Cerium photosensitizers: Structure-function relationships and applications in photocatalytic aryl coupling reactions

Article abstract of DOI:10.1021/jacs.6b02248

Two complete mixed-ligand series of luminescent Ce^III complexes with the general formulas [(Me₃Si)₂NC(NⁱPr)₂]_xCe^III[N(SiMe₃)₂]_3-x (x = 0, 1-N; x = 1, 2-N, x = 2, 3-N; x = 3, 4) and [(Me₃Si)₂NC(NⁱPr)₂]_xCe^III(OAr)_3-x (x = 0, 1-OAr; x = 1, 2-OAr, x = 2, 3-OAr; x = 3, 4) were developed, featuring photoluminescence quantum yields up to 0.81(2) and lifetimes to 117(1) ns. Although the 4f ? 5d absorptive transitions for these complexes were all found at ca. 420 nm, their emission bands exhibited large Stokes shifts with maxima occurring at 553 nm for 1-N, 518 nm for 2-N, 508 nm for 3-N, and 459 nm for 4, featuring yellow, lime-green, green, and blue light, respectively. Combined time-dependent density functional theory (TD-DFT) calculations and spectroscopic studies suggested that the long-lived ²D excited states of these complexes corresponded to singly occupied 5d 2 orbitals. The observed difference in the Stokes shifts was attributed to the relaxation of excited states through vibrational processes facilitated by the ligands. The photochemistry of the sterically congested complex 4 was demonstrated by C-C bond forming reaction between 4-fluoroiodobenzene and benzene through an outer sphere electron transfer pathway, which expands the capabilities of cerium photosensitizers beyond our previous results that demonstrated inner sphere halogen atom abstraction reactivity by 1-N.

Full text of DOI:10.1021/jacs.6b02248

Subscriber access provided by UNIV LAVAL
Article
Cerium Photosensitizers: Structure-Function Relationships
and Applications in Photocatalytic Aryl Coupling Reactions
Haolin Yin, Patrick J. Carroll, Brian C Manor, Jessica M. Anna, and Eric J. Schelter
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b02248 • Publication Date (Web): 08 Apr 2016
Downloaded from http://pubs.acs.org on April 10, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Cerium Photosensitizers: Structure-Function Relationships
and Applications in Photocatalytic Aryl Coupling Reactions
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Haolin Yin, Patrick J. Carroll, Brian C. Manor, Jessica M. Anna* and Eric J. Schelter*
P. Roy and Diana T. Vagelos Laboratories, Department of Chemistry, University of Pennsylvania, 231 South 34 Street, Philꢀ
adelphia, Pennsylvania 19104, United States
Supporting Information
ABSTRACT: Two complete mixedꢀligand series of luminescent CeIII complexes with the general formulae
[(Me3Si)2NC(NiPr)2]xCeIII[N(SiMe3)2]3ꢀx (x = 0, 1ꢀN; x = 1, 2ꢀN, x = 2, 3ꢀN; x = 3, 4) and [(Me3Si)2NC(NiPr)2]xCeIII(OAr)3ꢀx (x = 0, 1ꢀ
OAr; x = 1, 2ꢀOAr, x = 2, 3ꢀOAr; x = 3, 4) were developed, featuring photoluminescence quantum yields up to 0.81(2) and lifetimes to
117(1) ns. Although the 4f→5d absorptive transitions for these complexes were all found at ca. 420 nm, their emission bands exhibited
large Stokes shifts with maxima occurring at 553 nm for 1ꢀN, 518 nm for 2ꢀN, 508 nm for 3ꢀN and 459 nm for 4, featuring yellow, limeꢀ
green, green and blue light respectively. Combined timeꢀdependent density functional theory (TDꢀDFT) calculations and spectroscopic
2
2
studies suggested the longꢀlived D excited states of these complexes corresponded to singly occupied 5dz orbitals. The observed differꢀ
ence in the Stokes shifts was attributed to the relaxation of excited states through vibrational processes facilitated by the ligands. The phoꢀ
tochemistry of the sterically congested complex 4 was demonstrated by C−C bond forming reaction between 4ꢀfluoroiodobenzene and
benzene through an outer sphere electron transfer pathway, which expands the capabilities of cerium photosensitizers beyond our previous
results that demonstrated inner sphere halogen atom abstraction reactivity by 1ꢀN.
undergo intersystem crossing (ISC) to afford longꢀlived triplets. In
1.INTRODUCTION
order to take full advantage of solar energy, the ideal absorption
energy for visible light photosensitizers is ca. 2.8 eV (440 nm,
blue light), defined by the intensity maximum in the sunlight
spectrum.4,37ꢀ38 However, the intersystem crossing (ISC) process
inevitably results in losses of absorbed energy.
Photoluminescent complexes find direct applications in lightꢀ
emitting materials,1 but have also drawn significant attention for
applications in photovoltaic devices2ꢀ3 and photoredox catalysis.4
The luminescence phenomenon implies the existence of longꢀ
lived electronic excited states that are energetically poised to enꢀ
gage in single electron transfer (SET) reactions, much more so
than in their ground states. The development of photoꢀreduction
chemistry relies on the availability of powerful photoreductants as
well as the ability to tune their optical and electrochemical properꢀ
ties using simple modifications.5ꢀ7 Molecular complexes, includꢀ
In contrast to the transition metal photosensitizers, the luminesꢀ
cence of the CeIII cation is metalꢀcentered and does not involve a
change in spin state.39ꢀ40 The interconfigurational f→d transition
for CeIII cations is electron dipole allowed based on ꢁS = 0 and |J
− Jʹ| ≤ 3 ≤ |J + J′| selection rules, where J and J′ are the total anguꢀ
lar momentum quantum numbers of the fn groundꢀ and fnꢀ1d1 exꢀ
cited state configurations, respectively.41 The promise of applying
CeIII cations for photochemistry was supported by early photolysis
studies in aqueous solutions; H2 gas evolution was observed upon
irradiation of aqueous solutions containing Ce(III) cations with
UV light.42ꢀ45 Recently, we reported a physical basis for visible
light promoted reductive photochemistry in organic media with
wellꢀdefined molecular complexes, Ce[N(SiMe3)2]3 (1ꢀN) and
[(Me3Si)2NC(NiPr)2]Ce[N(SiMe3)2]2 (2ꢀN).46 Both complexes
2+
ing Ru(bpy)3 (bpy = 2,2′ꢀbipyridine), facꢀIr(ppy)3 (ppy = 2,2′ꢀ
phenylpyridine) and their derivatives, have been widely applied as
visible light photoꢀredox catalysts to promote C−C bond forming
reactions8ꢀ18 and CO2 reduction.19 Similarly, dimeric [Au2(µꢀ
dppm)2]2+ (dppm = bis(diphenylphosphanyl)methane) was also
shown to promote C−C bond formation under UV light or sunꢀ
light.20ꢀ21 As powerful reductants with excitedꢀstate reduction
potentials as low as −3.0 V versus Cp2Fe+/0, the excited states of
luminescent tungsten(0) isocyanide complexes were shown to be
effectively quenched by very weak oxidants such as benzopheꢀ
none and anthracene.22ꢀ23 Moreover, developments in C−N,24ꢀ26
C−O27, C−S28 and C−C29ꢀ31 bond coupling reactions with lumiꢀ
nescent CuI complexes indicate SET mechanisms are viable
pathways for Ullmannꢀtype coupling chemistry. And the use of
CuI photosensitizers has contributed to establishing the applicaꢀ
tion of earth abundant elements for photoꢀredox catalysis.32ꢀ35
Single photon excitation of transition metal photosensitizers
leads to excited states through metalꢀtoꢀligand charge transfer
transitions (MLCT).36 The resulting singlet excited states quickly
2
absorbed blue light to access D excited states47 featuring singly
2
occupied 5dz orbitals. The metalloradical character of the resultꢀ
ing 2D excited states promoted inner sphere halogen atom abstracꢀ
tion reactions toward PhCH2Cl, aryl iodides and aryl bromides in
stoichiometric and catalytic fashions. These photoꢀreduction reacꢀ
tivities demonstrated the energy of light absorbed by these comꢀ
plexes can be effectively directed into chemical transformations.
