Synthesis, separation, and circularly polarized luminescence studies of enantiomers of iridium(III) luminophores

Article abstract of DOI:10.1021/ic701747j

A family of heteroleptic (C∧N)₂Ir(acac) and homoleptic fac-Ir(C∧N)₃ complexes have been synthesized and their photophysical properties studied (where C∧N = a substituted 2-phenylpyridine and acac = acetylacetonate). The neutral Δ and ∧ complexes were separated with greater than 95% enantiomeric purity by chiral supercritical fluid chromatography, and the solution circular dichroism and circularly polarized luminescence spectra for each of the enantio-enriched iridium complexes were obtained. The experimentally measured emission dissymmetries (g_em) for this series compared well with predicted values provided by time-dependent density functional theory calculations. The discovered trend further showed a correlation with the dissymmetries of ionic, enantiopure hemicage compounds of Ru(II) and Zn(II), thus demonstrating the applicability of the model for predicting emission dissymmetry values across a wide range of complexes.

Full text of DOI:10.1021/ic701747j

Inorg. Chem. 2008, 47, 2039-2048
Synthesis, Separation, and Circularly Polarized Luminescence Studies
of Enantiomers of Iridium(III) Luminophores
Frederick J. Coughlin,† Michael S. Westrol,† Karl D. Oyler,† Neal Byrne,‡ Christina Kraml,‡
Eli Zysman-Colman,† Michael S. Lowry,† and Stefan Bernhard*,†
Department of Chemistry, Princeton UniVersity, Princeton, New Jersey 08544, and Lotus
Separations, LLC, 201 Frick Laboratory, Princeton, New Jersey 08544
Received September 5, 2007
A family of heteroleptic (C^N)2Ir(acac) and homoleptic fac-Ir(C^N)3 complexes have been synthesized and their
photophysical properties studied (where C^N ) a substituted 2-phenylpyridine and acac ) acetylacetonate). The
neutral ∆ and Λ complexes were separated with greater than 95% enantiomeric purity by chiral supercritical fluid
chromatography, and the solution circular dichroism and circularly polarized luminescence spectra for each of the
enantio-enriched iridium complexes were obtained. The experimentally measured emission dissymmetries (gem) for
this series compared well with predicted values provided by time-dependent density functional theory calculations.
The discovered trend further showed a correlation with the dissymmetries of ionic, enantiopure hemicage compounds
of Ru(II) and Zn(II), thus demonstrating the applicability of the model for predicting emission dissymmetry values
across a wide range of complexes.
Introduction
synthetically stable, and possess a facile synthesis.11 While
racemic mixtures of Ir(III) luminophores emit light with no
net polarization, enantio-enriched complexes emit circularly
polarized light with a polarization bias. Possible uses for such
compounds include three-dimensional electronic displays12
and biological assays.13
Enantiomeric resolution of racemic, bidentate transition
metal complexes into their optically active Λ and ∆ isomers
can be quite difficult and is typically attempted through one
of several methods. The most commonly employed technique
is cocrystallization with an appropriate chiral counterion;14
this strategy can be effective for ionic metal complexes but
is not conducive for neutral complexes as is the case here.
A second means is through synthetic predetermination of
the helicity about the metal center, as demonstrated by our
group15 as well as von Zelewsky and co-workers,16,17 using
Organic light-emitting diodes (OLEDs)1 incorporating
cyclometalated iridium(III) complexes have received in-
creased attention recently.2–10 These compounds hold par-
ticular promise as they are readily color-tuned through
systematic modification of the ligands, possess high phos-
phorescent quantum efficiencies, are photochemically and
* Author to whom correspondence should be addressed. E-mail: bern@
princeton.edu.
†
Princeton University.
Lotus Separations, LLC.
‡
(1) Hung, L. S.; Chen, C. H. Mater. Sci. Eng. Rep. 2002, 143.
(2) Dixon, I. M.; Collin, J.-P.; Sauvage, J.-P.; Flamigni, L.; Susana, E.;
Barigelletti, F. Chem. Soc. ReV. 2000, 29, 385–391.
(3) Adachi, C.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Appl. Phys.
2001, 90, 5048.
(4) Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest,
S. R. Appl. Phys. Lett. 1999, 75, 4.
(5) Nazeeruddin, M. K.; Humphry-Baker, R.; Berner, D.; Rivier, S.;
Zuppiroli, L.; Grätzel, M. J. Am. Chem. Soc. 2003, 125, 8790.
(6) Tsuboyama, A.; Iwawaki, H.; Furugori, M.; Mukaide, T.; Kamatani,
J.; Igawa, S.; Moriyama, T.; Miura, S.; Takiguchi, T.; Okada, S.;
Hoshino, M.; Ueno, K. J. Am. Chem. Soc. 2003, 125, 12971–12979.
(7) Laskar, I. R.; Hsu, S.-F.; Chen, T.-M. Polyhedron 2006, 25, 1167–
1176.
(11) Lowry, M. S.; Hudson, W. R.; Pascal, R. A., Jr.; Bernhard, S. J. Am.
Chem. Soc. 2004, 126, 14129–14135.
(12) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.; Lee, H.-
E.; Adachi, C.; Burrows, P. E.; Forrest, S. R.; Thompson, M. E. J. Am.
Chem. Soc. 2001, 123, 4304–4312.
(13) Lo, K. K.-W.; Hui, W.-K.; Chung, C.-K.; Tsang, K. H.-K.; Ng, D. C.-
M.; Zhu, N.; Cheung, K.-K. Coord. Chem. ReV. 2005, 249, 1434–
1450.
(8) Hsu, N.-M.; Li, W.-R. Angew. Chem., Int. Ed. 2006, 45, 4138–4142.
(9) Slinker, J. D.; Gorodetsky, A. A.; Lowry, M. S.; Wang, J. J.; Parker,
S.; Rohl, R.; Bernhard, S.; Malliaras, G. G. J. Am. Chem. Soc. 2004,
126, 2763–2767.
(14) Sagüés, J. A. A.; Gillard, R. D.; Smalley, D. H.; Williams, P. A. Inorg.
Chim. Acta 1980, 43, 211–216.
(10) Slinker, J. D.; Koh, C. Y.; Malliaras, G. G.; Lowry, M. S.; Bernhard,
S. Appl. Phys. Lett. 2005, 86, 173506.
(15) Oyler, K. D.; Coughlin, F. J.; Bernhard, S. J. Am. Chem. Soc. 2007,
129, 210–217.
10.1021/ic701747j CCC: $40.75  2008 American Chemical Society
Inorganic Chemistry, Vol. 47, No. 6, 2008 2039
Published on Web 02/14/2008

Coughlin et al.
caged, CHIRAGEN-type pineno-polypyridines. Although
successful for both ionic and neutral complexes, the general-
ity and applicability of this method is limited as it requires
a multistep synthesis of the chiral ligand. Such a process
does not lend itself well to systematic studies of subsituent
effects for the purposes of excited-state tuning. Finally,
chromatographic resolution of Λ and ∆ isomers is theoreti-
cally possible using a chiral stationary phase but to our
knowledge has not been demonstrated or developed for
neutral iridium complexes.
