63Cu NMR spectroscopy of copper(I) complexes with various tridentate ligands: CO as a useful 63Cu NMR probe for sharpening 63Cu NMR signals and analyzing the electronic donor effect of a ligand

Article abstract of DOI:10.1021/ic060745r

⁶³Cu NMR spectroscopic studies of copper(I) complexes with various N-donor tridentate ligands are reported. As has been previously reported for most copper(I) complexes, ⁶³Cu NMR signals, when acetonitrile is coordinated to copper(I)

Full text of DOI:10.1021/ic060745r

Inorg. Chem. 2007, 46, 541−551
63Cu NMR Spectroscopy of Copper(I) Complexes with Various Tridentate
Ligands: CO as a Useful 63Cu NMR Probe for Sharpening 63Cu NMR
Signals and Analyzing the Electronic Donor Effect of a Ligand
Masato Kujime, Takuya Kurahashi, Masaaki Tomura, and Hiroshi Fujii*
Institute for Molecular Science and Okazaki Institute for IntegratiVe Bioscience, National Institutes
of Natural Sciences, Myodaiji, Okazaki 444-8787, Japan
Received May 2, 2006
63Cu NMR spectroscopic studies of copper(I) complexes with various N-donor tridentate ligands are reported. As
has been previously reported for most copper(I) complexes, 63Cu NMR signals, when acetonitrile is coordinated to
copper(I) complexes of these tridentate ligands, are broad or undetectable. However, when CO is bound to tridentate
copper(I) complexes, the 63Cu NMR signals become much sharper and show a large downfield shift compared to
those for the corresponding acetonitrile complexes. Temperature dependence of 63Cu NMR signals for these copper-
(I) complexes show that a quadrupole relaxation process is much more significant to their 63Cu NMR line widths
than a ligand exchange process. Therefore, an electronic effect of the copper bound CO makes the 63Cu NMR
signal sharp and easily detected. The large downfield shift for the copper(I) carbonyl complex can be explained by
a paramagnetic shielding effect induced by the copper bound CO, which amplifies small structural and electronic
changes that occur around the copper ion to be easily detected in their 63Cu NMR shifts. This is evidenced by the
correlation between the 63Cu NMR shifts for the copper(I) carbonyl complexes and their
ν(CtO) values. Furthermore,
the 63Cu NMR shifts for copper(I) carbonyl complexes with imino-type tridentate ligands show a different correlation
line with those for amino-type tridentate ligands. On the other hand, 13C NMR shifts for the copper bound 13CO for
these copper(I) carbonyl complexes do not correlate with the
ν(CtO) values. The X-ray crystal structures of these
copper(I) carbonyl complexes do not show any evidence of a significant structural change around the Cu
−
CO
moiety. The findings herein indicate that CO complexation makes 63Cu NMR spectroscopy much more useful for
Cu(I) chemistry.
Introduction
electronic states, and reactivity of copper complexes have
been widely investigated with various spectroscopic methods.
UV-visible absorption, resonance Raman, and electron
paramagnetic resonance spectroscopy are powerful tools for
studying the structures and electronic states of copper(II)
complexes because of their characteristic absorptions result-
ing from d-d transitions, ligand-metal charge transfers, and
an unpaired electron in the copper(II) ion.3-9 On the other
hand, these spectroscopic methods have not been applied
extensively to studies of copper(I) complexes because of the
Copper complexes are known to play important roles in
the active site of many copper proteins in vivo as well as in
homogeneous and heterogeneous catalysis for organic chemi-
cal reactions.1-13 For the past decades, the structures,
* To whom correspondence should be addressed. E-mail: hiro@
ims.ac.jp. Tel.: +81-564-59-5578. FAX: +81-564-59-5600.
(1) Handbook of Metalloproteins; Messerschmit, A., Huber, R., Poulas,
T., Wieghardt, K., Eds.; John Wiley & Sons: Chichester, U.K., 2001.
(2) Handbook on Metalloproteins; Bertini, I., Sigel, A., Sigel, H., Eds.;
Marcel Dekker: New York, 2001.
(3) Solomon, E. I.; Chen, P.; Metz, M.; Lee, S.-K.; Palmer, A. E. Angew.
Chem., Int. Ed. 2001, 40, 4570-4590.
(9) Kim, E.; Chufa´n, E. E.; Kamaraj, K.; Karlin, K. D. Chem. ReV. 2004,
104, 1077-1133.
(4) Solomon, E. I.; Szilagyi, R. K.; Geroge, S. D.; Basumallick, L. Chem.
ReV. 2004, 104, 419-458.
(10) Nakamura, E.; Mori, S. Angew. Chem., Int. Ed. 2000, 39, 3750-3771.
(11) Thomas, A. W.; Ley, S. V. Angew. Chem., Int. Ed. 2003, 42, 5400-
(5) Rorabacher, D. B. Chem. ReV. 2004, 104, 651-697.
(6) Henkel, G.; Kreb, B. Chem. ReV. 2004, 104, 801-824.
(7) Mirica, L. M.; Ottenwaelder, X.; Stack, T. D. P. Chem. ReV. 2004,
104, 1013-1045.
5449.
(12) Beletskaya, I. P.; Cheprakov, A. V. Coord. Chem. ReV. 2004, 248,
2337-2364.
(13) Pintauer, T.; Matyjaszewski, K. Coord. Chem. ReV. 2005, 249, 1155-
(8) Lewis, E. A.; Tolman, W. B. Chem. ReV. 2004, 104, 1047-1076.
1184.
10.1021/ic060745r CCC: $37.00
Published on Web 12/23/2006
© 2007 American Chemical Society
Inorganic Chemistry, Vol. 46, No. 2, 2007 541

Kujime et al.
featureless spectroscopic properties resulting from the closed
shell d10 electron configuration of copper(I) ions.
For diamagnetic copper(I) complexes, 63Cu NMR spec-
troscopy would appear to have the greatest potential for
characterizing their structures and electronic states. 63Cu
NMR spectroscopy has been used in studies of copper(I)
complexes both in solution and in the solid state.14-32 For
example, bimolecular rate constants for the electron-exchange
processes, equilibrium constants for ligand substitution
reactions, and π acceptabilities of copper bound ligands have
been determined by 63Cu NMR spectroscopy in solution.16
However, the application of 63Cu NMR spectroscopy has
been limited to copper(I) complexes with a rigorous tetra-
hedral, Td, symmetry around the copper(I) ion. This is due
to fast quadrupole relaxation of the 63Cu (I ) 3/2) nucleus,
63
which lacks rigorous Td symmetry. As a result, only the
-
Cu NMR signals of copper complexes with a Td symmetry,
such as CuL4-type complexes, are observable, and 63Cu NMR
signals of many copper(I) complexes with reduced symmetry,
such as CuL3L′- and CuL2L′2-type complexes, are extremely
broad and frequently not even detectable.
The 63Cu NMR spectrum of a copper(I) complex lacking
rigorous Td symmetry was first reported for a copper(I)
carbonyl complex with a Cu4O4 cubane core in which each
copper(I) ion binds to three tert-butoxides and CO with a
trigonally distorted tetrahedral geometry.22 The broad 63Cu
NMR signal observed at 49 ppm at 25 °C was split into a
doublet (J(63Cu-13C) ) 800 Hz) at 60 °C. We also
successfully detected very sharp 63Cu NMR signals for
copper(I) carbonyl complexes with various hydrotris(1-
pyrazolyl)borate (TPB) ligands even at room temperature,
in spite of their C3 symmetry around copper(I) ions.30 In
addition, the 63Cu NMR chemical shifts of these complexes
Figure 1. Structures of the N-donor tridentate ligands used in this study.
correlated with the ν(CtO) values of the copper bound CO.
These findings suggest that, for other copper(I) complexes,
the copper bound CO may facilitate the detection of 63Cu
NMR signals by sharpening the signals and that CO has
considerable potential as a probe in 63Cu NMR spectroscopic
studies intending to characterize the nature of the environ-
ment around copper ions in copper complexes.
(14) Ramsey, N. F. Phys. ReV. 1950, 78, 699-703.
(15) Mann, B. E. Specialist Periodical Reports: Spectroscopic Properties
of Inorganic and Organometallic Compounds; Royal Society of
Chemistry: London, 1990; Vol. 23, pp 1-170.
(16) Malito, J. Annu. Rep. NMR Spectrosc. 1999, 38, 265-287.
(17) Yamamoto, T.; Haraguchi, H.; Fujiwara, S. J. Phys. Chem. 1970, 74,
4369-4373.
(18) Lutz, O.; Oehler, H.; Kroneck, P. Z. Naturforsch. 1978, 33A, 1021-
1024.
