Effect of ligand deuteration on the decay of Eu3+(5D0) in tris(2,2,6,6-tetramethyl-3,5-heptanedionato)europium(III)

Article abstract of DOI:10.1021/jp982180t

Contributions to the first-order rate constant for the decay of the Eu³⁺(⁵D₀) state of tris(2,2,6,6-tetramethyl-3,5-heptanedionato)europium(III), Eu(thd)₃, from radiative and multiphonon mechanisms are evaluated independently by measuring the luminescence decay rates in undeuterated Eu(thd)₃, fully deuterated Eu(thd-d₁₉)₃, and α-deuterated Eu(thd-d₁)₃, which is also designated Eu(α-D,thd)₃. In the latter case, deuterium substitution is at the α-carbon, between the carbonyl groups of the β-diketonate ligand. These measurements yield a multiphonon contribution of 332 s^-1, of which 157 s^-1 is attributed to relaxation via the C_α-H stretching vibration and 175 s^-1 to relaxation via the C-H stretching modes of the tert-butyl groups. By use of the measured total Eu³⁺(⁵D₀) relaxation rate constant and the multiphonon rate constants given above, the Eu³⁺(⁵D₀) radiative relaxation rate constant is inferred to be 1930 s^-1. It is suggested that this unusually high radiative rate constant may be due to an increased allowedness in the ⁵D₀ → ⁷F_J transitions due to contributions to the predominantly 4f crystal-field wave functions from a low-lying ligand-to-metal charge-transfer state.

Full text of DOI:10.1021/jp982180t

8690
J. Phys. Chem. A 1998, 102, 8690-8694
3
+ 5
Effect of Ligand Deuteration on the Decay of Eu ( D0) in
Tris(2,2,6,6-tetramethyl-3,5-heptanedionato)europium(III)
Todd C. Schwendemann, Paul S. May, and Mary T. Berry*
Department of Chemistry, UniVersity of South Dakota, Vermillion, South Dakota 57069-2390
Yuqing Hou and Cal Y. Meyers
Department of Chemistry and Biochemistry, Southern Illinois UniVersity, Carbondale, Illinois 62901-4409
ReceiVed: May 8, 1998; In Final Form: August 31, 1998
3
+ 5
Contributions to the first-order rate constant for the decay of the Eu ( D
,5-heptanedionato)europium(III), Eu(thd) , from radiative and multiphonon mechanisms are evaluated
independently by measuring the luminescence decay rates in undeuterated Eu(thd) , fully deuterated Eu(thd-
, and R-deuterated Eu(thd-d , which is also designated Eu(R-D,thd) . In the latter case, deuterium
substitution is at the R-carbon, between the carbonyl groups of the â-diketonate ligand. These measurements
0
) state of tris(2,2,6,6-tetramethyl-
3
3
3
d
19
)
3
1
)
3
3
-1
-1
yield a multiphonon contribution of 332 s , of which 157 s is attributed to relaxation via the C
R
-H stretching
-
1
vibration and 175 s to relaxation via the C-H stretching modes of the tert-butyl groups. By use of the
3
+ 5
0
measured total Eu ( D ) relaxation rate constant and the multiphonon rate constants given above, the
3+ 5
-1
Eu ( D
radiative rate constant may be due to an increased allowedness in the D
to the predominantly 4f crystal-field wave functions from a low-lying ligand-to-metal charge-transfer state.
0
) radiative relaxation rate constant is inferred to be 1930 s . It is suggested that this unusually high
5
7
0
J
f F transitions due to contributions
1
. Introduction
There is a considerable body of work assessing the importance
of O-H oscillators in providing a nonradiative path for
5
3+
relaxation of the D0 state of Eu ions in both solid state and
in solution.1 The high-energy O-H stretching mode provides
an efficient path for electronic-to-vibrational energy transfer,
an electronic-state relaxation mechanism generally referred to
in solid-state materials as “multiphonon relaxation”. Much less
work has been done to assess the importance of C-H groups
on organic ligands to nonradiative relaxation. Notable exceptions
-3
1
4
include the work of Kropp and Windsor , Dickens et al., and
5
,6
May et al.
