Cu(i) and Ag(i) complexes of 7-azaindolyl and 2,2′-dipyridylamino substituted 1,3,5-triazine and benzene: The central core impact on structure, solution dynamics and fluorescence of the complexes

Article abstract of DOI:10.1039/b814393e

The interactions of Cu(i) and Ag(i) ions with four star-shaped ligands, 2,4,6-tris(N-7-azaindolyl)-1,3,5-triazine (tat), 1,3,5-tris(N-7-azaindolyl) benzene (tab), 2,4,6-tris(2,2′-dipyridylamino)-1,3,5-triazine (tdat) and 1,3,5-tris(2,2′-dipyridylamino)ben

Full text of DOI:10.1039/b814393e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Cu(I) and Ag(I) complexes of 7-azaindolyl and 2,2¢-dipyridylamino substituted
1,3,5-triazine and benzene: the central core impact on structure, solution
dynamics and fluorescence of the complexes†
Elizabeth Wong, John Li, Corey Seward and Suning Wang*
Received 19th August 2008, Accepted 14th November 2008
First published as an Advance Article on the web 27th January 2009
DOI: 10.1039/b814393e
The interactions of Cu(I) and Ag(I) ions with four star-shaped ligands, 2,4,6-tris(N-7-azaindolyl)-
1,3,5-triazine (tat), 1,3,5-tris(N-7-azaindolyl)benzene (tab), 2,4,6-tris(2,2¢-dipyridylamino)-1,3,5-
triazine (tdat) and 1,3,5-tris(2,2¢-dipyridylamino)benzene (tdab) have been investigated by X-ray
diffraction, NMR and fluorescent spectroscopic analyses. Eight new complexes [Cu(PPh3)(tat)][BF4]
(1), [Cu(PPh3)(tab)][BF4] (2), [Cu(PPh3)(tdab)][BF4] (3), {[Cu(PPh3)2]2(tdat)]}[BF4]2 (4), (AgNO3)1.5(tab)
(5), (AgNO3)2(tat) (6), (AgNO3)4(tdab) (7), and (AgNO3)3(tdat)(H2O)2 (8) have been isolated from the
reactions of [Cu(PPh3)2(CH3CN)2][BF4] and AgNO3 with the corresponding ligands. The structures of
compounds 1–7 have been established by single-crystal X-ray diffraction analyses, which show that the
central core in the chelate ligand results in distinct structures for both Cu(I) and Ag(I) complexes. All
Cu(I) complexes are discrete molecules while all Ag(I) complexes are polymeric with helical, sandwich,
or chair-like structures. A variable temperature 1H NMR study established that the Cu(I) complexes
display dynamic exchange in solution. Fluorescent titration experiments showed that the four ligands
have distinct responses toward Cu(I) and Ag(I) ions, which may be correlated to the distinct structures
of the complexes and the electronic property differences of the ligands.
Introduction
The study of metal–ligand interactions is the core of modern
inorganic chemistry. Understanding metal–ligand interactions
allows us to design and synthesize new complexes and materials
that have the desired properties for targeted applications.1 Our
group has been particularly interested in the interactions of
metal ions with luminescent ligands because of their relevance in
the development of phosphorescent materials for optoelectronic
devices,2 photochemically active materials3 and sensor materials.4
Among the ligand systems that have been examined by our
group, 7-azaindolyl and 2,2¢-dipyridylamino derivative ligands
with a central triazine or benzene core and a 3-fold symmetry
such as 2,4,6-tris(N-7-azaindolyl)-1,3,5-triazine (tat), 1,3,5-tris(N-
Chart 1
7-azaindolyl)benzene (tab), 2,4,6-tris(2,2¢-dipyridylamino)-1,3,5-
triazine (tdat) and 1,3,5-tris(2,2¢-dipyridylamino)benzene (tdab)
(shown in Chart 1), are the most attractive because of the
and luminescent properties.4d,6 We have also shown that the tab
presence of multiple binding sites and the highly emissive nature
ligand can readily form phosphorescent N,C,N-cyclometallated
of the ligands.5 Previously, we have shown that the tdab and
complexes with either a Pd(II) or Pt(II) center.7 Our current
investigation focuses on the interactions of Cu(I) and Ag(I) ions
with this group of ligands. Our interests in Cu(I) complexes are
tdat ligands can bind readily to Zn(II), Pd(II) and Pt(II) via
either the 2,2¢-dipyridylamino chelate or N,C,N-cyclometallation
to form mononuclear, dinuclear or trinuclear coordination or
motivated by several recent reports of highly efficient organic light-
organometallic compounds that have distinct structural features
emitting devices based on phosphorescent Cu(I) complexes with
N,N-chelate ligands.8 Ag(I) is in the same family as Cu(I), but
has a greater tendency to form polymeric species than Cu(I).1,9 To
Department of Chemistry, Queen’s University, Kingston, Ontario, K7L 3N6,
Canada. E-mail: wangs@chem.queensu.ca
compare the difference between these two metal ions toward the
† Electronic supplementary information (ESI) available: 1H NMR, UV-Vis
7-azaindolyl and 2,2¢-dipyridylamino functionalized ligands, we
and fluorescence spectra of complexes. CCDC reference numbers 699116–
also examined the interactions of Ag(I) with tat, tab, tdab and tdat
ligands. The details are presented herein.
699122. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/b814393e
1776 | Dalton Trans., 2009, 1776–1785
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
◦
1
in 43% yield. Mp >250 C (decomposition). H NMR (CD2Cl2,
400 MHz, 298K, d, ppm): 8.14 (d, J = 4.0 Hz, 6H, py), 7.53 (dd,
J = 8.7 Hz, J = 8.0 Hz, 6H, py), 7.37 (m, 15H, ph), 7.01 (dd, J =
8.7 Hz, J = 4.0 Hz, py), 6.81 (d, J = 8.0 Hz, 6H, py), 6.31 (s, 3H,
central phenyl). Anal. calcd. for C54H45CuN9PBF4·THF: C, 64.97;
H, 4.24; N, 12.63. Found: C, 63.81; H, 4.46; N, 11.63%.
Experimental
General procedures
All starting reagents and reagent grade solvents were purchased
from Aldrich Chemical Company and used without further
purification. The starting material [Cu(CH3CN)2(PPh3)2][BF4] was
prepared from literature procedures,10 while tat, tab, tdab and
tdat ligands were prepared according to the procedures reported
previously by our group.5 The 1H NMR and13C NMR spectra were
recorded on either Bruker Avance 300 MHz, 400 MHz or 500 MHz
spectrometers. Emission and excitation spectra were recorded on
a Photon Technologies International QuantaMaster Model C-60
Spectrometer. UV-Vis spectra were recorded on an Ocean Optics
UV-Vis spectrophotometer. Elemental analyses were performed by
Canadian Microanalytical Service Ltd., Delta, British Columbia,
Canada and by the Analytical Laboratory for Environmental
Science Research and Training, University of Toronto, Toronto,
Ontario, Canada.
Synthesis of [(Cu(PPh3)2)2(tdat)][BF4]2, (4). Under nitrogen, a
solution of tdat (0.025 g, 0.043 mmol) in CH2Cl2 was layered
with [Cu(CH3CN)2(PPh3)2][BF4] (0.097 g, 0.14 mmol) dissolved
in the same solvent. The molar ratio of the two compounds is
1 : 3. A subsequent layer of hexanes was then added. The solutions
were allowed to slowly diffuse over several days. Colorless crystals
1
of 4 were isolated in 41% yield. H NMR (CD2Cl2, 500 MHz,
298 K, d, ppm): 8.21 (br, 6H, py), 7.62 (br, 6H, py), 7.38 (t, J =
8.0 Hz, 12 H, ph), 7.22 (t, J = 8.0 Hz, 24 H, ph), 7.10 (broad and
overlapping peaks, 24 H, Ph; 12 H, py). Mp 240 ◦C. Anal. calcd.
for C105H84N12Cu2P4B2F8·0.5CH2Cl2: C, 63.92; H, 4.29; N, 8.48.