In this study, to examine the applicability of copper bound
CO as a 63Cu NMR probe, we obtained 63Cu NMR spectra
of copper(I) complexes with various N-donor tridentate
ligands, in which the ligands bind to copper ions in a κ3
manner (Figure 1). 63Cu NMR signals, when CO is bound
to most of the copper(I) complexes prepared in this study,
become sharper and show a large downfield shift compared
to those for the corresponding acetonitrile complexes. The
large downfield shift in the 63Cu NMR signal for the copper-
(I) carbonyl complexes is interpreted by a paramagnetic
shielding term induced by the copper bound CO, which
amplifies small structural and electronic changes that occur
around the copper ion to be easily detected in their 63Cu
NMR shifts. On the other hand, the 13C NMR shifts of the
copper bound 13CO of these copper(I) carbonyl complexes
are nearly constant. The findings herein indicate that CO
complexation makes 63Cu NMR spectroscopy much more
useful for Cu(I) chemistry.
(19) Kroneck, P.; Lutz, O.; Nolle, A.; Oehler. H. Z. Naturforsch. 1980,
35A, 221-225.
(20) Marker, A.; Gunter, M. J. J. Magn. Reson. 1982, 47, 118-132.
(21) Kroneck, P.; Kodweiâ, J.; Lutz, O.; Nolle, A.; Zepf, D. Z. Naturforsch.
1982, 37A, 186-190.
(22) Geerts, R. L.; Huffman, J. C.; Folting, K.; Lemmen, T. H.; Caulton,
K. G. J. Am. Chem. Soc. 1983, 105, 3503-3506.
(23) Kitagawa, S.; Munakata, M. Inorg. Chem. 1984, 23, 4388-4390.
(24) Nakatsuji, H.; Kanda, K.; Endo, K.; Yonezawa, T. J. Am. Chem. Soc.
1984, 106, 4653-4660.
(25) Kitagawa, S.; Munakata, M.; Sasaki, M. Inorg. Chim. Acta 1986, 120,
77-80.
(26) Endo, K.; Yamamoto, K.; Deguchi, K.; Matsushita, K. Bull. Chem.
Soc. Jpn. 1987, 60, 2803-2807.
(27) Mohr, B.; Brooks, E. E.; Rath, N.; Deutsch, E. Inorg. Chem. 1991,
30, 4541-4545.
(28) Gill, D. S. J. Chem. Soc., Faraday Trans. 1995, 91, 2307-2312.
(29) Black, J. R.; Champness, N. R.; Levason, W.; Reid, G. Inorg. Chem.
1996, 35, 1820-1824.
(30) Imai, S.; Fujisawa, K.; Kobayashi, T.; Shirasawa, N.; Fujii, H.;
Yoshimura, T.; Kitajima, N.; Moro-oka, Y. Inorg. Chem. 1998, 37,
3066-3070.
(31) Kroeker, S.; Wasylishen, R. E.; Hanna, J. V. J. Am. Chem. Soc. 1999,
121, 1582-1590.
(32) Irangu, J. K.; Jordan, R. B. Inorg. Chem. 2003, 42, 3934-3942.
542 Inorganic Chemistry, Vol. 46, No. 2, 2007

63Cu NMR Spectroscopy of Copper(I) Complexes
colorless powder. The copper acetonitrile complex was further
purified by re-crystallization from CH3CN/Et2O. Exceptionally, [(Et-
TIC)Cu(CH3CN)]ClO4, [(TPME)Cu(CH3CN)]ClO4, [(TPYM)Cu(CH3-
CN)]ClO4, and [(iPr-BITC)Cu(CH3CN)]ClO4 were used for the
preparation of carbonyl complexes without characterization. [(3,5-
Experimental Section
General Methods. The manipulation of copper(I) complexes was
performed under an Ar atmosphere using standard Schlenk tech-
niques and an inert-atmosphere glovebox. Anhydrous solvents were
purchased from commercial sources and were degassed before use.
The tridentate ligands, 3,5-iPr2-TPB,33 iPr-TIC,34 Ph-TIC,34 3,5-
iPr2-TPM,35 3,5-Me2-TPM,36 TPC,37 TPME,38 TPYM,39 Me-
TACN,40 iPr-TACN,41 Bn-TACN,42 and Me-TACD,43 were syn-
thesized as described previously. Synthesis of iPr-BITC will be
published elsewhere. The metal complexes, [(CH3CN)4Cu]ClO4,44
(3,5-iPr2-TPB)Cu,33 (3,5-iPr2-TPB)Cu(CH3CN),35 (3,5-iPr2-TPB)-
Cu(PPh3),33 [(3,5-iPr2-TPM)Cu(CH3CN)]ClO4,35 [(3,5-iPr2-TPM)-
CuCO]ClO4,35 [(iPr-TACN)Cu(CH3CN)]ClO4,45 and [(Bn-TACN)-
CuCO]ClO4,45 were prepared by a previously published method.
All other reagents were obtained from commercial sources and used
as received.
1
Me2-TPM)Cu(CH3CN)]ClO4. H NMR (CD2Cl2): δ 7.77 (1H, s),
6.08 (3H, s), 2.60 (9H, s), 2.38 (9H, s), 1.99 (3H, s). IR (KBr,
cm-1): 3091, 2994, 2949, 2919, 2253, 1568, 1466, 1415, 1394,
1306, 1247, 1088, 1036, 974, 902, 853, 827, 812, 792, 707, 625,
480. [(3,5-Ph2-TPM)Cu(CH3CN)]ClO4. 1H NMR (CD2Cl2): δ 8.46
(1H, s), 8.00 (6H, d, J ) 6.6 Hz), 7.63-7.61 (9H, m), 7.55 (3H, t,
J ) 7.6 Hz), 7.23 (6H, t, J ) 7.9 Hz), 7.07 (6H, d, J ) 8.2 Hz),
6.86 (3H, s), 2.16 (3H, s). IR (KBr, cm-1): 3127, 3099, 3061, 2927,
2292, 1638, 1558, 1486, 1460, 1438, 1412, 1371, 1269, 1215, 1152,
1077, 1030, 1005, 959, 925, 849, 824, 761, 701, 667, 636, 580,
517. [(TPC)Cu(CH3CN)]ClO4. 1H NMR (CD3CN): δ 8.62 (3H, d,
J ) 4.9 Hz), 8.09 (3H, d, J ) 8.2 Hz), 7.94 (3H, d, J ) 8.0 Hz),
7.40 (3H, d, J ) 7.7 Hz), 5.94 (1H, s), 2.14 (3H, s). IR (KBr,
cm-1): 3082, 2275, 1593, 1464, 1437, 1364, 1308, 1287, 1188,
1158, 1148, 1118, 1055, 932, 887, 776, 755, 660, 623, 517, 502.
Safety Note. Caution! Although we haVe not encountered any
problems, it should be noted that perchlorate complexes are
potentially explosiVe and should be handled with great care.
Synthesis of 3,5-Ph2-TPM. 3,5-Ph2-TPM was synthesized by
the same procedure as was used for 3,5-Me2-TPM using 3,5-
1
[(Me-TACN)Cu(CH3CN)]ClO4. H NMR (CD3CN): δ 2.64 (6H,
m), 2.62 (9H, s), 2.10 (3H, s). IR (KBr, cm-1): 2947, 2882, 2819,
2252, 1493, 1455, 1362, 1300, 1253, 1216, 1092, 1020, 982, 941,
1
diphenylimidazole as a starting material in 37% yield. H NMR
(CDCl3): δ 7.82 (1H, s), 7.80 (6H, d, J ) 7.0 Hz), 7.34 (6H, t, J
) 7.5 Hz), 7.29-7.22 (6H, m), 7.14 (6H, t, J ) 7.7 Hz), 6.81 (6H,
d, J ) 7.1 Hz), 6.57 (3H, s). 13C NMR (CDCl3): δ 151.5, 145.2,
133.1, 129.3, 129.0, 128.8, 128.5, 128.4, 128.2, 128.1, 126.3, 126.0,
105.3, 105.0. IR (KBr, cm-1): 3056, 1604, 1559, 1484, 1460, 1436,
1407, 1321, 1300, 1275, 1212, 1082, 1026, 1006, 954, 917, 835,
809, 759, 691, 579.
Synthesis of Et-TIC. Et-TIC was synthesized by the same
procedure for iPr-TIC and Ph-TIC using 2-ethylimidazole as a
starting material in 15% yield. 1H NMR (CDCl3): δ 6.76 (3H, s),
3.98 (1H, br), 3.46 (9H, s), 2.64 (6H, q, J ) 7.8 Hz), 1.20 (9H, t,
J ) 7.6 Hz). 13C NMR (CDCl3): δ 147.4, 143.7, 117.3, 69.8, 31.4,
19.4, 11.9. IR (KBr, cm-1): 3368, 2977, 2935, 2874, 1559, 1506,
1470, 1410, 1375, 1292, 1169, 1073, 1028, 971, 887.
