the multi-tw scale future for pv - university of chicago multi-tw future... · • 1905 - albert...

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

University of Chicago Physics of Sustainable Energy Short Course Greg Wilson Director, Center for Materials Applications & Performance Co-Director, National Center for Photovoltaics National Renewable Energy Laboratory Golden, Colorado - USA 17 June, 2016

The Multi-TW Scale Future for PV

Innova&on for Our Energy Future NATIONAL RENEWABLE ENERGY LABORATORY

Overview

1.  Brief introduction of NREL, the MCST Directorate and NREL’s PV Program.

2.  PV101 3.  Energy and climate change. 4.  Cell efficiency and module cost - 39 years of

progress. 5.  Enabling PV as a global carbon emissions

reduction tool. 6.  Final comments.

2


National Renewable Energy Laboratory

•  Physical Assets Owned by the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy

•  Operated by the Alliance for Sustainable Energy under Contract to DOE

•  1700 permanent staff and world-‐class faciliGes •  More than 350 acGve partnerships annually

•  Campus is a living model of sustainable energy

Dedicated Solely to Advancing Energy Efficiency and Clean Energy Research toward Enabling Deployment onto a Modernized Grid

3


NREL’s Program Portfolio

Efficient Energy Use •  Buildings Technologies •  Vehicle Technologies

Delivery & Storage •  Battery and Thermal

Storage •  Hydrogen •  Smart Grid and Grid

Integration

Renewable Resources •  Solar •  Wind and Water •  Biomass •  Geothermal

Foundational Science

Strategic Analysis

4


NREL’s PV Program: Originating as SERI in 1977, today a key component of the U.S. DOE SunShot Initiative.

5

6¢/kWh

SunShot 2020 Goal


• 20x-100x • 500x • Cu(In,Ga)Se2 ~ 1-2 um • c-Si ~ 180 um

Multiple PV Technologies

•  Ground mount with 2-axis tracking.

•  Concentration of up to 1000x. •  Utilizes the most advanced III-V

multi-junction cells. •  Highest efficiency terrestrial PV

solution - ~35%.

6

•  CdTe and CIGS are most common today. Future could be Perovskites.

•  CIGS can be very light so focus is rooftop and BIPV applications.

•  CdTe panels today are widely used in ground mount applications.

•  Many new inorganic, organic and hybrid PV absorbers are possible.

•  Silicon PV controls >90% of terrestrial PV market.

•  Consists of 2 forms, multi-crystalline and single crystal.

•  Highest efficiency rooftop PV technology is cSi - ~22.5%.


PV Research at NREL

7

NREL’s PV Research Portfolio

Measurements & Characteriza7on

Silicon cSi 1J & Tandems

III-‐V MJ & 1J

Module Reliability & Systems Engineering

PV Technologies PV Cross-‐CuDng R&D

Thin Film PV CIGS / CdTe / Novel

Perovskites & OPV

Extensive Capabilities and PV Experience Under One Roof

PV Conversion Technology R&D Ü  III-V Multijunction Ü  Crystalline Si Ü  cSi Tandem Ü  CdTe Ü  CIGS Ü Organic PV Ü  III-V 1J via HVPE Ü  Novel PV Absorbers Ü  Perovskites

Demonstrate > 50% 4J device

Demonstrate cSi based 2J cells > 30%

Materials questions, enable >16% production module

Materials questions, enable >16% production module

Build on BES prog., demonstrate market viability

Develop low-cost, 1-sun III-V cells

Build on EFRC, identify new absorbers

7"

Outdoor performance!

Power and temperature !During our test, the GaAs module produced more power when its temperature was higher

This relationship includes temperature and spectral effects

Module Performance

7

Answer industrial processing and stability questions

Develop low cost, industrial n-CZ cells >23%


Overview





8


A Brief History of Photovoltaics •  1839 - French scientist Edmond Becquerel discovers the photovoltaic

effect while experimenting with an electrolytic cell made up of two metal electrodes placed in an electricity-conducting solution—electricity generation increased when exposed to light.

