Proc. Indian Acad. Sci. (Chem. Sci.), Vol. 93, No. 3, April 1984, pp. 283-293, 9 Printed in India.
Structural research on non-transition metal double nitrites: Crystal structure, optical absorption and emission of TIBa2(NO2)5
G FAVA GASPARRI, M NARDELLI* and F F E R M I t
Istituti di Chimica Generale ed Inorganica e di Strutturistica Chimica dell'Universita di Parma. via M. D'Azeglio, 85-43100 Parma, Italy
t Istituto di Fisica della Universita' di Parma, Gruppo Nazionale di Struttura della Materia del C.N.R. ~ia M. D'Azeglio 85, 43100 Parma, Italy
Abstract. The thallium-barium double nitrite, TIBa2(NO2)5, is pyroelectric in the 77-600 K range and crystallizes in the Pca21 space group. The lattice constants at 293 K are:
a = 17.868(12), b =4.934(3), c = 13.426(11) A (MoK~, 2 = 0.71069A). There are four stoichiometric units in the unit cell of volume V = 1184(1) A 3 (Do=3-98, D x
= 3.979 Mg m-a), F (000)= 1232, /z = 20-36 m m - i . The crystal structure was solved by Patterson and Fourier methods and refined by least-squares to a final conventional agreement index R = 0.053 for 1371 independent reflections collected in a 0 range of 3-30 ~ using MoKcc radiation. There are two independent barium atoms surrounded by NO2 groups, both with coordination number 10 and distances in the ranges Ba-O = 2.69(4)-3-18(4) A and Ba-N
= 3.01(4)-3.18(4) A. The environment of thallium is clearly affected by the lone-pair stereoactivity and involves 12 TI-O and TI-N contacts less than 3"5A, but only four T1-O distances are shorter than 3 A (min. 2.76(2), max. 2.85(3) A), with a pyramidal coordination and thallium at the apex of the pyramid. All these coordination polyhedra are joined in chains running along the shortest lattice vector [010].
The single crystal electronic spectra, studied in absorption with polarized light and in photostimulated emission, are interpreted as due to transitions involving NO2 electronic levels perturbed by TI~-, whose spin-orbit interaction makes probable also the forbidden singlet-triplet transitions, in agreement with the interpretative picture given for post-transition metal nitrites.
Keywords. Thallium-barium double nitrite; crystal structure; optical absorption; emission;
single crystal electronic spectra.
1. Introduction
The crystal structure of TIBa2(NO2)5 has been determined as part of a research programme on non-transition metal double nitrites in progress at our laboratories with the aim of giving the structural information necessary for the interpretation of single crystal physical properties. Previous research in this field includes the crystal structure analysis of K2Pb(NO2)3 ( N O 3 ) ' H 2 0 (Nardelli and Pelizzi 1980), a term of the potassium and lead nitrite-nitrate variable composition series (Nardelli et al 1955), whose spectroscopic properties can be interpreted on the basis of the interactions Pb(li) exe~ ts on the nitrite ions allowing electronic transitions otherwise forbidden.
Crystals of non-transition metal nitrites exhibit interesting physical properties, like NaNO2, KNO2, NaAg(NO2)2 which are ferroelectric (Hellwege 1975) and Ba(NO2)2"H20 which shows a strong pyroelectric effect (Abrahams et al 1980).
The title compound was first prepared by Cuttica and Paciello (1922), who defined its
* To whom all correspondence should be addressed.
(Chem. Scio) - - 7
283
composition, and the morphology of the crystals was studied by Cavalca et al (1955), hereafter CNB, who found, from the ternary Ba(NO2)2-TINO2-H20 phase system at 25~ that no other double salt is formed.
In the present paper the crystal structure of TIBa2(NO2)5 is described which shows some interesting aspects concerning coordination of the metal ions, particularly TI(I) whose lone pair is found to be stereoactive. The single crystal electronic spectral properties, studied both in absorption and in photostimulated emission, can be interpreted on the basis of the electronic levels of N O 2 perturbed by the presence of T1 +, in agreement with the interpretative picture which Maria et al (1968) give for the nitrite salts of the post-transition-series metals.