The C-H stretch is generally considered to be inefficient in
relaxing the 12 000 cm
particularly when compared to the O-H stretch. This inef-
ficiency is due largely to the lower energy of the vibration.
-
1
5
7
3+
D0- F6 energy gap of Eu ,
-1
Relaxation due to the C-D stretch ( νj ) 2100 cm ) is expected
to be negligible. Dickens et al. have assessed the contribution
Figure 1. Structures for the (a) substituted 1,4,7,10-tetraazacyclodode-
cane, (b) oxydiacetate, and (c) 2,2,6,6-tetramethyl-3,5-dionate ligands.
7
4
3+ 5
of ligand C-H oscillators to the decay of Eu ( D0) in two
complexes with octadentate ligands based on 1,4,7,10-tetraaza-
cyclododecane (see Figure 1a). For the first ligand, with the
more closely associated methylene group, replacement of the
methylene hydrogens with deuterium led to a decrease in the
rate constant by 26 s per C-H oscillator. Replacing the more
distant hydrogens at the benzylic position of the second ligand
led to a decrease of only 5 s per C-H oscillator. Kropp and
Windsor, in an earlier work, found that, for 0.1 M aqueous
-
1
calculated radiative rate constant of 94 s , estimated that the
3
+
ligand contribution to the decay rate of Eu in Eu(ODA)3‚
-
1
6D2O was 210 s . If the ligand multiphonon decay can be
attributed to the C-H stretch, which is the highest-energy
-1
-1
vibration (see Figure 1b), an average contribution of 18 s from
each of the 12 C-H oscillators is suggested, again consistent
-
1
4
with the findings of Dickins et al.
1
The â-diketonate complex, tris(2,2,6,6-tetramethyl-3,5-hep-
tanedionato)europium(III) (Eu(thd)3) is a particularly promising
system in which to investigate the role of C-H oscillators in
(
D2O) solutions of europium acetate, replacing acetate by
5
perdeuterioacetate led to a decrease in the D0 decay constant
from 560 to 420 s . Since the dominant solution species at
that concentration should have been Eu(OAc)2 , their results
suggest a contribution of about 23 s per C-H group, which
is quite consistent with the findings of Dickins et al. May et
al., using a measured decay constant of 304 s and a
-1
3+ 5
the nonradiative deactivation of Eu ( D0). Although Eu(thd)3
+
is hygroscopic and can form a monoaquo adduct, Eu(thd)3‚H2O,
the anhydrous form is used in this work, and therefore, no O-H
oscillators are present in the crystal. The highest-energy vibration
-
1
4
5,6
-1
-1
available is the C-H stretch at ∼2900 cm . There is one C-H
1
0.1021/jp982180t CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/21/1998

Effect of Ligand Deuteration
J. Phys. Chem. A, Vol. 102, No. 45, 1998 8691
oscillator at the R position and there are 18 on the tert-butyl
groups of each of the ligands. Owing to their relative proximity,
the R oscillators might be expected to interact more effectively
per oscillator with Eu3 compared to the tert-butyl oscillators,
but the large relative number of tert-butyl oscillators might result
in their making a significant, collective contribution to nonra-
diative decay. In this work, the roles of the two different kinds
+
3+ 5
of C-H oscillators in the deactivation of Eu ( D0) are assessed
independently by selectively eliminating alternative paths of
multiphonon relaxation by means of ligand deuteration. Total
deuteration permits measurement of the radiative rate constant
5
for decay of the D0 state.
In this work, we report the effect of deuteration of the thd
3+ 5
ligand on the rate constant for Eu ( D0) relaxation in Eu(thd)3.
