2630
J. Pbys. Cbem. 1980, 84, 2630-2634
is a particularly interesting aspect of the chemistry of these complexes. Note Added in Proof. Cox and Maas (Chem. Phys., 71, 330 (1980)) have reported observations similar to those described above. Acknowledgment. We thank Mr. J. Eberhardt and Mr.
R. B. Knott for the construction of the laser spectrometer. We also thank Mr. R. N. Whittem for his assistance in designing and constructing the equipment and for his helpful comments and criticisms and Mr. R. Brockman for his assistance in the construction of the UV-visible spectrophotometer.
References and Notes (1) J. C. Taylor, A. Ekstrom, and C. H. Randall, Inorg. Cbem., 17, 3285 (1978). (2) A. Ekstrom and C. H. Randall, J. Pbys. Cbem., 82, 2180 (1978). (3) A. Ekstrom, C. H. Randall, H. Loeh, J. C. Taylor, and L. Szego, Inorg. Nucl. Chem. Lett., 14, 301 (1977). (4) D. M Cox, R. B. Hall, J. A. Horsley, G. M Kramer, P. Rabinowltz,and A. Kaldor, Science, 205, 390 (1979). (5) A. Kaldor, R. B. Hall, D. M. Cox, J. A. Horsley, P. Rabinowltz, and 0. M. Kramer, J . Am. Cbem. Soc., 101, 4465 (1979). (6) A. Ekstromand C. H. Randall, submHted for publicationIn I m g . Cbem. (7) For a summary of work on the spectroscopy of the uranyl Ion see H. D. Burrows and T. J. Kemp, Chem. SOC. Rev., 3, 139 (1974). (8) The absorption cross sectlon is defined as E = 2.303ART/LPN, where A is the absorbance (log (I,,/I)),R = gas constant, T = absolute temperature, L = cell length (cm), P = sample pressure, and N = Avogadro’s number.
(9) M. H. Lietzke, US. Atomic Energy Commission Report No. ORNL 3259, Oak Ridge National Laboratory, Oak Ridge, TN, 1962. (10) H. Ogoshi and K. Nakamoto, J. Cbem. Phys., 45, 3113 (1966). (11) A. Ekstrom, H. Hurst, C. H. Randall, and H. Loeh, unpublished observation. The v3 vibrations of 180-labeledUOdHFA)?TMPwere found at 954 cm-‘(16/16), 938 cm-’ (16/18), and 908 cm- (18/18). These p o s b s agree very well with values calculated from normal coordinate ampiis of a linear triatomic molecule (ref 12). For similar Observations on ‘0-labeled UO,(HFA),THF see ref 4. (12) J. I.Steinfeld, ”Molecules and Radiation”, MIT Press, Cambridge, MA, 1978, p 181. (13) An approximate calculation (ref 12) gives a value of 841 cm-‘ for the v 1 frequency for U180‘80. A nonlinear least-squares treatment of the three v3 frequencies (16118, 18/18, 16/16)and the v, ( W l 8 frequency gives force constants of 7.35 m dyn/A and -0.24 m dyn/ for the uranyl band and the uranyl-uranyl band interaction. (14) The absorptlon cross section as determined from the peak helght of this band was found to be significantly temperature dependent. Thus a similar experiment at 210 OC again gave an excellent linear plot of absorbance vs. dlmer pressure, but resulted in an absorption cross section of 2.2 X lo-’’ cm2/dlmer molecule. This decrease in cross sectlon was accompanied by considerable broadening of the absorption band, particularly on the low-energy slde. (15) R. Colton, R. Levitus, and G. Wilklnson, Nature(London), 186, 233 (1960). (16) G. J. Bullen, R. Mason, and P. Pauling, Inorg. Chem., 4, 1145 (1965). (17) C. S. Erasmus and J. C. A. Boyen, Acta Crystallogr.,Sect. 6 , 26, 1843 (1970). (18) The simpllfled analysis examined the symmetry of the vibrational frequencies for extensions of the linear uranyl bond, for the followlng symmetries of the postulated dlmer species-(I) D,,, (il) C Z h ,(lli) C2,,,(iv) C , , f o examine whether they were Infrared allowed and to determine the effect of isotoplc substltution. (19) G. H. Dieke and A. B. F. Duncan, “Spectroscopic Propertles of Uranium Compounds”, McGraw-Hill, New York, 1949, p 36.