When discussing chiral luminescent compounds, a primary
property of interest is a complex’s emission dissymmetry
factor (gem), which indicates the degree to which one type
of circularly polarized light (right-handed or left-handed) is
preferentially emitted over the other. This value can be
measured through circular polarized luminescence (CPL)
spectroscopy and is described as the ratio of differential
emission intensity (∆I) to the total intensity (I) in eq 1, where
IL and IR represent the intensity of left and right circularly
polarized luminescence, respectively.
by CPL, and then using this data to establish a predictable
trend in emission dissymmetry using density functional
theory (DFT) calculations of the excited state.
For the compounds to be tested, a family of heteroleptic
(C^N)2Ir(acac) and homoleptic fac-Ir(C^N)3 complexes were
chosen. bis-Cyclometalated iridium complexes containing the
ꢀ-diketonato ancillary ligand, acac (where acac ) acetylaceto-
nate), have been and continue to be an area of great interest.
The acac ligand’s orbitals have been shown, both computa-
tionally and experimentally, to be spectroscopically passive,
serving only to affect the ligand field splitting of the complex
and thus decrease the highest occupied molecular orbital-lowest
unoccupied molecular orbital (HOMO–LUMO) energy gap
relative to tris-cyclometalated, homoleptic analogues.22 Herein,
we report the synthesis and enantiomeric separation of this series
of six structurally diverse iridium complexes along with their
chiroptical properties. Subsequently, computational methods
were employed to predict chiroptical absorption (circular
dichroism, CD) as well as emission behavior (circular polarized
luminescence); these computer simulations were found to agree
strongly with the experimental data and provide a heretofore
unprecedented means of predicting emission dissymmetry from
transition metal complexes.
IL - IR
∆I
gem(λ) )
)
(1)
1
1
2
I
(IL + IR)
Experimental Section
2
General. The syntheses of all ligands and their pyridinium salt
precursors have been previously reported11 with the exception of
3-(4-fluorophenyl)-5,6,7,8-tetrahydroisoquinoline (FCyppy, 1) and
3-(4-bromophenyl)-5,6,7,8-tetrahydroisoquinoline (BrCyppy, 2). All
solvents and reagents were purchased commercially from either
Aldrich or Alfa Aesar and used without any further purification.
1H NMR spectra were recorded on Varian/INOVA 400 and 500
MHz spectrometers. 13C NMR spectra were performed on a Bruker
Avance II 500 with TCI cryoprobe with APT spectra processing.
Electrospray mass spectra were performed on a Hewlett-Packard
5898B MS Engine. Elemental analysis was carried out by the
University of Illinois at Urbana–Champaign Microanalysis Labora-
tory on an Exeter CE440. HRMS analysis was conducted at the
University of Iowa High Resolution Mass Spectrometry Facility.
3-(4-Fluorophenyl)-5,6,7,8-tetrahydroisoquinoline (1). To a
solution of 2′-pyridinium bromide-4-fluoroacetophenone (1.51 g,
5.1 mmol, 1 equiv) in 23 mL MeOH was added 1-cyclohexenecar-
boxaldehyde (680 µL, 5.1 mmol, 1 equiv) and NH4OAc (2.2 g,
28.5 mmol, 5.6 equiv). The solution was refluxed under N2 for 20 h.
Upon cooling, the reaction was quenched with H2O and extracted
5× hexanes. The organic layer was washed 2× with H2O followed
by 3× with 2 M HCl, and then the aqueous layer was basified
with NaOH (s) and subsequently extracted with 4× hexanes. The
organic layer was dried over MgSO4 and filtered through a celite
plug. The organic layer was concentrated under reduced pressure
then in Vacuo to yield a light-brown liquid. Yield: 86%. Rf (10%
For most practical applications, it is desirable to maximize
the absolute value of gem as much as possible in order to
produce a signal strong enough to be readily detectable.
Typically, values reported in the literature are on the order
of 10-3 in solution for transition metal complexes.18–20
Although some lanthanide compounds have exhibited higher
emission dissymmetries21 as a result of their metal-centered
f f f transitions, their luminescence is not readily color-
tunable, as is the case for certain d-block complexes.
Owing largely to the difficulties associated with enantio-
meric resolution along with the lack of commercial CPL
spectrometers, little to no work has been conducted to
establish any firm relationship or trend between chemical
structure and the magnitude of emission dissymmetry, which
can subsequently be used as a means to predict the gem value
for a given compound. Such a model would serve as a
practical tool in the discovery of transition metal complexes
possessing both useful luminescent properties and high
emission dissymmetry factors. For this reason, we chose to
investigate a series of neutral Ir(III) metal complexes by
resolving each into its ∆ and Λ isomers through chiral
chromatography, experimentally measuring their gem factors
1
EtOAc/hexanes on silica): 0.28. H NMR (500 MHz, CDCl3): δ
(16) Hamann, C.; von Zelewsky, A.; Neels, A.; Stoeckli-Evans, H. J. Chem.
Soc., Dalton Trans. 2004, 402–406.
8.33 (s, 1H), 7.90 (dd, J ) 5.5, 8.5 Hz, 1H), 7.33 (s, 1H), 7.10 (t,
J ) 8.5 Hz, 2H), 2.77 (m, 4H), 1.82 (m, 4H). 13C NMR (125 MHz,
CDCl3): δ 164.1, 162.1, 153.3, 150.2, 146.7, 135.8, 131.6, 128.3,
128.2, 120.3, 115.4, 113.3, 28.9, 25.9, 22.6, 22.4. EI-LRMS calcd
for C15H14FN: 227.28. Found: 228. Anal. calcd for C15H14FN [M+]:
C, 79.27; H, 6.21; N, 6.16. Found: C, 79.02; H, 6.24; N, 6.23.
(17) Schaffner-Hamann, C.; von Zelewsky, A.; Barbieri, A.; Barigelletti,
F.; Muller, G.; Riehl, J. P.; Neels, A. J. Am. Chem. Soc. 2004, 126,
9339–9348.
(18) Riehl, J. P.; Richardson, F. S. Chem. ReV. 1986, 86, 1–16.
(19) Gunde, K. E.; Credi, A.; Jandrasics, E.; von Zelewsky, A.; Richardson,
F. S. Inorg. Chem. 1997, 36, 426–434.
(20) Field, J. E.; Muller, G.; Riehl, J. P.; Venkataraman, D. J. Am. Chem.
Soc. 2003, 125, 11808–11809.
(21) Petoud, S.; Muller, G.; Moore, E. G.; Xu, J.; Sokolnicki, J.; Riehl,
J. P.; Le, U. N.; Cohen, S. M.; Raymond, K. N. J. Am. Chem. Soc.
2007, 129, 77–83.
