The luminescence of the rare earth cryptates [Tb ⊂ 2.2.1]3+ and [Sm ⊂ 2.2.1]3+

The luminescence of the rare earth cryptates [Tb ⊂ 2.2.1]3+ and [Sm ⊂ 2.2.1]3+

CHEMICAL PHYSICS LETTERS Volume 129, number 6 19 September 1986 THE LUMINESCENCE OF THE RARE EARTH CRYPTATES [Tb c 2.2.113+ AND [Sm c 2.2.1j3+ N. S...

405KB Sizes 2 Downloads 13 Views

CHEMICAL PHYSICS LETTERS

Volume 129, number 6

19 September 1986

THE LUMINESCENCE OF THE RARE EARTH CRYPTATES [Tb c 2.2.113+ AND [Sm c 2.2.1j3+ N. SABBATINI, S. DELLONTE Istituto Chimico “G. Ciamicran” deN’Universitb and Istituio FRAE-CNR,

Bologna, Ita&

and G. BLASSE Physical Laboratory,

State University Utrecht, Utrecht, The Netherlandr

Received 6 June 1986

The spectroscopic and photophysical properties of [Tb C 2.2.11 3* and [Sm C 2.2.11 3+ in aqueous solution and the solid state are reported. The role played by multiphonon emission and non-radiative deactivation via the excited configuration state in determining the observed luminescence properties is examined.

1. Introduction The spectroscopic and photophysical properties of the [EuC2.2.1J3+ cryptate (2.2.1 is the 4,7,13,16,21pentaoxa-I ,IOdiazabicyclo-8,8,5-tricosane cryptand [I]) have recently been investigated in aqueous solution [2] and the solid state [3]. Interestingly, it was found that in the complex the cryptand ligand partly shields the metal ion from interaction with the water molecules, but simultaneously introduces ligand-tometal charge-transfer states at low energy. As a consequence, in going from the aquo ion to the cryptate, the non-radiative process via coupling with OH oscillators [4,5] is strongly reduced, while an efficient nonradiative deactivation via the low-lying charge-transfer state [3,6] is introduced. It is well known [7-91 that for the Tb3+ ion the excited 4f75d configuration lies at high energies, so that it should not influence the luminescence properties. Thus, in the presence of H,O molecules in the coordination environment of Tb3+ ions, the vibronic coupling of the luminescent state with the high-energy OH vibrations is expected to be the only efficient path [4,5] for radiationless deactivation. Encapsulation of Tb3+ into the cryptand cavity should result in a high 0 009.2614/86/$ 03.50 OElsevier Science Publishers B.V. (North- Holland Physics Publishing Division)

value of the quantum efficiency and a long decay time. In the case of Sm3+ the charge-transfer state is at higher energies than in the case of Eu3+ [7]. This is also expected to reduce the radiationless decay rate via the charge-transfer state. Because of the small energy gap in the case of Sm3+ (~7000 cm-‘), the radiationless decay rate via the OH and l&and vibrations is expected to dominate the radiative decay. In this paper we report on the luminescence properties of the [Tb C 2.2.11 3+ and [Sm C 2.2.11 3* cryptates. Experiments have been performed in aqueous solution as well as in the solid state for the sake of comparison with the previous study on the analogous Eu3+ cryptate [2,3].

2. Experimental 2. I. Materials The 2.2.1 cryptates of Tb3+ and Sm3+ were prep,ared following the same rocedure described in ref. [2] for the analogous Eu z compound. For the Tb3+ 541

Volume 129, number 6

CHEMICALPHYSICS LETTERS

is [Tb C 22.11 Cl, -2H,O. Calculated for [Tb C 2.2.11 Cl,.2H20: Tb, 25.1; C, 30.3; H, 5.7; N, 4.4; Cl, 16.8; found: Tb, 25.2; C, 30.2; H, 5.5; N, 4.3; Cl, 17.0. The presence of two water molecules was confirmed for the case of Tt? by luminescence decay measurements (see below). For the Sm’+ cryptate the chemical composition is [Sm C 2.2.11 Cl,.2H20. Calculated for [Sm C 2.2.11 C13*2H20: Sm, 24.1; C, 30.7; H, 5.8; N, 4.5; Cl, 17.0; found: Sm, 24.0; C, 30.7; H, 5.8; N, 4.4; Cl, 16.9. cryptate the chemical composition