Our interests in developing CeIII photosensitizers are also
grounded in the earth abundance of the element. Cerium has a
similar abundance to copper in Earth’s upper continental crust
with relative abundance of ~101.5 (atoms per 106 atoms of Si) and
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 11
is more abundant than tungsten (100), gold (10ꢀ3), ruthenium (10ꢀ3)
[(Me3Si)2NC(NiPr)2]2Ce[N(SiMe3)2]
(3ꢀN)
and
and iridium (<10ꢀ5), respectively.48 Cerium is also readily separatꢀ
ed from other lanthanide elements using oxidation chemistry49
and is a waste byproduct in the separations chemistry of light rare
earth elements.
[(Me3Si)2NC(NiPr)2]3Ce (4) in the gas phase at B3LYP level of
theory. The results from the groundꢀstate DFT calculations were
consistent with experimental observations; ꢁG1 at 298.15 K for
the first carbodiimide insertion into 1ꢀN was found to be essentialꢀ
ly thermoꢀneutral (0.79 kcal molꢀ1) while the second and third
insertions were less favorable on thermodynamic grounds (ꢁG2 =
3.87 kcal molꢀ1, ꢁG3 = 8.46 kcal molꢀ1) (Figure S83).
1
2
3
4
5
6
7
Solid materials doped with Ce3+ were important in the develꢀ
opment of white light LEDs due to their broad emission bands
2
originating from the D excited states of Ce3+ cations.50 For exꢀ
ample, the yellow emitter YAG:Ce3+ absorbs blue light and exhibꢀ
its a broad emission band spanning from 450 nm to over 700
nm.51 Thus, the combination of emitted light from YAG:Ce phosꢀ
phor and transmitted blue light from blue LEDs produces white
light. Although CeIII cations exhibit intrinsic 4f→5d absorptive
transitions,52 luminescence has not typically been observed from
molecular CeIII compounds likely due to rapid quenching of their
excited states by nonꢀradiative relaxation pathways.39 In fact, only
a small number of luminescent molecular CeIII complexes have
been reported.52ꢀ58 Among these, a large variation was observed in
the luminescence properties; a photoluminescence quantum yield
of 0.7 was reported for [K(THF)2][(C5Me5)2CeI2] compared to
0.01 for [(C5Me5)2CeI(NCMe)2].54 In order to rationally design
CeIII photosensitizers for varied purposes, a working model of the
photophysical properties of CeIII luminescence is required.59 Hereꢀ
in, we report two complete mixedꢀligand series of cerium(III)
complexes that display varied emission wavelengths correspondꢀ
ing to yellow, green and blue light, despite basically invariant
absorption bands at ca. 420 nm. The photophysical properties and
excitedꢀstate reduction potentials of the series were evaluated
based on electrochemical and spectroscopic data. We also demonꢀ
strate that a sterically congested homoleptic trisꢀguanidinate CeIII
complex (4) can be applied as photocatalyst by outer sphere elecꢀ
tron transfer, a fundamental expansion from the exclusively inner
sphere photosensitizers demonstrated in our initial report.46
8
9
Scheme 1. Synthesis of Luminescent Ce(III) Complexes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.RESULTS AND DISCUSSION
2.1. Synthesis of Luminescent Cerium Complexes. The
promise of applying rigid guanidinate ligands toward accessing
highly luminescent Ce(III) complexes was initiated from a simple
insertion reaction of Ce[N(SiMe3)2]3 (1ꢀN) with excess N,N'ꢀ
diisopropylcarbodiimide (Scheme 1).46 The resulting monoꢀ
insertion product [(Me3Si)2NC(NiPr)2]Ce[N(SiMe3)2]2 (2ꢀN) was
found to be a much brighter emitter than 1ꢀN; the photoluminesꢀ
cence quantum yields (ΦPL) were determined to be 0.46 for 2ꢀN
compared to 0.03 for 1ꢀN.46 To extend the scope of luminescent
In light of these observations, salt metathesis reactions were
performed
between
CeI3
and
2
equiv
complexes,
the
corresponding
aryloxide
complex,
[(Me3Si)2NC(NiPr)2]Na(THF) in THF to afford the pale yellow
bridging iodide complex, [(Me3Si)2NC(NiPr)2]2CeIII}2(ꢂ2‒I)2 (3ꢀI)
(Scheme 1). An Xꢀray crystallography study revealed the dimeric
structure of the product, featuring a [Ce2I2]4+ diamondoid core
(Figure S2). Four broad resonances were observed in the 1H NMR
spectrum obtained at 300 K for 3ꢀI in tolueneꢀd8 or C6D6 (Figure
S11). Heating the tolueneꢀd8 solution of 3ꢀI resulted in peak coaꢀ
[(Me3Si)2NC(NiPr)2]Ce(OAr)2 (2ꢀOAr), was synthesized through
protonolysis reaction of 2ꢀN with 2 equiv HOAr (Ar = 2,6ꢀdiꢀtertꢀ
butylphenol) (Scheme 1). Similarly, protonolysis of 1ꢀN with 3
equiv HOAr resulted in Ce(OAr)3 (1ꢀOAr) according to a previꢀ
ous report.60ꢀ61
With the monoꢀguanidinate bisꢀamide/aryloxide Ce(III) comꢀ
plexes in hand, we set out to prepare bisꢀ and trisꢀguanidinate
derivatives. However, treatment of 2ꢀN with excess N,N'ꢀ
diisopropylcarbodiimide did not lead to further insertion reactivity
up to 80 °C in benzene.46 In order to understand the energy reꢀ
quirements for the insertion reactions, density functional theory
(DFT) calculations were carried out for 1ꢀN and the established
1
lescence at 325 K (Figure S13). The H NMR spectrum, fully
resolved at 380 K (Figure S12), displayed three resonances at
14.20, 3.68 and −3.26 ppm in a 2:18:12 ratio, attributable to –
CH(iPr), –CH3(iPr) and –CH3(SiMe3) groups, respectively, from
the guanidinate ligands. Further salt metathesis reactions of 3ꢀI
with NaN(SiMe3)2 or Na(OAr)(THF)2 afforded bright yellow
and
expected
insertion
products:
2ꢀN,
2
ACS Paragon Plus Environment

Page 3 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Figure 1. Thermal ellipsoid plots of 2ꢀN(a), 3ꢀN(b), 4(c) at the 30% probability level.
solids of corresponding amide and aryloxide derivatives, heating a 1:1 mixture of 1ꢀN and 4 in C6D6 at 80 °C for 30 h
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[(Me3Si)2NC(NiPr)2]2Ce[N(SiMe3)2]
[(Me3Si)2NC(NiPr)2]2Ce[OAr] (3ꢀOAr) (Scheme 1).
(3ꢀN)
and
resulted in only negligible amounts of 2ꢀN and 3ꢀN as detected by
1H NMR spectroscopy (Figure S21). The preparation of the yelꢀ
low CeIII compounds described above completed two mixedꢀ
Conversely,
treatment
of
CeI3
with
3
equiv
[(Me3Si)2NC(NiPr)2]Na(THF) in toluene afforded pale yellow
solids of [(Me3Si)2NC(NiPr)2]3Ce (4) (Scheme 1). The identity of
4 was confirmed by Xꢀray crystallography (Figure 1c), constitutꢀ
ligand
series
with
the
general
formulae
[(Me3Si)2NC(NiPr)2]xCeIII[N(SiMe3)2]3ꢀx (x = 0, 1ꢀN; x = 1, 2ꢀN, x
= 2, 3ꢀN; x = 3, 4) and [(Me3Si)2NC(NiPr)2]xCeIII(OAr)3ꢀx (x = 0,
1ꢀOAr; x = 1, 2ꢀOAr, x = 2, 3ꢀOAr; x = 3, 4). All of these comꢀ
pounds were found to be luminescent, thus allowing a systematic
comparison by stepwise structural variations to study the photoꢀ
physical properties of luminescent CeIII complexes.
1
ing the first homoleptic CeIII guanidinate complex. The H NMR
spectrum of 4 in C6D6 at 300 K revealed three resonances at
10.02, 2.81 and −4.15 ppm, consistent with a C3v solution geomeꢀ
try. Complex 4 was robust toward ligand redistribution reactions;
iPr
iPr
iPr
SiMe3
SiMe3
SiMe3
SiMe3
N
N
N
N
Me3Si
Me3Si
Me3Si
Me3Si
N
Me3Si
Me3Si
Ce N
N
Ce
Ce N
N
N
Ce N
SiMe3
N
SiMe3
3
iPr
4
iPr
2
iPr
3
2
2-N
3-N
1-N
Figure 2. Absorption (solid line) and emission spectra (dashed line) of 1ꢀN (red), 2ꢀN (green), 3ꢀN (pink) and 4 (blue). Inset: images of
toluene solution of 1ꢀN, 2ꢀN, 3ꢀN, 4 in 1 cm path length quartz cuvettes (1.0 mM) under 365 nm UV irradiation.