Preparation for Copper(I) Acetonitrile Complexes. Typical
procedure: To a solution of the desired tridentate ligand in THF
was added 1 equiv of solid [Cu(CH3CN)4]ClO4. The resulting
mixture was stirred for several hours at ambient temperature and
then filtered. Pentane was added to the filtrate with rapid stirring,
causing the precipitation of the copper(I) acetonitrile complex as a
1
891, 770, 744, 625, 570, 494. [(Me-TACD)Cu(CH3CN)]ClO4. H
NMR (CD3CN): 2.73-2.68 (6H, m), 2.51-2.46 (6H, m), 2.41 (9H,
s), 1.95 (3H, s), 1.85-1.72 (6H, m).
Preparation of Copper(I) Carbonyl Complexes. Typical
procedure: The copper(I) acetonitrile complex was dissolved in a
minimum amount of CH2Cl2. The CH2Cl2 solution was stirred under
an atmosphere of CO for 1 h at ambient temperature. The diffusion
of Et2O into the solution resulted in the crystallization of the product
as a colorless solid. Because of its instability, [(iPr-BITC)CuCO]-
ClO4 was not isolated as a solid and used without crystallization.
1
[(3,5-Me2-TPM)CuCO]ClO4. Yield: 73%. H NMR (CD2Cl2): δ
7.87 (1H, s), 6.16 (3H, s), 2.67 (9H, s), 2.44 (9H, s). IR (KBr,
cm-1): 3141, 2989, 2929, 2111, 1560, 1465, 1400, 1303, 1252,
1092, 1043, 981, 907, 854, 780, 703, 663, 624. [(3,5-Ph2-TPM)-
CuCO]ClO4. Yield: 79%. 1H NMR (CD2Cl2): δ 8.51 (1H, s), 7.87
(6H, br), 7.67-7.59 (12H, m), 7.26-7.07 (12H, m), 6.88 (3H, s).
IR (KBr, cm-1): 3124, 3058, 2927, 2108, 1556, 1485, 1460, 1438,
1413, 1372, 1322, 1275, 1254, 1203, 1179, 1103, 1027, 1007, 960,
924, 841, 826, 811, 775, 758, 700, 623, 580. Anal. Calcd for
C
47.5H35N6Cl2CuO5 ([(3,5-Ph2-TPM)CuCO]ClO4·0.5CH2Cl2): C,
63.09; H, 3.90; N, 9.29. Found: C, 62.64; H, 4.21; N, 9.75. [(Et-
TIC)CuCO]ClO4. Yield: 58%. H NMR (CD2Cl2): δ 7.09 (3H,
(33) Kitajima, N.; Fujisawa, K.; Fujimoto, C.; Moro-oka, Y.; Hashimoto,
S.; Kitagawa, T.; Toriumi, K.; Tatsumi, K.; Nakamura, A. J. Am.
Chem. Soc. 1992, 114, 1277-1291.
(34) Kujime, M.; Fujii, H. Tetrahedron Lett. 2005, 46, 2809-2812.
(35) Fujisawa, K.; Ono, T.; Ishikawa, Y.; Amir, N.; Miyashita, Y.;
Okamoto, K.; Lehnert, N. Inorg. Chem. 2006, 45, 1698-1713.
(36) Julia´, S.; del Mazo, J.; Avila, L. Org. Prep. Proced. Int. 1984, 16,
299-307.
(37) Wibaut, J. P.; de Jonge, A. P.; Van der Voort, H. G. P.; Otto, H. L.
Recl. TraV. Chim. Pays-Bas 1951, 70, 1054-1066.
(38) Jonas, R. T.; Stack, T. D. P. Inorg. Chem. 1998, 37, 6615-6629.
(39) White, D. L.; Faller, J. W. Inorg. Chem. 1982, 21, 3119-3122.
(40) Wieghardt, K.; Chaudhuri, P.; Nuber, B.; Weiss, J. Inorg. Chem. 1982,
21, 3086-3090.
(41) Haselhorst, G.; Stoetzel, S.; Strassburger, A.; Walz, W.; Wieghardt,
K.; Nuber, B. J. Chem. Soc., Dalton Trans. 1993, 83-90.
(42) Beissel, T.; Della Vedova, B. S. P. C.; Wieghardt, K.; Boese, R. Inorg.
Chem. 1990, 29, 1736-1741.
1
s), 5.11 (1H, br), 3.64 (9H, s), 2.87 (6H, q, J ) 7.0 Hz), 1.36 (9H,
d, J ) 7.1 Hz). IR (KBr, cm-1): 3449, 2974, 2942, 2877, 2076,
1483, 1457, 1415, 1379, 1318, 1291, 1201, 1112, 1070, 932, 889,
779, 737, 625. Anal. Calcd for C23.3H34.6N6Cl1.6CuO6 ([(Et-TIC)-
CuCO]ClO4·0.3CH2Cl2): C, 45.50; H, 5.67; N, 13.66. Found: C,
1
45.55; H, 5.82; N, 13.33. [(iPr-TIC)CuCO]ClO4. Yield: 67%. H
NMR (CD2Cl2): δ 7.07 (3H, s), 5.21 (1H, br), 3.65 (9H, s), 3.28
(3H, sept, J ) 7.0 Hz), 1.51 (18H, d, J ) 7.1 Hz). IR (KBr, cm-1):
2970, 2930, 2872, 2069, 1483, 1421, 1383, 1362, 1310, 1283, 1089,
1
930, 902, 874, 844, 768. [(Ph-TIC)CuCO]ClO4. Yield: 69%. H
NMR (CD2Cl2): δ 7.53-7.48 (15H, m), 7.34 (3H, s), 5.32 (1H,
s), 3.65 (9H, s). IR (KBr, cm-1): 3411, 3162, 3125, 3070, 2954,
2076, 1579, 1474, 1440, 1403, 1366, 1317, 1291, 1239, 1186, 1105,
1021, 931, 921, 883, 781, 739, 713, 701, 677, 658, 624, 559.
[(TPYM)CuCO]ClO4. Yield: 45%. 1H NMR (CD2Cl2): δ 8.79 (3H,
d, J ) 4.7 Hz), 8.18 (3H, d, J ) 7.6 Hz), 8.04 (3H, t, J ) 7.7 Hz),
7.51 (3H, t, J ) 6.4 Hz), 6.47 (1H, s). IR (KBr, cm-1): 3073,
2976, 2091, 1600, 1577, 1474, 1442, 1355, 1159, 1089,1025, 912,
(43) Kallinen, K. O.; Pakkanen, T. T.; Pakkanen, T. A. J. Organomet. Chem.
1997, 547, 319-327.
(44) Hathaway, B. J.; Holah, D. G.; Postlethwaite, J. D. J. Chem. Soc.
1961, 3215.
(45) Mahapatra, S.; Halfen, J. A.; Wilkinson, E. C.; Pan, G.; Wang, X.;
Young, V. G., Jr.; Cramer, C. J.; Que, L., Jr.; Tolman, W. B. J. Am.
Chem. Soc. 1996, 118, 11555-11574.
Inorganic Chemistry, Vol. 46, No. 2, 2007 543

Kujime et al.
Table 1. Crystallographic Data for [(iPr-TIC)CuCO]ClO4, [(Me-TACN)CuCO]ClO4, and [(Bn-TACN)CuCO]ClO4
[(iPr-TIC)CuCO]ClO4‚2CH2Cl2
[(Me-TACN)CuCO]ClO4
[(Bn-TACN)CuCO]ClO4
empirical formula
C25H38N6Cl5O6Cu
orthorhombic
759.42
P212121
17.289(2)
22.316(4)
8.866(1)
3420(6)
4
1.475
10.74
-100
3322 (all data)
390
C10H21N3ClO5Cu
cubic
362.29
P213
11.3362(1)
11.3362(1)
11.3362(1)
1456.8(2)
4
1.652
17.04
-100
1101
C22H33N3ClO5Cu
orthorhombic
518.52
crystal system
mol wt
space group
a, Å
Pbca
16.205(3)
17.827(3)
18.731(3)
5411(1)
8
1.273
9.39
-100
6191 (all data)
343
b, Å
c, Å
V, Å3
Z
D
calcd, g cm-3
µ(Mo KR), cm-1
temperature, °C
no. of observations
no. of variables
R1, %
62
2.0
4.3
8.9
wR2, %
11.3
5.2
16.9
882, 850, 783, 762, 739, 648, 622, 503. [(TPC)CuCO]ClO4.
Yield: 72%. 1H NMR (CD2Cl2): δ 8.78 (3H, d, J ) 4.6 Hz), 8.48
(3H, d, J ) 8.3 Hz), 8.11 (3H, t, J ) 7.9 Hz), 7.52 (3H, t, J ) 6.1
Hz), 6.87 (1H, s). IR (KBr, cm-1): 3083, 2102, 1596, 1465, 1438,
1311, 1290, 1189, 1156, 1120, 1050, 929, 890, 774, 756, 660.
[(TPME)CuCO]ClO4. Yield: 61%. 1H NMR (CD2Cl2): δ 8.83 (3H,
d, J ) 5.2 Hz), 8.16-8.11 (6H, m), 7.59-7.56 (3H, m), 7.52 (3H,
t, J ) 6.1 Hz), 3.99 (3H, s). IR (KBr, cm-1): 3120, 3071, 2989,
2101, 1594, 1466, 1437, 1311, 1296, 1203, 1092, 1020, 979, 939,
775, 760, 672, 659, 641, 623, 524, 506. [(Me-TACN)CuCO]ClO4.