•  1883 - Charles Fritts, an American inventor, described the first solar cells made from selenium wafers.

•  1905 - Albert Einstein published his paper on the photoelectric effect (along with a paper on his theory of relativity).

•  1921 - Albert Einstein wins the Nobel Prize for his theories (1904 research and technical paper) explaining the photoelectric effect.

•  1954 - Photovoltaic technology is born in the United States when Daryl Chapin, Calvin Fuller, and Gerald Pearson develop the silicon photovoltaic (PV) cell at Bell Labs. Bell Telephone Laboratories produced a silicon solar cell with 4% efficiency and later achieved 11% efficiency.

•  1958 - The Vanguard I space satellite used a small (less than one watt) array to power its radios. Later that year, Explorer III, Vanguard II, and Sputnik-3 were launched with PV-powered systems on board.

9


Physics of a PV Cell

Source: R. Swanson, SunPower

10


Source: Philips Lumileds

Phosphor

LED vs. PV Cell LED •  Applying voltage across p-n junction excites

electrons that then produce photons (light) as they fall back to their ground state.

•  The wavelength of the light produced is directly related to the bandgap of the semiconductor that is used.

PV Cell •  Light (photons) passing through the

absorber knocks electrons from their atomic orbits creating electron-hole pairs.

•  Electrons are collected by contacts on the emitter and holes (absence of electrons) are collected by contacts on the absorber.

•  The wavelength of light absorbed to produce current is directly related to the bandgap of the semiconductor that is used.

Source: R. Swanson, SunPower

11


Energy and Light Candle Light Temperature: ~ 1850K Efficiency: ~ 0.3 lm/W

Incandescent Lamp Light Temperature: ~ 3000K Efficiency: 10 - 15 lm/W

Compact Fluorescent Lamp Light Temperature: ~ 3000K Efficiency: 60 - 75 lm/W

LED Lamp (Blue LED + Phosphors) Light Temperature: ~ 3000K Efficiency: 100 - 130 lm/W

12


Bandgap?? •  The bandgap of a semiconductor is

the energy required for an electron to “jump” from its lowest energy state (valence band) to its highest energy state (conduction band).

•  For metals, the conduction and valence bands overlap so they conduct electricity.

•  For insulators, the bandgap is very large so no electricity can flow.

•  For semiconductors, the bandgap is smaller and can be manipulated by the addition of impurities called “dopants”. This is the reason a semiconductor can act like a conductor and an insulator.

Source: “Handbook of Photovoltaic Science and Engineering”, edited by Antonio Luque and Steven Hegedus (2003).

13


PV Cell Nomenclature (-)

(+)

Ele ctron Hole

Top Electrical Contact n-Type Semiconductor (Exce ss Electrons) p-Type Semiconductor (Exce ss Hole s)

Back Ele ctrical Contact Subs tra te

Top Electrical Contact Emitter Region Absorber Region

Back Electrical Contact Substrate (thin film PV cell)

Absorber – Can be made from any number of semiconductors and is commonly p-type (collects positive charges). Emitter – Usually formed by doping the semiconductor used as the absorber in order to produce the opposite type, in this example n-type (collects negative charges). Substrate – Not used in the case of wafer silicon. For thin film PV cells the substrate is usually glass or metal.

Efficiency of a PV cell depends on many things but mostly the quality of the semiconductor material, the quality of the surface “passivation” and the quality of the electrical contacts.

14


I-V Curve and PV Cell Efficiency Cell Efficiency = η η = POUT/PIN

PIN is the power of the incoming light (usually 1000 W/m2) POUT is the product of the max power voltage and current. Fill Factor is defined as the ratio of the cell’s max power, IMPxVMP and ISCxVOC.

15


N-type

Load

Reflection and optical transmission loss through the glass and/or encapsulant

Factors Affecting CURRENT

Solar cell efficiency is determined by inherent losses (i.e., solar photons with energies below the bandgap of the host material) and those primarily due to optical or electrical losses (i.e., reflectance, series resistance, etc.). Jsc, Voc, and FF are intimately and fundamentally tied. It is difficult to affect one without changing the others.