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 11
2.2. Electronic Structures of Luminescent Ce(III) Complexes.
were assigned as the corresponding acceptor orbitals using NTO
We noted that both series of CeIII complexes defined in the current
work were exclusively yellow despite the difference in their ligꢀ
and coordination spheres. Indeed, the electronic absorption specꢀ
tra collected for 1ꢀN, 2ꢀN, 3ꢀN, 4 and 3ꢀOAr, 2ꢀOAr and 1ꢀOAr
in toluene displayed remarkable similarity; absorption features at
ca. 420 nm and ca. 360 nm were observed for all complexes with
absorptivities of ε ~ 102 Mꢀ1 cmꢀ1. These molar extinction coeffiꢀ
cients are consistent with reported f→d transitions.62ꢀ63 Absorpꢀ
tion and emission spectra for the representative guanidinateꢀamide
mixedꢀligand series are shown in Figure 2, featuring the yellow
emitting complex 1ꢀN, limeꢀgreen 2ꢀN, green 3ꢀN and blue 4.
Our previous work on 1ꢀN and 2ꢀN showed the origins of the
absorption features between 320−500 nm as metalꢀcentered interꢀ
configurational 4f→5d transitions.46 For the other members of the
series, 3ꢀN and 4, timeꢀdependent density functional theory (TDꢀ
DFT) calculations were carried out at the B3LYP level of theory
to assist in assignment of the observed transitions. Subsequent to
the TDꢀDFT calculations, natural transition orbital (NTO) compuꢀ
tations were performed to visualize compact orbital representaꢀ
tions for the vertical excitations of interest.64ꢀ65
calculations (Figure S81−82). In comparison, the degeneracy of
the 5dꢀmanifold was broken for complexes 2ꢀN and 3ꢀN in apꢀ
proximate C2v solution symmetry; only one of the 5dxz or 5dyz
orbitals was identified as the acceptor (Figure S77).
1
2
3
4
5
6
7
8
The TDꢀDFT assignment of the absorptive transitions was
further supported by experimental evidence. The lanthanum anaꢀ
logue of 1ꢀN, La[N(SiMe3)2]3 (4f0) showed no absorption feature
between 320−500 nm, further supporting both absorption bands
observed for 1ꢀN are 4f→5d transitions (Figure S42). We also
prepared the NEt4Cl adduct of 1ꢀN. The blue emitting compound,
[NEt4]{CeIIICl[N(SiMe3)2]3} (1ꢀNCl−) demonstrated a trigonal
pyramidal CeClN3 core with sum of N−Ce−N angles at 347.3(1)°
(Figure S8). The electronic absorption spectrum of 1ꢀNCl−
demonstrated only one absorption feature at 366 nm as compared
to two bands for 1ꢀN (Figure S42). The absence of the absorptive
transition at ca. 420 nm for 1ꢀNCl− is consistent with the assignꢀ
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ment as a transition to a state corresponding to the 5dz orbital in
1ꢀN that is driven to higher energies by crystal field arguments
upon coordination of Cl− in 1ꢀNCl−.
Despite the similar absorption energies for 1ꢀN, 2ꢀN, 3ꢀN and
4, the wavelengths of their emission maxima spanned a large
range from 553 nm (1ꢀN) to 459 nm (4) (Figure 2). This observaꢀ
tion was best described with the observed Stokes shifts calculated
from the difference in wavelength between the positions of emisꢀ
sion maxima and the lowest energy absorption maxima. Among
the guanidinateꢀamide mixedꢀligand series, complex 1ꢀN featured
the largest Stokes shift of 138 nm while complex 4 demonstrated
a much smaller Stokes shift of 35 nm. Interestingly, the observed
Stokes shifts were found to decrease with increasing numbers of
guanidinate ligands across the series (Figure 4, solid line). The
existence of such a trend was confirmed by the guandinateꢀ
aryloxide mixedꢀligand series (Figure 4, dashed line). The largest
Stokes shift in the guanidinateꢀaryloxide mixedꢀligands series was
found to be 112 nm for 1ꢀOAr, smaller than 138 nm observed for
1ꢀN.
Figure 3. Calculated acceptor natural transition orbitals (NTOs)
of the transition at ca. 420 nm for 1ꢀN, 2ꢀN, 3ꢀN and 4, demonꢀ
2
strating the primarily nonbonding 5dz orbital character.
For all the complexes, the donor and acceptor orbitals for the
2
transition at ca. 420 nm were identified to be 4f orbitals and 5dz
orbitals, respectively (Figure 3). This result was readily rationalꢀ
ized by inspection of the complex crystal structures. In the comꢀ
plexes with approximately C3v geometry (1ꢀN and 4), or those
with approximately C2v geometry (2ꢀN and 3ꢀN), the 5dz2 orbitals
are essentially nonꢀbonding in character. While the ligand field
has an only small impact on the optical energies of 4fꢀorbitals,66
the energy of metal centered 4f→5d transitions is dependent on
the relative energies of the acceptor 5dꢀorbitals.67 Therefore, the
2
nonꢀbonding nature of the 5dz orbitals is understood to be responꢀ
sible for the similar lowest energy absorption features in visible
range (400−800 nm) of these luminescent CeIII complexes, as well
as their yellow colors.
On the other hand, the identities of acceptor orbitals for the
second lowest energy absorptive 4f→5d transitions is dictated by
the ligand field splitting of the 5d manifold. For 1ꢀN and 4 with
C3v solution symmetry, twoꢀfold degenerate sets of 5dꢀorbitals
Figure 4. Observed Stokes shift of guanidinateꢀamide mixedꢀ
ligand series (solid line, including 1ꢀN, 2ꢀN, 3ꢀN and 4) and guanꢀ
idinateꢀaryloxide mixedꢀligand series (dashed line, including 1ꢀ
OAr, 2ꢀOAr, 3ꢀOAr, 4).
4
ACS Paragon Plus Environment

Page 5 of 11
Journal of the American Chemical Society
In an effort to understand the Stokes shifts, we attempted to
0.81 for 4). For both series, the quantum yields reached maxima at
0.81 for the blueꢀemitting complex 4. This quantum yield was
higher than the reported quantum yield of 0.55 in ethanol for the
1
2
3
4
5
6
7
8
assign the ground and excited states giving rise to the emissive
transitions. We determined that the lifetime for 4 was invariant
when measured at different emission wavelengths. This result
suggested that only one longꢀlived excited state was involved in
the emission process. Excitation spectra collected in toluene for
both series of complexes showed intense bands, overlapping with
their lowest energy absorption bands at ca. 420 nm (See Figure
S44−S51). This observation allowed us to conclude that lowest
lying 5dz2 orbitalꢀbased excited state corresponded to the longꢀ
lived excited state for these complexes. All the emission spectra
were fit with two overlapping Gaussian functions (on an energy
scale) separated by 1200 to 1700 cmꢀ1 (Figure S52−S59). The
energy difference between the two Gaussian functions was conꢀ
single
other
blue
emitting
CeIII
complex,
tris(Nꢀ
[CeIII(triRNTB)2](CF3SO3)3 complex (triRNTB
=
propylbenzimidazolꢀ2ꢀylmethyl)amine).58 On the other hand, the
low quantum yield of 0.03 for 1ꢀN indicated the presence of seꢀ
vere nonꢀradiative relaxation processes that quenched the longꢀ
lived 2D excited state.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
sistent with the spinꢀorbit splitting the F ground state manifold
2
into the 2F5/2 ground state and F7/2 excited state for the Ce3+ catiꢀ
on.
The Stokes shifts represent the degree of energy loss between
excited state following the absorptive vertical transition and longꢀ
lived excited state. Since complexes 1ꢀN, 2ꢀN, 3ꢀN and 4 demonꢀ
strated a common longꢀlived 2D excited state with singly occupied
nonꢀbonding 5dz2 orbital and 4fꢀorbital based ground states, very
little geometric changes were expected for the excitedꢀstate strucꢀ
tures compared to the groundꢀstate ones. This was consistent with
the findings for reported complexes [(C5H4SiMe3)3LnII]− (Ln =
La, Ce, Pr, Nd, Gd, Tb, Dy, Ho, Er)68ꢀ70 with 5dz orbitalꢀbased
2
ground
states
and
[(C5H4SiMe3)3UII]−71
and
{[C5H3(SiMe3)2]3ThII}−68 with 6dz ꢀorbitals based ground states;
the bond lengths and geometries of these divalent species were
essentially identical compared to their LnIII, UIII and ThIII counterꢀ
2
2
2
parts, resulting from the nonꢀbonding character of 5dz and 6dz
orbitals.