Found: C, 63.89; H, 4.52; N, 8.04%.
Synthesis of [(CuPPh3)(tat)][BF4], (1). Under nitrogen, a solu-
tion of [Cu(CH3CN)2(PPh3)][BF4] (0.106 g, 0.140 mmol) in CH2Cl2
was added to a solution of tat (0.030 g, 0.070 mmol) in the same
solvent. The reaction was carried out in a molar ratio of 1 : 2. The
mixture was stirred for 1 min and was then layered with hexanes.
The crystals formed with difficulty and the solution required
repeated cycles of dissolution and hexanes layering. Crystals of
1 were isolated in 72% yield. Mp >282 C (decomposition). H
NMR (CD2Cl2, 500 MHz, 298K, d, ppm): 8.65 (dd, J = 5.0 Hz,
J = 1.6 Hz, 2H, 7-aza), 8.60 (d, J = 4.0 Hz, 2H, 7-aza), 8.55 (dd,
J= 4.8 Hz, J = 1.6 Hz, 1H, 7-aza), 8.23 (d, J = 4.4 Hz, 1H, 7-aza),
8.18 (dd, J = 8.0 Hz, J= 1.6 Hz, 2H, 7-aza), 8.00 (dd, J= 7.6 Hz,
J = 1.6 Hz, 1H, 7-aza), 7.48 (dd, J = 8.0 Hz, J = 5.2 Hz, 1H,
7-aza), 7.32 (dd, J = 7.0 Hz, J = 4.8 Hz, 1H, 7-aza), 7.28 (m,
3H, ph), 7.17 (m, 6H, ph), 6.92 (m, 6H, ph), 6.82 (d, J = 4.4 Hz,
1H, 7-aza). 13C NMR (CD2Cl2, 300 MHz, 298K, d, ppm): 206.7,
161.5, 149.1, 146.9, 145.7, 145.1, 132.8, 132.6, 132.4, 132.3, 132.0,
130.5, 130.0, 129.2, 129.1, 128.2, 127.2, 125.7, 125.1, 121.2, 120.2,
108.6. Anal. calcd. for C42H30N9CuPBF4: C, 59.90; H, 3.60; N,
14.97. Found: C, 60.13; H, 3.32; N, 14.94%.
Synthesis of (AgNO3)1.5(tab), (5). To a solution of tab (0.025 g,
0.059 mmol) in CH2Cl2, AgNO3 (0.040 g, 0.234 mmol) dissolved
in methanol (5 mL) was slowly added. AgNO3 and tab were in
a molar ratio of 1 : 4. The vial was covered with aluminum foil
to prevent light-induced decomposition of the silver salt. The
solvents were allowed to diffuse slowly over several days. The
room temperature reaction ◦produced brown transparent crystals
◦
1
1
of 5 in 82% yield. Mp 240 C. H NMR (DMSO-d6, 300 MHz,
298K, d, ppm): 8.46 (dd, J = 4.8 Hz, J = 1.2 Hz, 6H, 7-aza), 8.32
(s, 3H, benzene), 8.25 (dd, J = 8.0 Hz, J = 1.2 Hz, 6H, 7-aza),
8.19 (d, J = 3.6 Hz, 6H, 7-aza), 7.35 (dd, J = 8.0 Hz, J = 4.8 Hz,
6H, 7-aza), 6.87 (d, J = 3.6 Hz, 6H, 7-aza). 13C NMR (DMSO-d6,
300 MHz, 298K, d, ppm): 146.5, 144.9, 138.7, 131.4, 130.1, 123.3,
118.3, 116.8. Anal. calcd. for C27H18N7.5O4.5Ag1.5·H2O: C, 46.33;
H, 2.86; N, 15.01. Found: C, 45.73; H, 2.34; N, 14.50%.
Synthesis of (AgNO3)2(tat), (6). On top of a solution of tat
(0.025 g, 0.058 mmol) in CH2Cl2 (10 mL), AgNO3 (0.040 g,
0.234 mmol) dissolved in excess methanol was layered. The molar
ratio of the two components is 1 : 4. As in the reaction for 5,
the vial was covered with aluminum foil to prevent light-induced
decomposition of the silver salt. After two days, the reaction at
room temperature yielded crystals of compound 6 in 43% yield.
Mp >300 ◦C (decomposition). 1H NMR (methanol-d4, 300 MHz,
298K, d, ppm): 8.62 (dd, J = 6.0 Hz, J = 2.1 Hz, 2H, 7-aza), 8.51
(d, J = 4.2 Hz, 2H, 7-aza), 8.35 (d, J = 5.1 Hz, 4H, 7-aza), 8.26 (d,
J = 6.0 Hz, 2H, 7-aza), 8.22 (d, J = 8.1 Hz, 4H, 7-aza), 7.51 (d,
J = 3.9 Hz, 4H, 7-aza), 7.48 (d, J = 5.4 Hz, 2H, 7-aza), 7.26 (dd,
J = 8.0 Hz, J= 5.1 Hz, 4H, 7-aza), 6.91 (d, J = 4.2 Hz, 2H, 7-aza),
6.64 (d, J= 3.6 Hz, 4H, 7-aza). Anal. calcd. for C24H15N11O6Ag2:
C, 37.47; H, 1.97; N, 20.03. Found: C, 37.21; H, 1.91; N, 19.35%.
Synthesis of [(CuPPh3)(tab)][BF4], (2). Under nitrogen, tab
(0.023 g, 0.059 mmol) was mixed with [Cu(CH3CN)2(PPh3)][BF4]
(0.177 g, 0.234 mmol) in CH2Cl2 in a molar ratio of 1 : 4. Hexanes
were then layered on top. The mixture was stirred for 1 min
and the overnight reaction at room temperature yielded colorless,
rectangular crystals of compound 2 in 65% yield. Mp >280 ◦C
1
(decomposition). H NMR (CD2Cl2, 300 MHz, 298K, d, ppm):
8.50 (s, 3H, 7-aza), 8.22 (d, J = 7.5 Hz, 3H, 7-aza), 7.72 (s, 3H,
benzene), 7.37 (dd, J = 7.8 Hz, J = 4.5 Hz, 3H, 7-aza), 7.20
(m, 3H, ph), 7.02 (m, 6H, ph), 6.88 (d, J = 3.6, 3H, 7-aza),
7.79 (m, 6H, Ph), 13C NMR (CD2Cl2, 298K, d, ppm): 143.6,
133.9, 133.1, 132.9, 131.5, 130.8, 129.6,129.1. Anal. calcd. for
C45H32N6CuPBF4·CH2Cl2: C, 59.85; H, 3.72; N, 9.10. Found: C,
60.32; H, 3.69; N, 8.81%.
Synthesis of (AgNO3)4(tdab), (7). A solution of tdab (0.051 g,
0.087 mmol) in CH2Cl2 (6 mL) was prepared and it was layered
with benzene (4 mL), methanol (4 mL) and a sonicated solution
of AgNO3 (0.043 g, 0.251 mmol) in methanol (6 mL). The vial was
covered with aluminum foil. The solvents were allowed to diffuse
slowly yielding a yellow crystal in 35% yield. 1H NMR (DMSO-
d6, 300 MHz, 298K, d, ppm): 8.25 (d, J = 4.5 Hz, 6H), 7.70 (dd,
J ª 7.7 Hz, 6H), 7.05 (m, 12H), 6.50 (s, 3H). Anal. calcd. for
Synthesis of [(CuPPh3)(tdab)][BF4], (3). Under nitrogen gas,
tdab (0.025 g, 0.043 mmol) and [Cu(CH3CN)2(PPh3)2][BF4].
(0.097 g, 140 mmol) were combined in a molar ratio of 1 : 1
in CH2Cl2. Hexanes and THF were then layered on top. After
several days, transparent light yellow crystals of 3 were isolated
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 1776–1785 | 1777
©

View Article Online
C37H31N13O12Ag4: C, 34.26; H, 2.41; N, 14.04. Found: C, 33.93; H,
2.24; N, 13.88%.
