• Aucun résultat trouvé

Chemical reactions and transport processes in lead iodide single crystals

N/A
N/A
Protected

Academic year: 2021

Partager "Chemical reactions and transport processes in lead iodide single crystals"

Copied!
5
0
0

Texte intégral

(1)

HAL Id: jpa-00220113

https://hal.archives-ouvertes.fr/jpa-00220113

Submitted on 1 Jan 1980

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Chemical reactions and transport processes in lead

iodide single crystals

T. Hagihara, K. Iwamoto, K. Fukumoto, N. Ayai

To cite this version:

(2)

JOURNAL DE PHYSIQUE Colloque C6, supplément au n° 7, Tome 41, Juillet 1980, page C6-297

Chemical reactions and transport processes in lead iodide single crystals

T. Hagihara, K. Iwamoto, K. Fukumoto and N. Ayai Department of Physics, Osaka Kyoiku University, Osaka, Japan

Résumé. — Le mouvement ionique anisotrope et les propriétés de dipôle (K + - V , ) , de l'ion du potassium et la

vacance de l'ion d'iode, ont été étudiés en mesurant la d.c. conductivité ionique, la perte diélectrique (DL) et le thermocourant ionique (ITC). Les énergies migratoires de la V,- par l'application du champ électrique paral-lèle (cas I) ou perpendiculaire (cas II) à l'axe c des cristaux se sont révélés être la même valeur 0,26 eV dans les échantillons dopés K.+ ions. D'autre part, l'énergie migratoire de la vacance des ions de plomb dans les cristaux

dopés telle que 0,58 eV dans le cas I et 0,35 eV dans le cas II. Le maximum de DL près de 1,3 x 103 Hz se sont

révélés être à 402 K. D'après les mesurages d'ITC deux courants maximums près de 167 et 185 K. ont été observés. Les résultats des mesures de DL et d'ITC ont été examinés en tenant compte des facteurs préexponentiels et de l'énergie d'activation pour le saut de V,- du dipôle. Il a été trouvé que le relâchement du dipôle se compose de deux degrés.

Abstract. — An anisotropic ionic motion and properties of potassium ion and iodine ion vacancy dipole ( K+- V | )

have been studied by measuring d.c. ionic conductivity, dielectric losses (DL) and ionic thermocurrent (ITC). The migration energy of V,- by the application of the electric field parallel (case I) or perpendicular (case II) to c-axis of crystals were found to be the same value of 0.26 eV in K+-doped crystals. On the other hand, the

migra-tion energy of Pb + + ion vacancy in Bi+ + +-doped crystals for cases I and II were 0.58 and 0.35 eV, respectively.

The DL peak near at 1.3 x 103 Hz was found at 402 K. From ITC measurements two current maxima at 167

and 185 K have been observed. The results from DL and ITC measurements were examined with respect to the preexponential factors and the activation energies for jump of V, of dipoles. It is found that the relaxation pro-cess of dipole is composed of two stages.

1. Introduction. — In the field of material science

the properties of lead iodide have been investigated extensively by many authors so far. There have, for example, been publications on the ionic conducti-vity [1, 2], photoconducticonducti-vity [3], thermostimulated current [4] and the structure and/or phase-transition of crystals [5, 6]. But, the transport processes and chemical reactions of point defects in Pbl2 crystals

have not been well established. Then, the informa-tion as to defect behaviour in Pbl2 crystals has been

becoming essential.

The purposes of the present report are (a) to study an anisotropic transport process of anion and cation vacancies and (b) to investigate the properties of potassium ion and iodine ion vacancy dipole by using the dielectric loss (DL) measurements as well as the ionic thermocurrent (ITC) techniques.

2. Experimental procedure. — A purification by vacuum distillation at about 10 "4 torr was carried

out at around 630 K. Single crystallizations were performed in iodine vapour by using the Bridgman method. The introduction of K+ or Bi+ + + ions

were done by adding an appropriate amount of K.I or Bil3, respectively, to purified Pbl2 powder. The

crystals of type 12 R belonging to a hexagonal space group were obtained, which do not change its

struc-ture up to near melting point. The crystal slabs of 10 x 10 x 0.5 mm3 for measurements were arranged

by cleaving with a thin blade (perpendicular to crys-tallographic c-axis) as well as by cutting with carbo-random disc blade cutter (parallel to crystallographic c-axis). The measurements for d.c. conductivity were done by using an electrometer having a sensitivity of 1.0 x 10~1 5A. For DL measurements a model

TR-1C dielectric loss measurement system which can coverall the frequency range from 30 Hz to 5 MHz was used. Moreover, to study the detailed behaviour of K+-V[- dipole the ITC were measured with a

specially designed cryostat and a model TR-84M vibrating reed electrometer whose current detection-limit was 1.0 x 10"1 7 A. The specimen was polarised

in a static electric field of 5.0 x 102 V/cm and then

cooled down to 120 K, after which the field was switched off. The specimen was warmed up to at the nearly constant rate of 0.1 K/s.