The difference in the Stokes shifts for the guanidinateꢀamide
and guanidinateꢀaryloxide series reflects the difference in energy
2
relaxation of longꢀlived D excited states in the series. Such deꢀ
crease of relaxed excitedꢀstate energy is likely facilitated by couꢀ
pled vibrational modes. We expected the dominant relaxation
pathway occurred through interactions of the metal cation with
ligand C−H oscillators. The geometry of complex 1ꢀN had previꢀ
ously been determined to be trigonal pyramidal using gas phase
electron diffraction studies.72 The deviation of 1ꢀN from trigonal
planar geometry implied the presence of C−H agostic interactions
between the Ce3+ cation and methyl groups of the silyl amide
ligands. The replacement of amide ligands with guanidinate ligꢀ
ands removed C−H agostic interactions and thus resulted in less
dissipation of energy through vibrational process in excited state.
In addition, we postulate the rigidity of bidentate guanidinate
ligands compared to monodentate amide ligands restricted vibraꢀ
tion relaxation process through Ce−N and N−Si bonds.
Figure 5. Quantum yields and lifetime data for 1ꢀN, 2ꢀN, 3ꢀN, 4,
3ꢀOAr, 2ꢀOAr and 1ꢀOAr. Standard deviations of data are proꢀ
vided.
Timeꢀresolved luminescence data were collected near the emisꢀ
sion maxima upon excitation at 380 nm (Figure S60−S68). Single
exponential fits were applied to the decay data to afford lifetimes
To quantitatively determine the efficiency of emission, photoꢀ
luminescence quantum yields (ΦPL) were obtained in toluene
2
(τ). The lifetime for the D excited state of 1ꢀN was found to be
shorter (τ = 24 ns) compared to 2ꢀN (τ = 65 ns), 3ꢀN (τ = 117 ns)
and 4 (τ = 83 ns). These data are within the range of reported
lifetimes for molecular Ce3+ emitters.54 The difference in lifetime
was attributable to the nonꢀradiative decay processes imposed by
ligand oscillators. The ability of X−H (X = C, N, O, B) oscillators
to quench lanthanide longꢀlived excited states had been widely
established in the literature.75ꢀ80 With the partial replacement of
amide ligands for guanidinate ligands, the quenching processes
through
a
comparative method73 referenced to 9,10ꢀ
diphenylanthracence (Φstd = 0.97) in cyclohexane.74 The calculatꢀ
ed quantum yields (Figure 5) were found to increase with increasꢀ
ing number of guanidinate ligands in both the guanidinateꢀamide
mixedꢀligands series (from 0.03 for 1ꢀN, 0.46 for 2ꢀN, 0.79 for 3ꢀ
N to 0.81 for 4) and the guanidinateꢀaryloxide mixedꢀligands
series (from 0.18 for 1ꢀOAr, 0.24 for 2ꢀOAr, 0.32 for 3ꢀOAr to
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 11
from low energy ligand vibrational modes were suppressed. This
Stokes shifts from 138 nm for 1ꢀN to 35 nm for 4 implied less
1
2
3
4
5
6
7
8
phenomenon was reflected in the nonꢀradiative decay rates (knr)
extracted from the photoluminescence quantum yields and lifeꢀ
times; knr was determined to be 41.0(1) × 106 sꢀ1 for 1ꢀN, larger
than the value 8.3(1) × 106 sꢀ1 determined for 2ꢀN, and the value
of 1.8(2) × 106 sꢀ1 and 2.4(3) × 106 sꢀ1 for 3ꢀN and 4, respectively.
Similarly, a decreasing trend of knr was also noted for the guaniꢀ
dinateꢀaryloxide series, ranging from 9.0(1) × 106 sꢀ1 for 1ꢀOAr,
8.9(1) × 106 sꢀ1 for 2ꢀOAr, 5.8(1) × 106 sꢀ1 for 3ꢀOAr and 2.4(3) ×
106 sꢀ1 for 4. The decrease in knr can also be attributed to the supꢀ
pression of vibrational quenching processes; the pyramidal solidꢀ
state structure of 1ꢀOAr61 indicates the presence of C−H agostic
interactions. The fiveꢀfold smaller knr value found for 1ꢀOAr
compared to 1ꢀN can be understood based on differences in the
ligands and their sterics.81ꢀ82 For example, the spatial orientation
of methyl groups in the OAr− ligand prohibits C−H bonds from
engaging in agostic interactions as efficiently as the [N(SiMe3)2]−
ligands. In addition, the trend in observed knr for the mixedꢀligand
series is consistent with the expectation from the energy gap
law:83ꢀ84 compounds with lower emission energies demonstrate
higher nonꢀradiative decay rates (Figure S72−S73). On the other
hand, the observed radiative decay rates, kr ranged from 1.12(1) ×
106 sꢀ1 for 1ꢀN to 9.7(3) × 106 sꢀ1 for 4. The magnitudes of kr valꢀ
ues were consistent with predictions made from the StricklerꢀBerg
equation85 (Table 1)
energy loss in the excited state of complex 4 compared to 1ꢀN,
affording more powerful photoꢀreductants. One the other hand,
the steric congestion around the Ce3+ cation in 4 was increased
compared to 1ꢀN, making Ce3+ center less accessible for subꢀ
strates.
In order to estimate the excitedꢀstate reduction potentials (Ee.s.
for 3ꢀN and 4, the RehmꢀWeller formalism was used:18,86
Ee.s. = Eg.s. – E0,0 + ω
)
In this relationship, the energy difference between 0th vibrational
electronic ground state and excited state (E0,0) can be approximatꢀ
ed by the emission band energy. The electrostatic interaction due
to the separation of charges is represented by the work function,
ω. The work function is a small contribution compared to E0,0 and
usually omitted from the equation for simplicity.18 The ground
state reduction potential (Eg.s.) was obtained from electrochemisꢀ
try data collected in CH2Cl2.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Predicted radiative decay rates krSB from StricklerꢀBerg
analysis and observed radiative decay rates krobs and nonꢀradiative
obs
.
decay rates knr
krSB (×106 sꢀ1)a
krobs(×106 sꢀ1)b knrobs(×106 sꢀ1)b
1ꢀN
2ꢀN
3ꢀN
1.7
1.7
1.8
2.4
1.7
2.1
2.7
1.12(1)
7.0(1)
6.8(2)
9.7(3)
2.8(1)
2.8(1)
2.0(1)
41.0(1)
8.3(1)
1.8(2)
2.4(3)
5.8(1)
8.9(1)
9.0(1)
4
3ꢀOAr
2ꢀOAr
1ꢀOAr
a. krSB was calculated from StricklerꢀBerg equation.
b. krobs, knrobs was obtained from the equation krobs = ΦPL/τ, knr
(1ꢀΦPL)/τ.
=
obs
As with 4, the pale yellow dimeric complex, 3ꢀI, was found to be
blue emitting, with a quantum yield of 0.53 and a lifetime of 69 ns
in toluene. The colorless complex 1ꢀNCl− emitted blue light under
UV light with a quantum yield of 0.80 and a lifetime of 76 ns. The
small Stokes shift (38 nm for 3ꢀI, 88 nm for 1ꢀNCl−), high quanꢀ
tum yield and long lifetime of 3ꢀI and 1ꢀNCl− further supports our
thesis about the role of vibrational relaxation processes in these
luminescent CeIII complexes. With the above photoꢀphysical inꢀ
terpretation of CeIII longꢀlived excited states obtained from specꢀ
troscopic and computational results, we were prompted to investiꢀ
gate the photochemical properties for the guanidinateꢀamide
mixedꢀligand series.
Figure 6. Cyclic voltammetry of 2ꢀN, 3ꢀN and 4 at a scan rate of
0.5 V sꢀ1 in CH2Cl2, with 0.1 M [nPr4N][BarF ] supporting electroꢀ
4
lyte.