X-Ray crystallographic analysis
Single crystals of complexes 1–7 were mounted on glass fibers
for data collection. Data for 1, 2 and 4–6 were collected on a
Bruker Smart 1000 CCD X-ray diffractometer while data for 3
and 7 were collected on a Bruker Apex II single crystal X-ray
diffractometer with graphite-monochromated Mo Ka radiation,
operating at 50 kV and 30 mA, at either 298 K or 180 K. No
significant decay was observed for any of the crystals. Data were
processed on a PC using Bruker Apex II software and corrected
for absorption effects. The structural solution and refinements
were performed using the Bruker SHELXTL software package
(version 6.14).11a All structures were solved by direct methods.
Crystals 3, 4, and 6 belong to the triclinic space group P-1 while
1, 2, 5 and 7 belong to the monoclinic space groups P21/n, P21/c,
C2/c, and Cc, respectively. Crystals 2 and 4 contain disordered
CH2Cl2 solvent molecules (one per molecule of 2 and four per
molecule of 4) while crystals of 3 contain a disordered THF
solvent molecule (one per molecule). To improve the quality of the
structural data, the solvent molecules and their contributions were
removed for these three molecules using the SQUEEZE routine
in the PLATON program.11b The crystal data for 2–4 in Table 1
excluded contributions from solvent molecules in the lattice. The
Synthesis of (AgNO3)3(tdat), (8). Upon a solution of tdat
(0.050 g, 0.0849 mmol) in CH2Cl2, benzene (4 mL), methanol
(4 mL) and a sonicated solution of AgNO3 (43 mg, 0.255 mmol)
in methanol (6 mL) were successively layered. The vial was covered
in aluminum foil and the solvents were allowed to diffuse slowly
over several days, affording a white cotton ball-like powder of 8.
1H NMR (DMSO-d6, 298K, 300 MHz, d, ppm): 8.27 (d, J =
5.0 Hz, 6H), 7.76 (dd, J = 8.0 Hz, 6H), 7.47 (d, J = 8.0 Hz,
6H), 7.21 (dd, J = 5.0 Hz, J = 8.0 Hz, 6H). Anal. calcd. for
C33H24N15O9Ag3·2H2O: C, 34.85; H, 2.39; N, 18.47. Found: C,
34.84; H, 2.20; N, 18.43%.
Fluorescent titrations
All fluorescence titrations were performed at ambient temperature.
Batch solutions of ligands tab and tdab (1 ¥ 10-5 M in acetonitrile),
tat and tdat (1 ¥ 10-5 M in 4 : 1 acetonitrile–CH2Cl2) were prepared.
The metal salts (AgNO3 and [Cu(CH3CN)2(PPh3)2][BF4]) were
prepared in the same solvents used for the ligands. Prior to the
addition of the metal salt solution to the ligand solution, the
emission and excitation spectra for the free ligand solution were
measured. Titrations were performed at one minute intervals to
ensure that the solutions were at equilibrium. The emission spectra
were recorded after each addition. For the titrations with AgNO3,
the vial was kept covered with aluminum foil to prevent any photo-
induced decomposition.
-
BF4 anions in the Cu(I) complexes 1–4 all display some degrees
of disordering which were modeled and refined successfully. Some
-
of the NO3 anions in the crystals of 5–7 are disordered which
contributed to the poor quality of the structural data for 5 and
6 and as a consequence, the bond distances and angles for 5
and 6 are not as accurate as other molecules reported here. The
crystals of 5 and 7 contain one CH3OH solvent molecule per
asymmetric unit, which was refined successfully. For 7, there is a
Table 1 Crystal data for complexes 1–7
1
2
3
4
5
6
7
Formula
C42H30BCuF4- C45H33BCuF4N6- C54H42BCuF4N9 C105H84B2Cu2F8N1P4· C54H36Ag3N15O9· C48H30Ag4-
C37H31Ag4N13O13
N9P
P·CH2Cl2
924.02
P21/c
12.115(4)
14.578(4)
23.231(7)
90
P·C4H8O
1070.39
P-1
4CH2Cl2
2273.09
P-1
2CH3OH
1426.67
C2/c
N22O12
1538.42
P-1
Fw
842.07
P21/n
15.009(4)
14.840(4)
17.565(5)
90
98.363(6)
90
3870.8(19)
4
1297.23
Cc
15.0865(3)
17.9907(4)
15.2317(3)
90
90.281(1)
90
4134.09(15)
4
2.092
20.84
296(2)
52.00
8043
Space group
˚
a/A
11.2570(1)
15.4097(2)
15.5266(2)
81.520(1)
85.3200(1)
69.1880(1)
2488.85(5)
2
14.823(4)
17.746(5)
21.116(5)
83.863(5)
81.498(6)
77.153(5)
5340(2)
2
29.138(11)
9.961(4)
19.959(8)
90
112.053(7)
90
10.365(5)
10.661(5)
12.142(6)
100.933(8)
94.262(8)
115.538(7)
1170.0(9)
1
˚
b/A
˚
c/A
a/◦◦
b/◦
92.189(5)
90
4100(2)
4
g /
3
˚
V/A
5369(3)
4
Z
D
calc/g cm-3
1.445
6.70
1.497
7.64
1.428
5.39
1.414
7.26
1.765
11.62
2.183
17.46
m/cm-1
T/K
293(2)
52.00
180(2)
52.00
19058
180(2)
54.30
23805
296(2)
50.00
24845
180(2)
52.00
13863
180(2)
52.00
7037
◦
2qmax
/
No. of reflns measd 20499
No. of reflns used
7602 (0.0273) 8002 (0.0476)
10992 (0.0183)
16884 (0.0277)
5260 (0.1073)
4514 (0.0942) 5738 (0.0201)
(Rint
)
No. of params
532
553
667
1215
389
388
607
R [I > 4s(I)]:
0.0443
0.0931
0.0770
0.1109
1.060
0.0528
0.1143
0.0863
0.1237
1.029
0.0391
0.1120
0.0471
0.1162
1.079
0.0560
0.1413
0.0881
0.1544
0.957
0.0806
0.1794
0.2034
0.2342
1.020
0.0729
0.1180
0.1821
0.1375
0.847
0.0492
0.1184
0.0525
0.1219
1.034
a
b
R1 wR2
a
R (all data) R1
b
wR2
GOF on F2
2
2
a R1 = R[|Fo| - |Fc|]/R|Fo|. b wR2 = {R[w(Fo - Fc2)]/R(wFo2)}1/2. w = 1/[s (Fo2) + (0.075P)2], where P = [max. (Fo2, 0) + 2Fc2]/3.
1778 | Dalton Trans., 2009, 1776–1785
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
◦
-3
˚
˚
˚
Table 2 Selected bond lengths (A) and angles ( )
large residual electron density (3.99 e A ) which is at 1.01 A away
from Ag(1), attributable to the partial twinning of the crystal. All
non-hydrogen atoms were refined anisotropically. The positions
of hydrogen atoms were calculated, and their contributions in
structural factor calculations were included. The crystal data for
all complexes are summarized in Table 1. Important bond lengths
and angles for all the compounds are listed in Table 2.