(22) Ragni, R.; Plummer, E. A.; Brunner, K.; Hofstraat, J. W.; Babudri,
F.; Farinola, G. M.; Naso, F.; De Cola, L. J. Mater. Chem. 2006, 16,
1161–1170.
2040 Inorganic Chemistry, Vol. 47, No. 6, 2008

Enantiomers of Iridium(III) Luminophores
3-(4-Bromophenyl)-5,6,7,8-tetrahydroisoquinoline (2). To a
solution of 2′-pyridinium bromide-4-bromoacetophenone (10.6 g,
29.6 mmol, 1 equiv) in 100 mL of MeOH was added 1-cyclohex-
enecarboxaldehyde (3.4 mL, 31.2 mmol, 1.05 equiv) and NH4OAc
(10.8 g, 140.3 mmol, 4.7 equiv). The solution was refluxed under
N2 for 20 h. Upon cooling, the red-colored reaction was quenched
with H2O. The tan precipitate was filtered through a Buchner funnel
and washed copiously with H2O. The tan solid was dried under
reduced pressure then in Vacuo to yield a light-tan solid. Yield:
76%. Rf (10% EtOAc/hexanes on silica): 0.35. 1H NMR (500 MHz,
CDCl3): δ 8.33 (s, 1H), 7.79 (d, J ) 8.8 Hz, 1H), 7.53 (d, J ) 8.8
Hz, 2H), 7.34 (s, 1H), 2.76 (m, 4H), 1.81 (m, 4H). 13C NMR (125
MHz, CDCl3): δ 153.0, 150.2, 146.8, 138.5, 132.0, 131.6, 128.1,
122.6, 120.3, 28.8, 26.0, 22.5, 22.3. EI-LRMS calcd for C15H14BrN
[M+]: 288.19. Found: 287.0. Anal. calcd for C15H14BrN: C, 62.52;
H, 4.90; N, 4.86. Found: C, 62.20; H, 4.89, N 4.91.
7.41 (d, J ) 8.4 Hz, 2H), 6.35 (dd, J ) 2.8, 8.8 Hz, 2H), 5.73 (d,
J ) 2.4 Hz, 2H), 5.18 (s, 1H), 3.52 (s, 6H), 2.34 (s, 6H), 1.77 (s,
6H). 13C NMR (125 MHz, CDCl3): δ 184.43, 165.61, 159.29,
148.70, 147.66, 138.29, 137.55, 129.78, 124.51, 118.34, 117.25,
105.78, 100.45, 54.43, 28.88, 18.45. ESI-LRMS calcd for
C26H23IrN2O2 [M+ - C5H8O2]: 587.7. Found: 588. Anal. calcd for
C31H31IrN2O4 ·2H2O: C, 51.44; H, 4.87; N, 3.87. Found: C, 51.63;
H, 4.57; N, 3.85.
1
(Phmppy)2Ir(acac) (6). H NMR (400 MHz, CDCl3): δ 8.34
(s, 2H), 7.75 (d, J ) 8.0 Hz, 2H), 7.54 (dd, J ) 2.0, 8.4 Hz, 2H),
7.53 (d, J ) 8.0 Hz, 2H), 7.29 (m, 4H), 7.24 (m, 4H), 7.17 (m,
2H), 7.00 (dd, J ) 2.0, 8.0 Hz, 2H), 6.44 (d, J ) 1.6 Hz, 2H), 5.21
(s, 1H), 2.39 (s, 6H), 1.80 (s, 6H). 13C NMR (125 MHz, CDCl3):
δ 184.52, 165.66, 148.02, 146.84, 144.51, 142.05, 140.48, 137.65,
131.46, 131.02, 128.22, 127.21, 127.14, 126.51, 123.48, 120.10,
118.06, 28.91, 18.59. ESI-LRMS calcd for C38H30IrN2O [M+
-
Synthesis of Tetrakis-(C^N)-µ-(dichloro)-diiridium(III). The
syntheses of the ligands11 and chloro-bridged dimers have been
reported previously.11,12,23 The IrCl3 ·nH2O is combined with 2.1
equiv of cyclometalating ligand (C^N) in a 3:1 mixture of
2-methoxyethanol and water. The reaction was heated to 125 °C
for 18 h. Upon cooling, the reaction was quenched with H2O. The
precipitate was isolated by vacuum filtration through a Buchner
funnel and washed copiously with water and hexanes. The yellow
solid was vaccuum-dried to yield the product, [(C^N)2Ir-µ-Cl]2.
Yield: 70–85%.
Synthesis of bis-(C^N)-(Acetylacetonato)-iridium(III) Com-
plexes. The synthesis of each complex was performed on the basis
of a previously reported procedure.5 To a solution of corresponding
dimer (0.03 mmol, 1 equiv) in 5 mL of CH2Cl2 was added
acetylacetone (8.23 µL, 0.08 mmol, 2.7 equiv) in 0.5 mL of EtOH.
To this mixture was added tetrabutylammonium hydroxide as a 40%
w/w aqueous solution (53 µL, 0.51 mmol, 16.9 equiv). The reaction
mixture was refluxed for 18 h. Upon cooling, the crude mixture
was run through a short silica plug using CH2Cl2 as the eluent to
remove baseline impurities. The yellow solution was then concen-
trated 90%, and ca. 2 mL of EtOH was added. The remaining
CH2Cl2 was removed under reduced pressure to form a yellow
crystalline precipitate. The EtOH suspension was then cooled in
an ice bath for 10 min to promote additional crystallization. The
solid was collected by vacuum filtration through a Buchner funnel,
washed with cold EtOH, and vaccuum-dried to yield the desired
heteroleptic (C^N)2Ir(acac) product. Yield: 80–90%.
C3H5O]: 722.89. Found: 723. Anal. calcd for C41H35IrN2O2 ·H2O:
C, 61.71; H, 4.67; N, 3.51. Found: C, 61.35; H, 4.62; N, 3.71.
1
(BrCyppy)2Ir(acac) (7). H NMR (400 MHz, CDCl3): δ 8.04
(s, 2H), 7.46 (s, 2H), 7.30 (d, J ) 8.4 Hz, 2H), 6.92 (dd, J ) 2.0,
8.4 Hz, 2H), 6.28 (d, J ) 2.0 Hz, 2H), 5.17 (s, 1H), 2.99 (m, 4H),
2.76 (m, 4H), 1.88 (m, 8H), 1.76 (s, 6H). 13C NMR (125 MHz,
CD2Cl2): δ 184.97, 163.46, 149.04, 149.02, 148.31, 145.27, 135.59,
132.95, 124.67, 124.26, 122.81, 119.29, 100.81, 29.41, 28.76, 26.76,
22.67, 22.62. ESI-LRMS calcd for C32H28Br2IrN2O [M+ - C3H5O]:
808.61. Found: 808. Anal. calcd for C35H33Br2IrN2O2: C, 48.56;
H, 3.84; N, 3.24. Found: C, 48.59; H, 4.27; N, 3.25.