2. Experimental 2.1 X-ray data
Crystals suitable for the x-ray study were from the preparation described by CNB. The sample used for the analysis was a fragment defined by the following cleavage faces (in parentheses are their distances in mm from an arbitrary point inside the crystal): _+ 100 (0"023), + 010 (0"032), _+ 001 (0-024), 110 (0"032), 011 (0"029), 101 (0-023), 111 (0.032).
All reflections with 0 ~< h ~< 26, 0 ~ k ~< 7, 0 ~< l ~< 20 in the range 0-07 ~< (sin 0)/2
~< 0.70A -1 were measured, by use of a Philips P W 1100 computer-controlled diffractometer, with graphite monochromatized MoKct radiation. Accurate unit cell parameters were obtained by least-squares from the angular coordinates of 20 reflections measured in a 0 range of 9-30 ~ Final lattice ~nstants, as well as other information concerning data collection and refinement, are listed in table 1. The new
Table 1. Data collection and processing parameters.
Stoichiometric unit Formula weight Space group Cell constants
Unit cell volume
Stoichiometric units per cell Observed density (pycnometer) Calculated density
Absorption coefficient
Max-rain absorption corrections Radiation (MoKct)
(000)
Temperature of measurement Scan speed
Scan width Total data collected
Independent data with I > 3a (I) Total variables
R = ~llfol--If, Hl~.lfol
Rw = [ywClFol-IFcD2/~WVoq ''~
s = I z w ( I r o l - I F ~ l ) ~ / ( n - m ) Y '2 Weights
Tl}~t2 (NO2)5 709.1 Pca21
a = 17.868(12) A b = 4-934(3) c = 13.426(11) V = 1184(1) A 3 Z = 4
D o = 3"98 Mg m - a Dx = 3.979 Mg m - 3
# = 20.36 ram-1 1.18-0.78 2 = 0.71069 A 1232 20~
frO5 degrees/sec A0 = (0.60 + 0.20 tan 0) ~ 2031
n = 1371 m = 162 0-0530 0.0666 0.6083
w = 1/(a 2 (F) + 0-005 F 2)
Crystal structure, optical absorption and emission of TIBa2(NO2)5 285 lattice p a r a m e t e r s c o m p a r e well with those published by CNB, the differences between the two sets o f c o n s t a n t s being o f the o r d e r o f 0-4 %.
The Laue g r o u p s y m m e t r y , D2h, a n d the systematic absences, hOl for h # 2n and hkO for k 4: 2n, confirmed the assignment o f CNa for the possible space g r o u p s D~]-Pcam and C 5,v-Pea21. Intensities were m e a s u r e d using the in-20 scan technique.
O n e standard reflection (412) was m o n i t o r e d every 30 min during the course o f the d a t a collection, and s h o w e d no significant deviations f r o m the initial measurements. T h e integrated intensities were corrected for Lorentz, polarization, a n d a b s o r p t i o n effects.
2.2 Solution and refinement of the structure
T h e interpretation o f the three-dimensional P a t t e r s o n function to localize the heaviest a t o m s was successful in the Pea21 acentric space group, in a g r e e m e n t with the indication o f the E-statistics which gives ( E 2 - 1 )h = 0.689. A pyroelectric effect, m e a s u r e d when the structure was already solved a n d refined, confirmed the assignment o f the acentric space group.
F r o m the c o o r d i n a t e s o f the two i n d e p e n d e n t b a r i u m a t o m s a n d thallium, deduced f r o m the P a t t e r s o n function (R = 12.38 %), a F o u r i e r m a p was o b t a i n e d revealing the positions o f all the o t h e r a t o m s (R = 10"56 ~o). S o m e difficulty arose in defining the position o f the O 9 - N 5 - O 10 ion, p r o b a b l y as a consequence o f s o m e kind o f disorder, as indicated by the final, exceptionally high t h e r m a l p a r a m e t e r s o f its a t o m s and their standard errors (see the Beq values o f table 2).
