论文英文翻译

Temperature

Dependence

of

Si-Based

Thin- Film

Solar

Cells

Fabricated

on

Amorphous

to

Microcrystalline

Silicon

Transition Phase

Kobsak

SRIPRAPHA Ihsanul

Afdi

YUNAZ Seung

Yeop

MYONG Akira

YAMADA

and

Makoto

KONAGAI

Department

of

Physical

Electronics,

Tokyo

Institute

of

Technology,

2-12-1-S9-9,

O-okayama, Meguro-ku, Tokyo 152-8552, Japan Quantum Nanoelectronics Research Center,

Tokyo

Institute

of

Technology,

2-12-1-S9-9,

O-okayama,

Meguro-ku,

Tokyo

152-8552,

Japan

(Received June 5, 2007; accepted August 20, 2007; published online November 6, 2007)

The

temperature

dependence

of

silicon

(Si)-based

thin-film

single-junction

solar

cells,

whose intrinsic

absorbers

were fabricated on the transition

phase

between hydrogenated

amorphous silicon (a-Si:H) to hydrogenated microcrystalline silicon (mc-Si:H), was

investigated. By varying the hydrogen dilution ratio, wide

-band-gap protocrystalline

silicon (pc-Si:H) and

mc-Si:H absorber layers were obtained. Photo-current density

–

voltage

(Photo-J

–

V) characteristics

were

measured

under

AM1.5

illumination at

ambient temperatures in the range of 25

–

75

C. We found that the solar cells with

pc-Si:H, which

exists just

below the a-Si:H to mc-Si:H transition boundary, showed the

lowest

temperature coefficient (TC) for conversion efficiency and

open-circuit voltage

(Voc), while the solar cells fabricated at the onset of the a

-Si:H to mc-Si:H phase

transition exhibited a

relatively high TC for and Voc.

Experimental

results indicated

that pc-Si:H is a promising material for the absorber

layer of

the single junction

or

the top cell of tandem solar cells that operate in high temperature regions.

KEYWORDS: temperature dependence, amorphous silicon, protocrystalline silicon, Si thin-film

solar cell, solar cells

1.

Introduction

In general, the solar cell

performance is measured

under the standard

test conditions

(STC) of a cell temperature of 25

摄氏度

and an irradiance of 100mWcm

2 with AM 1.5

spectral distributions.

However,

in an outdoor installation,

the operating

temperature

of solar

cells considerably changes depending on the environment, i.e., the climate

in

the installed a tropical region, the operating temperature often reaches more

than

70

摄

氏

度

.

The

increase

in the

operating

temperature

leads

to a decline in

conversion

efficiency

mainly

due

to

the

drop

in

open-circuit

voltage(Voc)

Among

Si-based

solar cells, bulk crystalline Si

solar

cells

which

include

single-crystalline- Si

(c-Si)

and

polycrystalline-Si

(poly-Si)

solar cells show higher than thin-film solar cells at room temperature. However,

η

of

c-Siand poly-Si solar cells seriously decreases with an increase

in the operating

temperature,

while

hydrogenated amorphous

Si (a-Si:H)-based thin-film

solar cells

exhibit relatively small variation in.

The main reason for the lower

temperature

coefficient

(TC)

of

a-Si:H-based

solar

cells

is

their

wide-band-gap

intrinsic absorber or high

Voc

compared with

those of bulk crystalline

Si-solar cells.

Taking the real

output power affected

by the operating temperature

and

production cost

into

account,

a-Si:H-based

thin-film

solar

cells

have

advantages

over

bulk

crystalline-Si solar cells for

use in high

temperature

areas

such as a tropical region.

However,

it

is

well

known

that

a-Si:H-based

thin-film

solar

cells

exhibit

light-induced degradation

after light

exposure, the so-called Staebler

–

Wronski effect

(SWE).The SWE in a-Si-based

thin-film solar cells is also a veryim portant factor that

must

be

considered

for

outdoor

installation.

During

the

past

30

years,

extensive

research

has been conducted to suppress the SWE. As a result, two kinds of edge materials near

the

phase

boundary

have

been

developed

as

stable

intrinsic

absorbers:

one

is

the

wideband-

gap

protocrystalline

silicon

(pc-Si:H)

existing

just

below

the

a-Si:H-to-microcrystalline

silicon

(mc-Si:H)

transition

transition

and

the

other

is

the

narrow- band-gap

mc-Si:H with crystalline silicon volume fraction (Xc) of 30

–

50%

obtained near the onset of the

phase

transition.

The pc-Si:H

material

nucleate

from the

deposition of the a-Si:H at just before the transition boundary of a-Si:H to a-Si:H t

mc-Si:H

mixedphase. Once, the a-Si:H t mc-Si:H

transition is detected,which can be

observed by a real time spectroscopic ellipsometry (RTSE), the growing material is no

longer

considered

pc-Si:H.