( )2/1000 mW

VJFF ocsc ⋅=ηh = solar cell efficiency Jsc = short-‐circuit current Voc = open-‐circuit voltage FF = fill factor

Electrical resistance and transmission loss of the transparent conducting materials and contacts

Charge carrier recombination (related to carrier mobility, lifetime, defect density)

Back contact and bulk series resistance

Factors Affecting VOLTAGE

Bulk and surface carrier recombination

Absorber and emitter semiconductor choice, doping, and crystallinity

Carrier Recombination in the depletion region

Factors Affecting FILL FACTOR

Series and shunt resistance

P-type

1 2

3 7 5

4 6 8

1

2

3

4

5

6

7

8

PV Cell Efficiency

16


Definitions: Energy & Power

Energy Basic unit is Joule 3.6e6 J = 1 kW-hr

Power Basic unit is Watt 1 W = J/s 1 W = V*A

∴Power is energy transferred per unit time.

Solar Constant = 1,366 W/m2

Average Earth Insolation = 1,000 W/m2 (sea level, normal incidence)

17


PV Industry Growth – By Technology

18


Overview





19

Innova&on for Our Energy Future NATIONAL RENEWABLE ENERGY LABORATORY 20

Earth Human Population = 7.3 B

Annual Electricity Demand = 23,300 TWh Annual CO2 Emissions = 32.2 Gt

Fraction of GHG Emissions from Energy Use ≈ 68%


Motivation is Clear – Energy Needs vs. CO2

21

•  Earth uses ~6 TW of electricity, ~2/3 from fossil fuels.

•  [CO2] ~402 ppm and rising. Source: 2014 U.S. National Climate Assessment

Nature - 31 March, 2016


Motivation is Clear – Energy Needs vs. CO2

22

•  Earth uses ~6 TW of electricity, ~2/3 from fossil fuels.

•  [CO2] ~402 ppm and rising. Source: 2014 U.S. National Climate Assessment

Nature - 31 March, 2016


Latest Earth System Model Results

23

Source: Tokarska et al., Nature Clim Change, 23MAY16

•  Predicted global average surface temperature has gone up, not down, with increasingly sophisticated earth system models.

•  Warming and CO2 emissions are in fact roughly linear.

•  Burning all of earth’s fossil fuel resources, projected by ~2200 given current trends, would result in global average warming of 6.4–9.5° C, mean Arctic warming of 14.7–19.5° C.


•  Must hold overall energy consumption essentially flat for the next 25 years through major gains in energy efficiency.

•  Must begin a major shift to zero carbon generation immediately, with measurable reductions by 2020.

450ppm Goal: Dramatic Change Needed

Source: 2015 IEA World Energy Outlook

24

© Fraunhofer ISE

25

2 – 6 per year

COAL

Uranium

900 Total reserve

90-300 Total

Petroleum

310 Total

Natural Gas

330 Total

WIND 60-120 per year

OTEC

Biomass

3 -11 per year

HYDRO 3 – 4 per year

TIDES

SOLAR

23,000 per year

Geothermal 0.3 – 2 per year

© R. Perez et al.

0.3 per year 2050: ~28 TW

finite renewable

Waves 0.2-2 per year

World Energy Resources (TW-yr)

25

2010 World required ~16 TW


Solar Insolation – U.S.

PV farm: 20% panels, 2-axis

trackers, Denver solar resource.

356 mi on a side supplies all of the

world’s energy needs in 2035.

26

PV farm: 20% panels, 2-axis

trackers, St. Louis solar resource.

388 mi on a side supplies all of the

world’s energy needs in 2035.


PV & Carbon Emissions Reduction

•  This paper focused on industry growth constrained by technology, mfg. capacity and demand. Energy storage, demand shifting and grid modernization not considered directly but incorporated into assumed demand curve.