3. Results and discussion. — On the assumption that K+ and Bi+ + + ions in Pbl2 crystals are located

substitutionally to P b+ + ion sites, from the charge

neutrality requirement, following two chemical reac-tions are considered :

K I ^ KPV . + I " + V,- ,

(3)

C6-298 T. HAGIHARA, K . IWAMOTO, K . I.UKUMOT0 AND N. AYAl

and

2 B i 1 3 + 2 B i A t t

+

6 1 -

+

V , , - - ,

where K,',+ + and V1- mean the potassium ion at

lead ion site and the vacancy a t iodine ion site, res- pectively. Consequently, by doping with K + and Bi+ ++ ions we can examine the properties of cation

and anion vacancies as well as the impurity-vacancy dipole. To study an anisotropic transport of isolated I - and Pbi + ion vacancy in crystals d.c. conductivity

was measured for both cases where the applied electric field is parallel to c-axis (case I) and perpen- dicular to c-axis (case 11). As for K+-V,- dipole the measurements of DL and ITC were performed.

Fig 1. - The anisotropic d.c conductivity of PbI, singlc crystals where the electric field ( E ) is parallel to c-axis ( E 11 c) and perpen- dlcular to c-axis ( E I c).

Figure 1 shows the typical example for the d.c. conductivity of crystals as zone-refined as well as doped with K + ions. There exist two or three charac-

teristic regions. The activation energies calculated from the gradient of log oT vs. 1/T curves [7] are given in the figure. The conductivity for case I1 is about two orders of magnitude greater than that in case I. The regions exhibiting the largest activation energy in each curves, i.e., 1.40 and 1.49 eV, have been well known as an intrinsic region of conducti- vity of PbI, crystals since Seith [l] has reported his works, where both cation and anion contribute to

the conductivity. Furthermore, below 543 K a domi- nant charge carrier is I - ion according to the measu- rements of transport numbers by Tubandt [8], there- fore, the regions characterised hy 0.26 eV is due to the migration of I - ion vacancy. Therefore, it is considered that the region of 0.26 eV in case 1 refers to the mode-(a) in figure 2. In case 11, we can also

Fig. 2. - Thc anisotropic migration of iodine and lead ion vacancy in Pbl, crystals. Vacancies are shown by dotted circles.

find out the region labelled 0.26 eV in the corres- ponding temperature range, then, it may be concluded that V, - in case I1 move between the nearest neigh-

bouring iodine layers (ZIG-ZAG motion) as shown in figure 2(c). Both regions labelled 0.46 in case 1 and 0.61 eV in case I1 may be attributed to the migration of free V1- dissociated from dipole. The activation energies responsible for modeib) in figure 2 were not directly observed in d.c. conductivity measurements,. however, the energy can be evaluated to be 0.44 eV from the results of DL and ITC measurements as described below. On the other hand, d.c. conductivity as to the specimen doped with Bif + + ions are shown

in figure 3. The characteristic energies for both cases I

and I1 are I .29 and 0.58 eV and 0.35 eV, respecti- vely. The high temperature region labelled 1.29 eV for case I should be identified as that intrinsic in nature. By taking into account of the directions of the fields in cases I and I1 and the structure of crystal the regions labelled 0.58 and 0.35 eV may be refered to the mode-(d) and (e) in figure 2, respectively.

The frequency dependence of loss tgS of speci- mens doped with 0.30 and 2.7 mol% K + ions are shown in figure 4. In these curves remarkable humps are observed near at 3 x 10' and 2 x lo3 Hz. By raising temperature of specimens these humps shift to high frequency side, e.g., 1.26 kHz a t 402 K and 2.5 kHz at 423 K. The existence of the relaxation of Debye type at 1.25 kHz was confirmed by analysing the Cole-Cole plots [9]. From the Arrhenius plots of peak frequencies of loss t g 8 [lo] the activation energy (E,,) for the crystals doped with 0.30 and

(4)

CHEMICAL REACTIONS AND TKANSI'ORT I'Rc DCESSES IN LEAD IODIDE SlNGLE CRYSTALS C6-299

Fig. 3. - - The anisotropic d.c. conductivity of Pbl, crystals doped with bismuth ions.

Fig. 4. - The frequency dependence of the dielectric loss tg d For Pb12 crystals doped with potassium ions.

0.46 eV, which is closely related to the energy of V, -

that is required for jumping at another iodine ion site as shown in figure S(a). Moreover, the characte- ristic relaxation time 7, in T = 7, exp(Ea,/kT) was

calculated to be 9.2 x 10-lo s at 402 K [I I]. Never- theless, we could not detect any humps in dielectric

Fig. 5 . - The configurat~ons and relaxation processes of d~pole ( K ' - V , - ) in Pb12 crystals.

loss tg 6 in specimens for case

I1

and in the pure crystals. Another details as to dipoles were obtained from ITC measurements.