-
1
The multiphonon contribution to the rate constant is 332 s ,
-
1
of which 157s is attributed to relaxation via the CR-H
-
1
stretching vibration and 175 s to relaxation via the C-H
stretching modes of the tert-butyl groups. Using the measured
3+ 5
total Eu ( D0) relaxation rate constant and the multiphonon
3+ 5
rate constants given above, we have inferred that the Eu ( D0)
-
1
radiative relaxation rate constant is 1930 s . That the residual
rate constant is indeed radiative and not attributable to rapid
energy migration to traps is confirmed by measurements in 0.5%
3+
Eu doped into Gd(thd)3. This unusually high radiative rate
constant may be associated with an increased allowedness in
5
7
D0 f FJ transitions due to contributions to the predominantly
f crystal-field wave functions from a low-lying ligand-to-metal
4
charge-transfer state. The nature of this charge-transfer state in
5
Eu(thd)3 and its role in D0 relaxation at temperatures above
2
00 K have been discussed previously.8
Figure 2. Synthetic route to thd-d18.
washed with 0.64 M H2SO4, and the hexane solution evaporated
by passing dry N2 through it.
2. Experimental Section
2
.1. Synthesis of 2,2,6,6-Tetramethyl-3,5-heptanedione-d18
2.2. Synthesis of the Europium Complex Eu(thd-d19)3. The
synthesis of the complex was adapted from the method of
(
thd-d18). The synthesis of thd-d18 was accomplished as outlined
in Figure 2. The preparations of pinacolone-d12 and trimethyl-
d9-acetic acid, starting from acetone-d6, were adapted from
standard methods. Following the general method for converting
carboxylic acids into their chlorides, trimethyl-d9-acetic acid
was converted into pivaloyl chloride-d9, which was isolated and
purified by fractional distillation; bp 104-105 °C [lit. bp 105-
06 °C for pivaloyl chloride]. To a solution of freshly distilled
THF (50 mL) and LDA/THF-heptane solution (Acros Organics;
0.4 mL, 2 M, 20.8 mmol) maintained in an ice-water bath,
pinacolone-d12 (2.6 mL, 20.8 mmol) was added dropwise. After
the mixture was stirred at 0 °C for 20 min, pivaloyl chloride-d9
1.3 mL, 10.4 mmol) was syringed in dropwise. The bath was
removed, and the mixture was stirred at room temperature for
0 min and then transferred to a separatory funnel. The residue
was washed with ether and the washings also transferred. The
mixture in the funnel was washed with dilute HCl, and the
organic layer was separated and dried over anhydrous MgSO4.
The mixture was filtered and the filtrate evaporated in vacuo,
leaving a yellow oil (3.42 g). Column chromatography (silica
1
3
Eisentraut and Sievers. The ligand (6 mmol) was dissolved
in 3 mL of EtOD. To this was added 0.245 g (6 mmol) of NaOD
dissolved in 5 mL of 50/50 EtOD/D2O. The solution was stirred
at room temperature for 1 h. To this solution was added 0.733
g EuCl3‚6H2O (2 mmol). A yellow precipitate formed im-
mediately. To reduce the solubility of the complex in the EtOD/
D2O mixture, 10 mL of D2O was added. The precipitate was
filtered and dried at 50 °C under vacuum for 10 min, and the
crude product was sublimed under vacuum at 125-175 °C for
purification. The sublimed material showed a melting point at
187.5-188.0 °C (cor). The Aldrich Catalog lists 187-189 °C
for Eu(thd)3 (Resolve-Al).
9
10
11
1
1
(
2.3. Synthesis of the Europium Complex Eu(r-D,thd)3. The
complex with the ligand deuterated only at the methylene
position was prepared as was the fully deuterated complex
except that the ligand was the undeuterated form purchased from
Aldrich. Efficient D/H exchange at the R-position occurs in the
NaOD/EtOD/D2O solution prior to the addition of EuCl3.