d
Luminescence Spectra of the Uranyl Ion in Two Geometrically Similar Coordination Environments. Uranyl Nitrate Hexahydrate and Di-p-aquo-bis[dioxobis(nitrato)uranium(VI)] Diimidazole Harry G. Brittaln* Department of Chemistry, Seton Hail University, South Orange, New Jersey 07079
and Dale L. Perry*’ Earth Sciences Division, Lawrence Berkeley Laboratory, University of California, Berkeley, California 94 720 (Received: March 7, 1980)
The luminescence spectra of the uranyl ion have been obtained at room and liquid-nitrogen temperatures in the crystal hosts of uranyl nitrate hexahydrate (UNH) and di-p-aquo-bis[dioxobis(nitrato)uranium(VI)]diimidazole (UNI). In both coordination spheres, the uranyl ion lies at the center of similar, distorted coordination hexagons consisting of two bidentate nitrate groups and two water molecules; the only difference in the coordination geometries is that the water molecules are terminal in the UNH complex and bridging in the UNI complex. The uranium-uranium “bond” distance in the UNI complex is 3.93 A. At room temperature, the emission spectra of the two compounds are essentially identical, but significant differencesappear upon cooling to 77 K. Vibronic structure is observed in the crystal of the UNI complex but not in the crystal of uranyl nitrate hexahydrate; this implies that the geometry of the uranyl ion in the two excited states is somewhat different. An energy level sequence is presented in which the various emission lines arise from a slightly split excited state (splitting approximately 80-85 cm-l) to several vibrational levels of the ground electronic manifold. The energy spacing of the ground vibrational levels (approximately 860 cm-’) was found to vary when changing crystal systems.
Luminescence Spectra of the Uranyl Ion
The Journal of PhVsfcal Chemistry, Vol. 84, hb. 20. 1980 2631 Uranyl Bonds
U - 05 = 1.77 U -06= 1.75 LOgUOe= 179.1'
01
v
o
7
W A V E L E N G T H lnrn) Flguo 3. Roomtenperalum lumhescenca spectra of me uranyl ion in the UNH and UNI complexes. The Intensity scales are wmpleteiy
arbitrary and are not comparable.
Experimental Section Single crystals of uranyl nitrate hexahydrate (Baker Chemicals) were grown by slow evaporation from an aqueous solution, and the crystals of di-r-aquo-bis[dioxobis(nitrato)uranium(VI)] diimidazole were the same as those used in the single-crystal X-ray diffraction structural study? Luminescence spectra were recorded on a high-resolution emission spectrometer constructed in the laboratory of one of the authors (H.G.B). The 365-nm output of a 200-W Hg-Xe arc Lamp was selected by a combination of a 0.1-m grating monochromator (Model H-10-UV-V, Instrument SA) and a UV-transmitting "black glass" filter Flpuro 2. The coordination sphere of d l - ~ - a q u ~ b l s [ d l o x o b l ~ n ~ a t o and ~ then focused onto the sample crystals. For low-temuranlum(VII] dllmidazole (UNII. For the sake of clarity, the structural perature work, the crystals were contained in a Suprasil details of me imidazole (im) molecules that are hyarogsn bonded to quartz tube and immersed in liquid nitrogen. The emission h e orldglng water ligands have been omitted. was collected at 90° to the exciting beam, analyzed by a has no well-defined S and A, and has a quantum number 0.5-m grating monochromator (Model 1870, Spex Indusof R = 4 or 1. tries), and detected by an EM1 9798B pbotomulti lier tube 6 - 2 0 response). An emission bandpass of 15 yielded Perry and co-workers5have recently reported the synmaximum resolution at room temperature, but