INTERFACE PHENOMENA IN THE FIELD ION AND FIELD ELECTRON ENERGY SPECTROSCOPY OF HIGH-Tc SUPERCONDUCTORS

HAL Id: jpa-00229978

https://hal.archives-ouvertes.fr/jpa-00229978

Submitted on 1 Jan 1989

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.

INTERFACE PHENOMENA IN THE FIELD ION AND FIELD ELECTRON ENERGY SPECTROSCOPY

OF HIGH-Tc SUPERCONDUCTORS

N. Ernst, G. Bozdech, W. Schmidt, M. Naschitzki, A. Melmed

To cite this version:

N. Ernst, G. Bozdech, W. Schmidt, M. Naschitzki, A. Melmed. INTERFACE PHENOM- ENA IN THE FIELD ION AND FIELD ELECTRON ENERGY SPECTROSCOPY OF HIGH- Tc SUPERCONDUCTORS. Journal de Physique Colloques, 1989, 50 (C8), pp.C8-471-C8-475.

�10.1051/jphyscol:1989880�. �jpa-00229978�

COLLOQUE DE PHYSIQUE

C o l l o q u e C8, Suppl6ment au n Q l l , Tome 50, novembre 1989

INTERFACE PHENOMENA IN *HE FIELD ION AND FIELD ELECTRON ENERGY SPECTROSCOPY OF HIGH-T, SUPERCONDUCTORS

N. ERNST, G. BOZDECH, W.A. SCHMIDT, M. NASCHTTZKI a n d A.J. MELMED' Fritz-Haber-Institut der Max-~lanck-Gesellschaft, Faradayweg 4-6,

?-I000 Berlin 3 3 , F.R.G.

Custom Probes Unlimited, Box 3938, Gaithersburg, MD 20878, U.S.A.

Abstract - Retarding potential analyses of ions, Ar+ and Dz+, and electrons field emitted from RBa2Cu307-, (R = Eu, Gd, Y) tips have been carried out for surface conditions characterized by field evaporation (FEV) and field ion imaging (FIM). Energy distributions o f ions and electrons, taken between 20 and 78 K, are shifted t o values ranging between 1 and 1 9 eV above and below EF, suggesting the presence of an insulating layer. Controlled FEV, applied t o decrease t h e thickness o f this "dead" layer, leads t o decreasing energy shifts. An upper limit o f about 7 eV for the band gap in the insulating layer was estimated from combined field-ion and field- electron energy distribution measurements.

1 - INTRODUCTION

Tunneling experiments may yield valuable data for an experimental foundation o f the physical mechanisms responsible for electrical transport i n oxide-based superconductors, such as the width o f the band gap at the Fermi energy, EF, and i t s temperature dependence in the superconducting regime 111. Experiments using metal-insulator-(high-Tc)superconductor sandwiches or point-contact (STM) arrangements have already indicated gap-widths ranging between 2 5 and 6 5 meV. Since field- electron-emission spectroscopy (FES), especially at low field strength, is essentially confined t o a small energy ran e (= 1 0 0 meV) at and below EF, it should be a sensitive probe f o r the transition from the normal to t f e superconducting phase 12.31 However, one drawback for STM and FES experiments on oxide-based high-Tc specimens appears t o be the presence o f insulating ("dead") layers which might be opaque for tunneling electrons /1,4,5/. The present experimental approach attempts t o establish a (quantum mechanically) transparent tip surface for tunneling electrons, originating from bulk states close t o EF, by use o f low temperature field evaporation (FEV).

2 - EXPERIMENTAL

Two measurement systems were used, each including a probe-hole-channel-plate FIM. Firstly, a magnetic-sector atom-probe combined with a retarding-potential analyzer, RPA, enabled mass and charge-resolved ion-energy analysis and liquid-He cooling of high-Tc specimens (G.B.&N.E.) /41.