After correction for secondary extinction following Zachariasen (1963): g = 5.4(9) x 10- 7, the anisotropic least-squares refinement converged to the final values for R, Rw a n d S q u o t e d in table 1.
T o define the a b s o l u t e configuration o f the a t o m i c a r r a n g e m e n t in the structure, a complete refinement was carried out a s s u m i n g the c e n t r o s y m m e t r i c ~c~-~coordinates for
Table 2. Final positional parameters (x 105 for TI, Ba(1), Ba(2) and x 104 for the remainder) and equivalent isotropic thermal parameters Beq (AZ).
x/a y/b z/c B~
TI 38599 (7) 47656 (23) 41537 (14) 2.96 (3) Ba 1 40539 (6) - 524 (23) 9762 [14) 1.00 (2) Ba2 15550 (6) 14335 (2 1 ) 25000(-) 0.96 (2) N1 2529 (16) 1063 (42) 4984 (20) 2.0 (4) N2 330 (16) 5964 (60) 1425 (18) 2.8 (6) N3 2806 (13) 5513 (44) 2055 (18) 2.0 (5) N4 5006 (19) - 55 (73) 3454 (18) 3.2 (6) N5 1291 (19) -4091 (61) 4357 (32) 4.9(1.0) Ol 2609 (13) 1518 (54) 4032 (17) 4.2 (6) 02 3070(12) 1660(57) 5520(16) 3-7(6) 03 227 (12) 3480 (37) 1446 (21) 3.2 (6) 04 935 (11) 6788 (52) 1634 (21) 3.9 (7) 05 3397 (16) 4580 (51) 1848 (20) 3.6 (6) 06 2735 (12) 7992 (42) 2031 (18) 3-0 (5) 07 4451 (13) 508 (46) 2966 (13) 2.8 (5) 08 4921 (14) 485 (53) 4356 (15) 3.6 (6) 09 1495 (15) - 2977 (74) 3671 (30) 7.5 (1.1) O10 1041 (24) - 5929 (72) 4615 (34) 8-6 (1.4)
all the atoms. At the end of this refinement the values R = 0-0556, Rw = 0-0715, S = 0.6547 were obtained, and applying Hamilton's (1965) significance test, the ratio Rw(~yz)/R~ (xyz) was calculated as 1.074, while the theoretical value, for 1371 independent Fo's and 162 variables, at the half-percent confidence level, is 1.12o9,0.oos = 1.003. So the xyz atomic coordinates of table 2 correspond to the correct absolute configuration.
In the final refinement cycle the ratio o f maximum least-squares shift to error was 0-4 for the z coordinate of 09, and the maximum and minimum heights in the final difference Fourier synthesis were + 2"6 and - 3.9 e A - 3.
Atomic scattering factors for the neutral atoms were taken from Ibers and Hamilton (1974), as were the corrections for anomalous dispersion. Calculations were made on the CDC-CYBER 76 computer of the "Consorzio per la gestione del Centro di Calcolo Elettronico Interuniversitario dell'Italia Nord-Orientale"
(CINECA,
Casalecchio, Bologna, Italy) with the financial support of the University of Parma, and on the OOtJLD-SEL 77/22 computer of the "Centro di Studio per !a Strutturistica Diffrattometrica del CNR (Parma)", using the SHELX-76 program (Sheldrick 1976) for solution and refinement of the structure, SECEXT (Nardelli 1983a) for the secondary extinction correction, PAaSX (Nardelli 1983b) for the geometrical description o f the structure, PLUXO (Motherwell 1976) for the structural drawings.Tables of anisotropic thermal parameters and observed and calculated structure factors have been deposited with the Editor.
2.3. Thermal and optical spectral data
The measurements were performed on single crystals of T1BaE(NO2)5.
The thermal depolarization current was measured by putting the sample in a cryostat between two aluminium plates. The sample was cooled down to 77 K and then the temperature was increased at a constant rate (dT/dt = 0.1 K sec- 1) up to 600 K. The depolarization current was recorded by a CAaV 401 vibrating reed electrometer.