1
(dFmppy)2Ir(acac) (8). H NMR (300 MHz, CDCl3): δ 8.23
(s, 2H), 8.13 (d, J ) 8.5 Hz, 2H), 7.60 (d, J ) 8.26 Hz, 2H), 6.31
(ddd, J ) 2.4, 6.9, 9.3 Hz, 2H), 5.62 (d, J ) 9.1 Hz, 2H), 5.25 (s,
1H), 2.41 (s, 6H), 1.82 (s, 6H). 13C NMR (125 MHz, CDCl3): δ
184.88, 163.22, 162.54, 150.40, 147.75, 138.63, 137.76, 131.52,
128.70, 122.15, 114.90, 100.77, 97.19, 28.83, 18.56. ESI-LRMS
calcd for C24H15F4IrN2 [M+ - C5H8O2]: 599.61. Found: 599.
(dF(CF3)ppy)2Ir(acac) (9). 1H NMR (400 MHz, CDCl3): δ 8.63
(s, 2H), 8.34 (dd, J ) 2.0, 8.4 Hz, 2H), 7.98 (dd, J ) 2.4, 8.8 Hz,
2H), 6.70 (m, 2H), 5.61 (dd, J ) 2.0, 8.4 Hz, 2H), 5.29 (s, 1H),
1.81 (s, 6H). 13C NMR (125 MHz, CDCl3): δ 185.80, 168.63,
162.70, 160.87, 152.40, 144.92, 135.30, 127.30, 124.21, 122.27,
121.41, 115.46, 101.17, 98.04, 28.57. ESI-LRMS calcd for
C29H17F10IrN2O2 [M+]: 807.67. Found: 806. ESI-LRMS calcd for
C24H9F10IrN2 [M+ - C5H8O2]: 707.553. Found: 706.
Synthesis of fac-Ir(C^N)3 Complexes. A solution of Ir(acac)3
(100 mg, 0.25 mmol, 1 equiv) and the C^N ligand (0.77 mmol,
3.1 equiv) in 4 mL of ethylene glycol was heated to 250 °C for
42 h under a stream of N2. Upon cooling, the reaction was diluted
with H2O and CH2Cl2. The aqueous phase was extracted 4× with
CH2Cl2. The organic phase was concentrated under reduced pressure
and then purified via flash column chromatography (CH2Cl2 on
silica) to yield the desired product.
fac-Ir(FCyppy)3 (10). Yield: 83%. Rf (CH2Cl2 on silica): 0.73.
1H NMR (400 MHz, CDCl3): δ 7.50 (dd, J ) 5.6, 8.8 Hz, 1H),
7.43 (s, 1H), 7.14 (s, 1H), 6.53 (dt, J ) 2.8, 8.8 Hz, 1H), 6.35 (dd,
J ) 2.8, 10.4 Hz, 1H), 2.80 (m, 2H), 2.45 (m, 2H), 1.71 (m, 4H).
13C NMR (100 MHz, CDCl3): δ 164.8, 163.2, 162.1, 146.8, 146.6,
140.3, 131.2, 124.6, 124.5, 122.1, 121.9, 118.5, 107.1, 106.8, 28.9,
26.2, 22.3, 22.2. EI-LRMS calcd for C45H39F3IrN3 [M+]: 871.038.
Found: 873.0. EI-HRMS for C45H39F3191IrN3: 869.273. Found:
869.270.
fac-Ir(BrCyppy)3 (11). Yield: 80%. Rf (CH2Cl2 on silica): 0.81.
1H NMR (500 MHz, CDCl3): δ 7.47 (s, 1H), 7.38 (d, J ) 8.5 Hz,
1H), 7.08 (s, 1H), 6.98 (d, J ) 8.5 Hz, 1H), 6.76 (s, 1H), 2.82 (m,
2H), 2.43 (m, 2H), 1.72 (m, 4H). 13C NMR (125 MHz, CDCl3): δ
(ppy)2Ir(acac) (3) (ppy ) 2-phenylpyridine)23. 1H NMR (400
MHz, CDCl3): δ 8.47 (d, J ) 6.0 Hz, 2H), 7.80 (d, J ) 8.0 Hz,
2H), 7.68 (t, J ) 7.6 Hz, 2H,), 7.50 (d, J ) 7.6 Hz, 2H), 7.09 (t,
J ) 7.6 Hz, 2H), 6.76 (t, J ) 7.6 Hz, 2H), 6.64 (t, J ) 7.6 Hz,
2H), 6.22 (d, J ) 7.6 Hz, 2H), 5.17 (s, 1H), 1.74 (s, 6H).
1
(Fmppy)2Ir(acac) (4). H NMR (500 MHz, CDCl3): δ 8.23 (s,
2H), 7.65 (d, J ) 8.2 Hz, 2H), 7.53 (d, J ) 8.2 Hz, 2H), 7.45 (dd,
J ) 8.2 Hz, JHF ) 6.1 Hz, 2H), 6.50 (dd, J ) 8.6 Hz, JHF ) 7.0
Hz, 2H), 5.82 (d, JHF ) 9.8 Hz, 2H), 5.22 (s, 1H), 2.38 (s, 6H),
1.79 (s, 6H). 13C NMR (125 MHz, CDCl3): δ 184.71, 164.86,
161.56, 149.45, 147.77, 141.17, 138.08, 131.17, 124.83, 119.02,
118.01, 108.00, 100.63, 28.84, 18.52. ESI-LRMS calcd for
C24H17F2IrN2[M+ - C5H8O2]: 563.62. Found: 564. Anal. calcd for
C29H25F2IrN2O2: C, 52.48; H, 3.80; N, 4.22. Found: C, 51.96; H,
3.83; N, 4.25.
(MeOmppy)2Ir(acac) (5). 1H NMR (400 MHz, CDCl3): δ 8.22
(s, 2H), 7.58 (d, J ) 8.4 Hz, 2H), 7.46 (dd, J ) 1.6, 8.4 Hz, 2H),
(23) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.; Kwong,
R.; Tsyba, I.; Bortz, M.; Mui, B.; Bau, R.; Thompson, M. E. Inorg.
Chem. 2001, 40, 1704–1711.
Inorganic Chemistry, Vol. 47, No. 6, 2008 2041

Coughlin et al.
162.2, 146.9, 146.8, 143.1, 138.7, 131.9, 124.7, 124.5, 123.1, 118.9,
100.0, 29.0, 26.2, 22.3, 22.2. EI-LRMS calcd for C45H42Br3IrN3[M+]:
1056.78. Found: 1054.0. Anal. calcd for C45H42Br3IrN3 ·4H2O: C,
48.01; H, 4.21; N, 3.73. Found: C, 47.71; H, 3.82, N 3.65.
