bond lengths and angles, and anisotropic displacement parameters;
(2) X-ray powder patterns (experimental and calculated) for all
structurally characterized compounds; (3) TGA traces for all
compounds; (4) additional luminescence spectra for all luminescent
compounds; and (5) luminescence photos for all luminescent
compounds. This material is available free of charge via the Internet
at http://pubs.acs.org.
William and Mary. The research at the University of Maine
was supported by the National Science Foundation. We are
indebted to NSF (CHE-0443345) and the College of William
and Mary for the purchase of X-ray equipment.
Supporting Information Available: (1) Full details of the crystal
structure determinations for 1b, 2, 3b, 4, 5, 6a, 6d, 7, 8b, 9b and
9c in CIF format and tables of crystal data, atomic coordinates,
IC8005184
Inorganic Chemistry, Vol. 47, No. 15, 2008 6947