Cyclic voltammetry of 2ꢀN, 3ꢀN and 4 in CH2Cl2 demonstrated
quasiꢀreversible redox features (Figure 6). The observed large
separations between the cathodic and anodic waves were intrinsic
to the compounds and attributed to slow heterogeneous electron
transfer kinetics between the electrodes and solution species.87ꢀ88
To better establish their thermodynamic potentials, the cyclic
voltammograms were simulated using reversible CeIII/IV couples
(Eg.s.) and standard heterogeneous electron transfer rates (Figure
S32). The simulated potentials of the CeIII/IV couples (Eg.s.) for 2ꢀ
N, 3ꢀN and 4 were determined at +0.09, −0.15 and –0.22 V versus
Cp2Fe0/+, respectively. The decreasing Eg.s. from 2ꢀN to 4 indicatꢀ
ed that guanidinate ligand was more electron donating than the
amide ligand. In the case of 1ꢀN, no feature was observed in
CH2Cl2. Electrochemical data for 1ꢀN obtained from THF89 were
2.3. Photoꢀinduced Inner and Outer sphere electron transfer.
In our previous work, we demonstrated that the 2D excited state of
1ꢀN and 2ꢀN could effectively participate in chemical reactions
through single electron transfer(SET).46 The presence of an accesꢀ
sible coordination site in 1ꢀN and 2ꢀN allowed inner sphere haloꢀ
gen abstraction reactivity. In this work, the decrease of observed
6
ACS Paragon Plus Environment

Page 7 of 11
Journal of the American Chemical Society
applied for the estimation of Ee.s., after correcting for the potential
loss for 3ꢀN and 4 in excited state following vertical transitions at
difference between those solvents with Cp2Fe0/+ couple.
ca. 420 nm. However, no reaction was observed for 3ꢀN or 4 with
PhCH2Cl under the same reaction conditions (Figure S22−23).
The lack of reactivity for 3ꢀN and 4 toward PhCH2Cl can be raꢀ
tionalized based on the accessibility of Ce3+ cation; the steric
congestion in 3ꢀN and 4 around Ce3+ cations do not favor subꢀ
strate association in their 2D excited states.
1
2
3
4
5
6
7
8
Thus, the excitedꢀstate reduction potentials Ee.s. were estimated
to be –2.19, −2.30, −2.59 and –2.92 V for complex 1ꢀN, 2ꢀN, 3ꢀN
and 4, respectively (Table 2). These results put complex 4 among
the strongest known photoꢀreductants, comparable to [(PNP)CuI]2
(PNP− = bis(2ꢀdiisopropylphosphino)phenyl)amide)32 with its
estimated Ee.s. of −3.2 V versus Cp2Fe+/0 and W(CNIph)6 (Iph =
We have also demonstrated the photocatalytic coupling reacꢀ
tions between 4ꢀFꢀC6H4X (X = Br, I) and benzene mediated by 1ꢀ
N to afford 1ꢀ(4ꢀfluorophenyl)benzene.46 In addition to the obserꢀ
vation of corresponding CeIV−Br species in stoichiometric reacꢀ
tions between 1ꢀN and 1ꢀbromoꢀ4ꢀfluorobenzene,46 the inner
sphere SET mechanism of 1ꢀN with aryl bromides is also reflectꢀ
ed in the relative conversions of catalytic reactions toward differꢀ
ent paraꢀsubstituted substrates (Table 3). The conversion of 4ꢀRꢀ
C6H4Br (R = Me, H, F) was found to decrease with the electron
donating ability of the substituents under same reaction condition
(entry 1ꢀ3). Such a pattern of reactivity is opposite to that exꢀ
pected for an outer sphere SET mechanism where electron poor
aryl rings would be more efficient electron acceptors. Rather, the
observed Me > H > F sequence is consistent with an inner sphere
SET pathway where more electronꢀrich aryl bromide associates
with Ce(III) cation more readily. Interestingly, 2ꢀfluorophenyl
bromide reacts much faster and reaches complete conversion unꢀ
der the same reaction condition (entry 4). We tentatively attributꢀ
ed this result to the chelation effect of 2ꢀfluorophenyl bromide to
the CeIII cation through C−F→CeIII interaction90ꢀ95 that facilitated
substrate binding to the CeIII cation. The involvement of
C−F→LnIII (Ln = La, Sm, Yb) interactions has also been postulatꢀ
ed for stoichiometric and catalytic functionalization of C−F bonds
mediated by lanthanide complexes.96ꢀ98
2,6ꢀdiisopropylphenyl)22 with an Ee.s. of −2.8 V versus Cp2Fe0/+
.
9
The guanidinateꢀamide series complexes are much more reducing
than [Ru(bpy)3]2+ (bpy = 2,2'ꢀbipyridine) and Ir(ppy)3 (ppy = 2,2'ꢀ
phenylpyridine) with values of Ee.s. = −1.31 V and –2.23 V versus
Cp2Fe+/0, respectively.7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
Table 2. Estimation of CeIV/III reduction potential in D excited
state for 1ꢀN, 2ꢀN, 3ꢀN and 4
Eg.s.a /eV
+0.05b
+0.09c
−0.15c
−0.22c
E0,0 /eV
+2.24
+2.39
+2.44
+2.70
Ee.s. /eV
−2.19
−2.30
−2.59
−2.92
1ꢀN
2ꢀN
3ꢀN
4
a. Cyclic voltammetry in 0.1 M [nPr4N][BArF ]/CH2Cl2.
4
b. No features observed for 1ꢀN in CH2Cl2, E1/2 of 1ꢀN in THF
was used and corrected for the difference of Cp2Fe+/0 in THF and
CH2Cl2.
c. The quasiꢀreversible cyclic voltammetry spectra were simulated
with CeIV/III couple (E1/2) and heterogeneous
Previously, we demonstrated the inner sphere halogen abstracꢀ
2
tion reactivity of benzyl chloride for 1ꢀN and 2ꢀN in their D exꢀ
cited states. Irradiating C6D6 solutions of 1ꢀN or 2ꢀN with excess
PhCH2Cl using commercially available compact fluorescent
lamps (CFLs) led to the formation of corresponding CeIV−Cl
compounds, 1ꢀNCl and 2ꢀNCl, accompanied by the formation of
PhCH2CH2Ph (Scheme 2).
Table 3. Catalytic photoꢀinduced arylation reactions of benzene
with 1ꢀN.
Scheme 2. Reaction of 1ꢀN, 2ꢀN, 3ꢀN and 4 with excess PhCH2Cl
in C6D6 under compact fluorescent lamp irradiation.
Entry
X =
Br
Br
Br
Br
I
R =
4ꢀMe
H
Conversiona /%
Yielda /%
1
2
92
76
72
32
86
76
85
91
87
88
23
80
3
4ꢀF
69
4
2ꢀF
> 99
> 99
> 99
> 99
> 99
> 99
> 99
5
4ꢀMe
H
6
I
7
I
4ꢀF
8
I
2ꢀF
9
I
3ꢀMe
2ꢀMe
10
I
a. All percentage conversion to product (percentage conversion of aryl
halides starting materials) are determined by GC.
The arylation reaction of benzene catalyzed by 1ꢀN was found
to be faster for aryl iodides compared to aryl bromides; quantitaꢀ
tive conversion can be achieved for aryl iodides (entries 5ꢀ10)
under the same reaction condition for aryl bromides. Notably, a
significantly lower yield was observed for 2ꢀmethylphenyl iodide
(entry 10, 23 %) compared to 4ꢀmethylphenyl iodide (entry 5, 76
In the current work, complex 3ꢀN and 4 were more readily oxiꢀ
dized than 1ꢀN and 2ꢀN in their ground state. The smaller Stokes
shifts of 3ꢀN and 4 compared to 1ꢀN and 2ꢀN indicate less energy
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 11
%) and 3ꢀmethylphenyl iodide (entry 9, 88 %), despite the comꢀ
cal anion to give iodide and 4ꢀFꢀC6H4•, (3) addition of 4ꢀFꢀC6H4•
1
2
3
4
5
6
7
8
plete conversion for all three substrates. This observation can also
be reasoned based on our proposed reaction pathway (Scheme
4A); direct Hꢀatom abstraction of bulky •N(SiMe3)2 radical from
radical adduct [Ar−C6H6]• is expected to be unfavorable for the 2ꢀ
methylphenyl iodide substrate.
to benzene forming a radical adduct, (4) oxidation of the radical
adduct by 4+ to regenerate 4; (5) reaction of
(THF)Na[(iPr)2CN(SiMe3)2] with H+ to afford the biphenyl prodꢀ
uct. Similar outer sphere single electron transfer pathways for
catalytic phenylation reactions of aryl halides have been reported
with Group I and transition metal 1,10ꢀphenanthroline complexes
at elevated temperatures.100ꢀ104 More recently, homolytic aromatic
substitution reactions mediated by the photoredox catalyst,
Ir(ppy)3 have also been demonstrated.14
Scheme 3. Catalytic photoꢀinduced phenylation of 4ꢀ
fluorophenyliodide with 4.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Proposed cycle for photoꢀinduced arylation of benzene
through inner sphere SET pathway and outer sphere SET pathꢀ
way.