2.4. FTIR Analysis of Materials. FTIR spectra (Midac, M
series) of undeuterated Eu(thd)3 (Aldrich, Resolve-Al), Eu(R-
D,thd)3, and Eu(thd-d19)3 were measured in KBr (Aldrich, FTIR
grade) pellets. These were used to confirm the level of
deuteration in the complex.
3
2
gel, hexanes) provided thd-d18, 2.15 g, in about 75% yield ( H
NMR). A portion, 1.72 g, was dissolved in methanol and
aqueous CuSO4-NaOAc solution was added. The precipitated
purple thd-d18 Cu complex was extracted with hexanes, and the
combined extracts were evaporated, leaving a purple solid, 1.63
g, which was recrystallized from 95% ethanol; 1.46 g, mp
5
7
2.5. Luminescence Measurements. The D0 r F0 laser
5
7
excitation spectrum and the D0 f F2 emission spectrum were
1
92.5-193.5 °C [cor; lit.12 192-193 °C for thd-Cu complex].
measured at room temperature and compared to previously
8
This complex was dissolved in hexane (3 mL), 5 mL of 0.64
M H2SO4 was added, and the mixture was stirred for 2 h. The
aqueous layer was removed and extracted with hexane, and the
extract was added to the organic layer. The latter was then
measured spectra for undeuterated Eu(thd)3. For both spectra,
the excitation source was an N2-laser/dye-laser combination
(Laser Photonics UV-12/DL-14), using rhodamine 590 dye.
Luminescence from the sample was collected using a single

8692 J. Phys. Chem. A, Vol. 102, No. 45, 1998
Schwendemann et al.
f-matching lens focused at the entrance slit of a 0.46 m
monochromator (Jobin Yvon HR460). Detection was achieved
with a Hamamatsu R2949 photomultiplier connected to a
Stanford Research Systems model SR430 multichannel scaler.
For the excitation spectrum, the monochromator was set for
zeroth-order collection and the dye laser was scanned from 5750
to 5820 Å. Cutoff filters (610 and 590 nm) were used to
eliminate laser scatter from the detection system. A 1.0 OD
neutral density filter and narrow (2 Å) monochromator slits were
used to reduce total signal. The laser was operated with the
oscillator only (amplifier removed) for the same reason.
For the emission spectrum, excitation was fixed at 5787 Å,
and the monochromator was scanned from 6050 to 6400 Å with
3
Å resolution. A 590 nm cutoff filter was used to eliminate
laser scatter.
5
The experimental setup for measuring the D0 luminescence
decay profiles and, hence, the D0 decay rates was similar to
5
that used for measuring the excitation and emission spectra. The
5
7
excitation was fixed on the D0 r F0 transition (5787-5790
Å, shifting to longer wavelengths with lower temperatures), and
5
7
emission was monitored on the D0 f F2 transition at 6103 Å
with 2 Å resolution. The time-resolved emission decay profiles
were averaged over 1500 laser shots at the multichannel scaler
and then downloaded to the controlling PC for analysis.
Temperature control (156-280 K) was achieved by mounting
the sample on a heated copper rod in an optical Dewar, through
which cooled N2 gas was passed. The temperature was
controlled by adjusting the N2 flow rate and the voltage applied
to the heater at the copper rod. Temperature was monitored using
a chromel-alumel thermocouple.
5
f 7F
Figure 3. Comparison of the D
and undeuterated Eu(thd)
0
2
emission for fully deuterated
5
3
. Excitation was set at 5787 Å on the D
F transition. Emission resolution was 3 Å.
