INJECTION LUMINESCENCE IN AMORPHOUS SILICON p+-i-n+ JUNCTIONS

HAL Id: jpa-00220955

https://hal.archives-ouvertes.fr/jpa-00220955

Submitted on 1 Jan 1981

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.

INJECTION LUMINESCENCE IN AMORPHOUS SILICON p+-i-n+ JUNCTIONS

T. Nashashibi, I. Austin, T. Searle, R. Gibson, P. Lecomber, W. Spear

To cite this version:

T. Nashashibi, I. Austin, T. Searle, R. Gibson, P. Lecomber, et al.. INJECTION LUMINESCENCE

IN AMORPHOUS SILICON p+-i-n+ JUNCTIONS. Journal de Physique Colloques, 1981, 42 (C4),

pp.C4-467-C4-470. �10.1051/jphyscol:1981498�. �jpa-00220955�

INJECTION LUMINESCENCE IN AMORPHOUS SILICON

p+-i-n+

JUNCTIONS

T.S. Nashashibi, I.G. Austin, T.M. Searle, R.A. ~ibson*, P.G. ~ e ~ o m b e r * and W.E. spear*

Department of Physics, The University, Sheffie Zd 53, U . K .

*

Department of Physics, The University, Dundee, Scotland

Abstract We present a detailed study of electroluminescence (EL) spectra and

+ +

EL quantum efficiency in well characterised a-Si p - i a junctions under forward bias. The PL characteristics of the i region were probed using laser excitation. Some factors controlling EL efficiency are discussed. EL and PL recombination models are compared and discussed.

Introduction EL in amorphous silicon has received little attention apart from brief reports by Pankove and Carlson (1) and Street et a1 (2). These authors observed a broad emission peak near 1.3eV in p-i-n and Schottky junctions. In this paper we present a detailed study of EL spectra and quantum efficiency in a-Si junctions deposited on stainless steel substrates. The PL behaviour of the i region was also probed using a laser beam.

Experimental All the junctions were prepared at Dundee by the glow discharge method and their electrical characteristics have been fully described elgekkere (3). A typical device configuration is shown in Fig 1. A boron doped p layer about 150nm thick (diborane-silane ratio 2 0.01) was grown onto polished stainless steel, followed by 5 700nm of undoped material, and a phosphorus doped n+ layer (% 50nm thick, phosphine-silane ratio

2

^0.003). A semitransparent Au electrode (% O.lcm 2 ) was evaporated onto the n+ layer to provide electrical contact. Junction thick- nesses were also measured in some cases by hydrogen profiling (4). EL spectra were measured under forward bias, and PL spectra under zero bias, using a cooled Ge detector.

Results EL emission was measured for forward dark currents in the range 50-1000pA and forward voltages V between 6 and 12V, at temperatures between 100 and 200K.

In this temperature an3 voltage range, the forward current (iinj) varies as V where p varies from 6 to 8 as T decreases. The high power law implies space charge : and contact effects. i.

.

is much larger than the thermal equilibrium current and is due to carriers injePfJd from both contacts (9).

Figure 2 shows a typical EL spectrum from a junction at 140K. The EL emission peak is weakly structured and is centred around 0.9eV. In contrast the PL spectrum is strongly modulated by interference fringes and the centroid is dispfaced % 0.2eV to higher energies. PL excitation was carried out through the top (n ) contact using Argon laser light at 2.4eV. The EL emission shows a small spectral shift to lower energies between 125 and 200K but no observable change in linewidth.

Figure 3 shows a log-log plot of the total EL intensity as a function of i.

.

^{; the}

intensity varies as iV

.

^{where v} varies from 1.1 to 1.4 as T increases (~ik"~3, inset). Fig 4 comparJs the T dependence of in

nEL

and qpL

,

using 2.4eV laser excitation for the latter.