at liquidthesis and crystal structure of di-rr-aquo-bis[dioxobis(ninitrogen temperature, a bandwidth of 5 A was necessary trato)uranium(VI)] diimidazole (UNI), a water-bridged to achieve optimum resolution. No attempt was made to dimer of uranyl nitrate. The central coordination sphere correct the emission spectra for monochromator or phoin this dimer consists of two bidentate nitrate groups in tomultiplier response. Excitation at wavelengths other addition to the two bridging water molecules that occupy the fifth and sixth equatorial coordination sites of the two than 365 nm did not lead to any differences in the luminescence spectra, except for intensity changes that were uranyl ions. The uranium atoms in the dimer are sepaconsistent with variations in the degree to which the exrated by a distance of 3.93 A, and each water molecule in turn has one imidazole unit hydrogen bonded to it. The citation energy was absorbed. coordination sphere of donor ligands about the uranyl ion Results and Discussion in this complex is identical with rhat present in uranyl At room temperature. the luminescence spectra of U" nitrate hexahydrate (UNHP except that the water moleand UNI are almost identical, as is shown in Figure 3. The cules in the latter complex are terminally bonded. The uranyl emission recorded for both systems closely resemuranyl ion in both complexes lies at the center of a disbles the spectrum obtained for the uranyl ion in water: torted hexagon of oxygen donor atoms associated with the nitrate and water groups (see Figures 1 and 2). which indicates that all emission lines found in the crystal originate from the uranyl ion alone. No luminescence is In the present work, single crystal luminescence spectra ohsewed in the UNI crystal spectrum that can be assigned have been obtained for uranyl nitrate hexahydrate and to imidazole. For both crystal systems, a total of four di-rr-aquo-bis[dioxobis(nitrato)uranium(VI)] diimidazole luminescence bands are observed, and the wavelength at room and liquid-nitrogen temperatures. The low-temmaxima of these bands correspond closely in the UNI and perature, high-resolution spectra have enabled an experUNH systems. Three of the bands are located at the same imental mapping of energy levels. and the close correwavelength (510,533, and 558 nm), while the band at the spondence of crystal strucrures allows certain conclusions highest energy is found to exhibit a slight crystal system to be drawn regarding the influence of both the water and dependence (488 nm for the UNH complex and 491 nm hydrogen-bonded imidazole molecules on the energy levels for the UNI complex). of the uranyl ion in the two complexes.
A:.
2632
The Journal of Physical Chemistty, Vol. 84, No. 20, 1980
Brittain and Perry
Ln v,
-
z W
1
I
I
1
480
485
490
495
WAVELENGTH (nm) Figure 4. Luminescence at 77 K for the UNI and UNH complexes; this emission Is from band system A. The intensity scales are arbitrary.
WAVELENGTH (nm) Figure 6. Luminescence at 77 K for the UNI and UNH complexes; this emission is from band system C. The intensity scales are arbitrary.
TABLE I: Wavelengths and Energies of the Luminescence Bands of Uranyl Nitrate Hexahydrate (UNH) at 77 K wavelength band maximum, i i energy, cm-' A 4855 20 597 4875 sh 20 515 B 5068 19 732 19 650 5089 C 5301 18 864 5324 18 783
WAVELENGTH (nm) Figure 5. Luminescence at 77 K for the UNI and UNH complexes; this emission is from band system B. While the intensity scales are arbitrary, the UNI peak at 506.8 nm has been reduced by a factor of 3.