IR-460 spectrometer, as KBr pellets. The accuracy of frequency
for the IR bands was (1 cm-1. Elemental analyses were performed
at the Research Center for Molecular-Scale Nanoscience, Institute
for Molecular Science.
X-ray Data Collection. Crystallographic data and the results of
refinements are summarized in Table 1. Diffraction measurements
for the single crystals of [(iPr-TIC)CuCO]ClO4, [(Me-TACN)-
CuCO]ClO4, and [(Bn-TACN)CuCO]ClO4 were made using Mo
KR radiation (λ ) 0.71070 Å) on a Rigaku RAXIS IV imaging
plate area detector or Rigaku MSC mercury charge coupled device
at 173 K. The data were corrected for Lorentz polarization effects.
Empirical absorption corrections were applied. The structural
analysis was performed using the teXsan crystallographic software
package.46 The structures were solved by a combination of direct
method (SIR92)47 and Fourier synthesis (DIRDIF94).48 Least-
squares refinements were carried out using teXsan or SHELXL-
97.49 All non-hydrogen atoms were refined anisotropically. Hy-
drogen atoms were placed at the calculated positions except for
the O-H moiety in [(iPr-TIC)CuCO]ClO4.
1
Yield: 71%. H NMR (CD2Cl2): δ 2.99 (9H, s), 2.98(6H, m). IR
(KBr, cm-1): 2988, 2972, 2943, 2862, 2825, 2081, 1490, 1474,
1433, 1426, 1362, 1301, 1214, 1165, 1090, 1015, 986, 891, 776,
1
747, 623, 573. [(iPr-TACN)CuCO]ClO4. Yield: 57%. H NMR
(CD2Cl2): δ 3.33 (3H, sept, J ) 6.5 Hz), 3.05-2.99 (6H, m), 2.88-
2.83 (6H, m), 1.36 (18H, d, J ) 6.6 Hz). IR (KBr, cm-1): 2969,
2937, 2871, 2084, 2062, 1489, 1470, 1455, 1394, 1380, 1373, 1348,
1335, 1325, 1276, 1160, 1137, 1092, 1006, 962, 950, 841, 775,
721, 623. [(Me-TACD)CuCO])ClO4. Yield: 65%. 1H NMR (CD2-
Cl2): δ 2.94-2.89 (6H, m), 2.74-2.69 (6H, m), 2.71 (9H, s), 2.16-
2.09 (3H, m), 2.06-1.99 (3H, m). IR (KBr, cm-1): 2926, 2871,
2070, 1473, 1317, 1285, 1210, 1175, 1092, 1023, 976, 904, 874,
794, 744, 726, 622. [(iPr-BITC)CuCO])ClO4. Yield: 55%. 1H NMR
(CD2Cl2): δ 7.32-7.28 (5H, m), 7.25 (2H, s), 7.05 (2H, d, J )
7.3 Hz), 5.16 (1H, br), 3.73 (6H, s), 3.61 (2H, s), 3.24 (2H, hept,
J ) 7.0 Hz), 1.43 (6H, d, J ) 7.0 Hz), 1.29 (6H, d, J ) 7.0 Hz).
IR (in CH2Cl2, cm-1): 2093.
Results
Crystal Structures of Copper(I) Carbonyl Complexes.
The crystal structures of [(iPr-TIC)CuCO]ClO4, [(Me-
TACN)CuCO]ClO4, and [(Bn-TACN)CuCO]ClO4 were de-
termined by X-ray crystallography (Figure 2 and Table 1),
and the selected bond distances and angles are listed in Table
2. These complexes have a distorted tetrahedral geometry
around the copper ion. The coordination structures of these
copper(I) carbonyl complexes are very close to those of the
corresponding acetonitrile complexes reported previously.34,50
The copper bound CO does not induce any significant
structural change around the copper ion. To investigate the
effect of tridentate ligands on their structure, we further
compared these complexes to other copper(I) carbonyl
Preparation of 13C Enriched Copper Carbonyl Complexes.
The 13C-labeled compounds of copper carbonyl complexes were
prepared in a similar manner as described above, except 13C-labeled
carbon monoxide was used. The preparations were carried out in
NMR tubes, and the products were used without further purification.
The clean formation of the desired copper(I) carbonyl complex was
1
confirmed by comparison of its H NMR spectrum with that of a
non-labeled sample.
Instruments. NMR spectra were recorded on a JEOL JNM-
LA500 spectrometer. 13C NMR spectra were obtained at sweep-
widths of 34 kHz at 125.78 MHz using 32K data points. 63Cu NMR
spectra were obtained at sweep-widths of 200 kHz at 132.66 MHz
using 16K data points. The pulse repetition time and pulse width
were 0.1 s and 9 µs, respectively. Typically, 20 000 transients were
collected for all copper(I) complexes. The sample concentrations
of the copper(I) complexes were ∼20 mM. The chemical shift
values of 13C and 63Cu NMR spectra are referenced to external
tetramethylsilane (TMS) in chloroform and tetrakis(acetonitrile)-
copper(I) haxafluorophosphate, [(CH3CN)4Cu]PF6, in acetonitrile-
d3, respectively. FT-IR spectra were obtained using a JASCO FT/
(46) teXsan; Crystal Structure Analysis Package, version 1.11; Rigaku
Corp.: Tokyo, Japan, 2000.
(47) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Burla,
M. C.; Polidoli, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435.
(48) Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bosman, W. P.; de
Gelder, R.; Israel, R.; Smits, J. M. M. The DIRDIF-94 Program
System; Technical Report for the Crystallography Laboratory; Uni-
versity of Nijmegen: Nijmegen, The Netherlands, 1994.
(49) Sheldrick, G. M. SHELXL-97: Program for Crystal Structure Refine-
ment; University of Go¨ttingen: Go¨ttingen, Germany, 1997.
(50) Lam, B. M. T.; Halfen, J. A.; Young, V. G., Jr.; Hagadorn, J. R.;
Holland, P. L.; Cucurull-Sa´nchez, L.; Novoa, J. J.; Alvarez, S.; Tolman,
W. B. Inorg. Chem. 2000, 39, 4059-4072.
544 Inorganic Chemistry, Vol. 46, No. 2, 2007

63Cu NMR Spectroscopy of Copper(I) Complexes
Figure 2. ORTEP drawing of (a) [(iPr-TIC)CuCO]ClO4, (b) [(Me-TACN)CuCO]ClO4, and (c) [(Bn-TACN)CuCO]ClO4 with thermal ellipsoids at the 50%
probability level.
Table 2. Selected Bond Lengths (Å) and Angles (deg) for Copper(I) Carbonyl Complexes
complex
Cu-C
C-O
Cu-N (av)
Cu-C-O
N-Cu-C (av)
N-Cu-N (av)
ref
(3,5-iPr2-TPB)CuCO
1.769(8)
1.797(11)
1.76(1)
1.118(10)
1.110(6)
1.14(1)
2.018(4)
2.062(5)
2.059(1)
2.059(6)
2.041(10)
2.075(4)
2.092(13)
2.097(5)
178.6(9)
177.6(6)
178(1)
180.0(1)
175.6(6)
176.7(7)
180.0(4)
177.7(6)
124.7(4)
123.3(18)
123.2(2)
125.0(2)
126.7(3)
124.9(2)
128.24(4)
127.5(2)
90.8(9)
92.7(9)
92.9(8)
90.4(2)
87.9(14)
90.5(2)
85.72(6)
86.3(2)
33
30
35
30
35
this work
this work
this work
(3-tBu-5-Me-TPB)CuCO
(3-tBu-5-iPr-TPB)CuCO
(3,5-Ph2-TPB)CuCO
[(3,5-iPr2-TPM)CuCO]ClO4
[(iPr-TIC)CuCO]ClO4
[(Me-TACN)CuCO]ClO4
[(Bn-TACN)CuCO]ClO4
1.78(1)
1.08(1)
1.777(5)
1.783(6)
1.778(3)
1.771(6)
1.127(6)
1.130(7)
1.127(4)
1.130(7)
complexes with tridentate ligands, such as (3,5-iPr2-TPB)-
CuCO,33 (3-tBu-5-Me-TPB)CuCO,30 (3,5-Ph-TPB)CuCO,30
and [(3,5-iPr2-TPM)CuCO]ClO4.35 As shown in Table 2, no
significant change in the Cu-C and C-O bond lengths or
the Cu-C-O angles were found for these copper carbonyl
complexes. However, the bond lengths between the copper
ion and the N atoms of the tridentate ligand, Cu-N, were
changed slightly, depending on the tridentate ligand. The
Cu-N bond lengths for complexes with imino tridentate
ligands (TPB, TPM, and TIC) are shorter than those with
the amino tridentate ligands (TACN). The average Cu-N
bond lengths for the TPB derivatives are the shortest in these
complexes because TPB ligands with a negative charge can
bind positively charged copper(I) ion stronger than neutral
ligands such as TPM and TIC.