10)

The

unique

properties

of

pc-Si:H

are

the

optical

band

gaps

(Eopt) and the Urbach tail. The Eopt of pc-Si:H is larger than conventional material

and increases with increasing H2 dilution ratio. Besides, the narrower Urbach tail in

pc-Si:H causes the higher hole drift mobility than conventional materials. The key

feature of the pc- Si:H

material is its relative stability to light induced degradation as observed in the

electron-mobility lifetime product and similarly in the solar cell fill factor. These

two kinds of materials are attractive for application to Si-based thin-film

solar cells because of their low SWE.

Although

the

pc-Si:H

solar

cell

has

shown

a

good

temperature

dependence

among

Si-based

thin-film solar cells,

the behavior of TC for pc-Si:H solar cells has not yet been

clarified. In this work, we investigated the temperature

dependence of a-Si:H-based

solar cells fabricated in the pc-Si:H to mc-Si:H

transition regime. The TC values

after

lightinduced degradation were also investigated in order to find

the optimal absorber layer for the use at high operating temperatures.

2.

Experimental Procedure

The p

–

i

–

n single-junction solar cells were fabricated on Asahi U-type glass

substrates in a multi chamber system with the structure of glass/SnO2:F/hydrogenated

p-type

amorphous

silicon

carbide

(p-a- SiC:H)/buffer/intrinsic

(i-)absorber/n-type

amorphous silicon (n-a-Si:H)/boron- doped zinc oxide (ZnO:B)/Ag/Al with the cell area

of

0.086 thicknesses of

p, buffer, i-,

and n-layers

were

kept

constant at around

12, 4, 320

–

340, and 2 nm, solar cells were fabricated at the

substrate

temperature

of

around

200

C

with

deposition

pressures

of

50

–

70

Pa.

Thevery

high frequency (60 MHz) plasma-enhanced chemical vapor deposition

(VHF-PECVD)

was used

to

deposit

the i-layer.

The

i-layers were deposited

at different

silane concentrations,

SC ? SiH4=e SiH4 t

H2T,

by varying

SC

from 6.0

to

2.4% in

order

to

obtain material with

the phase transition

from amorphous to microcrystalline a

decrease

in SC,

the deposition rate of the i-layer

declined from 1.6 to 0.9. The doped (p- and n-layers) and buffer layers were deposited

by a radio-frequency(13.56MHz) PECVD technique. ZnO was deposited by

metal organic

chemical vapor deposition (MOCVD)

as a back reflector,

while Ag

and Al were evaporated

as back electrodes for all samples.

The

Raman

spectroscopy

was

performed

using

a

JASCONRS-1000

system

with

a

semiconductor laser at a wavelength of 532 nm.

Ex-situ

spectroscopic ellipsometry

(SE)measurements

(J.

A.

Wollam)

were

used

with

a

variableangle

spectroscopic

ellipsometer. The temperature dependence of the solar cell parameters were measured

using a solar simulator in a chamber at ambient temperatures (T) in the range of 25

–

75

C with a step increment of 10

C under 1-sun (AM1.5, 100mW

cm

2) irradiation. The

temperature

of

the

sample was

regulated

by

a

temperature-controlled

airflow.

The

temperature dependence of solar

cells

was obtained from photo-current density

–

voltage

(photo-J

–

V) measurements. The value of TC can be expressed as

where Z denotes the solar cell parameters, i.e.,

η

, Voc, shortcircuit current density

(Jsc), and fill factor

(FF). The normalized temperature Tn

is

chosen to be

25

C because

it

corresponds

to

the

standard

reference

condition

for

solar

cell

measurement.

The

1-sun

standard light-soaking test was performed in a climate chamber at the temperature of

50

C for 100 h.

3.

Results and Discussion

3.1 Characterization of intrinsic absorbers

In the first series of experiments, we inspected the Raman spectra for the solar cells

fabricated in the phase transition regime. Figure 1 shows the Raman spectra for solar

cells prepared with different SCs which are measured from the nside

(rear-side of the solar cells). The Raman spectra were deconvoluted to four Gaussian

peaks centered at the Raman shift areas at around 430, 480, 510, and 520 cm

1, which

correspond to the longtitude optical (LO) mode of a-Si:H, the transverse optical (TO)

mode of a-Si:H, the defective crystalline phase and

the TO mode of

c-Si, respectively.

The defective part of the crystalline phase is included in the

crystalline fraction.

15)

The Xc calculated from the Raman spectrum is expressed as

where Ii is the area under the Gaussian peak centered at the Raman shift of i cm

1 and

I480

t I510 t I520 is the total integrated area. By decreasing

SC from 6.0 to 4.0%, the

peak position of Raman spectra of a-Si:H in the TO mode increased from the Raman shift

of 475 to 480 cm

1, as shown by the solid line in the figure, which means that the

a-Si:Hmicrostructure improved when SC decreased, leading to

further improvement of

stability against illumination.

16)

For SC ?3:2%, the Raman spectra exhibited

two

peaks

at

480 and

517 cm

1, which correspond to the onset of mc-Si:H growth. With further

decrease of SC, SC < 3:2%, the peak position of Raman spectra became that of c-Si in