To clarify these spectral differences further, low-temperature emission studies have been conducted. As would be expected, the emission intensity increases, and each feature at room temperature splits into at least two components. The wavelength maximum of each band shifts toward higher energy, implying that the room-temperature emission consists primarily of transitions containing a considerable vibronic component. A t liquid-nitrogen temperature, the three bands at highest energy persist, but the last band at 558 nm completely disappears. The
emission spectra for the three remaining bands (which are labeled A, B, and C in order of decreasing energy) are shown in Figures 4-6 for both the UNI and UNH crystal systems. The difference in the environment about the two types of uranyl ions in the UNI and UNH crystals is clearly apparent in the low-temperature spectra. An examination of Figures 1 and 2 reveals the close correspondence between the crystal structures, but the emission spectra reveal slight differences in wavelength maxima, and it would appear that vibronic phonon structure is present in the spectrum of the UNI complex while being absent in the spectrum of the UNH complex. Positions of all emission bands are found in Table I, while wavelength maxima and transition energies for the UNI spectral bands are shown in Table 11. A comparison of emission intensities associated with the UNI and UNH complex crystals is difficult because of the problem of attaining similar geometrical arrangements of the crystals, but it would appear that the uranyl emission in the UNI compound is approximately three times as intense as that in UNH. The detailed nature of the uranyl emission in the UNI crystal permits the assignment of all peaks in the A, B, and C systems. An examination of the energies of the emission lines leads to the conclusion that two electronic transitions are present in each band system; the remaining peaks are due to coupling of lattice phonons with these electronic
The Journal of Physical Chemistry, Vol. 84, No. 20, 1980
Lurnlnescence Spectra of the Uranyl Ion
TABLE 11: Wavelength and Energies of the Luminescence Bands of Di-M aquo-bis[ dioxobis(nitraio )uranium (VI)] Diimidazole (UNI)at 77 K band wavelength energy, system maximum, A cm-' assignment
-
A
B
C
4813 4829 4853 4873 4894 4912 4940 4955 5020 5036 5063 5085 5110 5129 5158 5178 5246 5263 5292 5315 5340 5364
20 777 20 708 20 606 20 520 20 433 20 358 20 243 20 182 19 920 19 857 19 751 19 666 19 570 19 497 19 387 19 312 19 062 19 001 18 896 18 813 18 727 18 643
2633
20.6 20.5
v, = 0
0 u,= 1 v, = 1 v,= 2 u,= 2 VI= 3 u,= 3 u,= 0 u,= 0 VI= 1 u,= 1 VI= 2 u,= 2 u, = 3 u, = 3 VI= 0 v,= 0 u,= 1 uz= 1 u, = 2 u, = 2 VI=
transitions. For each band system, two weak origins are noted at highest energy and a series of more intense peaks follows. It is iti general feature that the origin at highest energy has vibrational lines occurring every 170 cm-', while the other origin has vibrational spacings of approximately 190 cm-l. These vibrational energy spacings appear to remain relatively constant for corresponding origins in all emission band systems. Assignments for the vibronic structure are given in Table 11. It is highly interesting to note that no vibronic structure is present in ,the spectrum of the UNH compound. A comparison of the UNI and UNH transition energies would imply that all UNH emission bands actually correspond to = 1trmisitiona and that these are not very intense origins. Given the close structural similarity of the two uranyl complexes, this conclusion would appear resonable. A search at high sensitivity for the electronic origins corresponding to the u = 0 emission reveals nothing at 77 K. It is generally accepted that the absence of vibronic structure in solid-state luminescence spectroscopy indicates that the geometry (Le., bond lengths) and angles) of the excited state in essentially the same as that of the ground state. Conversely, a long progression indicates a large difference in the geometries. The absence of any vibronic structure for UNH implies that no geometry changes take place upon excitation, but the medium progression seen for UNI appears to indicate that the geometry of the excited uranyl ion in this crystal system changes upon excitation. Two other trends may be outlined from a consideration of the data in Tables I and 11. The energy separation between corresponding u1,2 lines in both the UNI and UNH emissions for tlhe A, B,and C band systems is very nearly constant for each type of crystal. In the case of the UNH compound, this energy separation is found to average 866 cm-l, while the separation is 853 cm-' for the UNI complex. It is well-known that the totally symmetric vibration of the uranyl ion has an energy of 856 cm-1,7 so it would appear that the A, B, and C emission systems correspond to emission from the excited state to several vibrational states of the ground-state manifold. The observation that the energy of this totally symmetric vibration is different in the UNH and UNI complex crystals is another indica-
h
Y Y v
>. c3 L11 W
z W
1.72
0.86
0
Flgure 7. Energy level diagram relating the luminescence peaks at 77 K to the energles of the various electronic states.