Polarized absorption spectra were taken at 293 K by means of a ZEISS POE UV microspectrophotometer whose bandpass was A2 = 4 nm.
The luminescence spectra were obtained by exciting with light of a high pressure Xe lamp (OSRAM XBO 150 W/1) filtered by means o f a 0.5 meter Jarrel-Ash monochromator.
The emitted light was analyze d, at right angle, by means o f an haS2 YOB!N-YVON monochromator, detected by an EMI 9558 QAM photomultiplier and processed by a picoammeter KEITHLEY 417. Excitation and emission bandpass were A).~ = 3 nm and A2~r ~ = 1"5 nm respectively.
3. Description of the structure
The structure can be described as essentially constituted by a set of metal coordination polyhedra formed and joined together by the N O 2 ions. There are three non- equivalent polyhedra: two involving barium, the third thallium. All these polyhedra form chains running along the shortest lattice repetition vector (figures 1, 2 and 3), as usually found in structures with chains o f coordination polyhedra. The projections of these polyhedra on the (010) plane are shown in figure 4 and the relevant bond distances and angles in them are quoted in table 3. The symmetry code listed in this table is used throughout the paper.
Crystal structure, optical absorption and emission of TlBa2(N02)5 287
Figure 2. Environment of Ba2, Environment of Bal.
Figure 1.
x • • 0
8m06 0 ' " , ~
N 3 ('I~ ',~, . ~ - ~ 0 4 i l
- ...(..)03i i
- - " ~ 08
9 ' YI
~ Z X
9 N I
0~
T~." , .5 90,o c/,/ . . . . ~:"-"~" " : , '"21
/ ,, '.', "-,,
,,,~,o~'~:
,' /aL--. ". .!' "'-kf-". ' '
V2,,'o~oV,~' ` . h r ' ', ; ,'
, OsV,ii N5 'v _ _'~;: . . .
i i
Figure 3. Environment of TI. Figure 4. Projection of the structure on the (010) plane.
Table 3. Selected interatomic distances (A) and angles (~').
Environment of Bal
Bal-O10 wi 2.74 (4) A [0.29 (3)] Bal-O3 Ix
Bal-O7 2.78 (2) [0.26 (1)] Bal-O5
Bal-O8 viii 2.85 (2) [0.22 (1)] Bal-O6 i
Bal-O5 i 3.12 (3) [0.12 (1)] Bal-N5 'v
BaI-N2 vi 3.10(3) [0.12(1)] Bal-N1 iv
Total valence around Ba(1): 1-96 (5)
O8viii-Bal-O7 125.2 (7) ~ O8viii-Ba I - N P v
NliV-Bal-O7 127.9 (6) O5-Bal-N2 vl
N2W-Ba 1-O 10 TM 72.1 (9) O 10vil-Ba 1-05
N5iv-Ba 1-O3 ix 85.1 (8) O3iX-Ba 1-O5 i
N51V-Ba 1-05 ~ 68"5 (8) O3iX-Bal-O6 i
N5iV-Bal-O6 i 88.1 (8) O5i-Bal-O6 i
Environment of Ba (2)
Ba2-O9 2"69 (4) A [0.33 (4)] Ba2-O2 iv
Ba2-O6 i 2'78 (2) [0.26 (1)] Ba2-O1
Ba2-O4 i 2-80 (3) [0.25 (2)] Ba2-O3
Ba2-O4 3.09 (3) [0-13 (1)] Ba2-O9 hj
Ba2-N3 3.07(2) [0.13(1)] Ba2-N4 v
Total valence around Ba (2): 2-05 (6)
O 1 -Ba2-O2 ~v 123-3 ( 7): O 1 -Ba2-N4 v
N4V-Ba2-O2 ~v 128.5 (7) O61-Ba2-O4 i
O4i-Ba2-O9 64-2 (9) O9-Ba2-O6 i
N3-Ba2-O9 tii 63'3 (8) N3-Ba2-O4
O9i"-Ba2-O4 55.4 (8) O9m-Ba2-O3
N3-Ba2-O3 105-6 (6) O3-Ba2-O4
Environment of TI.