Enantiomeric Separation. 1. Analytical Supercritical Fluid
Chromatography (SFC) Separations. Samples were analyzed on
a Berger Analytical SFC System equipped with a dual pump (FCM-
1200), an auto sampler (ALS 3100), a column oven (TCM-200),
and a diode-array detector (DAD-4100) (Berger Instruments,
Newark, DE). Analyses were carried out on a Chiralpak AD-H (0.46
× 25 cm) column under isocratic conditions using 30% or 40% of
the organic modifier (methanol or ethanol) in CO2 at a back pressure
of 100 bar with a flow rate of 3 mL/min and temperature of 40 °C.
2. Preparative SFC Separations. Separations were carried out
on a Berger Multigram II SFC equipped with two SD-1 Varian
pumps and a Knauer K-2501 Spectrophotometer (Mettler-Toledo,
Newark, DE). Samples were dissolved to an approximate concen-
tration of 0.1 mg/mL in ethanol/dichloromethane. A total of 1 mL
of solution was injected at intervals of 60 s onto a (2 × 15 cm)
Chiralpak AD-H column under isocratic conditions using 30% or
40% of the organic modifier (methanol or ethanol) in CO2 at a
back pressure of 100 bar with a flow rate of 50 mL/min and
temperature of 35 °C. Sample peaks were acquired using collection
“windows” at a wavelength of 280 nm.
Photophysical Characterization. UV–vis spectra were recorded
at room temperature in a 1.0 cm quartz cuvette using a Hewlett-
Packard 8453 spectrometer equipped with a diode-array detector.
The samples were prepared in toluene at a concentration of 25 µM,
with the exception of the ruthenium standard ([Ru(bpy)3](PF6)2),
which was prepared in acetonitrile.
Emission spectra were recorded using a Jobin-Yvon Fluorolog-3
spectrometer equipped with a double monochromator and a
Hamamatsu-928 photomultiplier tube (PMT) as the detector. Right
angle detection was used for the complexes, which were all excited
at 400 nm.
Excited-state lifetimes were measured using the emission mono-
chromator and PMT detector of the Jobin-Yvon Fluorolog-3
spectrometer. The samples were excited at 337 nm with a N2 laser
(Laser Science, Inc. VSL-337LRF, 10 ns pulse), and the emission
decay was recorded using a Tektronix TDS 3032B digital phosphor
oscilloscope. All samples were degassed using N2 for 10 min prior
to measurements.
Chiroptical Characterization. Upon receiving the resolved
enantiomers from the chiral SFC separation, chiroptical measure-
ments were performed. Solutions were made using toluene at
concentrations of 25 µM for the purposes of CD and CPL analysis.
CD measurements were performed on an AVIV-62DS spectrometer
with a 1.0 cm path length. Data points were collected at 1 nm
intervals using averaging times of 1 s/nm with a spectral bandwidth
of 1.5 nm.
complex [Ru(L3)](PF6)215 was measured. A slight bias was found,
as the average value of gem was -7.9 × 10-5. To adjust for this,
∆I for all CPL measurements was corrected using eq 2.
7.9 × 10-5
∆Icorr ) ∆Iexp
+
Iexp
(2)
2
Computational Methodology. All DFT calculations were carried
out with the Gaussian 03 suite.25 Ground-state and excited-state
geometries (both singlet and triplet) for all complexes were
evaluated computationally using Becke’s three-parameter exchange
functional (B3)26 in conjunction with the Lee, Yang, and Parr
(LYP)27,28 nonlocal functional using the default thresholds for
gradient convergence but with a slightly relaxed threshold for wave
function convergence [SCF)(CONVER)7)]. The LANL2DZ basis
set was employed for all calculations.29–31 For each complex, time-
dependent DFT (TD-DFT) calculations were performed at the
optimized ground-state geometry, calculating the energy, oscillator
strength, and rotatory strength for each of the 100 lowest singlet
excitations. The geometry of the lowest triplet state was also
obtained using unrestricted B3LYP calculations, and the energy
difference between the singlet and triplet states was calculated at
this geometry. Prediction of CD spectra was accomplished using
GaussSum 2.0.32 Electronic transitions were expanded as Gaussian
curves, with a full-width at half-maximum (fwhm) for each peak
set to 0.12 eV.
Results and Discussion
Synthesis of Iridium Complexes. The free ligands were
accessed via the Kröhnke condensation33 reaction from the
corresponding substituted phenacylpyridinium bromide, enal,
and ammonium acetate. The cyclohexyl-fused ppy ligands
present in complexes 7, 10, and 11 were chosen to both
increase solubility and extend the helicity of the complexes
and enhance their interaction with the chiral stationary phase
in the SFC separation. The heteroleptic complexes,
(C^N)2Ir(acac), were synthesized through a sequential two-
step procedure (Scheme 1, top). Dimer formation occurred
readily, the product of which precipitated out of the reaction.
The chloride-bridged iridium dimer was then cleaved under
basic conditions (TBAOH) to form the desired com-
plexes 3–9 (Scheme 1, bottom) in good yield.5 Removal of
polymeric side products through a short silica plug followed
by precipitation with cold ethanol resulted in pure hetero-
leptic iridium complexes.
The preparation of the corresponding homoleptic fac-
Ir(C^N)3 complexes 10 and 11 occurred in excellent yield
under a modified literature34 procedure using ethylene glycol
as the solvent. Good conversion was possible with the
The CPL measurements were performed as follows: The emission
of the samples was detected at a 90° angle to the excitation in a
Jobin-Yvon Fluorolog-3 and was passed through a photoelastic
modulator (Hinds International PEM-90, operating at a modulation
frequency of 50 kHz), followed by a linear polarizer. The light
was then incident upon the emission monochromator. A Hamamatsu-
928 PMT in conjunction with a gated photon counter (Stanford
Research SR 400) was used to acquire the CPL spectrum.24 To
ensure that there was no bias toward one type of circularly polarized
light, a solution containing a racemic mixture of bipyridyl hemicage
(25) Frisch, M. J. et al. Gaussian 03, revision C.02; Gaussian Inc.:
Wallingford, CT, 2004 (see Supporting Information for full citation).
(26) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
(27) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B: Condens. Matter Mater.
Phys. 1988, 37, 785–789.
(28) Miehlich, B.; Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989,
157, 200.
(29) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299–310.
(30) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270–283.
(31) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284–298.
(32) O’Boyle, N. M. GaussSum 2.0; Dublin City University: Dublin Ireland,
2006. Available at http://gausssum.sf.net.
(33) Kröhnke, F. Synthesis 1976, 1.
(24) Schippers, P. H.; van den Beukel, A.; Dekkers, H. P. J. M. J. Phys.
E: Sci. Instrum. 1982, 15, 945–950.
(34) Dedeian, K.; Djurovich, P. I.; Garces, F. O.; Carlson, G.; Watts, R. J.
Inorg. Chem. 1991, 30, 1685–1687.