2.2. Measurements in solution Abortion spectra were recorded with a Kontron Uvicon 860 spectrophotometer and luminescence spectra with a Perkin-Elmer 65040 spectrofluorometer. Corrected excitation spectra were obtained on a Perky-Ever LS5 spectrofluorometer. L~~escence lifetimes were measured by a JK neodymium YAG System 2000 laser, exciting with the third harmonic at 355 nm or Lambda Physik M 1OOAnitrogen laser ‘(337 nm). The emission quantum yield for excitation at 370 nm was evaluated by the method described by Haas and Stein [lo] using as standard the Tb3+ aquo ion [ 11] . We verified that the emission quantum yield of Tbzi for excitation at 370 nm is the same as that reported by Stein for excitation at 490 nm [ 111, 2.3. M~~~ents

19 September 1986

the Tb3’ ion. At short wavelengths (near to the instrumental cut-off) there is an intense absorption band which is ascribed to the spin-allowed transition to the ‘D term of the excited 4f7 5d configuration. This band carries a tail down into the visible region. The solution as well as the solid show efficient green Tb3+ luminescence. Its excitation spectrum corresponds to the absorptionlreflection spectrum except for the fact that excitation into the tail does not yield Tb3+ emission. This tail is ascribed to a Eu3+ impurity present in the starting materials. It coincides with the Eu3’ charge-transfer band [2,3]. From the absorption spectrum the Eu3+ concentration is estimated to be -0.1 at%. The emission spectrum under low resolution presents the usual ‘D, + 7FJ (J = 3,4,5,6) transitions, while rio ‘D, emission was detected (fig. 1). The absence of ‘D3 emission can easily be explained by a rapid 5D3 + ‘D, radiatio~ess transition due to the high vibrational frequencies available in the surroundings of the Tb3+ ion (cryptand and water molecules). At 4.2 K the emission transitions are split. The crystal-field splittings are, however, only some 50% of the values observed characteristically in solid oxides, viz. 300 versus 600 cm-l [8]. This shows that the crystal

in the solid state

The luminescence spectra and the diffuse reflection spectrum of the solid cryptates were obtained using the instrumentation described in ref. 131.These measurements were performed down to liquid-helium temperature.

3. Results and discussion 3.1. [n, c 2.2.1J3+ The experiments in aqueous solution were carried out at their natural pH (m6.5) where the [Tb C 2.2.11 3+ complex is inert [12]. The absorption spectrum of the solution and the diffuse reflection spectrum of the solid show the weak sharp-line transitions within the 4f8 configuration of 542

h,nm

Fig. 1. Emission spectrum of [Tb c 2.2.11 3* in aqueous s&ttion (A,,, = 370 nm).

CHEMICAL PHYSICS LETTERS

Volume 129, number 6

field on the rare-earth ions in the cryptand cage is relatively weak, as observed before [2,3]. Fig. 1 shows that the intensity of the ’ D4 + 7 F6 emission line has some 50% of the intensity of the 5D4 + 7F, line. As argued elsewhere [8], this points to a strong deviation of inversion symmetry at the rare-earth site (compare ref. [2] and ref. [3]). The quantum yield of the Tb3+ luminescence was found to be 30% for excitation in the 4f8 transitions. This value was measured at 300 K on the solid as well as the solution. The estimated error is some 20%. Upon cooling the solid sample to 4.2 K, no considerable increase of the luminescence efficiency was observed. This quantum efficiency is more than one order of magnitude larger than for the Eu3+ cryptate. The reasons for the higher efficiency are obvious: (i) The excited configuration state in the case of Tb3+ is at much higher energy than in the case of Eu3+. This reduces the non-radiative decay via such a state to a negligible process. (ii) The gap between the emitting level and the highest ground-state level is larger in Tb3+ than in Eu3+ (15000 versus 12000 cm-‘), reducing the probability for multiphonon emission. Nevertheless it is of importance in the case of Tb3+ cryptate, the quantum efficiency being only 30%. The temperature independence of this value shows that we are in fact dealing with this process [3 1. In order to check this picture we measured the lifetimes of the luminescence. The decay curves appeared to be simply exponential. Our results are given in table 1. First we note the temperature independence of