TI-O1 2.76 (2)A [0.19 (1)] T1-O2
T1-O7 2"84(2) [0'16(1)] T1-O8
T1-O5 3.21 (3) [0-080(4)] TI-O8 iii
T1-O7 iii 3.42 (2) [0-050 (2)] T1-NI
T1-N4 3.28 (3) [0.070 (4)] T1-N3
TI-N4 m 3 41 (3) [0.050(3)] T1-N2 ,i
Total valence around TI: 1.18
O7-T1-O8 42-1 (7) '~ N2ii-T1-N4tiz
O1-T1-O2 46.0 (7) N2"-T1-N 1
O1-TI-O7 80.7 (7) N2ii-T1-N4
O2-T1-O8 82.3(7) N3-T1-N4 ~ii
N4i'i-T1-N4 95.1 (8) N3-T1-N4
N 1 -T1-N4 98.6 (7) N3-T I-N 1
Symmetry code:
none x, y, z (v) x - 1/2; y, z (i) x , y - l , z (vi) 1/2+x,l-y,z (ii) 1/2-x,y. 1/2+z (vii) 1/2-x,l+y,z-1/2 (iii) x, 1 + y , z (viii) 1 - x , y, z - 1/2 (iv) 1/2-x,y,z-l/2 (ix) 1/2+x,y,z
2-77 (2)A 2.82 (3) 2.91 (2)
3.Ol (4)
3.17(3)
2'74 (2)A 2.79 (2) 2.94 (2) 3.18(4) 3.12(3)
2.78 (2)A 2-85 (3) 341 (3) 3"20 (3) 3.41 (2) 3-43 (3)
[0-27 (1)]
[0.24 (2)]
[0-19(1)]
[0.15 (1)]
[0.10 (1)]
1 0 5 . 4 (6) ~
72 5 (7) 69.5 (9) 71.4 (6) 107-4 (6) 4O-O (7)
[0,29 (2)]
[0.26 (1)]
[0.18(1)]
[0-10(1l]
[0-12(1)]
107-4 (7) ~ 72.9 (6) 70.6 (91 68.1 (6)
85.o(7)
40-2 (6)
[oq9(1)]
[0.16(1t]
[0.050(3)]
[0.080 (5)]
[0.050(2)]
[o.050(4)]
82.1 (7) ~ 95.9 (6) 96.7 (7) 91-3 (6) 100-8 (6) 86.5 (6)
Bond valences (in square brackets) are given in valence units and are equal to (bond length/Ro) -N with Ro = 2.297 A, N = 7 for BaOI)-O and Ba(II)-N. Ro = 2.100 A, N = 6 for TI(I)-O and TI(I)-N (Brown and Wu 1976).
Crystal structure, optical absorption and emission of TlBa2(N02)5 289 The two Bal and Ba2 coordination polyhedra are similar and both correspond to a coordination number C.N. = 6 if only the distances less than 3 A are considered, or C.N. = 10 if this limit is raised to 3.5 A. The number o f the coordinating nitrogen atoms is different in the two cases: 3 for Bal and 2 for Ba2, but the orientation o f the ligands is essentially the same. The main difference involves the N5-O9-O10 group which is bridging two adjacent Bal ions through N5 and O10, while at Ba2 it bridges two adjacent ions only through 09. So around Bal there are two B a . . . N - O . . . Ba bridges, while around Ba2 there is only one. One more difference involves the number of the bridging oxygen atoms which is one (05) for Bal and two ( 0 4 and 09) for Ba2.
In both cases these oxygen bridges are asymmetric ( 0 5 . . . Bal: 2.82(3) and 3.12(3) A;
0 4 . . . Ba2: 2.80(3) and 3.09(3) A; 0 9 . . . Ba2: 2.69(4) and 3" 18(4) A). There is only one chelating N O 2 group in each barium polyhedron, as it is in Ba(NO2)2" H20.
Worth noticing is the orientation o f the N O 2 groups, surrounding the two barium ions, with respect to the (010) plane (figure 4), as they are in turn approximately perpendicular or parallel to this plane in both environments: the perpendicular ions are bridging bidentate along the chains, while the parallel ones are monodentate.