2042 Inorganic Chemistry, Vol. 47, No. 6, 2008

Enantiomers of Iridium(III) Luminophores
Scheme 1. (Top) General Synthesis of Heteroleptic (C^N)2Ir(acac) Complexes Using 3 as an Examplea b and (Bottom) Structures of All Ir(III)
Complexes Synthesized, Including Homoleptic 10 and 11
,
a
b
IrCl3/3:1 2-methoxyethanol/water, 125 °C, 18 h. Acetylacetone (acacH)/10:1 CH2Cl2/EtOH, TBAOH, reflux, 18 h; TBAOH ) tert-butylammonium
hydroxide.
presence of electron-withdrawing groups on the phenyl ring
(Br, F) para to the pyridine. Their presence facilitates C-H
activation in the cyclometalation of the substituted ppy (2-
phenylpyridine) ligands. The facial homoleptic complexes
were formed exclusively, as indicated by SFC and 1H NMR,
which showed magnetically equivalent phenylpyridines, an
indication of the C3 symmetry of the complex.
Enantiomeric Separation. Iridium complexes (3–7 and
10) were readily resolved on chiral stationary phases (specif-
ically the amylose-based Chiralpak AD-H columns) using
SFC. The characteristically low solubility of these com-
pounds in the mobile phases, consisting of supercritical
carbon dioxide mixed with 30-40% methanol or ethanol,
resulted in relatively low throughputs. Nonetheless, employ-
ing an iterative approach, milligram quantities (up to 200
mg) were resolved using a Multigram II Berger SFC system.
Automated “stacked” injections and collections, as well as
the addition of 5% dichloromethane to the mobile phase were
required to generate sufficient quantities of the desired
enantiomers, with enantiomeric excesses (ee) ranging from
95 to 100% (Figure 1). Not all complexes could be resolved
in this manner. For instance, highly electron-deficient
(dFmppy)2Ir(acac) (8) could not interact sufficiently with the
chiral stationary phase to enable enantiomeric separation,
while fac-Ir(BrCyppy)3 (11) proved too insoluble to ef-
fectively load onto the column using the devised eluent
conditions, despite the presence of the solubility-enhancing
cyclohexyl ring. Of the nine iridium (III) complexes syn-
thesized, six (3–7 and 10) could be separated by SFC and
chiroptically analyzed.
Figure 1. Representative chromatogram for 5. Chiralpak AD-H (0.46 ×
25 cm), 40% ethanol in CO2 (100 bar), 3 mL/min, 280 nm. (A) Racemic
mixture of 5. (B) Isolated Λ enantiomer peak 1 (ee > 99%). (C) Isolated
∆ enantiomer peak 2 (ee > 99%).
region, while much lower intensity bands are present between
372 and 411 nm and 427 and 480 nm, which likely
correspond to mostly spin-allowed singlet metal-to-ligand
charge-transfer (1MLCT) transitions and spin-forbidden
3MLCT transitions, respectively.7 These bands may also be
partially associated with ligand-centered transitions, as well.35
These transitions are clearly resolved in the (C^N)2Ir(acac)
series but are not as readily distinguished for the two
Ir(C^N)3 complexes. The relatively high intensity of the more
Photophysical Observations. Spectroscopic data for the
six complexes that were successfully resolved into their
separate enantiomers are summarized in Table 1. The UV–vis
absorption spectra closely resemble those of typical bis-
cyclometalated Ir(III) acac complexes. Intense bands belong-
ing to spin-allowed, singlet ligand-centered (1LC) transitions
localized on the phenylpyridine ligands are visible in the UV
3
red-shifted, spin-forbidden MLCT bands compared to the
1MLCT bands is an indication of the large extent of
(35) Wu, F.-I.; Su, H.-J.; Shu, C.-F.; Luo, L.; Diau, W.-G.; Cheng, C.-H.;
Jiun-Pey, D.; Lee, G.-H. J. Mater. Chem. 2005, 15, 1035–1042.
Inorganic Chemistry, Vol. 47, No. 6, 2008 2043

Coughlin et al.
Table 1. Photophysical Data for Chromatographically Resolvable Ir(III) Complexesa b c
,
,
a
b
c
Measured in N2-saturated toluene solvent at 25 °C. ꢀ, Φ, and τ are (10% or better. The emission quantum yields (ΦPL) were measured vs
[Ru(bpy)3](PF6)2 (ΦPL ) 0.062).
complexes possess multiple discernible transitions, suggesting
3
the presence of significant LC character. Indeed, cyclom-
etalated iridium compounds have been well-documented to
emit from a mixed excited state, displaying characteristics
12
3
3
of both LC and MLCT emission.
The emission quantum yields in degassed toluene solutions
excited at 400 nm, ranging from 0.28 to 0.36, are similar to
those previously reported in 2-methyltetrahydrofuran for
analogous (C^N)2Ir(acac) complexes.12,36 Lifetime measure-
ments for the majority of the complexes ranged from 1 to 2
µs, which is also consistent with data collected for similar
neutral Ir(III) compounds.12,23 DFT calculations suggest that
the longer lifetime (3.08 µs) observed for (Phmppy)2Ir(acac)
(6) results from less probable relaxation from the more
twisted excited-state conformation (ꢁbiphenyl ) 30°) of the
biphenyl moiety to that of the ground state (ꢁbiphenyl ) 20°).
Chiroptical Characterization. The CD spectra of the
enantiomers of 5 are representative of the series of neutral
iridium complexes and are depicted in Figure 3 (the spectra
of 3–7 and 10 are reported in the Supporting Information).
From the selected spectra of 5, it is evident that each resolved
species absorbs essentially opposite polarization but with
equal intensity.
Figure 2. Normalized emission spectra of the Ir(III) complexes 3–7 and
10. The solutions were 25 µM in toluene.
spin–orbit coupling present in these complexes, as previous
studies of similar compounds have also shown.12,23,35
Normalized room-temperature emission spectra of the
complexes are depicted in Figure 2. The emission ranges
from blue to green, with Ir(FCyppy)3 (10) possessing the
highest energy light at 471 nm and (Phmppy)2Ir(acac) (6)
possessing the lowest at 537 nm; the former is hypsochro-
mically shifted change due to the inductive effects of fluorine
substitution, while the latter is bathochromically shifted as
a result of the extended π network found on the phenylpy-
ridine ligands. Replacement of one of the phenylpyridine
ligands for an acac bathochromically shifts the emission
spectrum by ca. 20 nm (10 vs 4). With the exception of
(ppy)2Ir(acac) (3), whose emission spectrum is relatively
broad and featureless (an indication of predominant 3MLCT
character), the emission spectra of the remaining iridium
The CPL analysis further supports the clean resolution of
the two enantiomeric complexes by chiral SFC. Like the CD
spectra, the two peaks exhibit a “handedness” for the
polarization of the emitted light. The CPL spectra for 5
(Figure 4A) shows the characteristic symmetry indicative of
(36) Li, J.; Djurovich, P. I.; Alleyne, B. D.; Yousufuddin, M.; Ho, N. N.;
Thomas, J. C.; Peters, J. C.; Bau, R.; Thompson, M. E. Inorg. Chem.