Secondly, a probe-hole FIM equipped with liquid-N2 cooling of the specimen was coupled w i t h a RPA for combined ion and electron-energy spectroscopy (M.N.,W.A.S.&N.E.). High-T, specimens (RBazCu307-,,

R

= Y, Gd, Eu), in the form o f small sharp-pointed fragments from well characterized samples (Tc;=:90 K), were attached t o Pt tips with conductive epoxy (A.J.M.) 161. Field evaporation and field ion microscopy were exclusively used for cleaning and imaging.

The energy analysis of ions (from image gas-atoms) and electrons field emitted from superconducting tips is schematically outlined in figs. l a and b assuming that the superconducting phase (A = o n e half of t h e energy gap) extends t o the surface. This means that no insulating layer exists which could prevent electron-tunneling into electronic levels located at and above EF

+

A, or from states at and below EF-A. Due t o the formation of the gap around EF, onset energies, e-So, o f ion as well as electron energy distributions are expected t o shift t o higher values by about 2 0 t o 3 0 meV at TeT,, within the frame work of the model outlined in fig. 1 131.

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1989880

I ^{( b )} field electron emission

( a field ion emission

% energy distribution

m

L

C

C

B

t U

@ Q

b) 10 H 12 13 14 15

-

e . 6 ( V l

disthnce

C

C

.-

0

Fig. 1

-

Experimental method: Retarding potential analysis of (a) ions and (b) electrons field emitted from a superconductor. One expects, e-So = I

-

@ R + , A for ions, and e-So

=

@ R + A for electrons (e =elemental charge, So = onset voltage of retardat~on curve, I = ionization energy and @R = w o r k function o f retarding electrode).

3

-

RESULTS

distance

I

Evac

I

A FIM pattern o f a Gd123 specimen, in the magnetic-sector atom probe with D2 as imaging gas (left photo i n fig. Za), shows surface conditions during retarding-potential analysis. Ion emission analyzed i n this case was restricted t o one or t w o bright spots within a single stripe (c-axis perpendicular to t h e stripes). The result is shown i n fig. 2b (upper left diagram).

n

8 ,

I 5 --- ---- L --- T

[ q r e t a r d i n g meshes

1 I

,

detector.

I

(a) D2-FIM o f Gd Ba2 Cu3 Opx at 20 K

U F E V = ~ . ~ kV exposure time: 8 sec UFEv-8.3 kV exposure time:8 sec '

(b) differen'tiated ion-retardation curves

10 12 14 16 18 20 22 10 15 20 25 30 35 40

e m i t t e r - r e t a r d e r v o l t a g e , 6 ( V 1

Fig. 2 - Experimental results o f magnetic-sector atom probe retarding-potential analysis: (a) Probe- hole FIM and (b) mass-resolved ion-energy spectrosocpy of high-T, and ordinary metal specimens. A trial function was fitted t o measured retardation curves t o yield normalized, differential distribution curves i n

(b).

Note the different &scales on the left and on the right. The white spot i n t h e probe hole indicatesthe area on the surface analyzed during measurementson a Gd123 t i p (diagram upper left).

As t h e field evaporation voltage approached 8.3 kV, the FIM pattern (right side o f fig. 2a) showed a clearer "striping" which was accompanied by a considerable shift o f the D2+ energy distribution towards the metal reference curve (thin line in fig. 2b). This behavior was also observed w i t h other specimens.and when Ar was used as an imaging gas (fig. Zb, lower left).

For one particular Gd 123 specimen, ion energy distributions first shifted towards the metal reference, and later, after FEV above 12.4 kV, again away from the reference position (fig. 2b, upper right).

Moreover, leaving one Eu123 specimen at room temperature overnight in UHV increased t h e energy shifts, as demonstrated i n fig. 2b (lower right corner). To summarize the ion-spectroscopic results, w e observed shifts o f onset energies for D2 + and Ar', which ranged between 1 and 19 eV above EF. Field evaporation o f (a few) surface layers caused decreasing ecergy shifts.