If bond distances in the two barium polyhedra are considered, not significantly different values are obtained for the weighted averages (L = O or N): (Bal-L)a,
= 2"90 (5), (Ba2-L)av = 2"86 (6) A, which agree quite well with the value, 2.87 A, o f the
"effective ionic radii" sum: 1.52 A, for Ba 2+ (C.N. = 10) and 1-35 A for 0 2- (C.N. = 3) (Shannon 1976). This agreement indicates that in these environments the approxima- tion can be made o f considering the metal-ligand distances regardless o f the nature (O or N) of the ligand atom. The bond valences calculated assuming for nitrogen the same parameters given by Brown and Wu (1976) for oxygen, are quoted in table 2. The agreement of the atomic valences, obtained by summing these bond valences at each atom, with the expected values confirms the proposed approximation.
Considering the thallium environment, the stereoactivity o f the lone pair appears quite evident. Indeed the most strictly bonded atoms are four oxygens at two N O 2 groups, and coordination shows a pyramidal geometry with thallium at the vertex of the pyramid. This arrangement agrees with what has been observed in other cases when the TI(I) lone pair is stereoactive (Brown and Faggiani 1980, and references therein).
The environment of thallium is completed by other longer contacts formed by other oxygen and nitrogen atoms producing chains of coordination polyhedra running along [010] (figure 3), and joining these chains with those formed by Bal and Ba2 (figure 4).
Probably the main cause of the lack o f centrosymmetry in this structure is just this
Table 4. Comparison of the average geometry of NO2 in T1Ba2(NO2)5 and in other compounds.
Compound R N-O O-N-O References
TIBa2(NO2)5 0.053 1-233(13) A 116.3(1.4) ~ Present paper Ba(NO2)2"H20 0.026-0-018 1.246(2) 114.2 (2) Abrahams et al (1980) K2Na [Co(NO2)6] 0.036 1.2316(9) 119.34(7) Ohba et al (1978) Li [(CH3),,N]2[CO(NO2)6] 0.065 1.237 (7) 117.8 (4) Nardelli and Pelizzi (1983) The values for N509010 have been omitted in the weighted averages.
asymmetry of the thallium environment which is responsible for the polar character of the coordination polyhedra chains.
Owing to the presence of the heavy Tl and Ba atoms and of some disorder affecting the N5-O9-O10 group, the accuracy o f the localization o f the N O ] groups is rather poor. Nevertheless, if the weighted averaged values are considered, a quite satisfactory description of the average geometry o f the N O 2 group is obtained, as shown by the data quoted in table 4 where these means are compared with the values found for the N O 2 ions in other compounds. These results assure that the considerations developed about the coordination o f these ions can be accepted at a quite good level o f confidence.
4. Thermal and optical spectral results
The thermal depolarization current shows a smooth path and some inversions of polarity. This behaviour cannot be explained simply, but its pyroelectric origin is certain. The reproducibility of the measurements was not good, however the current was rather intense and enabled us to estimate a pyroelectric coefficient that approached a maximum value o f 10- 8 C K - ~ cm -2
The polarized absorption spectra are shown in figure 5. These spectra were measured at 293 K with light propagating along the z axis of the crystal and polarized along the y or x axes (figure 3). The absorption spectra have the general features observed in non- transition metal nitrite salts (Maria et al 1968, and references therein). The rather intense peak at 2 = 460 nm must be attributed to the I A 1 -~ 3Bt transition o f the N O ~- ion allowed by the spin-orbit coupling due to the interacting Tl + cation. The broad band at shorter wavelengths is in the range o f the ~ A ~ -o ~ B~ and 1A 1 ~ ~ A 2 absorption o f the N O 2 , but it is quite different f r o m that observed for isolated N O 2 . This difference is certainly due to the contribution o f the interacting Tl + and Ba 2 + cations, however any speculation on this band is premature. The ~A~ ~ aB~ transition absorbs light polarized along the y crystal axis more efficiently. A simple calculation from the
350 400 450 500 550 600
WAVELENGTH ( n m )
Figure 5. Absorption spectrum of a single crystal of TIBa2(NO2)s observed at 293 K with light travelling along the crystal z axis. The light electric field is y-polarized (A) or x-polarized (o).