2005, 44, 1713–1727.
2044 Inorganic Chemistry, Vol. 47, No. 6, 2008

Enantiomers of Iridium(III) Luminophores
Table 2. Emission Dissymmetry Data
complex
λ
½
(nm)a
gem at λ (×10-3
)
½
Λ-(ppy)2Ir(acac) - 3
Λ-(Fmppy)2Ir(acac) - 4
Λ-(MeOmppy)2Ir(acac) - 5
Λ-(Phmppy)2Ir(acac) - 6
Λ-(BrCyppy)2Ir(acac) - 7
Λ-Ir(FCyppy)3- 10
525
505
515
545
510
495
450
610
+0.78
+1.51
+1.72
+0.60
+0.57
+1.99
+2.70
+0.61
Λ-Zn(hemicage)15
Λ-Ru(hemicage)15
a λ is defined as the wavelength at which the geometric centroid occurs
½
(see text).
ability to emit one direction of circularly polarized light as
compared to gem determined at λmax
.
Emission dissymmetry values are influenced by the
structure of the complexes. Compounds incorporating greater
helical twists should exhibit larger gem values by providing
a circular path for electron flow. This may explain the
relatively larger value (1.99 × 10-3) found for 10, which
possesses C3 symmetry, as compared to the (C^N)2Ir(acac)
complexes, which are C2 symmetric. Introduction of con-
formational flexibility may result in a lower gem value (6
and 7) in complexes of low symmetry, while the inclusion
of electron-donating or electron-withdrawing groups at the
4 position of the phenylpyridine ligand results in an increased
gem value (4 or 5 vs 3).
Figure 3. CD spectra for the resolved enantiomers of 5. The solutions
were 25 µM in toluene.
Computational Modeling. 1. Circular Dichroism. TD-
DFT methods are an established means of effectively
predicting CD spectra for metal complexes.37–40 This can
be conducted through the determination of the difference in
absorptivity of right and left circularly polarized light (∆ꢀ)
from the singlet transitions at a given wavelength, λ (eq 3):
A
(λ-Ei)
2σ
2
1
∆ꢀ(λ) )
E R e-
(3)
[
]
∑
i
i
2.297 × 10-39 × 2πσ
√
i
where Ei is the energy of the ith transition, Ri is the rotatory
strength, and σ is the line width (fwhm) of the absorption
band.41
TD-DFT-based predictions of the CD spectra were applied
to each of the resolved Ir(III) complexes and compared to
the experimental measurements. For the calculations, the Λ
geometry was applied, and a representative comparison of
the simulated and experimental CD spectra of 5 is shown in
Figure 5 (simulations of the remaining complexes are found
in the Supporting Information). Each CD spectrum consists
Figure 4. (A) Total emission and CPL spectra of 5. (B) Emission
dissymmetry (gem) of 5 as a function of wavelength.
a predominance of either ∆ or Λ isomerism. As the
spectrophotometer setup was designed to only detect circu-
larly polarized light, an achiral or racemic sample would not
display a signal, as it would have no directional preference
for emitting circularly polarized light.
1
of two regions: a weak, low-energy, mixed LC-1MLCT
transition associated with the metal ion as well as the π and
π* orbitals of the C^N ligands, and the more intense, higher-
1
Figure 4B depicts the determined emission dissymmetry value
(gem) for 5 as a function of wavelength. The overall gem values
in solution are on the order of 10-3 (Table 2) and are generally
above average for typical transition metal luminophores.18–20
Among the acac complexes, the gem ranges from 4.7 × 10-4
energy LC absorptions located exclusively on the C^N
ligands. The orbitals involved in the former transition are
shown in Figure 6 for complex 5. All of the simulations show
very good agreement to experimental measurements with
(37) Jorge, F. E.; Autschbach, J.; Ziegler, T. J. Am. Chem. Soc. 2005, 127,
for the (Phmppy)2Ir(acac) (6) to 1.72
×
10-3 for
975–985.
(38) Autschbach, J.; Jorge, F. E.; Ziegler, T. Inorg. Chem. 2003, 42, 2867–
2877.
(MeOmppy)2Ir(acac) (5). We report gem at λ , which is defined
½
as the wavelength at which half the area under the emission
curve has been integrated, in order to account for vibrational
effects present in the luminescence spectrum. In this manner,
(39) Jorge, F. E.; Autschbach, J.; Ziegler, T. Inorg. Chem. 2003, 42, 8902–
8910.
(40) Le Guennic, B.; Hieringer, W.; Gorling, A.; Autschbach, J. J. Phys.
Chem. A 2005, 109, 4836–4846.
gem at λ represents a more accurate depiction of a complex’s
(41) Diedrich, C.; Grimme, S. J. Phys. Chem. A 2003, 107, 2524–2539.
½
Inorganic Chemistry, Vol. 47, No. 6, 2008 2045

Coughlin et al.
Figure 5. Measured (blue) and simulated (red) Λ geometry (B3LYP/
LANL2DZ) CD spectrum of 5. Rotatory strengths (black) for each transition
are shown as unbroadened vertical lines.
Figure 8. (A) Emission dissymmetry vs calculated EiRi/osci for the Ir(III)
complexes (3–7, 10). (B) Emission dissymmetry (gem) vs calculated EiRi/
osci including enantiopure Ru(II) and Zn(II) hemicage complexes (Figure 8).15
See Table 3 for a summary of the individual values for each complex.
was, in fact, accurate at assigning this geometry, as verified
by performing calculations on a previously published chiral
complex of known configuration, fac-∆-Ir(pppy)3,16 and
comparing the spectrum to the experimentally measured CD
(Figure 7).
Figure 6. Orbitals of 5 associated with the absorption of a photon from
the singlet ground state (B f A) and the emission of a photon from the
triplet excited state (C f D). Both processes involve almost identical orbitals
and parameters obtained from TD-DFT calculations of the singlet state and
can be used as an approximation of this particular triplet-singlet transition
(phosphorescence).
2. Circularly Polarized Luminescence. The use of TD-
DFT methods significantly decreases the computational cost
required to accurately predict CD spectra; however, modeling
CPL spectra is far more difficult. Calculations of emission
spectra rely on recursions of normal modes that make
predictions computationally unfeasible for all but the simplest
molecules containing just a few heavy atoms. Our goal,
therefore, was to develop a computational methodology that
yields accurate approximations to experimental CPL spectra
within reasonable computation time.
The first step in this process is the calculation of the
absorptivity (ꢀ) of the compound, which can be determined
from the oscillator strength of the transitions involved (eq
4), where osci is the oscillator strength of the ith transition:
Figure 7. Measured (blue) and simulated (red) ∆ geometry (B3LYP/
LANL2DZ) CD spectrum of fac-∆-Ir(pppy)3.16 Rotatory strengths (black)
for each transition are shown as unbroadened vertical lines.