tion of some difference in the uranyl environments in the two systems. Another interesting feature observed is that two electronic transitions are found in each band. For both the UNI and UNH crystal systems, the energy separation between the components having the same value of u within each band system is essentially the same. For the UNH compound, this separation is 82 cm-l, while the separation in the UNI complex is 85 cm-l. This difference between UNI and UNH is probably not significant, and the authors believe that the same mechanism operates in both crystal systems. All of these conclusions enable the determination of an empirical energy level diagram relating the uranyl ion luminesence in both crystal systems,'and this diagram is shown in Figure 7 . The A, B, and C emission bands are observed to arise from emission of a pair of almost-degenerate excited states to the vibrational levels within the ground state. The lattice vibrations couple with each of the six transitions shown, and the observation that the phonon energy is the same for each transition band indicates that the excited state is split by the 80-cm-' value (another possibility that might be considered is that each ground vibrational level is split by 80 cm-l but has been rejected because of the phonon energy being so constant). The small differences that are detected in the energy levels of the uranyl ion when it is placed in the UNH and UNI compounds demonstrate the sensitivity of these levels to the coordination environment. The emitting state is almost certainly off orbital in origin: and the influence on these orbitals by chemical bonding is evident. Given the identical coordination donor set of ligands about the uranyl ion in both cases, the difference in the electronic properties of the uranyl species in the two crystal systems is most likely due to some long-range effect. In the UNI complex, two uranyl ions are bridged by water ligands with imidazole molecules hydrogen bonded to them, and it is possible that the observed effects are due to weak coupling between the two uranium(V1) metal centersS8It is equally possible that there exists a long-range effect of the hy-
2634
J. Phys. Chem. 1980, 84, 2634-2641
drogen-bonded imidazole molecules on the uranyl ion in the UNI crystal, and a similar effect of the noncoordinated lattice water molecules in the UNH complex can also be postulated. The inherently greater luminescence associated with the UNI system relative to that of UNH does suggest that some amount of energy transfer may be occurring. At the present time, it is not possible to partition the amounts of the contributions from these several potential effects. Acknowledgment. This work was supported by a grant from the Research Corporation to H.G.B. through the Cottrell Research Program (No. 8926). One of the authors (D.L.P.) wishes to thank the Miller Institute for Basic Research, University of California, Berkeley, for a Miller Research Fellowship which sponsored the initial synthetic portion of this research. The authors also thank Professor Paul N. Schatz for a most helpful discussion and the Department of Energy for support of this work under Contract No. W-7405-ENG-48.
References and Notes (1) National Science Foundation Postdoctoral Fellow, 1976-77; Miller
Fellow, 1977-79. Rabinowitch, E.; Belford, R. L. ”Spectroscopy and Photochemistry of Uranyl Compounds”; Pergamon Press: Oxford, 1964. Burrows, H. D.; Kemp, T. J. Chem. SOC. Rev. 1974, 3 , 139. Jorgensen, C. K.; Reisfeld, R. Chem. Phys. Lett. 1975, 35, 441. Perry, D. L.; Ruben, H.; Templeton, D. H.; Zalkin, A. Inorg. Chem. 1980, 79, 1067. Taylor, J. C.; Mueller, M. H. Acta Crystallogr. 1965, 79, 536. (a) Conn, G. K. T.; Wu, G. K. Trans. Faraday SOC.1938, 34, 1483. (b) Jones L. H.; Penneman, R. A. J . Chem. Phys. 1953, 21, 542. (c) Jones L. H. J . Chem. Phys. 1955, 23, 2105. The uranlum-uranium bond distance of 3.927 A in di-p-aquo-bis[doxobis(nitrato)uanium(VI)] dllmidazde is the shwtest reported length between two uanyl ion centers, and it thus affordsthe best opportunliy to date for detecting metal-metal coupllng in the UO ’; system. Two other bridging uranyl dlmers (which possess bridging hydroxy groups instead of water) that have quite similar bonding are the [(NO&U02(OH)2U02(H20)3]~Hz0e and CI(Hz0)3U02(OH)~U0z(H,0)3CI’o complexes which have uranlum-uranium distances of 3.939 and 3.944 A, respectively. Perrin, A. Acta Crystallorgr.,Sect. B 1976, 32, 1658. Aberg, M. Acta Chem. Scand. 1969, 23, 791.