A. Inner sphere SET pathway with 1ꢀN (X = Br, I):
We also investigated phenylation reactions of aryl halides with
complex 4. Such reaction using 4ꢀFꢀC6H4X (X = Br, I) and stoiꢀ
chiometric amount of NaN(SiMe3)2 in benzene with 1ꢀN as photoꢀ
redox catalyst had been shown by us to afford 1ꢀ(4ꢀ
fluorophenyl)benzene by an inner sphere halogen abstraction
pathway.46 In this case, irradiation of a benzene solution containꢀ
ing 4ꢀfluoroiodobenzene, (THF)Na[(iPr)2CN(SiMe3)2] and 10
mol%
4
for
6
days led to the isolation of 1ꢀ(4ꢀ
fluorophenyl)benzene in 35 % yield (Scheme 3) while no reacꢀ
tions were observed for the 4ꢀFꢀC6H4Br and 4ꢀFꢀC6H4Cl subꢀ
strates. The phenylation reaction of 4ꢀfluoroiodobenzene mediated
by 4 was also performed with intermittent light (Figure S84); no
1ꢀ(4ꢀfluorophenyl)benzene formation was observed during the
dark period, suggesting the product formation did not involve
radical propagation process.
B. Outer sphere SET pathway with 4 (X = I):
To further elucidate the difference in interactions of CeIII comꢀ
plexes with substrates, quenching kinetic data were obtained for
1ꢀN, 2ꢀN, 3ꢀN and 4 through SternꢀVolmer experiments99 using 4ꢀ
fluoroiodobenzene as quencher. The calculated quenching rates,
kq was found to be 3.5(1) × 107 Mꢀ1 sꢀ1 for 4, two orders of magniꢀ
tudes smaller than 2.12(2) × 109 Mꢀ1 sꢀ1 calculated for 1ꢀN (Table
4). The smaller quenching rate for 4 was attributed to the steric
encumbrance around the Ce3+ cation, preventing its direct interacꢀ
tion with aryl iodide substrates.
The oneꢀelectron oxidation product of 4 was synthesized.
Treatment of 4 with [Cp2Fe][BArF ] in THF followed by recrysꢀ
Table 4. Quenching rates of 1ꢀN, 2ꢀN, 3ꢀN and 4 with 4ꢀ
fluoroiodobenzene as quencher.
4
tallization from CH2Cl2/nꢀpentane layering afforded dark green
crystals of [4+][BArF ]. An Xꢀray diffraction study of [4+][BArF ]
1ꢀN
2ꢀN
3ꢀN
4
4
4
kq (×108 Mꢀ1 sꢀ1)
21.2(2) 2.18(3) 0.19(1)
0.35(1)
confirmed the proposed cationic trisꢀguanidinate structure with
average Ce−N bond length observed at 2.397(4) Å for 4+, comꢀ
pared to 2.526(4) Å observed for 4 (Figure S7). The ca. 0.1 Å
difference in Ce−N bond lengths was consistent with the differꢀ
ence in ionic radii between Ce3+ and Ce4+ cations.105 The dark
In addition, we also noted complex 4 was incapable of associating
a Cl− anion (Figure S28) or reacting with an inner sphere oxidant
Ph3C−Cl (Figure S29). In contrast, complex 1ꢀN reacts readily
with NEt4Cl to form [NEt4]{CeIIICl[N(SiMe3)2]3} (1ꢀNCl−) and
with Ph3C−Cl to afford CeIVCl[N(SiMe3)2]3 (1ꢀNCl). These difꢀ
ferences in reactivity between 1ꢀN and 4 suggest the steric enꢀ
cumbrance around the Ce3+ cation in 4 was sufficiently large to
prevent substrate binding. Considering the differences in quenchꢀ
ing kinetics and reactivity, we are prompt to propose the single
electron transfer of 4 occurs through an outer sphere pathway
instead of an inner sphere pathway. Therefore, the catalytic cycle
for the phenylation of aryl iodide with 4 would comprise followꢀ
ing steps (Scheme 4B): (1) photoꢀinduced reduction of 4ꢀFꢀC6H4I
by 4 to produce 4+ and [4ꢀFꢀC6H4I]•−, (2) dissociation of the radiꢀ
green colour of 4+ cation is uncommon for CeIV compounds.106
A
UVꢀvis spectrum of [4+][BArF ] collected in CH2Cl2 revealed an
4
intriguingly low energy absorption band at 1.60 eV (775 nm, Figꢀ
ure S41), indicative of small HOMO−LUMO gap.46,107 This
LMCT band was tentatively assigned to a transition from the
guanidinate nonꢀbonding π orbitals (πn) to a cerium 4fꢀorbital. The
isolation of [4+] suggests the outer sphere single electron transfer
of 4 is a viable process.
3.CONCLUSIONS
8
ACS Paragon Plus Environment

Page 9 of 11
Journal of the American Chemical Society
We have demonstrated, through comparison studies on two
National Science Foundation grant number OCIꢀ1053575. The E.
J. Petersson, D. M. Chenoweth and S. Park groups at UPenn are
thanked for use of their fluorometers. The Walsh group at UPenn
is acknowledged for use of their GC.
1
2
3
4
5
6
7
8
complete mixedꢀligand series, that vibrational relaxation process
2
played a key role in CeIII luminescence. The D excited states,
with singly occupied 5dz2 orbital were determined, using a combiꢀ
nation of TDꢀDFT and spectroscopic studies, to be the longꢀlived
excited state for the luminescent CeIII complexes, in approximate
C2v or C3v geometries. Although essentially the same energy of
light (ca. 420 nm, 2.95 eV) is required to promote the vertical
absorptive transitions for these complexes to their lowest energy
excited states, drastically different emission energies (from 553
nm, 2.24 eV to 459 nm, 2.70 eV) were found for the complexes.
Such energy variance reflected in the Stokes shifts indicates difꢀ
ferent degrees of vibrational relaxation associated with the local
ligand environments. The longꢀlived excited state also suffered
from vibrational quenching processes, leading to various lifetimes
and quantum yields of the complexes.
REFERENCES
(1) Gao, F. G.; Bard, A. J., J. Am. Chem. Soc. 2000, 122, 7426–7427.
(2) Li, J.ꢀC.; Wu, J.ꢀZ.; Gong, X., J. Phys. Chem. Lett. 2014, 5, 1017–
1021.
9
(3) Hussain, M.; ElꢀShafei, A.; Islam, A.; Han, L., Phys. Chem. Chem.
Phys. 2013, 15, 8401–8408.
(4) Schultz, D. M.; Yoon, T. P., Science 2014, 343, 1239176.
(5) Nguyen, J. D.; D'Amato, E. M.; Narayanam, J. M. R.; Stephenson, C.
R. J., Nat. Chem. 2012, 4, 854–859.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(6) Reckenthäler, M.; Griesbeck, A. G., Adv. Synth. Catal. 2013, 355,
2727–2744.
(7) Koike, T.; Akita, M., Inorg. Chem. Front. 2014, 1, 562–576.
(8) Zeitler, K., Angew. Chem., Int. Ed. 2009, 48, 9785–9789.
(9) Yoon, T. P.; Ischay, M. A.; Du, J., Nat. Chem. 2010, 2, 527–532.
(10) Narayanam, J. M. R.; Stephenson, C. R. J., Chem. Soc. Rev. 2011,
40, 102–113.
(11) Xuan, J.; Xiao, W.ꢀJ., Angew. Chem., Int. Ed. 2012, 51, 6828–6838.
(12) Maity, S.; Zheng, N., Synlett 2012, 23, 1851–1856.
(13) Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C., Chem. Rev. 2013,
113, 5322–5363.
(14) Cheng, Y.; Gu, X.; Li, P., Org. Lett. 2013, 15, 2664–2667.
(15) Hari, D. P.; König, B., Angew. Chem., Int. Ed. 2013, 52, 4734–4743.
