18.5% efficient n -type silicon solar cells made in pilot production. efficient n-type... · Over 19% efficient n -type silicon solar cells made in pilot production. Towards higher

www.ecn.nl

18.5% efficient n-type silicon solar cellsmade in pilot production.

Towards higher efficiency and lower cost PV

www.ecn.nl

Over 19% efficient n-type silicon solar cellsmade in pilot production.

Towards higher efficiency and lower cost PV

3 7-11-2010

A good example of collaborative effort …..

Newtechnology

Research Institute

PV manufacturer Equipment Supplier

4 7-11-2010

… towards bi-facial n-type solar cells !!!

• n-type: higher quality: higher efficiency• bi-facial: open rear: higher energy production

5 7-11-2010

The old days

ResearchInstitute

PVmanufacturer

EquipmentSupplier

New

Technology

6 7-11-2010

The now days

ResearchInstitute

PVmanufacturer

EquipmentSupplier

New

Technology

Scattering of technology

7 7-11-2010

The new days

Newtechnology

Research Institute

PV manufacturer Equipment Supplier

8 7-11-2010

Demand for innovation

• Most cell producers still use conventional p-type H-bar technology

• Strong need for high efficiency:- Leverages area-related wafer, module and system costs such

as materials and labour- Higher kWh/Wp

• Industrial scale

• ECN n-type cell technology meets requirements

… and speed is needed

9 7-11-2010

PANDA project: towards fabrication

• Yingli- Creating and operating pilot line- Debugging and optimizing pilot operation- Improving the technology- Cost assessment

• Tempress (subsidiary of Amtech Systems)- Diffusion and glass removal equipment- Optimization of diffusion processes

• ECN :- Cell technology- Process optimization

10 7-11-2010

Work horse of PV industry: Al-BSF cell

11 7-11-2010

What held n-type solar cells back?

• Historical: e.g., p-type base better radiation resistant in space

• High-temperature necessary for B-diffusion;but this is problematic only for p-type substrates

• Creating oppositely doped layers on front and back.

• B-emitter passivation.

ECN made breakthroughs in 2005

12 7-11-2010

• Major metal impurities affect n-type less• p-type dopant: boron-oxygen recombination centre.

n-type wafer offers better diffusion lengths

n-type Si

n++ BSF

p++ emitter

Feo/+

Tio/+

p-type Si

(n+) Phosphorous emitter

Fe+/o

Ti+/o

Back contact

B-O2i

Efficiency: record cells

13 7-11-2010

Cell IscA

UocV

FF%

Areacm2

JscmA/cm2

Eff.%

3-44 8.78 638 79.5% 240 36.6 18.58%*

3-12 8.90 635 78.2% 240 37.5 18.65%*

1-16 8.96 638 77.0% 237 37.8 18.59%*

#1 9.00 637 79.6% 237 38.0 > 19.0%

#17 8.98 637 79.5% 237 38.0 > 19.0%

*FhG-ISE, May 2010

Features of the Panda cell• Bifacial cell, on 6 inch CZ wafer.• n-type material for good efficiency

• Phosphorous BSF- good FF

- rear surface passivation.

- allows bi-facial design.• No aluminum BSF: no bending of cells, suitable for thin wafers

• Industrial processing• Metallization:

- industry compatible screen printing

- open front and rear: limited coverage and paste consumption- compatible with standard module manufacturing technology

14 7-11-2010

Pilot line for Panda cells

• Dedicated pilot line available at Yingli- Allowed very rapid development.- Experiment turn-around 1-2 days

• Objectives:- Demonstrate Panda technology- feasibility of production- Improve efficiency- Module manufacture and certification- Bottlenecks.- Specification of fab equipment

15 7-11-2010

Technology improvements in pilot line

• towards up-scalability.

- processing times- number of steps- use of consumables

• Large technological advances, e.g.

- metallization- rear surface passivation

16 7-11-2010

0.4 0.6 0.8 1.0 1.20.0

0.2

0.4

0.6

0.8

1.0

Boron emitter passivation:noneSiNx

NAOS+SiNx

IQE

wavelength [µµµµm]

17 7-11-2010

Breakthrough in passivation

New technique for n-type emitter passivation:15% relative gain in efficiency

Mihailetchi et al.,Appl. Phys. Lett. 2008.

Improved metallization

18 7-11-2010

Uoc Isc FF eta

- +0.14 A +0.6% +0.4%

• Paste slumping and shading• Contact resistance• Line resistance

Original Improved

Improved rear surface passivation

19 7-11-2010

Uoc Isc FF eta

+3 mV +.04A 0% +0.2%

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2

IQE

wavelength

front illum.

rear illum.

improved

reference

From lab to fab …2009-06:

start of Panda project2009-09:

Pilot line up and running

2010-03:Yingli announces 300MW Panda expansion.

2010-06:

Yingli shows first Panda modules2010-07:

Yingli announces initial production of Panda cells and 19% efficiencyreached in pilot line.

2010-10:Yingli announces additional 600MW Panda expansion.

20 7-11-2010

… in just over a year !

What does Panda mean for PV?

21 7-11-2010

• High-efficiency industrial silicon PV technology.

• New silicon cell technology in production

• A new mainstream technology in PV may have emerged

Towards 20% … , and beyond

• Young technology:- Further improvements are possible

• Combine n-type with benefits of MWT / BC technology

22 7-11-2010

23 7-11-2010

Conclusions• ECN’s n-type concept: high efficiency with an industrial process.

• 18.65% confirmed, but > 19% has been achieved

• Involvement of Yingli has hugely accelerated development.

- “From lab to fab” in a little over a year !

Thank you for your attention

24 7-11-2010

25 7-11-2010

n-type solar cell versus p-type solar cell

• Better EQE in infrared: Higher Isc

• Less recombination: Higher Voc and Fill Factor

• Results in at least 5 – 8% efficiency gain over p-type

0

0.2

0.4

0.6

0.8

1

300 500 700 900 1100

wavelength of light (nm)

exte

rnal

qu

antu

mef

fici

ency

(EQ

E)

N-type

p-type 0

200

400

600

800

1000

1200

1400

1600

1800

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2Wavelength (um)

Irradiance

Photon flux

bandgap Si1.1eV