Table 1 Lifetimes (in ms) of the [Tb cryptates

C

2.2.1J3+and

[Sm

C

C

2.2.113+

Physical state

T(K)

[Tb c 2.2.113+

[Sm

Hz0 solution

300 71 300 71 300 4.2

1.3 1.3 3.1 b) 3.1 1.3 1.5

0.005 a) 0.006 0.010 c) 0.010 0.007 d)

D20 solution solid

a) Compare Sm’+in H20: 0.002 ms [ll]. b, Compare TIJ” in D20: 3.88 ms [I1 1. C)Compare Sm3+inD20: 0.050 ms (111. d, Not measured because of instrumentation

limits.

2.2.113+

19 September 1986

the lifetimes. This confirms that we are dealing with multiphonon emission and excludes thermally activated non-radiative processes via the 4f7 5d excited states. Upon substitution of the reciprocal lifetimes obtained in H20 and D,O solutions in the empirical equation proposed by Horrocks [5] to estimate the number of metal-coordinated water molecules, one gets a value of 1.8 (estimated uncertainty -+0.5 [5 ) for the number of water molecules coordinated to Tb 3 + in the cryptate. The number of water molecules coordinated to Eu3+ in the analogous Eu3+ cryptate was found to be 3.2 [2]. A possible explanation for this difference could be found in the smaller ionic radius of Tb3+ compared to Eu3+ which could allow the ligand to contract slightly [ 131. As a consequence one water molecule, on the average, could be kept away from one of the three holes in the cryptate. The emission quantum yield in Hz0 solution, 0.30, indicates that [Tb C 2.2.11 3+ luminesces quite efficiently. The analogous value for the Tb3+ aquo ion is 0.08 [ll]. So, the encapsulation of Tb3+ into the cryptand cage results in a considerable enhancement of its luminescence. If we assume for the [Tb C 2.2.113+ cryptate the same radiative lifetime as for the Tb3+ aquo ion [4] and use that value together with the lifetime of the [Tb C 2.2.11 3+ cryptate in H20 solution to calculate the luminescence quantum yield, we get @ = 0.27, in good agreement with the value found experimentally. This indicates once again that no losses via higher states take place, The lifetimes at 300 K in H20 solution and the solid state are equal, as are the quantum efficiencies. This suggests that in both aggregation states the Tb3+ cryptate complex contains two water molecules. In the case of Eu3+ the cryptate complex in solution contains one molecule more. 3.2. (Sm C_2.2.113+ The Sm”+ cryptate shows only a very weak luminescence, independent of temperature, excitation wavelength and a regation state. The emission shows the usual red Sm9 + emission from the 4G,,2 level. Under whatever conditions, however, the quantum efficiency is considerable less than 1% and temperature-independent. The absorption spectrum of the solution, the diffuse reflection spectrum of the solid and the excitation 543

CHEMICAL PHYSICS LETTERS

Volume 129, number 6

400

300

500

A ( “In

Fig. 2. Absorption spectrum of a 1 X 10s2 M [Sm solution.

c

We assume that the total decay rate P is given by P = P, t P,(X,O) t P,(Cr), where P, is the radiative rate (=700 s-l), P,(Cr) the non-radiative rate for losses to cryptate vibrations and P,(X20) the nonradiative rate for losses to X20 (D20, H20) vibrations. In view of the data for water, we take P&120) = 25P,(D20 . In this way we estimate P,(D20) =3x103 s- 1 ,P,(H,0)=0.7X105 s-I and Prrr(Cr) = 10’ s-1. This shows that multiphonon emission via the D20 molecules is negligible and via the H20 molecules and the cryptand cage are of the same order of magnitude. The fact that the energy gap in the case of Sm3+ is smaller than in the case of Eu3+ is responsible for the importance of the cryptand vibrations in the case of radiationless decay of Sm3+ cryptate.