Compound 1
Cu(1)–P(1)
Cu(1)–N(2)
Cu(1)–N(4)
Cu(1)–N(7)
2.2106(10)
2.008(2)
2.050(2)
2.132(2)
P(1)–Cu(1)–N(2)
P(1)–Cu(1)–N(4)
P(1)–Cu(1)–N(7)
N(2)–Cu(1)–N(7)
N(2)–Cu(1)–N(4)
N(4)–Cu(1)–N(7)
127.64(7)
107.85(8)
111.04(7)
88.48(10)
121.25(10)
88.49(10)
Compound 2
Cu(1)–P(1)
Cu(1)–N(4)
Cu(1)–N(5)
Cu(1) ◊ ◊ ◊ C(26)
Compound 3
Cu(1)–P(1)
Cu(1)–N(1)
Cu(1)–N(2)
Cu(1) ◊ ◊ ◊ C(35)
Compound 4
Cu(1)–P(1)
Cu(1)–P(2)
Cu(1)–N(10)
Cu(1)–N(12)
Cu(2)–P(3)
Cu(2)–P(4)
Cu(2)–N(1)
Cu(2)–N(2)
Results and discussions
2.2279(11)
2.076(3)
2.018(3)
2.429(3)
P(1)–Cu(1)–N(5)
P(1)–Cu(1)–N(4)
N(4)–Cu(1)–N(5)
135.36(9)
102.50(8)
115.82(11)
Syntheses
The ligands tat, tab, tdab and tdat were obtained using procedures
reported previously by our group.5 The copper(I) complexes 1–4
were obtained by the reactions of the appropriate ligands with
the common starting material [Cu(CH3CN)2(PPh3)2][BF4] while
the Ag(I) complexes 5–8 were obtained by the reactions of the
appropriate ligands with AgNO3.
2.2027(5)
2.0779(15)
2.0609(17)
2.3110(18)
P(1)–Cu(1)–N(2)
P(1)–Cu(1)–N(1)
N(1)–Cu(1)–N(2)
122.17(5)
121.87(5)
107.69(6)
2.2721(12)
2.2667(13)
2.169(3)
P(1)–Cu(1)–N(10)
P(1)–Cu(1)–N(12)
P(1)–Cu(1)–P(2)
P(2)–Cu(1)–N(10)
P(2)–Cu(1)–N(12)
P(3)–Cu(2)–N(1)
P(3)–Cu(2)–N(2)
P(3)–Cu(2)–P(4)
P(4)–Cu(2)–N(1)
P(4)–Cu(2)–N(2)
110.20(9)
107.12(9)
120.27(4)
98.53(9)
125.23(9)
119.65(9)
113.56(9)
116.71(5)
106.36(9)
107.30(10)
Complexes 1–3 are mononuclear Cu(I) complexes with
the formula of [Cu(PPh3)(tat)][BF4], [Cu(PPh3)(tab)][BF4] and
[Cu(PPh3)(tdab)][BF4], respectively. The fact that all three com-
pounds were isolated in good yields from the reaction of the
chelate ligand with two or more equivalents of Cu(I) ions, seems
to indicate that these mononuclear Cu(I) complexes may be
thermodynamically favored products. The common feature among
the three compounds is that there is only one PPh3 ligand in
the complex, which is quite surprising since Cu(I) ions are well
known to have a high affinity toward phosphine ligands. The 1H
NMR spectra of these three complexes display broad peaks at
ambient temperature and complex patterns at low temperature
that are consistent with the presence of dynamic exchange in
solution. Complex 4 is a dinuclear Cu(I) complex with the
formula of {[Cu(PPh3)2]2(tdat)}[BF4]2 where each Cu(I) center
is associated with two PPh3 ligands. Trinuclear Cu(I) complexes
were not isolated from any of the ligand systems, despite the use
of a large excess of Cu(I) starting material. Although complexes
1–4 were isolated as the major products from the tat, tab, tdab
and tdat ligand systems, it is very likely that these are not the only
products from the reactions. The complex dynamic behavior of
the Cu(I) complexes in solution made it impossible to identify and
characterize the minor products formed in the reactions. Therefore
our investigation focuses on the major products that we can isolate
and fully characterize.
The Ag(I) complexes 5–7 (AgNO3)1.5(tab), (AgNO3)2(tat), and
(AgNO3)4(tdab), were obtained from the reactions of the corre-
sponding ligands with excess AgNO3 (typically 3 or 4 equivalents),
in good or modest yields. Ligand tdat forms cotton-ball-like
solids with AgNO3 that were found to have the formula of
(AgNO3)3(tdat)(H2O)2 (8). Attempts were made to vary the
stoichiometry of Ag(I) versus the ligand to isolate complexes
that have different stoichiometry from those of 5–8. However,
regardless of the ratio of AgNO3 : ligand used in the actual
reaction, complexes 5–8 were isolated consistently as the major
products. All Ag(I) complexes are insoluble in common organic
solvents such as methanol, CH2Cl2, and THF. The only solvent
in which they are soluble in is DMSO. As a result, the 1H
NMR spectra of all Ag(I) complexes were recorded in DMSO. To
establish the metal–ligand interactions and the bonding modes of
the tat, tab, tdat and tdab ligands with Cu(I) and Ag(I) ions, single
2.113(3)
2.2507(11)
2.2736(12)
2.102(3)
2.084(3)
N(10)–Cu(1)–N(12)
88.72(12)
Compound 5
Ag(1)–N(2)
Ag(1)–O(4)
Ag(2)–N(6)
Ag(2)–N(4)
Ag(2)–O(1)
Ag(2)–O(1A)
Ag(2) ◊ ◊ ◊ C(25)
Compound 6
Ag(1)–O(5)
Ag(1)–N(1)
Ag(1)–N(5)
Ag(1)–O(2A)
Ag(2)–O(2)
Ag(2)–N(2)
Ag(2)–N(7)
Ag(2)–N(9)
Ag(2)–O(4A)
2.201(11)
2.43(2)
2.260(9)
2.360(10)
2.417(9)
2.611(10)
2.685(13)
N(2)–Ag(1)–N(2A)
N(2A)–Ag(1)–O(4)
N(2)–Ag(1)–O(4)
N(6)–Ag(2)–N(4)
N(6)–Ag(2)–O(1)
N(4)–Ag(2)–O(1)
Ag(2)–O(1)–Ag(2A)
160.8(6)
82.9(7)
116.2(7)
118.5(3)
145.9(3)
95.5(3)
107.1(3)
2.520(9)
2.589(10)
2.204(9)
2.282(9)
2.318(8)
2.617(9)
2.271(9)
2.327(10)
2.511(10)
N(1)–Ag(1)–N(5)
N(1)–Ag(1)–O(2A)
N(1)–Ag(1)–O(5)
N(5)–Ag(1)–O(2A)
N(5)–Ag(1)–O(5)
N(2)–Ag(2)–N(7)
N(2)–Ag(2)–N(9)
N(2)–Ag(2)–O(2)
N(2)–Ag(2)–O(4)
N(7)–Ag(2)–N(9)
N(7)–Ag(2)–O(4A)
N(9)–Ag(2)–O(2)
N(9)–Ag(2)–O(4A)
75.5(3)
117.8(3)
112.5(3)
130.5(3)
118.2(3)
71.8(3)
71.1(3)
108.4(3)
157.3(3)
101.8(3)
123.2(3)
117.1(3)
88.2(4)
Compound 7
Ag(1)–N(1)
Ag(1)–N(4)
Ag(1) ◊ ◊ ◊ C(34)
Ag(2)–N(2)
Ag(2)–N(8)
Ag(2) ◊ ◊ ◊ C(36)
Ag(3)–N(5)
Ag(3)–O(2)
Ag(3)–O(4)
Ag(4)–N(9)
Ag(4)–O(7)
2.288(8)
2.288(8)
2.649(11)
2.316(8)
2.343(8)
N(1)–Ag(1)–N(4)
N(1)–Ag(1)–C(34)
N(4)–Ag(1)–C(34)
N(2)–Ag(2)–N(8)
N(2)–Ag(2)–C(36)
N(8)–Ag(2)–C(36)
N(5)–Ag(3)–O(2)
N(5)–Ag(3)–O(4)
O(2)–Ag(3)–O(4)
N(9)–Ag(4)–O(7)
N(9)–Ag(4)–O(8)
O(7)–Ag(4)–O(8)
173.3(3)
86.2(3)
87.1(3)
167.4(3)
86.6(3)
85.5(3)
142.1(3)
104.5(4)
111.9(4)
140.0(3)
158.1(3)
51.5(3)
2.635(11)
2.254(9)
2.447(10)
2.451(11)
2.237(9)
2.448(10)
2.540(11)
2.654(10)
2.655(10)
2.706(11)
2.761(10)
2.666(10)
2.603(10)
2.715(10)
2.752(10)
Ag(4)–O(8)
Ag(1) ◊ ◊ ◊ O(7A)
Ag(1) ◊ ◊ ◊ O(9A)
Ag(2) ◊ ◊ ◊ O(2A)
Ag(2) ◊ ◊ ◊ O(3A)
Ag(2) ◊ ◊ ◊ O(11)
Ag(3) ◊ ◊ ◊ O(1)
Ag(4) ◊ ◊ ◊ O(1A)
Ag(4) ◊ ◊ ◊ O(11)
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 1776–1785 | 1779
©

View Article Online
crystal X-ray diffraction analyses were conducted for complexes
1–7. For complex 8, repeated attempts failed to produce adequate
single crystals for X-ray study. Nonetheless, the cotton-ball-like
appearance of the crystals of 8 is an indication that 8 likely has a
polymeric structure in the solid state.