of copper(I) carbonyl complexes, and the ν(CtO) values
for copper(I) carbonyl complexes are summarized in Table
3. As shown in Table 3, the ν(CtO) values for copper(I)
carbonyl complexes are sensitive to the tridentate ligand. For
[(iPr-TACN)CuCO]ClO4, a main ν(CtO) band at 2062 cm-1
with a shoulder band at 2084 cm-1 was found in the solid
state (KBr pellet), but only a 2082 cm-1 band in dichlo-
romethane solution was observed (see Figure S1, Supporting
Information).53 The lower wavenumber band at 2062 cm-1
was consistent with a previous report for [(iPr-TACN)CuCO]-
SbF6 (2067 cm-1) in a solid state (KBr pellet).45 The ν(Ct
O) values for the tridentate ligands were increased roughly
in the order of hydrotris(pyrazolyl)borate (TPB) < tris(4-
imidazolyl)carbinol (TIC) < triazacyclododecacene (TACD)
<
triazacyclononane (TACN) < tris(pyridyl)carbinol
(TPC) < tris(pyrazolyl)methane (TPM). Contrarily, this
indicates that the π back-donation to the copper bound CO
is decreased in this order. The ν(CtO) values are also
changed by the substituents of the ligand. Interestingly, the
ν(CtO) values for the complexes containing tridentate
(53) Ratio of the signal intensities of 2062 and 2084 cm-1 bands was
changed by preparations. Although we do not have definite assign-
ments for these two bands, these two bands may be caused by a
conformational change of the isopropyl substituents in a solid state.
For example, content of solvent molecule in a microcrystal may change
a ratio of the conformational isomers.
IR Spectroscopy of Copper(I) Carbonyl Complexes.
The IR spectra of the copper(I) carbonyl complexes showed
intense absorptions in the range of 2050-2200 cm-1,
corresponding to stretching vibrations of a copper bound CO,
ν(CtO), which is sensitive to copper-CO bond charac-
ter.51,52 To investigate the electronic effect of a tridentate
ligand on the Cu-CO moiety, we obtained infrared spectra
(51) Strauss, S. H. J. Chem. Soc., Dalton Trans. 2000, 1-6.
(52) Willner, H.; Aubke, F. Angew. Chem., Int. Ed. 1997, 36, 2402-2425.
Inorganic Chemistry, Vol. 46, No. 2, 2007 545

Kujime et al.
Table 3. Vibrational Data and 63Cu and 13C NMR Chemical Shifts of Copper(I) Complexes
ν(CtO)a,
δ (63Cu NMR),b ppm
(line width W1/2, Hz)
δ (13C NMR),b ppm
complex
cm-1
(3,5-iPr2-TPB)CuCO
2061c,j
730
166
118
700
716
703
603
585
449
113
419
440
621
N.D.
609
543
504
N.D.
524
489
461
-62
419
394
585
-83
451
(205)d,h
177.2g
(3,5-iPr2-TPB)Cu(CH3CN)
(3,5-iPr2-TPB)Cu(PPh3)
(3-tBu-5-Me-TPB)CuCO
(3,5-Me2-TPB)CuCO
(3-tBu-5-iPr-TPB)CuCO
(3-Ph-5-iPr-TPB)CuCO
(3,5-Ph2-TPB)CuCO
[(3,5-iPr2-TPM)CuCO]ClO4
[(3,5-iPr2-TPM)Cu(CH3CN)]ClO4
[(3,5-Me2-TPM)CuCO]ClO4
[(3,5-Ph2-TPM)CuCO]ClO4
[(iPr-TIC)CuCO]ClO4
(12000)e
(11000)e
(70)d,h
2064c,j
2062c,j
2064c,j
2077c,j
2079c,j
2102
(110)d,h
(75)d,h
174.0h,k
174.7
(2900)d,h
(4200)d,h
(8200)
(9700)e
(4700)
(12000)
(1800)
v.b.e
2111
2108
2069
173.5
174.4
[(iPr-TIC)Cu(CH3CN)]ClO4
[(Et-TIC)CuCO]ClO4
[(Ph-TIC)CuCO]ClO4
2076
2080
2106
(1300)
(4500)
(6600)
v.b.e
[(TPC)CuCO]ClO4
[(TPC)Cu(CH3CN)]ClO4
[(TPCME)CuCO]ClO4
[(TPYM)CuCO]ClO4
[(iPr-TACN)CuCO]ClO4
[(iPr-TACN)Cu(CH3CN)]ClO4
[(Me-TACN)CuCO]ClO4
[(Bn-TACN)CuCO]ClO4
[(Me-TACD)CuCO]ClO4
[(Me-TACD)Cu(CH3CN)]ClO4
[(iPr-BITC)CuCO]ClO4
2102
2091
2062(2084)
(8700)
(6500)
(5500)
(14000)e
(4100)
(14000)
(410)
2081
2083
2070
175.2
173.3f,i
(6600)e
(10000)
2093b
a As KBr pellets. b In dichloromethane-d2. c In toluene. d In toluene-d8. e In acetonitrile-d3. f In 1,2-dichloroethane-d2. g Quartet, J ) 663 Hz. h Quartet,
J ) 655 Hz (in chloroform-d). i Quartet, J ) 586 Hz. j Reference 30. k Reference 35.
extremely sharp 63Cu NMR signal is observed for the
carbonyl complex, the 63Cu NMR signals for the acetonitrile
and triphenylphosphine complexes are quite broad. Further-
more, the carbonyl complex shows a large downfield shift
of the 63Cu NMR signal, but the acetonitrile and triph-
enylphosphine complexes do not.
Similar results were obtained for other tridentate ligands.
The 63Cu NMR signals for copper(I) acetonitrile complexes
with other tridentate ligands were also extremely broad or
undetectable and did not show a large downfield shift (Table
3). On the other hand, for most copper(I) carbonyl com-
plexes, the 63Cu NMR signals become significantly sharper
and are shifted far downfield, to around 400-800 ppm
(Figure 4 and Table 3).
To determine if the line width variations might be due to
differences in relaxation mechanisms, we examined the
temperature dependence of 63Cu NMR signals for copper(I)
Figure 3. 63Cu NMR spectra of (a) (3,5-iPr2-TPB)CuCO in CD2Cl2, (b)
complexes with various tridentate ligands. As shown in
Figure 5, the 63Cu NMR signals for all of these copper(I)
complexes become sharper with increasing temperature.54,55
This is in contrast to previous 63Cu NMR studies of
tetrahedral copper(I) acetonitrile and phosphite complexes,
[(CH3CN)4Cu]+ and [(P(OR)3)4Cu]+, whose line widths were
increased with increasing temperature.20,32 Although the
previous results could be interpreted to be due to a ligand
exchange process, the present results are consistent to a
quadrupole relaxation mechanism, for which the line width
of the 63Cu NMR signal is decreased with increasing
temperature. To further confirm the quadrupole relaxation
(3,5-iPr2-TPB)Cu(CH3CN) in CD3CN, and (c) (3,5-iPr2-TPB)Cu(PPh3) in
CD3CN at 298 K.
ligands with isopropyl substituents (3,5-iPr2-TPB, iPr-TPM,
iPr-TIC, and iPr-TACN) were smaller than those with other
substituents.
63Cu NMR Spectroscopy of Copper(I) Complexes. As
reported previously, the 63Cu NMR spectra of copper(I)
carbonyl complexes with TPB ligands exhibited very sharp
signals at around 600-700 ppm.30 To investigate the effect
of the fourth ligand on the 63Cu NMR signal, we obtained
63Cu NMR spectra of copper(I) 3,5-iPr2-TPB complexes with
various fourth ligands. As shown in Figure 3, while an
546 Inorganic Chemistry, Vol. 46, No. 2, 2007

63Cu NMR Spectroscopy of Copper(I) Complexes
Figure 5. Temperature dependent 63Cu NMR line widths for copper(I)
complexes. The solid lines show least-square fits by eq 2. These data are
summarized in Table S1, Supporting Information. O, (3,5-iPr2-TPB)CuCO
in toluene-d8; 0, (3,5-iPr2-TPB)Cu(CH3CN) in toluene-d8; 4, (3,5-iPr2-
TPB)Cu(PPh3) in toluene-d8; b, [(3,5-iPr2-TPB)CuCO]ClO4 in 1,2-C2D4-
Cl2; 9, [(iPr-TIC)CuCO]ClO4 in 1,2-C2D4Cl2; 2, [(iPr-TACN)CuCO]ClO4
in C6D5Cl; [, [(Me-TACD)CuCO]ClO4 in 1,2-C2D4Cl2.
re-orientation of the copper complex. Therefore, eq 1 is
simplified to eq 2.
exp(Ea/RT)
W1/2 ) A
(2)
T
where A is a proportionality constant. As shown in Figure
5, the temperature dependence of the line widths is fitted
properly by eq 2 with various Ea values (Table S3, Support-
ing Information). All of these results show the temperature
effect expected for quadrupole relaxation; however, as
can be seen in Figure 5 and Table S3, there are some
differences in the Ea values, with that for Me-TACD being
especially low.