Discussion -$t the doping levels used in the n+ and p+ regions of these junctions 'lpL is % 10 strongly suggesting the EL occurs in the i region. In spite of the complex dependence of i on Vf, the EL intensity varies almost linearly with i

(Fig 3) implying a cons&zat EL efficiency. in j

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1981498

JOURNAL DE PHYSIQUE

W -

^n*

i

-

^P+

STAINLESS STEEL

F i g 1 J u n c t i o n geometry and o p t i c a l system

i,,, ( m ~ l c m ' ) FORWARD BIAS CURRENT F i g 3 EL i n t e n s i t y v e r s u s i . m

.

a t v a r i o u s temperaturds

PHOTON ENERGY I e V ) F i g 2 EL and PL from a j u n c t i o n a t 140K

1 0 ~ 1 ~

Fig 4 Temperature dependence of nEL and nPL

( i ) EL and ?L S p e c t r a Since t h e s e j u n c t i o n s a r e d e p o s i t e d on h i g h l y r e f l e c t i n g s u b s t r a t e s , i n t e r f e r e n c e e f f e c t s must be taken i n t o account. Consider f i r s t t h e PL spectrum i n F i g 2, which i s s t r o n g l y modulated by f r i n g e s . A d e t a i l e d a n a l y s i s shows (5) t h a t almost a l l t h e modulation can b e e x p l a i n e d i n terms of two-beam i n t e r f e r e n c e , assuming s p e c u l a r r e f l e c t i o n a t t h e s t e e l s u b s t r a t e . The f r i n g e p e r i o d i c i t y depends on t h e

total

f i l m t h i c k n e s s (d) and t h e modulation depends on t / d , where t i s t h e p e n e t r a t i o n depth of t h e e x c i t i n g l i g h t . For e x c i t a t i o n a t 2.4eV i n a-Si, t << d and t h e modulation i s s t r o n g . Curve B i n F i g 2 shows t h e PL c o r r e c t e d f o r f r i n g e e f f e c t s i n t h i s way; t h e r e s i d u a l f r i n g e e f f e c t s probably a r i s e from weak multi-beam e f f e c t s , o r t h e n e g l e c t of d i s p e r s i o n .

I n c o n t r a s t , t h e measured EL s p e c t r a show weak f r i n g e s (Fig 2 ) . The

lack

of s t r o n g f r i n g e s w i t h p e r i o d i c i t y d i s c o n s i s t e n t with two p o s s i b l e s p a t i a l d i s t r i b - u t i o n s ; ( a ) uniform EL g e n e r a t i o n throughout t h e i r e g i o n , o r (b) EL g e n e r a t i o n c l o s e t o t h e p+-i boundary, a t a d i s t a n c e ^X from t h e s t a i n l e s s s t e e l s u b s t r a t e .

0

Fig 5 Comparison between EL and interference corrected PL, at 140K.

Fig 6 Dark EL and photo-enhanced EL

Since X % 150nm the fringe periodicity in this case is larger than the width of the EL spectrum.

The electrical evidence implies much shorter diffusion lengths for holes than for electrons, and favours (b). Model (b) may also explain the spectral shift between EL and PL (Fig 2) in terms of interference effects. If a fringe minimum lies close to the high energy side of the PL spectrum, it will be shifted to lower energies.

Curve AA' in Fig 5 shows the interference modulation corresponding to EL generation close to the p -i boundary, at a distance X = 150nm, giving a fringe minimum near 1.2eV. Curve C in Fig 5 shows the result& spectrum when the PL peak (B) in Fig 2 is corrected using the interference modulation curve AA' in Fig 5. The correction shifts the PL peak to lower energies where it corresponds well with the EL peak.

Thus the spectral shift between EL and PL can be accounted for, assuming that EL is generated in a narrow region 150nm from the substrate. An equally good fit can be obtained for X = 150 2 10nm. Recent hydrogen profiling data on similar junctions (4)

she$

that the p+ layer extends to a depth ?. 150nm. In support of this interpretation, Pankove and Carlson (1) observe no spectral shift between EL and PL in p-i-n junctions on a semitransparent substrate.

(ii) EL quantum efficiency In low defect density a-Si, n approaches unity;

little is known about nEL, through Pankove et a1 (1) estimaPEe comparable EL and PL efficiencies. For a junction, the .-internal EL efficiency is defined by the relation

where G is the rate of photon generation due to EL emission and iin. the forward current. Measurements were made on a number of our junctions by camparing the EL signal for a given forward current with the total PL signal using a known laser excitation power, and the same optical collection system. We observe EL efficiency vaiues ranging up to nEL = 0.1

n r p L ,

where

n'

is the effective PL efficiency for excitation light at Argon laser energy of 2.4egf: which has a penetration depth of

% 1OOnm.