and the central benzene ring are also notably longer than those in 1
(1.421(4) A). This suggests the presence of reduced p conjugation
˚
between the 7-azaindolyl and the benzene ring. Factors include
both steric repulsion between the C2-H of 7-azaindolyl and the H
atoms in the central phenyl ring and the reduced electronegativity
of the carbon atoms compared to the nitrogen atoms in the benzene
and triazine cores respectively. The tab ligand appears to wrap
around the Cu(I) center, encasing it, a consequence of the lack of
planarity. As in 1, intramolecular p–p stacking is evident between
the pyrrole ring of a 7-azaindolyl and a phenyl ring of PPh3.
However, additional intramolecular p–p stacking exists between
the non-coordinate 7-azaindolyl ring and another phenyl group
Crystal structures of 1–7
[Cu(PPh3)(tat)][BF4], (1). As shown in Fig. 1, the Cu(1) atom
in 1 is coordinated by two 7-azaindolyl groups and one PPh3.
In addition, the Cu(1) is bound to N(7) in the triazine ring with
˚
a normal bond length of 2.132(2) A, which is somewhat longer
than those of Cu(1)–N (7-azaindolyl) bonds. As a result, the
geometry around the Cu(1) center may be described as a distorted
tetrahedron. Notably, despite the coordination by a Cu(I) ion,
the tat ligand adopts a nearly planar conformation, as evidenced
by the small dihedral angles between the central triazine ring
and the three 7-azaindolyl rings (3.2◦, 8.0◦ and 23.2◦ for the
N(6), N(4) and N(2) ring, respectively). This can be attributed
to the enhanced p-conjugation between the 7-azaindolyl and the
N atoms of the triazine ring; another contributor is the favorable
H-bond interactions between the C2-H of the 7-azaindolyl and
the N atoms from the triazine. The average C–N distance between
˚
(the shortest separation distance is ~3.55 A). In principle it is
possible for the [Cu(PPh3)]+ group to migrate between different
7-azaindolyl groups, although the p–p interactions between the
aromatic rings may prevent it from doing so.
[Cu(PPh3)(tdab)][BF4], (3). The tdab ligand contains 6 pyridyl
groups which can potentially chelate to three Cu(I) ions. Our earlier
investigation of Zn(II) and Pt(II) complexes with tdab showed
that chelation always involved the two pyridyl rings bound to the
same nitrogen atom.6 In comparison with the earlier findings, the
structure of 3 shown in Fig. 2 is rather surprising and interesting.
First of all, as observed in complexes 1 and 2, the Cu(I) ion is
only bound by one PPh3 ligand. In addition, it is coordinated
by two pyridyl groups that are not bound to the same nitrogen
atom with a trigonal planar geometry, similar to that in 2. The
separation distance between a carbon atom (C(35)) in the central
˚
the 7-azaindolyl and the triazine (1.385(4) A) further suggests
the presence of strong p-conjugations. In addition, intramolecular
p–p stacking exists between the pyrrole ring of the 7-azaindolyl
and one phenyl ring of PPh3 as evidenced by the short atomic
˚
separation distances between them (~3.6 A on average). This
˚
conformation may prevent exchange processes from occurring,
and, more specifically, may prevent the Cu(I) from migrating to
the other uncoordinated sites.
benzene ring and the Cu(1) atom (2.311(1) A) is much shorter than
that observed in 2. Although various intramolecular interactions
between the phenyl groups of PPh3 and the pyridyls are present,
they are not as well defined as those in 1 and 2. Some of the
intramolecular interactions appear to be edge-on, not stacking.
One interesting feature is that the phosphorus atom is situated
directly above the centre of the benzene core ring (separation
˚
distance of 3.853(1) A). As observed in 2, the tdab ligand wraps
around the Cu(I) center in 3, which appears to provide good
protection of the Cu(I) center. Nonetheless, there are four non-
coordinating pyridyl groups in 3 that may compete for the Cu(I)
center in solution.
Fig. 1 Diagrams showing the structure of 1 (left) and 2 (right) with 30%
thermal ellipsoids.
[Cu(PPh3)(tab)][BF4], (2). The structure of 2 has some resem-
blance to that of 1, as shown in Fig. 1. The Cu(1) center is chelated
by two 7-azaindolyl groups and one PPh3 group. Although there
is a short contact distance between Cu(1) and C(26) in the
˚
central benzene ring in 2 (2.429(3) A), the Cu(1) atom is clearly
3-coordinate with a somewhat distorted trigonal planar geometry.
The coordination environment of the Cu(I) center in 2 resembles
that of {[Cu(PPh3)]2(ttab)}[BF4]2 (ttab = 1,2,4,5-tetrakis(N-7-
azaindolyl)benzene).12 Despite the use of excess Cu(I) starting
material in the reaction, as observed for 1, the 3rd 7-azaindolyl
group remains uncoordinated. In contrast to the structure of 1
where the entire tat ligand is coplanar, the three 7-azaindolyl rings
in 2 are not coplanar with the central benzene ring, but have a fairly
large dihedral angle (29.4, 45.1 and 48.4◦, for the N(2), N(4) and
N(5) rings, respectively.). The C–N bonds between the 7-azaindolyl
Fig. 2 Diagrams showing the structure of 3 (left) and 4 (right) with 30%
thermal ellipsoids. For clarity, all carbon atoms in 4 are shown in capped
stick styles.
[(Cu(PPh3)2)2 (tdat)][BF4]2, (4). The tdat ligand is an analogue
of tdab except that the central core is a triazine ring, instead of
a benzene ring. The structure of 4 shown in Fig. 2 is in sharp
contrast to those of 1–3. Instead of one PPh3 with each Cu(I)
1780 | Dalton Trans., 2009, 1776–1785
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
ion as observed in 1–3, both Cu(I) ions in 4 are bound by two
PPh3 groups. Furthermore, instead of the ‘atypical’ chelating
mode shown by 3, the Cu(I) ion in 4 is bound to two pyridyl
groups with a ‘normal” chelate mode, with both pyridyl groups
from the same dipyridylamino unit. As a result, the Cu(I) centers
in 4 display distorted tetrahedral geometry. With respect to the
triazine ring, the two [Cu(PPh3)2]+ units appear on opposite
sides, perhaps to avoid steric interactions. The distance between
˚
long Ag(1)–O(4) bond (2.43(2) A). Two CH3OH solvent molecules
are bound to the latter nitrate via hydrogen bonds. This Ag3 unit
is further linked together by a NO3 via the oxygen atom (O(1))
that acts as a bridging ligand for Ag(2) and Ag(2A) (Ag(2)-O(1)-
Ag(2A) = 107.1(3)◦), leading to the formation of a 1D chain shown
in Fig. 3.