•  High efficiency, advanced concept case based on 24% module efficiency, very thin absorber and ultra-low CAPEX. This case included to show potential impact of innovation alone on future PV market size.

Source: Needleman et al., E&ES, 21APR16

High efficiency with low CAPEX & mfg. costs.


Overview





28


PV Research – Dramatic Progress

29


$0.1

$1.0

$10.0

$100.0

0 1 10 100 1,000 10,000 100,000 1,000,000

Glo

bal

Ave

rage

Pow

er -M

odul

e P

rice

(20

10 $

/Wp)

Cumulative PV Module Shipments (MWp)

PV Module Experience Curve

Sources: Strategies Unlimited, Navigant Consulting & Paula Mints

SunShot 2020 Goal: 50¢/W

PV Module Cost Decline

30


US System Pricing – Utility Scale >5 MW

•  PV system costs at all scales declining rapidly. •  SunShot goal of $1/W corresponds to LCOE of 6 ¢/kWh.

31

26 MW

6 MW

685 MW

$-‐

$1

$2

$3

$4

$5

$6

$7

$8

Aug-‐07 Dec-‐08 May-‐10 Sep-‐11 Jan-‐13 Jun-‐14 Oct-‐15 Mar-‐17

Repo

rted

Pric

e (per W

DC)

Placed in Service

Tracking

Fixed Tilt

Original Source: NREL 3Q15 Solar Update, Feldman et al., 19JAN16


US System Pricing – Utility Scale >5 MW

•  PV system costs at all scales declining rapidly. •  SunShot goal of $1/W corresponds to LCOE of 6 ¢/kWh.

32

26 MW

6 MW

685 MW

$-‐

$1

$2

$3

$4

$5

$6

$7

$8

Aug-‐07 Dec-‐08 May-‐10 Sep-‐11 Jan-‐13 Jun-‐14 Oct-‐15 Mar-‐17

Repo

rted

Pric

e (per W

DC)

Placed in Service

Tracking

Fixed Tilt

Original Source: NREL 3Q15 Solar Update, Feldman et al., 19JAN16

0

50

100

150

200

250

300

Jan-‐06

Jan-‐07

Jan-‐08

Jan-‐09

Jan-‐10

Jan-‐11

Jan-‐12

Jan-‐13

Jan-‐14

Jan-‐15

PPA Execution Date

PV (7,234 MW, 100 contracts) CPV (35 MW, 2 contracts) Mix of PV/CPV (7 MW, 1 contract) CSP (1,342 MW, 5 contracts)

Levelized

PPA

Pric

e (Real 201

4 $/MWh)

550 MW

32 MW

210MW

150 MW

Original Source: Ryan Wiser, private communication, March 2016


PV Annual Production: 1990 - 2015

33


PV Industry Growth – By Technology

34


Global PV Capacity - Forecasts

35

0.0

1.0

2.0

3.0

4.0

5.0

6.0

Cumula7

ve PV Shipmen

ts (T

W)

BNEF

Energy [R]evoluGon 2015 -‐ med

WEO 2015 450 PV ProjecGon

CAGR = 15%

CAGR for last 6 years has been ~15%...Extending this to 2030 results in 2.7TW.


Overview





36


GA-SERI: The TW Workshop


0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

20.0

Cumula7

ve PV Shipmen

ts (T

W)

BNEF

Energy [R]evoluGon 2015 -‐ low

Energy [R]evoluGon 2015 -‐ med

Energy [R]evoluGon 2015 -‐ high

WEO 2015 450 PV ProjecGon

WEO 2015 450, PV at 40% of 2030 World Electricity

WEO 2015 450, PV at 20% of 2030 World Energy (TFC)

CAGR = 15%

PV a significant contributor to world electricity.

PV a significant contributor to world energy.

Low end of possible outcomes if no change (PV as “peak shaver”).