(16) Hopkinson, M. N.; Sahoo, B.; Li, J.ꢀL.; Glorius, F., Chem. – Eur. J.
2014, 20, 3874–3886.
(17) Tellis, J. C.; Primer, D. N.; Molander, G. A., Science 2014, 345,
433–436.
(18) Tucker, J. W.; Stephenson, C. R. J., J. Org. Chem. 2012, 77, 1617–
1622.
(19) Fei, H.; Sampson, M. D.; Lee, Y.; Kubiak, C. P.; Cohen, S. M.,
Inorg. Chem. 2015, 54, 6821–6828.
(20) Li, D.; Che, C.ꢀM.; Kwong, H.ꢀL.; Yam, V. W.ꢀW., J. Chem. Soc.,
Dalton Trans. 1992, 3325–3329.
(21) Revol, G.; McCallum, T.; Morin, M.; Gagosz, F.; Barriault, L.,
Angew. Chem., Int. Ed. 2013, 52, 13342–13345.
(22) Sattler, W.; Ener, M. E.; Blakemore, J. D.; Rachford, A. A.;
LaBeaume, P. J.; Thackeray, J. W.; Cameron, J. F.; Winkler, J. R.; Gray,
H. B., J. Am. Chem. Soc. 2013, 135, 10614–10617.
(23) Sattler, W.; Henling, L. M.; Winkler, J. R.; Gray, H. B., J. Am.
Chem. Soc. 2015, 137, 1198–1205.
In terms of rational CeIII photosensitizer design, we propose the
relaxation of the excited state should be minimized in order to
confer maximum energy from light into chemical or photovoltagꢀ
es. Our studies offer a design strategy for achieving powerful
molecular CeIII photoꢀreductants by using rigid ligands as well as
removing C−H oscillators from proximity with the metal cation,
in order to minimize vibrational relaxation of the excited state. On
the other hand, tuning the degree of relaxation and quenching
process could afford CeIII luminophores with tunable emission
colors and efficiency.
As an expansion to the inner sphere electron transfer reactivity
of 1ꢀN previously reported by us,46 we demonstrated complex 4,
with a less accessible metal cation than 1ꢀN, participated in an
outer sphere electron transfer pathway with 4ꢀfluoroiodobenzene
to afford 1ꢀ(4ꢀfluorophenyl)benzene. These combined results
revealed the possibility of using fꢀblock cations as photoꢀredox
catalysts. However, the efficiency of these catalytic reactions
mediated by CeIII photosensitizers is currently limited by the low
absorptivity intrinsic to f→d transitions. Further studies on sensiꢀ
tization of CeIII luminophores are underway to improve their light
harvesting capabilities.
(24) Creutz, S. E.; Lotito, K. J.; Fu, G. C.; Peters, J. C., Science 2012,
338, 647–651.
ASSOCIATED CONTENT
(25) Bissember, A. C.; Lundgren, R. J.; Creutz, S. E.; Peters, J. C.; Fu, G.
C., Angew. Chem., Int. Ed. 2013, 52, 5129–5133.
(26) Ziegler, D. T.; Choi, J.; MuñozꢀMolina, J. M.; Bissember, A. C.;
Peters, J. C.; Fu, G. C., J. Am. Chem. Soc. 2013, 135, 13107–13112.
(27) Tan, Y.; MunozꢀMolina, J. M.; Fu, G. C.; Peters, J. C., Chem. Sci.
2014, 5, 2831–2835.
(28) Uyeda, C.; Tan, Y.; Fu, G. C.; Peters, J. C., J. Am. Chem. Soc. 2013,
135, 9548–9552.
(29) Kern, J.ꢀM.; Sauvage, J.ꢀP., J. Chem. Soc., Chem. Commun. 1987,
546–548.
(30) Pirtsch, M.; Paria, S.; Matsuno, T.; Isobe, H.; Reiser, O., Chem. –
Eur. J. 2012, 18, 7336–7340.
(31) Bagal, D. B.; Kachkovskyi, G.; Knorn, M.; Rawner, T.; Bhanage, B.
M.; Reiser, O., Angew. Chem., Int. Ed. 2015, 54, 6999–7002.
(32) Harkins, S. B.; Peters, J. C., J. Am. Chem. Soc. 2005, 127, 2030–
2031.
Supporting Information
Crystallographic data (CIF), electrochemical data, electronic abꢀ
sorption data, excitation and emission data, computational details
and optimization data for catalysis. This material is available free
of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*jmanna@sas.upenn.edu *schelter@sas.upenn.edu
(33) Miller, A. J. M.; Dempsey, J. L.; Peters, J. C., Inorg. Chem. 2007,
46, 7244–7246.
ACKNOWLEDGMENT
(34) Lotito, K. J.; Peters, J. C., Chem. Commun. 2010, 46, 3690–3692.
(35) Deaton, J. C.; Switalski, S. C.; Kondakov, D. Y.; Young, R. H.;
Pawlik, T. D.; Giesen, D. J.; Harkins, S. B.; Miller, A. J. M.; Mickenberg,
S. F.; Peters, J. C., J. Am. Chem. Soc. 2010, 132, 9499–9508.
(36) Xiang, H.; Cheng, J.; Ma, X.; Zhou, X.; Chruma, J. J., Chem. Soc.
Rev. 2013, 42, 6128–6185.
We gratefully acknowledge the University of Pennsylvania and
the National Science Foundation (CHEꢀ1362854) for financial
support. This work used the Extreme Science and Engineering
Discovery Environment (XSEDE), which is supported by U.S.
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 11
(37) Ghosh, I.; Ghosh, T.; Bardagi, J. I.; König, B., Science 2014, 346,
(74) Suzuki, K.; Kobayashi, A.; Kaneko, S.; Takehira, K.; Yoshihara, T.;
725–728.
Ishida, H.; Shiina, Y.; Oishi, S.; Tobita, S., Phys. Chem. Chem. Phys.
2009, 11, 9850–9860.
(75) Browne, W. R.; Vos, J. G., Coord. Chem. Rev. 2001, 219–221, 761–
787.
1
2
3
(38) Thekaekara, M. P., Solar Energy 1976, 18, 309–325.
(39) Cotton, S., Lanthanide and Actinide Chemistry. john Wiley and
Sons, West Sussex, U. K., 2006, p 61–77.
4
5
6
7
8
9
(40) Aspinall, H. C., Chemistry of the f-Block Elements. CRC Press,
2001, p 34.
(41) Duan, C. K.; Reid, M. F., J. Solid. State Chem. 2003, 171, 299–303.
(42) Heidt, L. J.; Smith, M. E., J. Am. Chem. Soc. 1948, 70, 2476–2481.
(43) Heidt, L. J.; McMillan, A. F., J. Am. Chem. Soc. 1954, 76, 2135–
2139.
(44) Donohue, T., Chem. Phys. Lett. 1979, 61, 601–604.
(45) Matthews, R. W.; Sworski, T. J., J. Phys. Chem. 1975, 79, 681–686.
(46) Yin, H.; Carroll, P. J.; Anna, J. M.; Schelter, E. J., J. Am. Chem. Soc.
2015, 137, 9234–9237.
(76) Bunzli, J.ꢀC. G.; Piguet, C., Chem. Soc. Rev. 2005, 34, 1048–1077.
(77) Marks, S.; Heck, J. G.; Habicht, M. H.; OñaꢀBurgos, P.; Feldmann,
C.; Roesky, P. W., J. Am. Chem. Soc. 2012, 134, 16983–16986.
(78) Bünzli, J.ꢀC. G., Coord. Chem. Rev. 2015, 293–294, 19–47.
(79) Yanagida, S.; Hasegawa, Y.; Murakoshi, K.; Wada, Y.; Nakashima,
N.; Yamanaka, T., Coord. Chem. Rev. 1998, 171, 461–480.
(80) Doffek, C.; Seitz, M., Angew. Chem., Int. Ed. 2015, 54, 9719–9721.
(81) Edelmann, F. T., Chem. Soc. Rev. 2012, 41, 7657–7672.
(82) Edelmann, F. T., Chem. Soc. Rev. 2009, 38, 2253–2268.
(83) Griesser, H. J.; Wild, U. P., Chem. Phys. 1980, 52, 117–131.
(84) Kober, E. M.; Caspar, J. V.; Lumpkin, R. S.; Meyer, T. J., J. Phys.
Chem. 1986, 90, 3722–3734.
(85) Strickler, S. J.; Berg, R. A., J. Chem. Phys. 1962, 37, 814–822.