-
(AgNO3)2(tat), (6). In the asymmetric unit of 6, there are two
AgNO3 groups attached to the tat ligand. Two of the 7-azaindolyl
groups are chelated to Ag(2) while the 3rd one is bound to Ag(1)
in a manner similar to that observed in 5. Most significant are the
short separation distances between the two Ag(I) ions and two of
the nitrogen atoms in the triazine ring (Ag(1)–N(1) = 2.589(10)
˚
Cu(1) and Cu(2) is 8.94(1) A. The structure of 4 confirms that
the “normal” dipyridylamino chelate mode of tdat or tdab to a
[Cu(PPh3)2]+ unit is possible. Therefore, the unusual bonding mode
displayed by tdab and the loss of one phosphine ligand in 3 cannot
be attributed to simple steric or electronic effects of the ligand since
the chelate groups in tdab and tdat are identical. In 4, one of the
dipyridylamino groups does not coordinate. As a result, dynamic
exchange of 4 in solution is possible.
˚
˚
A, Ag(2)–N(2) = 2.617(9) A), suggesting some degree of binding
interaction. Compared to those in 5, the three 7-azaindolyls in
6 are more coplanar with the central triazine ring, as evidenced
by the much smaller dihedral angles (23.9◦, 30.2◦ and 19.5◦ for
the N(4), N(6) and N(8) rings respectively), consistent with the
trend observed for 1 and 2. Perhaps as a consequence of the
conformation difference of tab versus tat in 5 and 6, these two
molecules have distinct stoichiometry and extended structures.
(AgNO3)1.5(tab), (5). As shown in Fig. 3, in the asymmet-
ric unit of the crystal of 5, there are two types of Ag(I) environment:
one being chelated by two 7-azaindolyl groups of the tab ligand
(Ag(2)) and one being bound by a single 7-azaindolyl group
(Ag(1)). The Ag(2) atom has a short contact distance of 2.685(13)
-
The NO3 bound to Ag(2) forms a bond with Ag(1A) from
an inversion center-related neighboring asymmetric unit via an
˚
A with C(25) in the central benzene ring. In addition, one nitrate
anion is bound to the Ag(2) center via one oxygen atom, resulting
in an approximate trigonal planar geometry around the Ag(2)
center. The Ag(1) atom sits on a C2 axis and is further bound by one
7-azaindolyl group from the C2 symmetry related asymmetric unit
in a near linear fashion (N(2)–Ag(1)–N(2A) = 160.8(6)◦). As a
result, two tab ligands link three Ag(I) ions together in 5 to form
a helical structure as shown in Fig. 3. The three 7-azaindolyls are
not coplanar with the central benzene ring as is evident in the large
dihedral angles (58.0, 46.2 and 48.8◦ for N(1), N(3) and N(5) rings
respectively). The binding modes of tab and the helical structure
of 5 resemble that observed in Pd3(tab)2Cl4 except that the C–H
bond that is in close contact with Ag(2) in 5 is broken and becomes
part of a N,C,N-chelate with the Pd center in the Pd3 complex.7
The Ag(1) ion is also bound by a disordered nitrate anion with a
˚
oxygen atom bridge, O(2), with Ag(1A)–O(2) = 2.282(9) A and
Ag(2)–O(2)–Ag(1A) = 135.3(4)◦. The O(1) atom in the same
nitrate group has a weak bond with Ag(1A), as evidenced by
˚
the Ag(1A)–O(1) distance of 2.709(10) A. As a result, molecule
6 has an unusual sandwich structure with four Ag(I) ions being
sandwiched between two tat ligands, as shown in Fig. 4. This
sandwich structure is further extended by the nitrate on Ag(1)
that forms a bond with Ag(2A) in the neighboring unit via the
˚
O(4) atom, (Ag(2)–O(4¢) = 2.511(10) A), resulting in an intriguing
1D sandwich chain, as shown in Fig. 4. There are extensive p–p
stacking interactions between the 1D chains.
(AgNO3)4(tdab), (7). In the asymmetic unit of the crystal of
7, there are four AgNO3 groups attached to one tdab ligand. As
shown in Fig. 5, the 6 pyridyl groups in tdab display two different
bonding modes with the Ag(I) ions: chelating and terminal
binding. Four pyridyl groups are chelated to Ag(1) and Ag(2),
respectively, with a near linear geometry (N(1)–Ag(1)–N(4) =
173.3(3)◦, N(2)-Ag(2)-N(8) = 167.4(3)◦). Again, as observed in the
structure of 3, the two chelating pyridyl groups on Ag(1) or Ag(2)
center are not bound to the same nitrogen atom. For Ag(I) ions,
such an “atypical” chelate mode by the two pyridyl groups may
be explained by the preference for a linear geometry by the Ag(I)
ion. The remaining two pyridyl groups act as terminal ligands
to the Ag(3) and Ag(4) ions, respectively. Both Ag(1) and Ag(2)
have a short contact distance with a carbon atom in the central
˚
benzene ring (Ag(1) ◊ ◊ ◊ C(34) = 2.649(11) A, Ag(2) ◊ ◊ ◊ C(36) =
-
˚
2.635(11) A). Ag(4) and Ag(3) are bound by one and two NO3
anions respectively, with Ag–O distances ranging from 2.447(1)
th
-
˚
˚
A to 2.540(10) A. The 4 NO3 is bound weakly to both Ag(2)
˚
and Ag(4) via the O(11) atom (Ag(2) ◊ ◊ ◊ O(11) = 2.666(10) A,
˚
Ag(4) ◊ ◊ ◊ O(11) = 2.752(10) A). All three dipyridylamino units in
the tdab ligand are approximately perpendicular to the central
benzene ring. As a result, the molecule of 7 has a chair-like
structure. Most significantly, the NO3 anions that are bound to
Ag(3) and Ag(4) also form weak bonds with Ag(1) and Ag(2)
Fig. 3 Top: a diagram showing the (AgNO3)3(tab)2 unit in 5 with 30%
thermal ellipsoids. Bottom: A diagram showing the 1D structure of 5, the
central phenyl ring is shown in black color.
-
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 1776–1785 | 1781
©

View Article Online
Fig. 4 Diagrams showing the (AgNO3)4(tat)2 unit of 6 (top) with 30%
thermal ellipsoids and the 1D structure (bottom).
atoms in the neighboring asymmetric units (see Table 2). Ag(3)
and Ag(4) from two neighboring units are also linked together by
a nitrate anion as shown in Fig. 6. As a result of the extensive
binding interactions between the Ag(I) ions and the nitrate, 7
forms a 2D extended network that stacks to form a 3D layered
crystal lattice, as shown in Fig. 6. A methanol solvent molecule
was located in the asymmetric unit that is hydrogen-bonded to a
nitrate oxygen atom (O(10)).
Compared to the corresponding Cu(I) complexes, there are
two distinct features displayed by the Ag(I) complexes. First, all
binding sites on the chelate ligands are occupied by the Ag(I) ions
in 5–7, whereas in the copper complexes, only some of the binding
sites are occupied by metal ions. Second, all Ag(I) complexes are
polymeric while all the copper complexes are discrete molecules.
These differences can in part be attributed to the presence of
the phosphine ligand in the copper complexes that saturates the
coordination sphere of Cu(I), thus preventing the formation of
polymeric species. Furthermore, the steric bulk of the PPh3 ligand
also limits the number of Cu(I) ions that can access the same
chelate ligand. The highly flexible coordination sphere around the
Ag(I) ion, compared to the relatively rigid geometry/coordination
number around the Cu(I) center, also facilitates the formation of
polymeric structures.
Fig. 5 Diagrams showing the [(AgNO3)4(tdab)] unit of 7 along with
nitrate anions from neighboring asymmetric units (O(1A) and O(7A))
(top) and the 2D structure involving 4 Ag4 units (bottom) with the central
benzene rings shown in black color.
ion to undergo dynamic site exchange in solution. To determine
1
if this is indeed the case, we examined the H NMR spectra of
complexes 1–4 at various temperatures. For complex 1, due to the
coplanarity of the tat ligand and the chelation to a Cu(I) ion, each
7-azaindolyl group has a distinct chemical environment in the solid
state. Hence, if 1 retains the same structure in solution, three sets of
peaks for the three 7-azaindolyl groups should be observed in the
1H NMR spectrum. At ambient temperature, instead of three sets,
we observed two sets of well resolved 7-azaindolyl peaks in a 2 : 1
ratio which can be assigned to coordinated and non-coordinated
7-azaindolyl rings. The non-coordinated 7-azaindolyl ring is likely
to be rotating rapidly around the N–C (triazine) bond in solution,
resulting in the exchange of the two coordinated 7-azaindolyl rings.