0
0
r
7
Decay profiles were measured for Eu(thd-d19) between 156
and 271 K and for 0.5% Eu:Gd(thd)3 at 208 K. Measurements
for Aldrich Resolve-Al Eu(thd)3, which were reported in a
appears at 1540 cm-1 in undeuterated Eu(thd)3, disappears in
both Eu(R-thd)3 and in Eu(thd-d19)3. A strong band, which
appears at 1500 cm for the undeuterated material, is much
reduced for the deuterated species, and a new peak appears at
450 cm . By analogy to the acetylacetonate (acac) complexes,
8
previous work, were repeated in the range 155-271 K. Decay
-1
profiles for Eu(R-D,thd)3 were measured in the range 153-
1
83 K.
-1
1
-
1
the 1540 cm band may be tentatively assigned to a CR-H
out-of-plane bending combination and the 1500 cm band to
3
. Results and Discussion
-1
a ring deformation mode involving principally C-C and C-O
3
.1. FTIR Analysis. The FTIR spectrum of our synthesized
-
1
stretching. These coordinates are somewhat coupled to the
Eu(thd-d19)3 shows bands at 2211, 2124, 2074, and 2044 cm
in the C-D stretching region. The region around 2900 cm ,
which is characteristic of the C-H stretch, shows no absorbance.
The lack of absorbance in the 3500 cm region of the IR
spectrum indicates that the sample is free of water.
The FTIR spectrum of our synthesized Eu(R-D,thd)3 shows
structure in the C-H stretch region identical to that of
undeuterated Eu(thd)3. Furthermore, the spectrum is free of any
features in the 2100 cm region that could be attributed to the
C-D stretch. Two conclusions might be drawn from this
information. First, there is the possibility that the R-position
was not deuterated. However, if this were the case, there should
also be no R-deuteration in our synthesized Eu(thd-d19)3, which
would have resulted in a C-H stretch band near 2900 cm in
the IR spectrum. A second possibility is that the CR-H/D stretch
is not appearing in the expected region, or is simply too weak
to be of use in assessing the degree of deuteration. Normal-
mode analyses of the â-diketonate complexes Cu(acac)2, Pd-
-
1
C -H in-plane bend in the acac complexes. A second indication
R
is the in-plane CR-H bend, which is coupled to the C-CH3
-
1
-1
-
1
2
stretch and appears at 1189 cm in Cu(acac) , at 1199 cm
-1
in Pd(acac)2, and at 1188 cm in Fe(acac)3. An analogous band
appears at 1180 cm in Eu(thd)3 but is much reduced in Eu-
R-D,thd)3 and Eu(thd-d19)3. Comparison of the relative intensity
-
1
(
of this band in the deuterated and nondeuterated materials
indicates 77% deuteration at the R-position for both Eu(R-D,-
thd)3 and Eu(thd-d19)3. The hydrogens/deuteriums on the tert-
butyl groups are not labile, and it is assumed that the degree of
tert-butyl deuteration in Eu(thd-d19)3 is the same as in the
deuterated acetone (99.8% D) from which the ligand was
synthesized.
-
1
-
1
3.2. Emission and Excitation Spectra of Eu(thd-d19)3. Both
5
7
5
7
the D0 r F0 excitation spectrum and the D0 f F2 emission
spectrum of Eu(thd-d ) closely match the corresponding
1
9 3
8
spectra reported previously for undeuterated Eu(thd) . The
zeroth-order D r F excitation spectrum shows a single peak
0 0
at 5787 Å at room temperature. The D f F emission spectra
0 2
3
5
7
(
acac)2, and Fe(acac)3 assign the CR-H stretch to the bands
appearing at 3072, 3070, and 3062 cm , respectively, but
-
1
14
5
7
1
since the CR-H oscillator comprises only /19 or approximately
of the deuterated and undeteurated complex are shown for
comparison in Figure 3. The spectral similarities, similar melting
points, and similar sublimation characteristics strongly support
the conclusion that our Eu(thd-d19)3 is structurally similar to
the undeuterated Eu(thd)3.
5
% of the total C-H oscillators, the signal may be too weak to
observe in our spectrum.