2.2.11 j+

spectrum of the luminescence of both show the wellknown Sm3+ transitions within the 4f5 configuration and, in the ultraviolet region, a broad, intense band corresponding to the tertiary amine + Sm3+ chargetransfer transition (fig. 2). Its maximum is situated at about 230 nm. This places the charge-transfer level some 9000 cm-l above that for the Eu3+ cryptate, in good agreement with the difference observed in other hosts [ 141. The absorption and reflection spectra show the same tail as mentioned above for Tb3+. Again it is ascribed to a Eu3+ rmpurity ’ in the starting materials. What is of importance here is the fact that the quantum efficiency of the Sm3+ luminescence is the same for charge transfer as for 4fS level excitation. Effects like those observed for Eu3+ cryptate [3] do not appear due to the high position of the excited configuration. This confirms the model proposed in ref. 131. In table 1 the lifetimes of the Sm3+ luminescence are tabulated for several conditions. The results discussed show clearly that the nonradiative loss in the Sm3+ cryptate must be ascribed to multiphonon emission, which explains the temperature independence of its luminescence properties. The short lifetimes in D,O solution show that the ligand vibrations must play a role in this non-radiative decay, especially if we note that the radiative decay time of Sm3+ in aqueous solution is 1.4 ms [ 111. We can estimate the several decay rates as follows. 544

19 September 1986

4. Conclusion The luminescence efficiency of Eu3+ cryptate is low due to the simultaneous appearance of non-radiative losses via the charge-transfer state and via highenergy vibrations (H20). The former can be overcome by using ions with the excited configuration at high energy, as, for example, Tb3+. However, as soon as the energy gap decreases relative to that of Eu3+, the efficiency is low anyhow, irrespective of the energy position of the excited configuration. This is clearly shown by the case of Sm3+ cryptate.

Acknowledgement The authors are grateful to Professor V. Balzani for discussions. GB wishes to thank Mr. G.J. Dirksen for measuring the solid samples of Sm3+ cryptate and Mr. M. Buys for measuring the solid samples of Tb3+ cryptate. He is also grateful for a grant from the ItalianDutch cultural exchange treaty.

References [l] J.M. Lehn, Struct. Bonding 16 (1973) 1. [2] N. Sabbatini, S. Dellonte, M. Ciano, A. Bonazzi and V. Balzani, Chem. Phys. Letters 107 (1984) 212. 13] G. Blasse, M. Buys and N. Sabbatini, Chem. Phys. Letters 124 (1986) 538.

Volume 129, number 6

CHEMICAL PHYSICS LETTERS

[4] G. Stein and E. Wurzberg, J. Chem. Phys. 62 (1975) 208. [S] W. Dew. Horrocks Jr. and D.R. Sudnick, Accounts Chem. Res. 14 (1981) 384. [6] G. Blasse, in: Handbook on the physics and chemistry of rare earths, Vol. 4, eds. K.A. Gschneider Jr. and L. Eyring (North-Holland, Amsterdam, 1979) ch. 34. [7] J.L. Ryan and C.K. Jtirgensen, J. Phys. Chem. 70 (1966) 2845. [8] G. Blasse and A. Bril, Philips Res. Rept. 22 (1967) 481.

19 September 1986

[9] W.T. Carnall, P.R. Fields and K. Rajnak, J. Chem. Phys. 49 (1968) 4412. [lo] Y. Haas and G. Stein, J. Phys. Chem. 75 (1971) 3668. [ll] G. Stein and E. Wurzberg, J. Chem. Phys. 62 (1975) 208. [12] O.A. Gansow and K.B. Triplett, US Patent 4,257,955 (March 24,1981); Chem. Abs. 94 (1981) 194446j. [13] J.M. Lehnand J.P. Sausage, J. Am. Chem. Sot. 97 (1975) 6700. [14] G. Blasse and A. Bril, Phys. Letters 23 (1966) 440.

545