As temperature decreases, a new set of 7-azaindolyl peaks appear
as shown in Fig. 6, and the resulting NMR spectrum is consistent
with the crystal structure of 1. Thus, we can conclude that the
dynamic exchange in 1 is caused by the rotation of the non-
coordinated 7-azaindolyl group, not the migration of the Cu(I)
ion. The behavior of 2 resembles that of 1.
The behavior of complex 3 is quite different. At ambient
temperature, as shown in Fig. 6, the H NMR spectrum of 3
Dynamic exchange in solution by complexes 1–7
1
The presence of unoccupied binding sites in complexes 1–4 as
established by the crystal structures makes it possible for the Cu(I)
displays only one set of pyridyl peaks and one single peak for
the three protons in the central benzene ring, consistent with the
1782 | Dalton Trans., 2009, 1776–1785
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
the Ag(I) complexes in solvents other than DMSO, variable
temperature NMR spectra could not be recorded. Although the
crystal structure shows that there are clearly two different types of
pyridyl groups in 7, the 1H NMR spectrum of 7 recorded in DMSO
shows only one set of pyridyl peaks that resembles that of the free
ligand, tdab, an indication of dynamic exchange among the pyridyl
groups and the dissociation of the Ag(I) ions from the ligand in
solution. The dynamic exchange of the Ag(I) complexes is most
likely facilitated by the coordinating DMSO solvent molecules that
can compete for the Ag(I) center. Again, due to the poor solubility
of 7 in solvents other than the coordinating DMSO solvent, a
detailed NMR study on its dynamic behavior in solution could
not be performed.
Fluorescent response of tat, tab, tdab and tdat ligands towad
Cu(I) and Ag(I)
Fig. 6 1H NMR spectra of 1 (top) and 3 (bottom) recorded in CD2Cl2 at
ambient temperature and low temperature, respectively. The peaks from
the central benzene ring in 3 are marked by *. The full variable temperature
NMR spectra can be found in the ESI† (Fig. S1–S4).
Ligands tat, tab, tdat and tdab are all fluorescent with the emission
maxima ranging from 345 to 400 nm in solution. One convenient
way to study the interactions of Cu(I) and Ag(I) with these ligands
is to examine their fluorescent responses toward Cu(I) and Ag(I)
ions. We therefore carried out fluorescence titration experiments of
the four ligands with [Cu(CH3CN)2(PPh3)2][BF4] and AgNO3. For
each fluorescence titration experiment, we did the UV-Vis titration
in parallel. The UV-Vis titration data can be found in the ESI†
(S9 and S10), which support unequivocally that the fluorescent
change with the addition of metal ions is not due to the change of
absorbance because at the excitation wavelength (tat, 294 nm; tab,
298 nm; tdat, 295 nm; and tdab, 315 nm) used in the fluorescence
titration experiments, there is either little change in absorbance or
a slight increase in absorbance with the addition of metal ions.
presence of dynamic exchange between the two coordinated and
the four non-coordinated pyridyl rings. At low temperature, the
1H NMR spectrum of 3 displays a complex pattern with multiple
sets of pyridyl peaks that cannot be fully assigned due to extensive
peak overlapping. Nonetheless, the three protons in the central
benzene ring are resolved into two sets of peaks with a 2 : 1 ratio
at 215 K (their assignments are based on COSY data), which
1
is consistent with the crystal structure. Similarly, the H NMR
spectrum of complex 4 displays only one set of broad pyridyl
peaks at ambient temperature, indicating that all pyridyl rings in
4 also undergo a dynamic exchange process in solution. Thus, in
contrast to complexes 1 and 2, the Cu(I) ion in 3 and 4 is most
likely migrating between the different binding sites, causing the
dynamic exchange of the pyridyl groups.
Tat and tab. For the tat ligand, the addition of [Cu(CH3CN)2-
(PPh3)2][BF4] to its solution yields a dramatic fluorescent in-
tensity increase that reaches a plateau after the addition
of ca. one equivalent of Cu(I). Because the titration spec-
trum of tat with [Cu(CH3CN)2(PPh3)2][BF4] resembles that of
[Cu(PPh3)(tat)][BF4], (1), the fluorescent enhancement of tat can
be attributed to the formation of complex 1. The addition of
AgNO3 to the solution of tat initially results in a slight quenching
of the emission peak and little change is observed after the addition
of more than 2 equivalents of Ag(I) (see ESI† and Fig. 7). Due
to the poor solubility of the Ag(I) complexes in solvents other
than DMSO, the fluorescence spectra of complexes 5–7 were not
recorded. Comparisons between the titration spectra and those of
the complexes could not therefore be achieved for the Ag(I) ions.
Despite the structural similarity of tab and tat, the response of tab
toward Cu(I) and Ag(I) is quite different from that of tat. Instead
of the intensity gain, the addition of [Cu(CH3CN)2(PPh3)2][BF4] to
the solution of tab causes a steady quenching of the emission peak,
a response similar to that of ttab.12 The addition of AgNO3 to the
solution of tab also results in significant quenching of the emission,
indicative of the presence of binding interactions between the
metal ion and the tab ligand in solution. The distinct response
of tab and tat toward Cu(I) ions, albeit not fully understood,
suggests that the nature of the central core in the ligand does
have a significant impact on the photophysical properties of Cu(I)
complexes.
The distinct solution behavior of the tat and tab complexes
versus the tdab and tdat complexes may be explained by the fact
that in tat and tab, there are only three 7-azaindolyl groups that
can bind to the Cu(I) center while in tdab and tdat there are six
pyridyl groups available for Cu(I) binding. In addition, the non-
coordinating 7-azaindolyl group is further away from the Cu(I)
center in 1 and 2, compared to the non-coordinate pyridyl groups
in 3 and 4, thus making it more difficult for the Cu(I) ions in 1
and 2 to migrate from one site to another than those in 3 and 4.
Similar dynamic exchanges have been previously observed in the
mononuclear and dinuclear ZnCl2 complexes6a of tdab and tdat
and the complex6b of (PdCl2)(tdab). Hence, the dynamic exchange
observed for 3 and 4 can be attributed to the inherent propensity of
the tdab and tdat ligands for pyridyl site exchange in coordination
unsaturated complexes.
For the Ag(I) complexes 5 and 6, the three 7-azaindolyl groups
in the tab and tat ligands have two distinct environments: chelating
and terminal. In solution, however, only one set of 7-azaindolyl
peaks are observed in the 1H NMR spectrum in DMSO at 298 K
for both complexes. They are distinctively different from those of
their respective free ligands, indicating that the Ag(I) ions are still
bound to the tab or tat ligand with the three 7-azaindolyl groups
undergoing rapid dynamic exchange. Due to the insolubility of
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 1776–1785 | 1783
©

View Article Online
The fluorescence titration experiments demonstrate that the
central core of the ligands has a distinct impact on the luminescent
responses of the 7-azaindolyl or 2¢2-dipyridylamino, which may be
attributed to the distinct structures of the complexes. However, the
electronic properties of the central core and the conformation of
the ligand may also be key factors. Nonetheless, because of the
distinct response from the ligands with a triazine core versus that
with a benzene core toward Cu(I) and Ag(I) ions, the combination
of this group of ligands has the potential for use as fluorescent
sensors for metal ions.
In summary, our investigation has established unequivocally
that the tat, tab, tdat and tdab ligands form distinct structures with
Cu(I) and Ag(I) ions that result in distinct fluorescent responses.