13C NMR Spectroscopy of Copper(I) Carbonyl Com-
plexes. We also measured the 13C NMR spectra of the copper
bound 13CO for several copper(I) carbonyl complexes, in
order to examine the effect of the binding of CO to copper-
(I) ions from 13C NMR spectroscopy. 13C NMR signals of
copper bound CO were observed only for 13C-enriched
carbonyl complexes because of the low natural abundance
of 13CO and broadening of the signal. As shown in Figure
6, the 13C NMR signals of the copper bound 13CO are
observed around 175 ppm from TMS. The 13C NMR signal
of 13CO gas in dichloromethane is observed at 184 ppm, thus,
the 13C NMR signals show an upfield shift with the binding
of a copper(I) ion. This is in contrast to the 13C NMR signals
of heme-iron bound 13CO, which were detected more
downfield (∼200 ppm from TMS) compared to that of
13CO gas.56,57 Since the 13C NMR shifts of the copper bound
Figure 4. 63Cu NMR spectra of copper(I) carbonyl complexes,
(iPr-TPMCuCO)ClO4 (a), (iPr-TICCuCO)ClO4 (b), (TPCCuCO)ClO4 (c),
(Me-TACNCuCO)ClO4 (d), and (Me-TACDCuCO)ClO4 (e), in
CD2Cl2 at 298 K.
mechanism for these copper(I) complexes with various
tridentate ligands, the temperature dependence of the line
widths are simulated with the quadrupole relaxation rate,
given by eq 1
2
1
T2
η2 2πe2qQ
3(2I + 1)
40{I2(2I - 1)}
) πW1/2
)
1 +
(
τ )
Q
)(
)
3
h
3.9478(QCC)2τQ (1)
where W1/2 is the line width at half-height, η is the symmetry
parameter, τQ is the correlation time, and (QCC) is an
effective quadrupole coupling constant. From the Stokes-
Einstein relationship, the correlation time can be expressed
as τQ
ηsample/T, where ηsample is the viscosity of the sample
and T is temperature of the sample. At low concentration,
the viscosity of the sample is very close to the viscosity of
the solvent (ηsolvent), ηsample ≈ ηsolvent, and ηsolvent is given by
ηsolvent exp(Ea/RT), where Ea is the activation energy for
(55) We found solvent effect on the 63Cu NMR signal for (3,5-iPr2-TPB)-
CuCO (Table S2, Supporting Information). The 63Cu NMR chemical
shift correlated with the acceptor number of the solvent. Similarly,
the ν(CtO) value for (3,5-iPr2-TPB)CuCO was also changed by the
solvent. Interaction of a solvent with the copper bound CO would
change the strength of the Cu-CO bond, resulting in the change in
the 63Cu NMR chemical shift and the ν(CtO) value. On the other
hand, the 63Cu NMR line width did not correlate with the solvent
viscosity but did correlate roughly with its 63Cu NMR chemical shift.
It seems that the effect on the solvent interaction to the copper complex
is more significant to the 63Cu NMR line width than that on the solvent
viscosity.
(54) 63Cu NMR signals for copper(I) complexes slightly shifted upfield
with increasing temperature (see Supporting Information). Similar
upfield shifts were also reported in previous studies (e.g., ref 20). As
discussed in previous studies, interpretation of such temperature
dependence may be complicated by the lack of a suitable internal
reference, which would allow the temperature-induced changes in bulk
magnetic susceptibility to be accounted. On the other hand, the strength
of the Cu-CO bond would not be changed by temperature because
the ν(CtO) value for (3,5-iPr2-TPB)CuCO at 353 K was the same as
that at 298 K.
Inorganic Chemistry, Vol. 46, No. 2, 2007 547

Kujime et al.
Figure 7. Illustration of the d and p mechanism of the 63Cu NMR chemical
shift. (a) d contribution, (b) p contribution. Although this figure is written
for the p interactions, these mechanisms involve both σ and π interactions.
To further confirm a relationship of the Cu-13C hyperfine
coupling with the quadrupole relaxation process of copper,
we examined temperature dependence of 13C NMR signals
of the copper bound CO (Figure S3, Supporting Information).
With increasing temperature, the Cu-13C hyperfine couplings
of the 13C NMR signals of the copper bound CO are more
resolved. This is consistent to the idea that the loss of the
Cu-13C hyperfine coupling is due to rapid quadrupole
relaxation of the copper nucleus, because the quadrupole
relaxation rates of the copper nuclei for the copper(I)
carbonyl complexes are slowed with increasing temperature,
as shown in the previous section. On the other hand, as
shown in the temperature dependence of 63Cu NMR signal,
a rapid exchange process of the copper bound CO is not so
significant to the line shape of the13C NMR signal of the
copper bound CO because this process allows us to expect
that the Cu-13C hyperfine coupling is less resolved with
increasing temperature.
Figure 6. 13C NMR spectra of the copper(I) carbonyl complexes, (a) (3,5-
iPr2-TPB)Cu13CO, (b) [(3,5-iPr2-TPM)Cu13CO]ClO4, (c) [(iPr-TIC)Cu13-
CO]ClO4, (d) [(TPC)Cu13CO]ClO4, (e) [(Bn-TACN)Cu13CO]ClO4 in
CD2Cl2, and (f) (Me-TACDCu13CO)ClO4 (6) in 1,2-C2D4Cl2 at 298 K. Peaks
shown by asterisks in spectrum (a) are the 13C NMR signals of the copper
bound 13CO split by the hyperfine coupling with natural abundant 65Cu
nucleus (30.91%). Peaks shown by crosses are 13C NMR signals resulting
from the tridentate ligands.
13CO of the copper(I) carbonyl complexes are observed in a
narrow range (173.5-177.2 ppm), the electronic effect of
the tridentate ligand does not affect the 13C NMR shift
(Figure S2, Supporting Information). Interestingly, the 13C
NMR signals of the copper bound 13CO for (3,5-iPr2-TPB)-
CuCO and (Me-TDCN)CuCO are separated into a quartet
by the hyperfine coupling with 63Cu (J ) 663 and 586 Hz).
Furthermore, as shown by asterisks in Figure 6a, the 13C
NMR signals resulting from the hyperfine coupling of the
copper bound 13CO with the naturally abundant 65Cu nucleus
(30.91%) are observed adjacent to the most downfield and
upfield signals in the quartet signals of the copper bound
13CO coupled with the 63Cu nucleus. On the other hand, the
13C NMR signals of the copper bound 13CO for other copper-
(I) carbonyl complexes are observed as a singlet. The
hyperfine coupling of a 13C NMR signal can be observed
only for the complexes exhibiting extremely sharp 63Cu NMR
signals. For other copper carbonyl complexes exhibiting
broad 63Cu NMR signals, the Cu-13C hyperfine coupling
was not observed in their 13C NMR signals because of the
fast quadrupole relaxation processes of the copper(I) nuclei.
Discussion
63Cu NMR Spectroscopy. The copper(I) ion has a closed
shell d10 electron configuration. Thus, the observed 63Cu
NMR chemical shifts for the copper(I) complexes examined
here can be expressed by Ramsey’s equation and divided
into two terms; σ ) σD + σP.14-16 σD is a diamagnetic
shielding term, which depends on the electron densities of
the s orbitals in the ground state, and σP is a paramagnetic
shielding term representing an energy gap between the
ground and the excited states of p and/or d orbitals. Since
the diamagnetic shielding term is primarily determined by
the inner-shell molecular orbitals, it depends on the metal,
its oxidation state, and its coordination number. On the other
hand, the paramagnetic shielding term is caused by the
donation of electrons from the ligands to the outer p orbitals
of copper, a p contribution, and by the back-donation of
electrons from the copper d orbitals to the ligands, a d
contribution (Figure 7). A previous theoretical calculation
by Nakatsuji et al.24 showed that, for the 63Cu NMR chemical
(56) Moon, R. B.; Richards, J. H. Biochemistry 1974, 13, 3437-3443.
(57) Moon, R. B.; Dill, K.; Richards, J. H. Biochemistry 1977, 16, 221-
228.