A
(λ-Ei)
2.174 × 108
ꢀ )
osc e-2.772
(4)
[
]
σ
∑
i
σ
i
respect to the energies and relative strengths of the transi-
tions; currently, no computational methodology exists to
accurately predict the vibrational substructure of the spec-
trum, but it is clear that TD-DFT calculations at this level
of theory can successfully assign the absolute configuration
with reasonable confidence.40,42 Our computational model
For a given transition i, the absorption dissymmetry of
the transition is defined by eq 5:
EiRi
∆ꢀ
ꢀ
|gabs(i)| )
∝
(5)
osci
where Ri and osci are the rotatory and oscillator strengths
calculated for the singlet absorption transition and Ei is the
energy of the absorption.
(42) Rinderspacher, B. C.; Schreiner, P. R. J. Phys. Chem. A 2004, 108,
2867–2870.
2046 Inorganic Chemistry, Vol. 47, No. 6, 2008

Enantiomers of Iridium(III) Luminophores
Table 3. Summary of Experimentally Measured Emission Dissymmetry Values and Calculated Data at the B3LYP/LANL2DZ Level
Complex
gem (λ ) × 10-3
Ri (cgs)
osci
Ei (hartrees)
EiRi/osci
½
Λ-(ppy)2Ir(acac) - 3
Λ-(Fmppy)2Ir(acac) - 4
Λ-(MeOmppy)2Ir(acac) - 5
Λ-(Phmppy)2Ir(acac) - 6
Λ-(BrCyppy)2Ir(acac) - 7
Λ-Ir(FCyppy)3- 10
+0.78
+1.51
+1.72
+0.60
+0.57
+1.99
+2.70
+0.61
71.3486
86.3741
107.8558
26.5894
91.7900
116.3174
315.6177
82.9699
0.0561
0.0551
0.0567
0.0473
0.0801
0.0645
0.0759
0.1359
0.081513
0.092379
0.085590
0.079982
0.087211
0.099052
0.084666
0.075561
103.67
144.81
162.81
44.96
99.94
178.63
352.07
46.13
Λ-Zn(hemicage)15
Λ-Ru(hemicage)15
Because the geometry—and as a consequence the
electronic structure—of the excited state should not change
significantly during intersystem crossing from the singlet
to the triplet state, the relaxation back to the ground state
should thus possess equal magnitude oscillator and rotatory
strength to that of the absorption transition, the latter of
which is opposite in sign.43 Thus, an emission dissym-
metry value (gem) can be extrapolated from the following
relationship (eq 6):
a strong correlation (Figure 8B), indicating that this trend
holds true for a variety of complexes with diverse
coordinated metals. The source of the reduction in
effectiveness can be attributed to the poorer predictions
of the energies (Ei) between the triplet and singlet states
for different metal centers. Additionally, as can be seen
from Figure 6B, the phenyl-substituted complex, 6,
deviates slightly in agreement with the other heteroleptic
Ir(III) compounds. The discrepancy can be attributed to
the greater flexibility provided by the additional phenyl
ring, which leads to slight deviations in the excited-state
geometry of the triplet state. Owing to the fact that our
computational model requires that the excited triplet and
singlet state geometries be the same, this flexibility leads
to a breakdown in the prediction’s effectiveness. Never-
theless, within the framework of structurally similar
phosphorescent metal complexes, gem can now be predicted
extremely well. To our knowledge, this is the first example
of an accurate prediction of gem. Furthermore, this model
can be extended to disparate metal complexes and thus
provide a means of advance screening a wide variety of
complexes for high dissymmetry values.
EiRi
|gem(i)| ∝
(6)
osci
with Ei ) Etriplet - Esinglet calculated at the triplet geome-
try.
Equation 6 was solved by initially optimizing the
ground-state geometry using B3LYP/LANL2DZ with
constraints of C2 and C3 symmetry for (C^N)2Ir(acac) and
Ir(C^N)3 complexes, respectively. TD-DFT calculations
were then performed at the optimized singlet-state geom-
etry to obtain the rotatory (Ri) and oscillator (osci)
strengths. The triplet geometry was optimized using
unrestricted (open-shell) DFT calculations with the ground-
state geometry as a starting point. Energies of the triplet
and singlet states at the calculated triplet state geometry
were obtained at the same level of theory. To ensure that
the transitions chosen from the TD-DFT calculations were
correct, the orbitals of the triplet highest singly occupied
molecular orbital and the lowest singly occupied molecular
orbital were compared to the orbitals involved for absorp-
tion, Figure 8.
Conclusion
Herein, we have described the synthesis and character-
ization of a series of neutral iridium complexes and their
enantiomeric resolution by chiral SFC, as verified by the
characteristic mirror image motif in both the CD and CPL
spectra. Emission dissymmetry values were measured for
the series, all of which were on the order of 10-3, with
Ir(FCyppy)3 (10) exhibiting the largest gem value, 1.99 ×
10-3, for the iridium compounds. TD-DFT calculations
were then applied to accurately predict the CD spectra of
the resolved complexes. Subsequently, these calculations
were expanded and applied to the emissive excited state
in order to develop an unprecedented theoretical model
capable of predicting emission dissymmetry. The calcula-
tion correlates extremely well to experimental measure-
ments and can be generalized to other metals and ligand
frameworks. Work is ongoing to better explain the
relationship between substitution pattern and identity of
the complexes to their chiroptical properties, with the
ultimate goal of this research being the development of
systems that optimize the emission preference of either
right or left circularly polarized light.
Table 3 summarizes the determined experimental and
calculated values, and Figure 8A shows excellent agree-
ment between experiment and theory for the six resolvable
Ir(III) complexes (3–7, and 10). Furthermore, when ionic,
enantiopure Zn(II) and Ru(II) hemicage complexes15
(Figure 9) are included, the fit is less tight but still presents
Acknowledgment. This work was supported by the
National Science Foundation (DMR-0605621 and CAREER
Award No. CHE-0449755) and through a Camille and Henry
Figure 9. Structure of hemicage complexes. M ) Ru(II), Ru(hemicage);
M ) Zn(II), Zn(hemicage).
Inorganic Chemistry, Vol. 47, No. 6, 2008 2047

Coughlin et al.
corresponding calculated chiroptical predictions for 3–7 and 10.
This material is available via the internet at http://pubs.acs.org.
Dreyfus Foundation New Faculty Award. E.Z.-C. was
supported by the Princeton Center for Complex Materials
(PCCM) in the form of a postdoctoral fellowship. F.J.C.
acknowledges support through a National Science Foundation
Graduate Research Fellowship.
IC701747J
Supporting Information Available: The complete citation
(43) Nakanishi, K.; Berova, N.; Woody, R. Circular Dichroism: Principles
for ref 25 is included along with the CD and CPL spectra with
and Applications; VCH: New York, 2000; p 191.
2048 Inorganic Chemistry, Vol. 47, No. 6, 2008