The different central cores, benzene and triazine, in tat and tab,
tdat and tdab, are clearly responsible for the observed structural
and fluorescent differences.
Acknowledgements
Fig. 7 Fluorescence titration diagrams of tdat (lex = 295 nm) in
4 : 1 CH3CN–CH2Cl2 and tdab (lex 315 nm) in CH3CN with
[Cu(CH3CN)2(PPh3)2][BF4] and AgNO3, respectively. Titration diagrams
of tat (lex = 294 nm) and tab (lex = 298 nm) can be found in the ESI†.
=
We thank the Natural Sciences and Engineering Council of
Canada for financial support of this work. We thank Dr. Rui-Yao
Wang for his assistance in some of the crystal structural work.
Tdat and tdab. For the tdat ligand, both [Cu(CH3CN)2-
(PPh3)2][BF4] and AgNO3 cause significant and steady quenching
of the tdat emission. However, for Ag(I) ions, a large excess is
needed to cause significant quenching of tdat as shown by Fig. 7
and 8 Nonetheless, the fluorescent responses of Cu(I) and Ag(I)
toward tdat are similar. Despite being a structural analogue of
tdat, the response of tdab toward Cu(I) and Ag(I) is the opposite
of tdat: instead of quenching, the metal ions cause fluorescent
intensity increase. The addition of Cu(I) results in a fluorescence
turn-on response as shown in Fig. 8, which reaches a plateau after
the addition of ca. one equivalent of Cu(I). Again, the titration
spectrum of tdab by Cu(I) is similar to the fluorescence spectrum
of [Cu(PPh3)(tdab)][BF4], (3), thus the drastic emission intensity
increase can be attributed to the formation of complex 3. The
addition of AgNO3 to the solution of tdab also causes a significant
turn-on response that has a clear transition point at about three
equivalents of Ag(I). As demonstrated by the crystal structures,
both Cu(I) and Ag(I) ions chelate to the tdab ligand via two py
ligands that are not bound to the same amino nitrogen. Perhaps it
is this unique chelating mode that is responsible for the significant
fluorescent enhancement of tdab upon binding with Cu(I) and
Ag(I).
References
1 (a) Inorganic Materials, ed. D. W. Bruce and D. O’Hare, 2nd edition,
John Wliey & Sons, New York, 1996; (b) J. W. Steed, J. L. Atwood,
Supramolecular Chemistry, John Wiley & Sons, New York, 2000;
(c) Extended Linear Chain Compounds, ed. J. S. Miller, vol. I, Plenum,
New York, 1982; (d) D. Tanaka and S. Kitagawa, Chem. Mater., 2008,
20, 922; (e) S. Kitagawa and R. Matsuda, Coord. Chem. Rev., 2007,
251, 2490; (f) B. D. Chandler, G. D. Enright, K. A. Udachin, S. Pawsey,
J. A. Ripmeester, D. T. Cramb and G. K. H. Shimizu, Nat. Mater.,
2008, 7, 229; (g) S. A. Dalrymple and G. K. H. Shimizu, J. Am. Chem.
Soc, 2007, 129, 12114; (h) M. J. Katz, H. Kaluarachchi, R. J. Batchelor,
A. A. Bokov, Z.-G. Ye and D. B. Leznoff, Angew. Chem., Int. Ed., 2007,
8804.
2 (a) Z. Shen, P. E. Burrows, V. Bulovic, S. R. Borrest and M. E.
Thompson, Science, 1997, 276, 2009; (b) M. A. Baldo, S. Lamansky, P.
Burrows, M. E. Thompson and S. R. Forrest, Appl. Phys. Lett., 1999,
75, 5; (c) R. C. Kwong, S. Sibley, T. Dubovoy, M. Baldo, S. R. Forrest
and M. E. Thompson, Chem. Mater., 1999, 11, 3709; (d) Q. D. Liu, L.
Thorne, I. Kozin, D. Song, C. Seward, M. D’Iorio, Y. Tao and S. Wang,
J. Chem. Soc., Dalton Trans., 2002, 3234; (e) M. Buda, G. Kalyzuzhny
and A. Bard, J. Am. Chem. Soc., 2002, 124, 6090; (f) B. O’Regan and
M. Gra¨tzel, Nature, 1991, 353, 737.
3 (a) V. H. Houlding and V. M. Miskowski, Coord. Chem. Rev., 1991,
111, 145; (b) C. N. Pettijohn, E. B. Jochnowitz, B. Chuong, J. K. Nagle
and A. Vogler, Coord. Chem. Rev., 1998, 171, 85; (c) H. Yersin, W.
Humbs and J. Strasser, Coord. Chem. Rev., 1997, 159, 325; (d) W. Paw,
S. D. Cummings, M. A. Mansour, W. B. Connick, D. K. Gieger and R.
Eisenberg, Coord. Chem. Rev., 1998, 171, 125; (e) J. E. McGarrah,
Y. J. Kim, M. Hissler and R. Eisenberg, Inorg. Chem., 2001, 40,
4510.
4 (a) Y. Sun, N. Ross, S. B. Zhao, K. Huszarik, W. L. Jia, R. Y. Wang,
D. Macartney and S. Wang, J. Am. Chem. Soc., 2007, 129, 7510; (b) J.
Lefebvre, R. J. Batchelor and D. B. Leznoff, J. Am. Chem. Soc., 2004,
126, 16117; (c) J. C. Vickery, M. M. Olmstead, E. Y. Fung and A. L.
Balch, Angew. Chem., Int. Ed. Engl., 1997, 36, 1179; (d) J. Pang, E. J-,P.
Marcotte, C. Seward, R. S. Brown and S. Wang, Angew. Chem., Int.
Ed., 2001, 40, 4042.
5 J. Pang, Y. Tao, X.-P. Yang, M. D’Iorio and S. Wang, J. Mater. Chem.,
2002, 12, 206.
6 (a) C. Seward, J. Pang and S. Wang, Eur. J. Inorg. Chem., 2002, 6, 1390;
(b) C. Seward, W. L. Jia, R. Y. Wang and S. Wang, Inorg. Chem., 2004,
43, 978; (c) Q. D. Liu, W. L. Jia, G. Wu and S. Wang, Organometallics,
2003, 22, 3781.
Fig.
8
The Stern–Volmer plots of fluorescence titrations by
[Cu(CH3CN)2(PPh3)2][BF4] and AgNO3.
1784 | Dalton Trans., 2009, 1776–1785
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
7 (a) D. T. Song, Q. Wu, A. Hook, I. Kozin and S. Wang, Organometallics,
2001, 20, 4683; (b) Q. Wu, A. Hook and S. Wang, Angew. Chem., Int.
Ed., 2000, 39, 3933.
10 (a) G. J. Kubas, B. Monzyk and A. L. Crumbliss, Inorg. Synth., 1979,
19, 90; (b) J. Diez, S. Falagan, P. Gamasa and J. Gimeno, Polyhedron,
1988, 7, 37.
11 (a) SHELXTL, version 6.14, Bruker AXS, 2000; (b) A. L. Spek,
Acta Crystallogr., Sect. A, 1990, 46, C34; A. L. Spek, PLATON A
Multipurpose Crystallographic Tool, Utrecht University, Utrecht, The
Netherlands, 2006.
8 (a) W. L. Jia, T. McCormick, Y. Tao, J. P. Lu and S. Wang, Inorg. Chem.,
2005, 44, 5706; (b) Q. Zhang, Q. Zhou, Y. Cheng, L. Wang, D. Ma, X.
Jing and F. Wang, Adv. Mater., 2004, 16, 432.
9 (a) C. Seward, W. L. Jia, R. Y. Wang, G. D. Enright and S. Wang,
Angew. Chem., Int. Ed. Engl., 2004, 43, 2933; (b) C. Seward, J. Chan,
D. Song and S. Wang, Inorg. Chem., 2003, 42, 1112.
12 S. B. Zhao, R. Y. Wang and S. Wang, Inorg. Chem., 2006, 45,
5830.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 1776–1785 | 1785
©