Global PV Growth & CO2 Emission Goals


Barrier: Devaluation of PV Electrons

39


Barrier: Devaluation of PV Electrons

40

This barrier can be eliminated by the development of low cost energy storage technologies… •  Li Ion Batteries •  Flow Batteries •  Thermal Storage •  Pumped Hydro •  CAES & “LAES” •  H2 Generation – Electrolysis


Multi-TW Scale PV: Projected Module Price

41

$0.1

$1.0

$10.0

$100.0

0 1 10 100 1,000 10,000 100,000 1,000,000 10,000,000

Glo

bal

Ave

rage

Pow

er -M

odul

e P

rice

(20

10 $

/Wp)

Cumulative PV Module Shipments (MWp)

PV Module Experience Curve

$0.50/W ~1 TW

$0.25/W ~8 TW


Example – Path to 3¢/kWh LCOE Target

42

Module cost, efficiency and reliability will be the focus of major new R&D efforts.



43



-‐

Developing Next-‐Gen Si Solar Cells to Enable Higher-‐Efficiency Modules at Today’s Cost

44

Silicon

FOUNDATIONAL KNOWLEDGE

INNOVATION TO APPLICATION

INTEGRATION TO IMPACT

Developed passivated tunnel contacts for advanced cell architecture

Developed Tabula Rasa wafer treatment to prevent O-precipitation

High-efficiency >23%, low-cost industrial-size cell on n-Cz wafer by 2018; currently 21.5% Exploring novel transparent and conductive micro- composites

Advanced Computer Science,

Visualization & Data

Impact entire c-Si PV market

NREL as US-based, internationally recognized Si PV research hub Creates Si industry workforce

Systems Engineering &

Integration

Decision Science & Analysis

Applied Materials Science &

Engineering

Power Systems & Electrical Engineering

Chemical Engineering

b) Traditional IBC cell

%%%%%%%%%%n"CZ%Si%Ion"implanted%

Passivated%contacts%

Tunneling%SiO2%

Intrinsic%pc"Si%

a) Interdigitated Back Passivated Contact cell (IBPC)

n"type%%pc"Si%

p"type%%pc"Si%

Doped%regions%diffused%or%%Ion"implanted%into%the%wafer%

c"Si%

Ion implanted or printed passivated contacts

n-Cz Si wafer

MIT

GIT

CSM

ASU

NIST

FhISE

UNSW


Si Tandem Cells

Source: Stefan Glunz presentation, NREL Si Workshop in Vail, CO, July, 2012

45

•  Path to > 30% efficiency for Si wafer based cells. •  Top cell requirements:

•  Lattice & CTE match to Si if epitaxially grown •  ~1.7 eV band gap

•  Perovskites may evolve into good polycrystalline choice.

1.1 eV

dNph

dλ (a

.u.)

top junction

Silicon bottom junction

11.5234 0.70.80.9photon energy (eV)

G173 global spectrum


Development of World Record GaInP/Si Dual-Junction, One-Sun Solar Cell

The device structure integrates a 1.8-eV GaInP top junction with a silicon bottom junction, with a four-terminal interconnection. The resulting device is pictured at right.

Ø  Cost-effective solar cells with efficiency greater than possible with conventional silicon could enable a very large market for low-concentration photovoltaics.

Ø  A two-junction structure with a silicon bottom junction is an attractive path to this goal.

Ø  NREL developed a new device structure combining a III-V GaInP top junction and a silicon bottom junction, and demonstrated a world record 29.8% efficiency – significantly exceeding the best conventional silicon efficiency of 25.6%.

Ø  The four-terminal structure allows ease of construction, and optimal energy production under real-world operating conditions.

Ø  We are presently developing an improved, manufacturable bonding technique to enable transfer of this structure to industry.