As shown in figure 6, the ITC in specimens doped with 0.30 m o l x K + ions was found to be composed

of two peaks appearing at 167 and 185 K. The peak at 167 K was uniquely observed under operating the polarisation at 168 K. The magnitudes of both peaks increased in proportion to the strength of the applied field up to 600 V/cm. Regarding two peaks, E,, and

7, for orientation of dipoles were 0.24

+

0.04 eV ;

2.4 x s (peak at 167K and 0.42

+

0.06eV; 5.2 x 10-1°s (peak a t 185 K). respectively [12, 131).

The activation energy and obtained from DL

are in good agreement with those from ITC peak appearing at 185 K. Hence, it is considered that this coincidence comes from the same relaxation process of dipoles. Therefore, we propose a tentative model as shown in figure 5(b). The ITC peak at 185 K may be caused by the relaxation of stage 2 while that for 167

K

may be attributed to the relaxation through s t a B 1. The process of stage 2 is substantially the same process as that supposed from DL measure-

EllC

2 0 I I I I I I I

130 150 170 190 210

T (to

(5)

C6- 300 T . HAGIHARA, K . IWAMOTO, K. FUKUMOTO A N D N . AYAI

ments as shown in figure 5(a). On the other hand, the been other reports to compare with the present relaxation in stage 1 may be justified by considering results.

the migration energy, 0.26 eV, of V,- moving between

the nearest neighbouring iodine Iayers as shown in Acknowledgment. - The authors are very much figure 2(a). No discussion about the relaxation time indebted to Prof. Dr. M. Ikeya for making a specially with respect to stage 1 is given since there has not designed cryostat for ITC measurements.

DISCUSSION

Question. - Z. MORLIN. Reply. - T. HAGMARA.

In your conductivity diagram the conductivity of The measurements as to the same specimen after the pure PbI, did not change in a large temperature quenching from about 530 K to room temperature interval. Why ? suggested that iodine ion vacancies, a dominant charge carrier in this temperature range, were trapped by the lead ion vacancies introduced during a zone- refining treatment.

References

[I] SFI 1 H, W., 2. PIly.y.56 (1 929) 802.

[2] DAWOOD. R. I. and FORTY, A . J., Philos Mag. 7 (1962) 1633. 131 DUGAN, A . E. and HENISCH, H. K.. Phys. Rev. 171 (1968) 1047

[4] CONSI.ANTINECU. M , GHITA, C., GHITA, L. and RAL.IOG, I.,

Phvr. Status Solidi A 38 (1976) K 175. 151 M I ~ A G A W A , T.. Acra Crpst. A 31 (1975) 823.

16) RAO, M. and SRIVASTAVA, 0. N . , J . Ph-VS. D 116 (1978) 919 [7] FULLER, R. G.. in Pob~t Dejects in Solids. Vol. I , C R A W ~ O R D ,

Jr. J. H and SLIFKIN, L. M. ed. (Plenum Prcss, New York

and London, 1972).

[a] T~JB,\XDT, C.. Haltdb. ~uperimet1tal Physik 21 (1932) 383.

191 COLL, K. S. and COLE, R . H., J. Chem. Phys. 9 (1941) 341. 1101 DRYDEN, J. S. and MEAKINS, R. J., Discuss. Faraday Soc. 23

(1957) 39

[II] HAVEY, Y., J. Chetn. Phys 21 (1953) 171.

[I21 Bvccr, C and FIESCHI, R., PIcvs. Rev. Lett. 12 (1964) 16.

Références

Documents relatifs

Carbapenemase, plasmid-mediated AmpC (pAmpC), extended-spectrum β- lactamase (ESBL) and plasmid-mediated quinolone resistance genes were characterised by PCR.. Plasmids were

sacchari and provided a morphological criterion 51 to separate the species, based on the ratio between the hind tibia length and the antennal.. 52 processus

Le test qui indique le degré de la peroxydation lipidique a révélé que I'effet oxydant des molécules anticancéreuses est dû à un état de stress oxydant survenu

Les résultats révèlent que la pluparts des patients sont âgés de plus de 60 ans et plus de la moitié n’ont pas d’ATCD de cancer colorectal, leurs régime

CoO single crystals have been deformed down to 4.2 K ; observations are analogous to those obtained for other NaCI structure crystals. By annealing

This theory shows !hat any donor or acceptor impurity can ( i n principle) be incorporated in tlie lattice either as a neutral defect ~\itIiout compensation or

In the decay-time distribution (Fig. 4), most of the events at positive decay times are signal events, while events reconstructed at negative de- cay times are mainly background..

TABLE I: Summary of results; ε denotes the reconstruc- tion efficiency and ε tot the total efficiency including daugh- ter branching fractions, both in percent; N is the number