-
1
The 1500 cm region of the spectrum gives a more useful
signature for R-deuteration. A closely spaced doublet, which

Effect of Ligand Deuteration
J. Phys. Chem. A, Vol. 102, No. 45, 1998 8693
Figure 5. Eu3+(5D
) decay rate constants for undeuterated Eu(thd)
0
3
3
(square symbols), Eu(thd-d19
open cirlce).
)
3
(closed circles), and 0.5% Eu:Gd(thd)
(
TABLE 1: Low-temperature Rate Constants for Eu3 ( D
+ 5
)
0
Relaxation in Eu(thd)
.5% Eu:Gd(thd)
3
, Eu(thd-d19
)
3
3
, Eu(r-D Thd) , and
0
3
Eu(thd) a
262 ( 9 s-
Eu(thd-d19
)
a
Eu(R-D,thd) a
2141 ( 6 s-1
0.5% Eu:Gd(thd)
2312 s-1
b
3
3
3
3
1
1966 ( 4 s-1
2
a
Mean values and standard deviations for the lowest-temperature
measurements. b Value for measurement made at 208 K.
Figure 4. Luminescence decay profile for Eu(thd-d19
)
3
at 192 K. The
the highest-energy vibration available to the lanthanide, and R
and â are lattice-dependent parameters. Variation with host for
R and â is not large. Typical values for R are on the order of
residual shows the quality of fit to a single exponential of the form I
exp(-1972 × time).
0
-
2
-3
7
1
0 -10 cm. The parameter â is generally of the order 10
3
+ 5
3
.3. Eu ( D0) Relaxation Rates in Eu(thd)3, Eu(thd-d19)3,
-
1
3+ 5
s . Large energy gaps, like that between Eu ( D0) and the
5
Eu(r-D,thd)3, and 0.5% Eu:Gd(thd)3. All D0 luminescence
decay curves measured in this study were found to be single-
exponential. Figure 4 shows a decay profile for Eu(thd-d19)3 at
7
FJ manifolds, can be bridged efficiently only by high-energy
vibrations. In Eu(thd)3, the highest-energy vibration is the C-H
stretch of the thd ligand.
1
92 K with the residual of the fit to an exponential function.
The temperature dependence of the decay rate constants for
As mentioned previously, relaxation due to the C-D stretch
is expected to be negligible. Therefore, the measured low-
Eu(thd)3 and Eu(thd-d19)3 in the range 150-270 K is shown in
Figure 5. Also included in Figure 5 is the rate constant measured
for 0.5% Eu:Gd(thd)3 at 208 K. As previously reported for
-
1
temperature rate constant in Eu(thd-d19)3 of 1966 ( 4 s is
almost entirely radiative, except for a small nonradiative
contribution due to incomplete deuteration at the R position of
the thd ligands. For Eu(R-D,thd)3, the low-temperature rate
8
undeuterated Eu(thd)3, the rate constants are relatively constant
at temperatures up to 200 K, after which they begin to rise
rapidly. The strong temperature dependence of the lifetime of
-1
constant is 2141 ( 6 s , which includes nonradiative contribu-
tions from the C-H oscillators of the tert-butyl groups. The
5
3+
the D0 state of Eu in crystalline Eu(thd)3 above 200 K has
been attributed to relaxation through a low-lying ligand-to-metal
-
1
low-temperature rate constant of 2262 ( 9 s measured in
1
8
undeuterated Eu(thd)3 is due to radiative relaxation and
nonradiative relaxation involving all the C-H oscillators of the
thd ligand. The reported rate constants are averages of the three
measured values for temperaures below 200 K. The errors
represent the standard deviation of the mean.
8
charge-transfer state. Owing to the significant thermal barrier
to charge-transfer deactivation of D0 in Eu(thd)3, relaxation
5
via this mechanism is ineffective below 200 K. Thus, at low
temperatures, relaxation is due only to radiative and multiphonon
processes.