S. Essig et al., Energy Procedia 77, p. 464 (2015).


-‐

IntegraBng CapabiliBes to Bring Advanced Technology from Space to Earth

47

High-Efficiency Multi-junction Solar Cells

Superstructure ordering modifies the energy band and tailored opGcal properGes

Devices with “engineered” mulGlayered structures have ~46% conversion efficiency

All commercial cells for space and concentrator PV based on NREL mul7junc7on technology

4.2

4.0

3.8

3.6

3.4

3.2

3.0

Volta

ge (

V)

0.1 1 10 100 1000

Concentration (Suns)

48

46

44

42

40

38

36

Effic

ienc

y (%

)

90

88

86

84

82

Fill

Fact

or (

%)

MM966n5 NREL AIST Model

Voc

Vmax

Junc. #3 GaInAs 1.0 eV

Junc. #1 GaInP 1.8 eV

Junc. #2 GaAs 1.4 eV Graded buffer

Junc. #4 GaInAs 0.7 eV

Graded buffer

handle

understood, the basic results can be extended to other or-dered III-V alloys.

II. SAMPLE GEOMETRY

The sample configuration chosen for this study (Figure 1)consists of In-rich GaxIn1!xP grown compressively strained ona GaAs (001) substrate miscut toward !111ð Þ B. In this paper,the names “Bþ” and “B!” are linked to the asymmetry intro-duced by the sample miscut: a compressively strained epilayerrotates so as to have a higher average surface miscut angle forrelaxation on the Bþ plane, and a smaller average surface mis-cut angle for relaxation on the B–plane.11 These rotations are auseful probe of relaxation, because the net epilayer "tilt" inx-ray diffraction (XRD) plots can be used to determine relativerelaxation on the two planes.

Ordering introduces a second asymmetry. For the miscutdirection shown, a single CuPt ordered variant is obtained inwhich adjacent B! planes alternate between being In-richand Ga-rich. This ordering is stabilized near the growth sur-face by P-P dimers under certain conditions.12 Once formed,it is metastable in the bulk and can be destroyed by anneal-ing.13 As shown in Figure 1(b), dislocations which glideacross the ordered planes can disrupt the ordering pattern. Adislocation gliding along a Bþ plane creates an APB; glidealong a B! plane does not. Glide of dislocations on the!1!11ð Þ and 111ð Þ A planes (not shown) can also create APBs,

but only if the corresponding Burgers vector is also co-planar with the Bþ plane ( 011½ & or !101½ &). If the Burgers

vector is coplanar with the B! plane ð½01!1& or ½!10!1&Þ, noAPB will result. Later in this paper, we will show how theseasymmetries can cause the glide plane to switch between B!and Bþ, but to do this we must first compute EAPB.

III. ANTIPHASE BOUNDARY ENERGY (EAPB)

To compute EAPB, ab initio total-energy calculationswere performed for a supercell of fully ordered Ga0.5In0.5P,first without an APB, and then with APBs of various spac-ings. Total energies were calculated using the screenedHeyd-Scuseria-Ernzerhof (HSE) hybrid density functional,14

as implemented in the VASP package.15 Our DFT calcula-tions employed the projector augmented wave method16 withan energy cutoff of 270 eV for the plane wave part of thewave function. The atomic coordinates were fully relaxed.To obtain EAPB values for widely spaced APBs (for whichthe APB-APB interaction energy is negligible), the APBspacing was increased until EAPB reached a constant value;this occurred at an APB spacing of '40 A.

Figure 2 shows the resulting atomic structure near anAPB viewed along the plane of the APB. No dislocation isshown; this is the APB structure which remains after a dislo-cation glides along the Bþ plane. Because these calculationswere performed for fully ordered material (ordering parame-ter g¼ 1), the resulting EAPB value is an upper limit whichwill be denoted Emax

APB. However, even under growth condi-tions favorable for ordering, the order parameter forGa0.5In0.5P is typically 0.4, and decreases as the alloy com-position moves away from x¼ 0.5. Therefore, the APBenergy for partially ordered material will be approximatedwith an g2 scaling factor: EAPB¼Emax