Crystal structure, optical absorption and emission o f T l B a 2 ( N 0 2 ) 5 291
~
!2
'l o
14 15 16 17 18 19 ~ /103cm -~
0 17 18 19 20 21 22 23 24 E/eV
Figure 6. Luminescence spectrum of a single crystal of TIBa2(NO2) 5 observed at 77 K for excitation in the singlet-triplet (~A~ --, 3B~) absorption band at 2 = 460 nm. (zx) or (o) refer to normalized spectra taken on two different samples.
A A
experimental results shows that It~Ix = ,-~ 1.2, Ig and I~ are the absorbed intensities polarized along the y and x axes of the crystal, respectively.
In figure 6 is shown the luminescence band observed at 77 K for excitation in the singlet- triplet absorption band (2ex = 460 nm). This band resembles that observed by Maria et al (1968) on powder o f TINO2 both in energy range and shape. The luminescence shows a sequence of peaks, not well resolved, whose energy separation of about 780 cm-1 fits nicely the deformation vibration energy o f N O 2 in the fundamental electronic state.
There are three facts that support the idea that the thallium cation and the surrounding N O s anions are responsible for the absorption and luminescence properties in the region of the singlet-triplet transition. They are: (i) no luminescence was observed with barium nitrite (Maria et a11968), (ii) the observation that Ba 2 § does not produce any significant increase in the intensity of the singlet-triplet transition (Sidman 1957; Reznik 1976), (iii) the strong spin-orbit coupling constant and the polarizability of the TI § cation. Moreover the features of the absorption band and the shape of the luminescence band suggest that the N O 2 ion, notwithstanding its interaction with TI § keeps its molecular peculiarities.
On the basis of these arguments we can try to obtain some information about the
A A
absorbing centres. This can be done by a comparison of the experimental ratio I r ~Ix, i.e. of the absorbed intensity y-polarized to that x-polarized, with the same ratio computed on the basis o f different, simple geometrical models. We considered four models, I, II, III, IV, whose absorption transition moments are supposed to lie (a local pseudo C2~, symmetry is assumed for the N O 2 ions):
I: parallel to the lines joining the five nitrogens N1, N2 ii, N3, N4, N4 iii of the nitrite ions, with the TI § cation (figure 3),
II: perpendicular to the planes of the five N O 2 ions,
III: parallel to the local pseudo 2-fold axes o f the five N O 2 ions,
IV: perpendicular to the pseudo 2-fold axes but in the planes of the N O ; ions.
Table 5. Absorbed intensities and their ratio for the IA~ --, aBz transition, with light polarized along the y and x crystal axes.
Model lff 1~ . ly ~Ix A ,4
I 0.91 1"41 0"64 II 1"70 0"42 4.05
III 1' 11 2-43 0"45
IV 1-27 I'01 1"26
To obtain the intensities I~ and i A, we add the squared cosines of the angles between the electric field vector of the light and the postulated directions of the transition moments. The anions O7-N4-O8 and 07iii-N4 iii -08 iii (see figure 3) contribute one half.
The results of these calculations, shown in table 5, indicate that models I, II and IlI give
,4 A
results quite different from the experimental value of the ratio l y / l x -- ,-- 1.2, while model IV is in good agreement.
The results of model I suggest that anion-cation charge transfer absorptions do not enhance the singlet-triplet absorption intensity. In fact in AgNO2 and NaAg(NO2)2 such a transition measured at about 2 -- 400 nm, has been considered as the possible source of the high singlet-triplet absorption intensity completely polarized along the axis joining the Ag + cations with the nitrogen of the NO~ anions (Reznik and Rumyantseva 1975).
Model II enables us to exclude the involvement of transitions perpendicular to the planes of the N O 2 ions, in agreement with the selection rules for a C2v symmetry. At the same time it confirms that the transition 1A l --* 3BI is the right one.