The top (n+) layer in these junctions has negligible PL efficiency and hydrogen profiling data (4) show that it extends to a depth % 50nm. We estimate that the excitation light at 2.4eV is attenuated by a factor of 0.3 on passing through this layer, and this is confirmed by PL experiments on the change in PL efficiency (using 2.4eV excitation) on etching away the n+ layer. Thus "L = 0.03 where

npL

i is

JOURNAL DE PHYSIQUE

the PL efficiency in these junctions just below the n+ contact. From the interference modulation curve AA' in Fig 5, we estimate that the measured EL efficiency will be cut down by a factor of Q 2-3. We conclude that the maximum internal EL efficiency in these junctions is

(iii) EL recombination models The model discussed above implies that

nEL

depends on the fraction of injected electrons which are captured by holes in the i region at a finite distance from the p+ contact, where the recombination luminescence efficien~y ('l) is reasonably high. Thus

where y is a measure of the hole injection and U the capture cross-section. Our measurements indicate (eq 2) tha: qm is 0.1 times smaller than the PL efficiency

in the i region just below the n contact. We speculate that part of this

reduction in

n

may be due to field quenching of the luminescence in the i region.

We estimate thgk a quenching factor of Q 2 may arise in this way, from experiments on eleetric field quenching of ?L in these junctions (5). However, the main reduction in n probably arises because much of the recombination occurs in a region close to theEL p+ contact, where the luminescence efficiency is low because of doping effects associated with the p+ Contact.

Recently, we have observed a novel effect in these junctions (5), in which the EL +

efficiency can be enhanced by a factor of 4

-

8 by band-gap illumination at the n contact (Fig 6). The extra EL is directly proportional to the photocurrent. The mechanism for the enhancement is not clear, but may involve field induced changes in the luminescence efficiency ('l) in eq (3), or the hole injection efficiency (y).

(iv) Relation between EL and PL The similarity in EL and PL spectra suggested above is interesting in view of the different excitation processes involved. For PL in a-Si, Tsang and Street (6) propose geminate recombination and infer a distribution of electron-hole separations between 3 and 7 nm. Time resolved studies (6,7) show that the PL peak moves to lower energies with increasing time as the recombination path changes from near pairs to distant pairs. In EL in these junctions however, electrons and holes are created Q 700nm apart and drift or diffuse together. The similarity in EL and.PL spectra implies that electrically and optically injected carriers thermalise in much the same way before luminescing.

That is, geminate and non-geminate pairs give rise to the same luminescence spectrum.

In sunimary, the optical and electrical evidence on these junctions suggests that EL and PL have the same spectra: they also have the same T dependence (Fig 4) but the internal EL efficiency is v 10 times smaller (eq 2). We suggest that effects due to geminate pairs are not important in the PL process, in agreement with recent ODMR recombination studies on a-Si(8).

References

1 PANKOVE JI and CARLSON DE, Applied Physics Letters

2

(1976) 620

2 STREET RA, TSANG C and KNIGHTS J, 14th Int Conf Phys Semicond (1978) 1139 3 GIBSON RA, SPEAR WE, LeCOMBER PG and SNELL AJ. J Non-Cryst Sol

35

(1980) 1005 4 MULLER G, DEMOND F, KALBITZER S, DAMJANTSCHITSCH H, MANNSPERGER H, SPEAR WE,

LeCOMBER PG end GIBSON RA,--Phi1 Mag 41B (1980) 571 5 NASHASHIBI TS et a1 t 6 b e published

6 TSANG C and STREET RA, Phys Rev (1979) 3027

7 SEAEE TM, NASHASHIBI TS, AUSTIN IG, DEVONSHIRE R and LOCKWOOD G, Phil Mag 39 (1979) 3 8 F 8 DEPINNA et a1 (this conference)

9 SNELL A J SPEAR WE and LeCOMBER PG, Phil Mag

X,

^{(1981 ) 407}