The difference in the rate constants for Eu(R-D,thd)3 and Eu-
thd-d19)3 gives the rate constant for the multiphonon path
Values for the low-temperature rate constants for 5D0
relaxation in Eu(thd)3, Eu(thd-d19)3, Eu(R-D,thd)3, and 0.5% Eu:
Gd(thd)3 are given in Table 1.
(
involving the C-H oscillators of the tert-butyl groups such that
Multiphonon relaxation rates, wmp, for lanthanides in a given
host generally follow an exponential energy gap law,
-1
1
5-17
w (tert-butyl C-H) ) 175 ( 7 s
mp
wmp ) â exp(-R(E - 2hνmax))
The difference between the rate constants for Eu(R-D,thd)3
and undeuterated Eu(thd)3 gives the rate constant for the
multiphonon path involving the R-hydrogens (taking into
where E is the energy gap to be bridged, hνmax is the energy of

8694 J. Phys. Chem. A, Vol. 102, No. 45, 1998
Schwendemann et al.
account that the R-position is only 77% deuterated) such that
TABLE 2: Contributions to Total Rate Constant for
3
+ 5
0 3
Eu ( D ) Relaxation in Eu(thd)
-
1
a
-H)b
mp(tert-butyl C-H)c
175 s-1 ( 7
d
w (C -H) ) 157 ( 15 s
w
r
w
mp(C
R
w
w
CT
mp
R
1
13
1
930 s- ( 5 157 s-1 ( 15
1.2 × 10
(-4100 cm-
1/(RT))
-
1
This gives a total multiphonon rate constant of 332 s and an
implied radiative rate constant of 1930 s .
This radiative rate constant is unusually large for Eu ( D0).
The D0 radiative rate constants reported for other systems fall
mainly in the range 100-700 s . Such a large implied radiative
rate constant might lead one to suspect that the residual rate
constant is, in fact, not completely radiative and that some
additional relaxation pathway has been overlooked. That there
is no contribution to D0 relaxation due to rapid energy migration
to (impurity) traps was established by measuring the rate
constant in 0.5% Eu:Gd(thd)3 at 208 K. This measurement gave
e
-
1
a
Radiative contribution to the rate constant. b Multiphonon contribu-
3
+ 5
tion to the rate constant due to interaction with C-H oscillators at the
R-position of the thd ligand. c Multiphonon contribution to the rate
constant due to interaction with C-H oscillators on the tert-butyl groups
of the thd ligand. d Temperature-dependent contribution to the rate
constant due to relaxation through a ligand-to-metal charge-transfer
state. See ref 8.
5
-1
5
that this large radiative rate is due to a small degree of charge-
transfer character in the D0 state.
5
-
1
wtotal ) 2312 s (see Figure 5), which is very similar to wtotal
2303 s measured for Eu(thd)3 at 207 K. Eu ( D0) f Eu -
D0) energy migration is extremely unlikely in the doped
Acknowledgment. T.C.S. and M.T.B. thank Martin Hulse
Creighton University) and David Hawkinson (University of
-1
3+ 5
3+
)
(
5
(
South Dakota) for suggesting a synthetic route to thd-d18.
Acknowledgment is made to the donors of the Petroleum
Research Fund, administered by the American Chemical Society
for partial support of this project. Additional support was
provided by the Research Corporation and the University of
South Dakota.
3+
system, since the Eu ions are, on average, too widely spaced.
-
1
It is concluded, then, that the residual 1930 s rate constant is
indeed radiative.
A possible explanation for the unusually fast radiative rate
may be related to the same phenomenon that causes very fast
8
nonradiative decay at elevated temperatures. If there is a low-
lying ligand-to-metal-charge-transfer state, as postulated in ref
References and Notes
5
8
, it is possible that the D0 state will have some small degree
(
(
1) Kropp, J. L.; Windsor, M. W. J. Chem. Phys. 1965, 42, 1599.