548 Inorganic Chemistry, Vol. 46, No. 2, 2007

63Cu NMR Spectroscopy of Copper(I) Complexes
shift of various copper(I) complexes, the paramagnetic
contribution is a major factor and the diamagnetic contribu-
tion is almost constant and a minor factor. Furthermore, for
the 63Cu NMR shift, the d contribution is a major source of
the paramagnetic shielding term, so the 63Cu NMR chemical
shift is determined by the electron acceptability of the ligand
from a copper(I) ion. When a π-acceptor ligand is coordi-
nated to a copper(I) ion, electrons are donated from the
copper(I) ion to the ligand through the dπ(metal)-pπ*(ligand)
overlap, resulting in an electron density around copper(I)
from the original d10 to d10-n. The decrease in the electron
density around copper(I) produces a downfield shift of the
63Cu NMR signal. Therefore, the 63Cu NMR signal for a
copper(I) complex shifts downfield with increasing π ac-
ceptability of the ligand of the copper(I) ion. To the contrary,
from the viewpoint of a copper ion, the 63Cu NMR chemical
shift is a measure of the electron density of the copper ion
and indicates the extent to which the copper ion can donate
electrons to ligands.
Figure 8. Relation between 63Cu NMR chemical shifts and line widths
(W1/2) of the copper(I) carbonyl complexes. These data are summarized in
Table 3.
On the other hand, the 63Cu NMR signal is only observable
when the quadrupole relaxation is slow.14-16 This is because
the nuclear relaxation of 63Cu with a nuclear spin I ) 3/2
would be expected to be dominated by the quadrupole
relaxation mechanism. The quadrupole relaxation rate is
determined by the electronic field gradient around the
nucleus. A uniform spherical orbital does not produce an
electronic field gradient, leading to a slow quadrupole
relaxation, but an ellipsoidal orbital forms an electronic field
gradient around the nucleus resulting in a rapid quadrupole
relaxation. As a result, the 63Cu NMR signal is observable
when the environment around the copper ion approaches a
cubic symmetry and when the distorted environment around
the copper ion causes the 63Cu NMR signal to be unobserv-
ably broad. In fact, 63Cu NMR signals were observed only
for the copper complexes with a Td symmetry, tetrahedral
copper complexes binding with the same ligands.17-32 For
the quadrupole relaxation mechanism, the line width of the
63Cu NMR signal indicates the symmetry of the electronic
field around the copper nucleus. Other important factors
affecting 63Cu NMR line width are temperature and viscosity
of a NMR sample and a ligand exchange process of a copper
complex. The temperature and viscosity of the NMR sample
change the correlation time (τQ), shown in eqs 1 and 2. With
increasing the temperature or with decreasing the viscosity,
a molecular rotation of a copper complex is faster and the
correlation time is shorter. The viscosity of the sample is
changed by the concentration or temperature of the sample.
In general, the viscosity is decreased with increasing tem-
perature or with decreasing the sample concentration.
Consequently, at the same concentration, the 63Cu NMR
signal becomes sharper with increasing temperature, as
shown in eq 2. On the other hand, the 63Cu NMR line width
is increased when the ligand exchange process of the copper
complex is fast. The ligand-dissociated form, which lacks
rigorous Td symmetry, broadens the 63Cu NMR signal
because of the fast quadrupole relaxation rate. The ligand
exchange process becomes faster with increasing tempera-
ture. Therefore, when the ligand exchange process is
dominant for the 63Cu NMR relaxation process, the 63Cu
NMR signal becomes boarder with increasing temepertaure.
63Cu NMR Spectra of Copper(I) Complexes with
Tridentate Ligands. The present 63Cu NMR spectra of
copper(I) complexes with various tridentate ligands can be
interpreted on the basis of the above discussion. The copper-
(I) carbonyl complexes prepared in this study exhibit 63Cu
NMR signals in spite of the distorted tetrahedral geometries
around copper(I) ions. As shown for [(iPr-BITC)CuCO]ClO4,
the 63Cu NMR signal of copper(I) carbonyl complexes is
also detectable even with a much lower coordination sym-
metry around the copper ion. Furthermore, 63Cu NMR signals
of copper(I) carbonyl complexes with various tridentate
ligands are sharper than the corresponding acetonitrile
complexes. These results clearly indicate that the copper
bound CO sharpens the 63Cu NMR signal of a copper(I)
complex. The participation of CO in line sharpening is further
supported by the correlation between the 63Cu NMR shift
of the copper(I) carbonyl complex and its line width (W1/2),
as shown in Figure 8. Since the 63Cu NMR shift of a copper-
(I) carbonyl complex is mainly determined by the Cu-CO
bond, as discussed in detail below, the correlation suggests
that bond characteristics of Cu-CO change the electronic
field gradient around the copper ion. As suggested from the
63Cu NMR chemical shift and the ν(CtO) value, CO works
as a good π-acceptor ligand to the copper(I) ion, but
acetonitrile and triphenylphosphine do not. Therefore, the
copper bound CO can accept an electron around the copper-
(I) ion in its antibonding orbital. The π acceptability of CO
coincidentally cancels a donor effect of the tridentate ligand
and changes an asymmetric charge distribution around the
copper ion to one that is more symmetric. The symmetric
electronic field gradient around the copper ion leads to a
slow quadrupole relaxation, resulting in a sharp 63Cu NMR
signal.
Interestingly, copper(I) complexes having a phenyl sub-
stituent in their tridentate ligands exhibit broader 63Cu NMR
signals compared to the corresponding copper(I) complexes
Inorganic Chemistry, Vol. 46, No. 2, 2007 549

Kujime et al.
with other substituents. The electronic effect of the phenyl
group in a tridentate ligand, such as a ring current effect,
may interact with the electronic field gradient around the
copper ion. This would lower the symmetry of the charge
distribution around the copper ion, resulting in a broad 63Cu
NMR signal for the phenyl substituent.
The temperature dependence of the 63Cu NMR signals for
the present copper(I) complexes with a tridentate ligand
suggests that the quadrupole relaxation process, shown in
eq 2, mainly determines the line widths of the 63Cu NMR
signals and suggests that the ligand exchange processes for
these complexes are not so significant as to change the line
widths. The slow ligand exchange for the present copper
complexes may be due to a strong electron donor effect of
the tridentate ligand, which strengthens the binding of the
fourth ligand. Except for the Me-TACD complex, the
activation energies, Ea, for re-orientation of copper complexes
are reasonable compared to those expected from temperature
dependence of solvent viscosity.58 Thus, for these copper(I)
complexes, solvent viscosity controls the correlation time,
τQ, and 63Cu NMR line width. On the other hand, for the
Me-TACD complex, some intramolecular motion may
contribute to τQ and to the narrowness of the 63Cu NMR
signal because the activation energy for the Me-TACD
complex is much smaller than that for solvent viscosity.
Large coordination space of Me-TACD may be related to
the intramolecular motion. This is supported by the result
that the Me-TACD complex is an outlier in Figure 8.
Similarly, 3,5-Me2-TPM and Me-TACN complexes are also
outliers in Figure 8, maybe for the same reason as the Me-
TACD complex. Overall, various line widths for the present
copper complexes represent a symmetry of an electronic field
gradient around the copper ion, which is modulated by
electronic effects of the tridentate ligand and the fourth
ligand, as discussed above.
The 63Cu NMR signals of all copper(I) carbonyl complexes
in this study show large downfield shifts, compared to the
corresponding acetonitrile and triphenylphosphine complexes.
Since CO is a good π-accepter ligand, the large downfield
shift can be explained by the paramagnetic shielding effect
caused by π back-donation from the copper d orbitals to the
antibonding p* orbital of the copper bound CO, as discussed
above. The paramagnetic shielding effect caused by CO
expands the 63Cu NMR shift range, similar to what is seen
in NMR spectra of paramagnetic compounds. As a result,
63Cu NMR signals of the copper(I) carbonyl complexes can
be observed over a wide range depending on the tridentate
ligand, although no drastic shifts are detected for the 63Cu
NMR spectra of copper(I) acetonitrile complexes. The
difference in these 63Cu NMR shifts of copper(I) carbonyl
complexes is the result of the electron donor effect of the
tridentate ligand. Overall, the copper bound CO amplifies
small structural and electronic changes that occur around
the copper ion to be easily detected in their 63Cu NMR
shifts.
Figure 9. Correlation between NMR chemical shifts (63Cu and 13C) and
CtO stretching vibrations of copper(I) carbonyl complexes. Filled circles:
63Cu NMR for the TPB, TPM, TIC, TPC, TPCME, and TPYM complexes.
Filled squares: 63Cu NMR for TACN complexes. Filled triangle: 63Cu
NMR for the TACD complex. Filled diamond: 63Cu NMR for the BITC
complex. Empty circles: 13C NMR for (3,5-iPr2-TPB)Cu13CO, [(3,5-iPr2-
TPM)Cu13CO]ClO4, [(iPr-TIC)Cu13CO]ClO4, [(TPC)Cu13CO]ClO4, and
[(Bn-TACN)Cu13CO]ClO4. These data are summarized in Table 3.