Combined field ion and field electron energy distribution measurements, carried o u t at 78 K w i t h t w o Y123 specimens (fig. 3), suggest that the surface for these particular cases was n o t transparent f o r tunneling electrons, despite o f some field evaporation. This conclusion is drawn from measurements o f electron-energy shifts towards values of ==4 eV below EF as obvious from fig. 3b. We note that f o r conditions specified in fig. 3 ( U ~ ~ v z 5 . 7 kV), the Ar+ onset-energy is'about 3.5 eV above EF, which is s t i l l 2.5 eV larger than t h e minimum shift measured for a Gd123 specimen (fig. 2b). Experiments are underway t o quantifiy further the effect o f controlled FEV o n the electron energy distribution.

(a) field ion emission (b) field electron emission

energy above

EF

( eV

-

V) energy below

E,

( eV

I

emitter

-

retarder voltage, 6 ( V ) emitter

-

retarder voltage, 6

I V I

Fig. 3 - Experimental results of (a) ion and (b) electron retarding-potential analysis. Upper scales are referenced t o onsets o f distrrbution curves measured for W (thin lines). Thick lines represent data from an Y123 specimen field evaporated and imaged in Ar at 5.7 kV and 78 K.

4 - DISCUSSION

Although the physical origins o f the present observations are not fully understood, an explanation for the measured energy shifts can be given by use of an electron potential-energy diagram outlined i n fig. 4, where effects at the superconductor-insulator and insulator-vacuum interfaces have been neglected /7,8/. From measured energy shifts, e.(A6i

+

A&), one can derive an upper limit f o r t h e w i d t h of the energy gap in t h e insulating layer, e.y. from experimental results in fig. 3, Egapc7 eV.

The existence o f the insulating ("dead") interface is possibly connected with the change o f oxygen stoichiometry i n the near-surface region at the tip as proposed, e.g., by Bakhtizin et at. /5/, which was recently investigated quantitatively by Camus, Elswijk and Melmed using atom-probe techniques (included i n these proceedings).

( a ) field ion emission

I

^{( b}

^l

field electron emission

1

electron conductor insuulator

aJ

I

^super distribution

distance

- --

^{dNId6 distance}

- ^--

^dNld6

Fig. 4

-

Electron-energy diagram giving a possible explanation o f energy shifts measured for (a) field- ion and (b) field-electron emission from oxide-based superconductors.

In conclusion, the presenf results clearly demonstrate that FEV can be applied for reducing the thickness o f the "dead layer. Further experiments, especially electron-energy distribution measurements for well defined (FIMIFEV) specimen surfaces, are required before one may decide whether FES can yield data for the understanding o f electrical transport processes rn high-Tc materials.

ACKNOWLEDGEMENT

We gratefully acknowledge the preparation and characterization o f sample materials by C.K. Chiang, R.D. Shull and M. Hill, National Institute of Standards and Technology (once NBS). We would also like t o thank Prof. J.H. Block, Fritz-Haber-lnstrtute, Berlin for supporting this project.

REFERENCES

111 PHILLIPS,J.C., "Physics o f High-Tc Superconductors", Academic Press, San Diego 1989 I21 KLEIN, R.. AND LEDER, L.B., Phys. Rev. 124 (1961) 1050

I31 GADZUK; J.W., Surf. ~ c i .

15

(1969) 466-

I41 ERNST, N., BOZDECH, G., AND MELMED, A.J., J. de Physique

49

C6 (1988) 453

151 BAKHTIZIN, R.Z., GHOTS, S.S., MESYATS, V.G., SHKURATOV, S.I. AND YUMAGHUZIN, Yu.M., J. de Physique

49

C6 (1988) 495

161 MELMED, A.J., J. de Physique 49 C6 (1988) 67

171 KAO, K.C., AND HWANG, W.-;1'Electrical transport in solids with particular reference t o organic semiconductors", Int. Ser. Sol. State

14,

Pergamon, Oxford 1981, and references therein

181 Similar diagrams have been used for the explanation of field-ion energy measurements for SiO, (SAKURAI, T., Surf SCI.

86

(1979) 562) and field-electron energy measurements for dielectric tips on W (LATHAM, R.V., Vacuum

32

(1982) 137)