2) Lis, S.; Choppin, G. R. Mater. Chem. Phys. 1992, 31, 159.
of charge-transfer character. A small contribution from the
charge-transfer state to the predominantly 4f wave function of
(3) Horrocks, W. DeW.; Sudnick, D. R. J. Am. Chem. Soc. 1979, 101,
334.
5
5
7
D0 could significantly enhance the allowedness of D0 f FJ
transitions while having minimal effect on the position of the
(4) Dickins, R. S.; Parker, D.; de Souza, A. S.; Williams, J. A. G.
Chem. Commun. 1996, 697.
5
19
D0 energy level. This explanation is also consistent with the
(5) May, P. S.; Richardson, F. S. Chem. Phys. Lett. 1991, 179, 277.
(6) May, P. S.; Metcalf, D. H.; Richardson, F. S.; Carter, P. C.; Miller,
C. E. J. Lumin. 1992, 51, 249.
5
7
unusually strong relative intensity of the D0 f F0 emission
observed in Eu(thd)3.8
(
7) Haas, Y.; Stein, G. J. Chem. Phys. 1972, 76, 1093.
A summary of the contributions to the total rate constant for
(8) Berry, M. T.; May, P. S.; Xu, H. J. Phys. Chem. 1996, 100, 9216.
5
D0 relaxation in Eu(thd)3 is given in Table 2. These values
(9) Furniss, B. S.; Hannaford, A. J.; Smith, P. W. G.; Tatchell, A. R.
Vogel’s Textbook of Practical Organic Chemistry, 5th ed.; Longman
Scientific & Technical: New York, 1989; pp 527, 623, 669.
3
+ 5
confirm the expectation that the efficiency of Eu ( D0)
quenching by C-H vibrations is not as great as that by O-H
vibrations but illustrate that, in instances where the C-H
oscillators are closely associated, as in the CR-H, or numerous,
as in the tert-butyl C-H, the effect can be significant.
(
10) Fieser, L. F.; Fieser, M. Reagents for Organic Synthesis; Wiley &
Sons: New York, 1967; Vol. 1, pp 767-769.
11) Rappoport, Z. Handbook of Tables for Organic Compound
Identification, 3rd edition; CRC Press: Cleveland, OH, 1967; p 212.
12) Man, E. H.; Swamer, F. W.; Hauser, C. R. J. Am. Chem. Soc. 1951,
(
(
7
3, 901.
4
. Conclusion
Relaxation of the Eu ( D0) state in Eu(thd)3 at temperatures
(13) Eisentraut, K. J.; Sievers, R. E. J. Am. Chem. Soc. 1965, 87, 5254.
(
14) Mikami, M.; Nakagawa, I.; Shimanough, T. Spectrochim. Acta, Part
3+ 5
A 1967, 23, 1037.
(
15) Engleman, R.; Jortner, Mol. Phys. 1970, 18, 145.
above 270 K is dominated by a temperature-dependent mech-
anism that has been attributed to deactivation through a ligand-
to-metal charge-transfer state. The dominant relaxation mech-
anism at low temperatures is radiative, but multiphonon
relaxation via ligand C-H stretching vibrations is significant.
The 1930 s rate constant for radiative decay observed here is
(16) van Dijk, J. M. F.; Schuurmans, M. F. H. J. Chem. Phys. 1983,
7
8, 5317.
8
(17) Schuurmans, M. F. H.; van Dijk, J. M. F. Physica (Amsterdam)
1984, 12B, 131.
(
18) Reference 8 reports w(Te200K) ) 2120 ( 130 s-1, representing
the 95% confidence limit. The large error bars were due to baseline
anomalies in the digital oscilloscope used for that work.
(19) Blasse, G. Struct. Bonding 1976, 26, 43.
-
1
3+ 5
apparently the fastest reported for Eu ( D0). It is speculated