Correlation of 63Cu NMR Shift with CtO Stretching
Vibration. The utility of the copper bound CO is further
supported by a linear correlation between the 63Cu NMR shift
of the copper(I) carbonyl complex and its ν(CtO) value. In
a previous paper,30 we showed a linear correlation between
the 63Cu NMR shifts of copper(I) carbonyl complexes with
the various hydrotris(1-pyrazolyl)borate (TPB) ligands and
their ν(CtO) values. As shown in Figure 9, the findings
herein show that the 63Cu NMR signals of all copper(I)
carbonyl complexes other than those of TPB complexes are
also correlated with their ν(CtO) values and shift to a higher
field with an increase in ν(CtO) value. This correlation can
be explained by the electronic donor effect of the tridentate
ligand to the copper ion. With an increase in electron
donation from the tridentate ligand, the electron density of
the copper ion is increased, and the filled d orbitals of d10
copper(I) ion are destabilized by electronic repulsion. As a
result, the π back-donation from copper(I) to the antibonding
orbitals of the copper bound CO would be stronger.
Therefore, the CtO bond is weakened, and the ν(CtO)
value for the copper(I) carbonyl complex shifts to a lower
wavenumber with an increase in the donor effect of the
tridentate ligand. Similarly, with an increase in the donor
effect, the 63Cu NMR shift of the copper(I) carbonyl complex
is shifted downfield because the large paramagnetic shielding
effect is induced by the strong π back-donation. Conse-
quently, the 63Cu NMR shift of the copper(I) carbonyl
complex correlates with its ν(CtO) value. This correlation
indicates that the 63Cu chemical shift of the copper(I)
carbonyl complex is a sensitive measure of the electron donor
ability of the ligand of a copper(I) complex.
Interestingly, the correlation line for the copper carbonyl
complex with the imino-type, sp2 N ligand is different from
(58) CRC Handbook of Chemistry and Physics, 86th ed.; Lide, D. R., Ed.;
CRC Press: Boca Raton, FL, 2005.
550 Inorganic Chemistry, Vol. 46, No. 2, 2007

63Cu NMR Spectroscopy of Copper(I) Complexes
that for the amino-type, sp3 N ligand. This is due to the
difference in the characteristics of the ligand N-Cu bond
between the imino-type ligand and the amino-type ligand.
The imino-type ligand binds to a copper ion via the σ and π
orbitals of the imino nitrogen atom in the ligand while the
amino-type ligand binds with the σ orbital of the amino
nitrogen atom. The imino-type ligand can accept electrons
by π back-donation from a copper ion in a π* orbital, but
the amino-type ligand cannot. This is supported by the Cu-N
(ligand) bond lengths. As shown in the X-ray crystal
structures of copper(I) carbonyl complexes (Figure 2 and
Table 2), the Cu-N (ligand) bond lengths of the imino-type
ligands are shorter than those of the amino-type ligands.
Therefore, an additional paramagnetic shielding effect would
be induced at the copper ion for the imino-type ligand,
resulting in a downfield shift of the correlation line. This
would also explain the deviation for [(iPr-BITC)CuCO]ClO4
from the correlation lines for the imino-type ligand. The
decrease in the π acceptability of [(iPr-BITC)CuCO]ClO4
by replacing one imidazole ring with an sp3-type thioether
donor may result in smaller downfield shift of the 63Cu NMR
signal than the expected shift from the correlation line (Figure
9). On the other hand, we also found that [(Me-TACD)-
CuCO]ClO4 deviates from the correlation line for the amino-
type ligand. Although we were not able to obtain the crystal
structure of [(Me-TACD)CuCO]ClO4, the crystal structure
In summary, we report a 63Cu NMR spectroscopy for
copper(I) complexes with various tridentate ligands. Copper-
(I) carbonyl complexes with various tridentate ligands give
63Cu NMR signals in spite of their C3 symmetry around
copper(I) ions. Furthermore, the paramagnetic shielding
effect caused by the copper bound CO induces a large
downfield shift of the 63Cu NMR signal for the copper(I)
carbonyl complex, which amplifies small structural and
electronic changes caused by the ligand, making them easily
detected. The findings herein indicate that CO complexation
makes 63Cu NMR spectroscopy much more useful for Cu(I)
chemistry.
Appendix
Abbreviations. TPB, hydrotris(1-pyrazolyl)borate; 3,5-
Me2-TPB, hydrotris(3,5-dimethyl-1-pyrazolyl)borate; 3,5-
iPr2-TPB, hydrotris(3,5-diisopropyl-1-pyrazolyl)borate; 3-tBu-
5-Me-TPB,
hydrotris(3-tertiarybutyl-5-methyl-1-
pyrazolyl)borate; 3-tBu-5-iPr-TPB, hydrotris(3-tertiarybutyl-
5-isopropyl-1-pyrazolyl)borate; 3-Ph-5-iPr-TPB, hydrotris(3-
phenyl-5-isopropyl-1-pyrazolyl)borate;3,5-Ph2-TPB,hydrotris(3,5-
diphenyl-1-pyrazolyl)borate; TPM, tris(1-pyrazolyl)methane;
3,5-Me2-TPM, tris(3,5-dimethyl-1-pyrazolyl)methane; 3,5-
iPr2-TPM, tris(3,5-diisopropyl-1-pyrazolyl)methane; 3,5-Ph2-
TPM, tris(3,5-diphenyl-1-pyrazolyl)methane; TIC, tris(1-
methyl-4-imidazolyl)carbinol; Et-TIC, tris(1-methyl-2-ethyl-
4-imidazolyl)carbinol; iPr-TIC, tris(1-methyl-2-isopropyl-4-
imidazolyl)carbinol; Ph-TIC, tris(1-methyl-2-phenyl-4-
imidazolyl)carbinol; TPC, tris(2-pyridyl)carbinol; TPME,
tris(2-pyridyl)methyl methyl ether; TPYM, tris(2-pyridyl)-
methane; TACN, 1,4,7-triazacyclononane; Me-TACN,
1,4,7-trimethyl-1,4,7-triazacyclononane; Bn-TACN, 1,4,7-
tribenzyl-1,4,7-triazacyclononane; iPr-TACN, 1,4,7-triiso-
propyl-1,4,7-triazacyclononane; TACD, 1,5,9-triazacyclodode-
cane; Me-TACD, 1,5,9,-trimethyl-1,5,9-triazacyclododecane;
iPr-TACD, 1,5,9,-triisopropyl-1,5,9-triazacyclododecane; iPr-
BITC, bis(1-methyl-2-isopropyl-4-imidazaolyl)phenylthiome-
thylcarbinol.
50
of [(iPr-TACD)Cu]ClO4 showed a drastic change in
coordination structure from other copper complexes with
tridentate ligands. Because of the large 12-membered mac-
rocycle, the copper ion resides deep within the macrocycle
with a Cu-N (ligand) bond distance (2.019-2.026 Å) shorter
than those (2.09-2.15 Å) observed for other amino-type
copper complexes. Furthermore, the coordination structure
of iPr-TACD to copper ion adopts a unique pyramidal
geometry; N-Cu-N angles: 113-117°. These significant
structural changes observed for the iPr-TACD complex
would be applicable to the Me-TACD complex and increase
the electron density of the copper(I) ion, leading to a
downfield deviation from the correlation line.
The strong correlation between the 63Cu NMR shift of a
copper(I) carbonyl complex and its ν(CtO) value implies
that the 13C NMR shift of the copper bound 13CO of a copper-
(I) carbonyl complex is also correlated with its ν(CtO)
value. However, as shown in Figure 9 and Figure S2, the
13C NMR shift of the copper bound CO is not correlated
with its ν(CtO) value. Therefore, the 13C NMR shift of the
copper bound CO is not a good measure of the electronic
donor ability of the ligand. In contrast to the 63Cu NMR shift,
both paramagnetic and diamagnetic shielding effects for the
13CO are changed by the binding of the copper(I) ion. Since
the change in the diamagnetic shielding effect is comparable
to the change in the paramagnetic shielding effect on the
13C NMR shift and these effects can cancel each other, the
observed 13C NMR shifts of the copper bound 13CO are
nearly constant for all complexes. As a result, the 13C NMR
shift of the copper bound CO of the copper carbonyl complex
does not correlate with its ν(CtO) value, in a straightforward
manner.
Acknowledgment. This work was supported by grants
from the Ministry of Education, Culture, Sports, Science and
Technology, Japan, and from the Japan Science and Tech-
nology Agency, CREST.
Supporting Information Available: IR spectra of [(iPr-
TACN)CuCO]ClO4 in the solid state (KBr) and in dichloromethane
solution (Figure S1), plots of 13C NMR shifts of copper(I) carbonyl
complexes over their ν(CtO) value (Figure S2), temperature
dependence of 63Cu and 13C NMR signals for copper(I) complexes
(Figures S3 and S4), 63Cu and 13C NMR and ν(CtO) data for
copper(I) complexes (Tables S1 and S2), activation energies for
copper(I) complexes (Table S3), X-ray crystallographic files in CIF
format for [(iPr-TIC)CuCO]ClO4, [(Me-TACN)CuCO]ClO4, and
[(Bn-TACN)CuCO]ClO4. This material is available free of charge
via the Internet at http://pubs.acs.org.
IC060745R
Inorganic Chemistry, Vol. 46, No. 2, 2007 551