APB ) g2.1,17

Calculated values of EmaxAPB for three representative III-V

alloys are shown in Table I. Negative values indicate that anAPB-creating dislocation will lower the energy of the systemas it glides, thereby providing a driving force for dislocationglide sAPB¼ –EAPB, where sAPB is a force per unit disloca-tion line length and EAPB is an energy per unit area. Usingspecial quasi-random structures3,18 or random alloy models,the energy released by disordering a unit volume of eachalloy (Edisordering) was also calculated. It is interesting to notethat Emax

APB and Edisordering are roughly proportional to each

FIG. 1. (a) Sample geometry, with the size of the substrate steps exaggeratedfor the purposes of illustration. In the actual experiment, the epilayer thick-ness of 0.25 lm is much greater than the step height. (b) Schematic diagramof the ordering pattern corresponding to the miscut direction shown in (a).Black layers are In-rich; white layers are Ga-rich. APBs can be created byglide on the Bþ plane, but not the B! plane. The supercell used for calcula-tions is a longer version of the shaded rectangle, containing two equallyspaced APBs so as to be periodic, and located in bulk-like material far fromboth the surface and the dislocation core (Figure 2).

FIG. 2. Relaxed atomic structure of an APB on a Bþ plane in fully orderedGaInP2. Distortion of the bond angles can be seen near the APB. Blackatoms are In, white atoms are Ga, and green-shaded atoms are P. The (green)shaded rectangle is the central portion of the widest supercell used in ourcalculations. Each supercell contains two APBs so as to be periodic. Thisfigure is rotated counter-clockwise with respect to Figure 1.

203506-2 McMahon et al. J. Appl. Phys. 114, 203506 (2013)

[This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP:192.174.37.50 On: Mon, 25 Nov 2013 18:19:25

CaGon site order affects DislocaGon glide energeGcs in alloys – allows raGonal design of mulGjuncGons



R&D 100 Awards: Ø  2012, with Solar

Junction Ø  2008, with Emcore Ø  2001, with Spectrolab


Advanced Computer Science, Visualization

& Data


Integration



Engineering

Power Systems & Electrical

Engineering

Chemical Engineering


High-Efficiency III-V Solar Cells at Unprecedented Low Costs

Schematic of an in-line HVPE reactor with continual substrate reuse that eliminates metal-organic sources and uses cheap elemental metals. Our reactor is a major step to this ultimate goal.

Ø  High efficiency is critical to lowering photovoltaic costs. III-V PV is the most efficient, but most expensive. We aim to radically lower III-V costs to make III-V cells the preferred photovoltaic technology.

Ø  Our approach to reduction of III-V growth cost is use of hydride vapor-phase epitaxy (HVPE), which drastically lowers both capital and materials costs while maintaining high efficiency.

Ø  We also address cost of the expensive substrates, through strategies for reusing them.

Ø  We have developed and are operating a novel HVPE reactor capable of growing >25% solar cells; 20.6% already demonstrated.

Simon et al., IEEE J. Photovolt. v.6, p. 191 (2016); Schulte et al., J. Cryst. Growth v. 434 p. 138 (2016)

48


at Today’s Cost

-‐

CdTe Solar Technology Leading the Way to Grid Parity

Cadmium Telluride







Integration



Engineering

Power Systems & Electrical Engineering

Chemical Engineering Atomistic

calculations and experiment indicate that Te on Cd sites limit lifetime

After years of empiricism, NREL has been systematically studying the fundamentals of II-VI material

These improve-ments helped NREL break the 60-year voltage barrier

By shifting to Cd-rich stoichiometry and placing P on Te sites, we have attained defect-free lifetimes with ideal conductivity for solar cells

A small group of companies and research groups has led to CdTe competing directly with conventional energy sources

NREL contributed to CdTe technology for 20 years and held world-record efficiency for 10 years

First Solar REEL

Solar LucinTech

EPIR Calyxo

49


Organic-Inorganic Lead Halides Perovskite

50

Perovskite PV Cells



51



0

50

100

150

200

250

Freq

uency

Degradation Rate (%/year)