(86) Rehm, D.; Weller, A., Isr. J. Chem. 1970, 8, 259–271.
(87) Robinson, J. R.; Gordon, Z.; Booth, C. H.; Carroll, P. J.; Walsh, P.
J.; Schelter, E. J., J. Am. Chem. Soc. 2013, 135, 19016–19024.
(88) Robinson, J. R.; Carroll, P. J.; Walsh, P. J.; Schelter, E. J., Angew.
Chem., Int. Ed. 2012, 51, 10159–10163.
(89) Williams, U. J.; Robinson, J. R.; Lewis, A. J.; Carroll, P. J.; Walsh,
P. J.; Schelter, E. J., Inorg. Chem. 2014, 53, 27–29.
(90) Plenio, H., Chem. Rev. 1997, 97, 3363–3384.
(91) Lewis, A. J.; Yin, H.; Carroll, P. J.; Schelter, E. J., Dalton Trans.
2014, 43, 10844–10851.
(92) Yin, H.; Robinson, J. R.; Carroll, P. J.; Walsh, P. J.; Schelter, E. J.,
Chem. Commun. 2014, 50, 3470–3472.
(93) Yin, H.; Lewis, A. J.; Carroll, P.; Schelter, E. J., Inorg. Chem. 2013,
52, 8234–8243.
(94) Yin, H.; Lewis, A. J.; Williams, U. J.; Carroll, P. J.; Schelter, E. J.,
Chem. Sci. 2013, 4, 798–805.
(95) Yin, H.; Zabula, A. V.; Schelter, E. J., Dalton Trans. 2016,
doi:10.1039/C1036DT00108D.
(96) Janjetovic, M.; Träff, A. M.; Hilmersson, G., Chem. – Eur. J. 2015,
21, 3772–3777.
(97) Träff, A. M.; Janjetovic, M.; Ta, L.; Hilmersson, G., Angew. Chem.
Int. Ed. 2013, 52, 12073–12076.
(98) Janjetovic, M.; Traff, A. M.; Ankner, T.; Wettergren, J.; Hilmersson,
G., Chem. Commun. 2013, 49, 1826–1828.
(99) Lakowicz, J. R., Principles of Fluorescence Spectroscopy. Springer,
2007, p 54–55.
(100) Shirakawa, E.; Itoh, K.ꢀi.; Higashino, T.; Hayashi, T., J. Am. Chem.
Soc. 2010, 132, 15537–15539.
(101) Li, H.; Sun, C.ꢀL.; Yu, M.; Yu, D.ꢀG.; Li, B.ꢀJ.; Shi, Z.ꢀJ., Chem. –
Eur. J. 2011, 17, 3593–3597.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(47) Russell Saunders term symbols (2D or 2F) are used in this work to
represent the ground and excited states. In our previous communication,
we used crystal field states to represent a specific excited state (2A1) based
on 5dꢀorbital splittings under C2v and C3v crystal fields.
(48) Haxel, G. B., Hedrick, J. B., Orris, G. J. Rare Earth Elements—
Critical Resources for High Technology, U.S. Geological Survey: 2002.
(49) Mioduski, T.; Hao, D.; Luan, H., J. Radioanal. Nucl. Chem. 1989,
132, 105–113.
(50) George, N. C.; Denault, K. A.; Seshadri, R., Annu. Rev. Mater. Res.
2013, 43, 481–501.
(51) Blasse, G.; Bril, A., Appl. Phys. Lett. 1967, 11, 53–55.
(52) Hazin, P. N.; Bruno, J. W.; Brittain, H. G., Organometallics 1987, 6,
913–918.
(53) Rausch, M. D.; Moriarty, K. J.; Atwood, J. L.; Weeks, J. A.; Hunter,
W. E.; Brittain, H. G., Organometallics 1986, 5, 1281–1283.
(54) Hazin, P. N.; Lakshminarayan, C.; Brinen, L. S.; Knee, J. L.; Bruno,
J. W.; Streib, W. E.; Folting, K., Inorg. Chem. 1988, 27, 1393–1400.
(55) Kunkely, H.; Vogler, A., J. Photochem. Photobiol., A 2002, 151, 45–
47.
(56) Jiao, Y.; Wang, J.; Wu, P.; Zhao, L.; He, C.; Zhang, J.; Duan, C.,
Chem. – Eur. J. 2014, 20, 2224–2231.
(57) Zhao, L.; Liu, Y.; He, C.; Wang, J.; Duan, C., Dalton Trans. 2014,
43, 335–343.
(58) Zheng, X.ꢀL.; Liu, Y.; Pan, M.; Lü, X.ꢀQ.; Zhang, J.ꢀY.; Zhao, C.ꢀY.;
Tong, Y.ꢀX.; Su, C.ꢀY., Angew. Chem., Int. Ed. 2007, 46, 7399–7403.
(59) Vogler, A.; Kunkely, H., Inorg. Chim. Acta 2006, 359, 4130–4138.
(60) Steele, L. A. M.; Boyle, T. J.; Kemp, R. A.; Moore, C., Polyhedron
2012, 42, 258–264.
(61) Stecher, H. A.; Sen, A.; Rheingold, A. L., Inorg. Chem. 1988, 27,
1130–1132.
(62) Higashiyama, N.; Takemura, K.; Kimura, K.; Adachi, G.ꢀy., Inorg.
Chim. Acta 1992, 194, 201–206.
(63) KudaꢀWedagedara, A. N. W.; Wang, C.; Martin, P. D.; Allen, M. J.,
J. Am. Chem. Soc. 2015, 137, 4960–4963.
(64) Tretiak, S.; Saxena, A.; Martin, R. L.; Bishop, A. R., Chem. Phys.
Lett. 2000, 331, 561–568.
(65) Martin, R. L., J. Chem. Phys. 2003, 118, 4775–4777.
(66) Atwood, D. A., The Rare Earth Elements: Fundamentals and
Applications. Wiley, 2013, p 27–34.
(67) Rogers, E. G.; Dorenbos, P., J. Lumin. 2014, 155, 135–140.
(68) Langeslay, R. R.; Fieser, M. E.; Ziller, J. W.; Furche, F.; Evans, W.
J., Chem. Sci. 2015, 6, 517–521.
(102) Sun, C.ꢀL.; Li, H.; Yu, D.ꢀG.; Yu, M.; Zhou, X.; Lu, X.ꢀY.; Huang,
K.; Zheng, S.ꢀF.; Li, B.ꢀJ.; Shi, Z.ꢀJ., Nat. Chem. 2010, 2, 1044–1049.
(103) Sun, C.ꢀL.; Gu, Y.ꢀF.; Huang, W.ꢀP.; Shi, Z.ꢀJ., Chem. Commun.
2011, 47, 9813–9815.
(104) Liu, W.; Cao, H.; Zhang, H.; Zhang, H.; Chung, K. H.; He, C.;
Wang, H.; Kwong, F. Y.; Lei, A., J. Am. Chem. Soc. 2010, 132, 16737–
16740.
(105) Shannon, R., Acta Crystallogr., Sect. A: Found. Crystallogr. 1976,
32, 751–767.
(106) Piro, N. A.; Robinson, J. R.; Walsh, P. J.; Schelter, E. J., Coord.
Chem. Rev. 2014, 260, 21–36.
(69) Hitchcock, P. B.; Lappert, M. F.; Maron, L.; Protchenko, A. V.,
Angew. Chem., Int. Ed. 2008, 47, 1488–1491.
(70) MacDonald, M. R.; Bates, J. E.; Ziller, J. W.; Furche, F.; Evans, W.
J., J. Am. Chem. Soc. 2013, 135, 9857–9868.
(71) MacDonald, M. R.; Fieser, M. E.; Bates, J. E.; Ziller, J. W.; Furche,
F.; Evans, W. J., J. Am. Chem. Soc. 2013, 135, 13310–13313.
(72) Fjeldberg, T.; Andersen, R. A., J. Mol. Struct. 1985, 129, 93–105.
(73) Williams, A. T. R.; Winfield, S. A.; Miller, J. N., Analyst 1983, 108,
1067–1071.
(107) Walter, M. D.; Booth, C. H.; Lukens, W. W.; Andersen, R. A.,
Organometallics 2009, 28, 698–707.
10
ACS Paragon Plus Environment

Page 11 of 11
Journal of the American Chemical Society
1
2
3
4
TOC
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11
ACS Paragon Plus Environment