Model III tells us that transition moments along the pseudo 2-fold axes do not contribute to the absorption, but model IV points out that the prevailing transition absorption moments lie in the planes of the NO 2 's perpendicular to the pseudo 2-fold axes. Theoretically, the transitions involved in models III and IV are both allowed in a C2v symmetry. However, the singlet-triplet absorption intensity was found to be 90 ~o in NaNO2 (Hochstrasser and Marchetti 1969) as in model IV, while in Nal_~TIxNO2 with x < 0.01 the presence ofT1 + increased the activity of the spin-orbit function which enhances the transition moments in the direction considered in model IV (Reznik et al 1974). Theoretical calculations (Harris et al 1970) on NaNO2 and AgNO2 crystals were found also to be in agreement with these experimental results, supporting the acceptance of model IV.
Finally, we tried to introduce further refinement in the calculations by weighting the contributions of the various nitrites considering some overlapping (Mulliken et al 1949), but the effect on the calculated ratios was not meaningful.
From this analysis, even if the models considered here are the simplest ones, the results agree very well with the general picture for the optical spectral properties of the non-transition metal nitrite salts, showing a quite good correlation between structure and optical properties.
Acknowledgements
The authors are indebted to Professor G L Rossi and Professor R Capelletti for the use of the microspectrophotometer and iTc-apparatus respectively, and to Dr A Mozzarelli for his help in collecting absorption spectral data.
Crystal structure, optical absorption and emission of T1Ba2(N02)5 293 References
Abrahams S C, Bernstein J L and Liminga R 1980 J. Chem. Phys. 72 5857 Brown I D and Wu K K 1976 Acta Crystallogr. B32 1957
Brown I D and Faggiani R 1980 Acta Crystallogr. B36 1802
Cavalca L, Nardelli M and Braibanti A 1955 Gazz. Chim. Ital. 85 1544 Cuttica V and Paciello A 1922 Gazz. Chim. hal. 52 146
Hamilton W C 1965 Acta Crystallogr. 18 502
Harris L E, Maria H J and McGlynn S P 1970 Czech. J. Phys. B20 1007 Hochstrasser R M and Marchetti A P 1969 J. Chem. Phys..50 1727
Hellwege K H (ed.) 1975 Landolt-B6rnstein numerical data and functional relationships in science and technology (Berlin: Verk) Group III Vol. 9
Ibers J A and Hamilton W C (ed.) 1974 International Tables for X-ray crystallography (Birmingham: Kynoch Press) Vol. IV
Maria H J, Armstrong A T and McGlynn S P 1968 J. Chem. Phys. 48 4694 Motherwell W D S 1976 Pluto Cambridge Univ.
Mulliken R S, Rieke C A, Orloff G and Orloff H 1949 J. Chem. Phys. 17 1248 Nardelli M, Cavalca L and Braibanti A 1955 Gazz. Chim. Ital. 85 1535 Nardelli M and Pelizzi G 1980 Inorg. Chim. Acta 38 15
Nardelli M 1983a SECEXT: Programfo~, correction of secondary extinction Parma University
Nardelli M 1983b PARST: A system of computer routines for calculating molecular parameters from results of crystal structure analysis, University of Parma; 1988 7 95 Computers & Chemistry
Nardelli M and Pelizzi G 1983 Unpublished results
Ohba S, Toriumi K, Sato S and Saito Y 1978 Acta Crystallogr. B34 3535 Reznik L E 1976 Opt. Spectrosc. 40 164
Rezmk L E, Garber P R and Fesun A V 1974 Sot'. Phys. Solid State 15 2185 Reznik L E and Rumyantseva L M 1975 Opt. Spectrosc. 39 509
Shannon R D 1976 Acta Crystallogr. A32 751
Sheldrick G M 1976 SHELX-76: Pro~ram for crystal structure determination Cambridge University Sidman J W 1957 J. Am. (_'hem. Soc. 79 2669
Zachariasen W H 1963 Acta Crystallogr. 16 1139