Median: 0.5 %/yearAverage: 0.8 %/year# reported rates = 2128

Reducing Cost of PV by Increasing PV LifeBme and Confidence in Long-‐Term Performance

52

PV Reliability




Fundamental understanding of adhesion, UV aging, moisture ingress, thermal management, reverse bias

Materials Science & Engineering

Study field and accelerated tesGng of PV products to quanGfy performance and long-‐term reliability Regional Test Centers are managed jointly with Sandia

Mechanical Design & Engineering



Develop internaGonal standards in partnership with manufacturers, test labs, installers, and internaGonal standards organizaGons


Integration


Power Systems & Electrical

Engineering

Bankable PV

Consistent manufacturing

System verification

Stan

dard

s

Stan

dard

s

Sta

ndar

ds

Durable design

NREL is a national laboratory of the U.S. Department of Energy Office of Energy Efficiency & Renewable Energy Operated by the Alliance for Sustainable Energy, LLC

This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications.

Contract No. DE-AC36-08GO28308

Updated Proposal for a Guide for Quality Management Systems for PV Manufacturing: Supplemental Requirements to ISO 9001-2008 Govind Ramu,1 Masaaki Yamamichi,2 Wei Zhou,3 Alex Mikonowicz,4 Sumanth Lokanath,5 Yoshihito Eguchi,6 Paul Norum,7 and Sarah Kurtz8 1 SunPower 2 National Institute of Advanced Industrial Science and Technology (AIST) 3 Trina Solar 4 Powermark 5 First Solar 6 Mitsui Chemical 7 Amonix 8 National Renewable Energy Laboratory

Technical Report NREL/TP-5J00-63742 March 2015


Low Cost Energy Storage

53

For 2 growth scenarios beyond peak shaving, R&D directed at storage in all forms will also be needed.


Li Ion Batteries - Cost Declines

54

Source: Nykvist & Nilsson, Nature Climate Change, 23MAR15


Future Energy System – Commodity H2

The ESIF helps NREL contribute to the U.S. Department of Energy’s and our nation’s energy goals by providing the R&D capabilities needed to collaborate with private industry, academia, government, and public entities on utility-scale projects focused on solving issues in relation to grid integration of renewable energy and other efficiency technologies.

Future H2 at Scale Energy System

Wind

Solar

ElectricityGrid

Value AddedApplications

H2

+–

Nuclear

Concentrated Solar Power

Generator

Value AddedApplications

OtherEnd Use

MetalsRefining

UpgradingOil /

Biomass

NH3

VehicleFuel

SyntheticFuels

Hydrogen/Natural Gas

Infrastructure

HydrogenStorage/

Distribution

HydrogenGeneration

Fuel Cell,Combustion

INDUSTRIA

L

TRAN

SPORTATIO

N

N2

CO2

N2

CO2

Battery

WHY HYDROGEN?• Hydrogen is an ideal clean energy

carrier—connecting diverse energy sources to diverse applications

• It can play a unique and critical role in addressing many of the energy sector’s greatest challenges

TODAY’S ENERGY SYSTEM• Renewable energy—particularly wind

and solar—offer great promise but have challenges associated with variable and concurrent generation

• Options to achieve deep decarbonization while meeting society’s multi-sector energy demands are limited, particularly in the industrial and transportation sectors

FUTURE H2 AT SCALE ENERGY SYSTEM• Connects low-carbon energy sources to

all of the energy sectors

• Uses carbon-free, renewable inputs to service all of society’s energy needs, in particular the difficult to decarbonize sectors of industry and transportation

• Does not compete with other options—rather, it enables increased renewable penetration

• Can decrease 45% of all U.S. carbon emissions by 2050

55


Overview





56


PV & Storage R&D – Not the Only Problem

PV Cell & Module

Technology

Public Policy

Regulated U7lity

Markets &

Finance

Grid Integra7on & Energy Storage

57


Visit us online at www.nrel.gov

Thank You

58

the multi-tw scale future for pv - university of chicago multi-tw future... · • 1905 - albert...

Documents