Optical and Magnetic Properties of Substituted Benzophenones with Lowest %7r* States G. P. M. van der Velden, E. de Boer,* and W. S. Veeman Department of Physical chemistry, University of Nvmegen, Toernoolveld, 6525 Ed Nvmegen, The Netherlands (Received: February 25, 1980)
Phosphorescence and optically detected magnetic resonance (ODMR) experiments are reported for the lowest excited triplet state of some benzophenone derivatives, Le., 4-benzoylbiphenyl, 4,4’-diphenylbenzophenone, and 4,4’-dibenzoylbiphenyl. Moreover, we studied two compounds similar in structure, l-benzoylnaphthalene and 1,5-dibenzoylnaphthalene. All of these compounds have been studied in single-crystal form and have a lowest m* triplet state. The contribution of the magnetic dipole-dipole interactions to the zero-field splitting is determined and compared with data for benzaldehydes and acetophenones. An isotope effect has been seen in the ODMR spectrum of the triplet state of 4,4’-bis(4-~hlorophenyl)benzophenone, which appeared to be present in 4,4’-diphenylbenzophenone.
Introduction In the present study benzophenone-like systems have been investigated for which it was expected that the character of the lowest triplet state would be no longer, as in benzophenone, nr*. The state character is partly determined by the nature of the substituent attached to benzophenone (BP). Substituents which promote rr* lowest triplet states are amino, diethylamino, dimethylamino, and phenyl. We have chosen phenyl groups containing substituents because amino-, (dimethylamino)-,and (diethy1amino)benzophenone are too sensitive for solvent effects, display multiple emission,l and are difficult to purify, and moreover no crystallographic data are available for these compounds at present. The phenyl-containing groups have been substituted solely at the para position(s) of BP. Examples are 4-benzoylbiphenyl (BB), 4,4‘-diphenylbenzophenone (DBP), and 4,4’-dibenzoylbiphenyl (DB). Moreover, we have studied two compounds similar in structure to BB and DB except that the biphenyl group has been replaced by a naphthyl group. These two compounds are l-benzoylnaphthalene (1-BN) and 1,5-dibenzoylnaphthalene (1,5-DBN). All mentioned structures have been depicted in Figure 1,together with the structure of 2-benzoylnaphthalene (2-BN). The triplet state of BB has been studied in the gas phase,2in solution at room temperat~re,~ and in matrices 0022-3654/80/2084-2634$0 1.OO/O
at 77 KS4 Its naphthyl analogue (1-BN) has also been studied frequently in matrices at 77 K.516 The crystal structures of the compounds considered are unknown, except those of DB and 1,5-DBN which have been solved recently in the crystallographic department of our f a c ~ l t y . ’ ~ Hardly ~ ~ ~ any spectroscopic data are available for these two compounds? The systems, BB, DB, 1-BN, and 1,5-DBN, have been studied by us in neat crystal form. 4,4’-Bis(4-~hlorophenyl)benzophenone (CBP) or its monochloro analogue appeared to be present as a radiative impurity in 4,4’-diphenylbenzophenone.Phosphorescence and ODMR experiments carried out on extensively zonerefined material clearly indicate that the emission belongs to CBP, which apparently cannot be removed by zone melting. The above-mentioned systems have been investigated with the help of phosphorescence emission (PE), ODMR, and MIDP experiments. We have found that the To states of these five systems are best described as a?r* states. The contribution of the magnetic dipole-dipole interaction (&) to the zero-field splitting (ZFS) has been determined by calculating the spin-orbit coupling (SOC) contribution. Subsequently the values of Dss have been compared with data for benzaldehydes and acetophenones. It was found that a description of DB as a “double 0 1980 American Chemical Society
