Thin-Film vs. Crystalline 400W Solar Module | How Should you Choose

Thin-film efficiency 12-15%, 400W approx. 3-4㎡ (5-8kg), flexible and bendable, suitable for RV curved roofs;

Monocrystalline efficiency 20-23%, approx. 1.8-2㎡ (6-9kg), high efficiency saves space, first choice for fixed installation.

Record Efficiency

Crystalline silicon like SunPower IBC mass production efficiency 22.8% (400W model 21.2%), NREL data shows TOPCon lab efficiency 26.1%;

Thin-film First Solar CdTe mass production 19.7% (400W 18.1%), Oxford PV perovskite tandem lab 33.7% (2023 production line sample 28.3%).

Temperature coefficient: crystalline silicon -0.4%/℃, thin-film -0.25%/℃. Thin-film gains 10%-15% under low light, efficiency difference amplified by environment.

Mass Production Efficiency

What supports crystalline silicon's mass production efficiency?

Crystalline silicon 400W module mass production efficiency reaches 20%-23%, relying on mature process details.

First look at technology routes: PERC is still mainstream, but TOPCon and HJT are competing for market share.

SunPower's Maxeon 6 uses IBC (Interdigitated Back Contact) technology, placing both positive and negative electrodes on the back, front side not shaded, mass production efficiency 22.8%, 400W model efficiency 21.2%.

Their Maxeon 5 launched in 2019 had 21.5% efficiency, gained 1.3 percentage points in three years, mainly by optimizing silver paste printing precision for back electrodes.

REC's Alpha Pure-R follows the TOPCon route, mass production efficiency 22.3%, 400W efficiency 20.9%.

Their factory in Norway uses LPCVD (Low-Pressure Chemical Vapor Deposition) for tunnel oxide layer, thickness controlled at 1.2 nanometers, absorbs less light than traditional 1.5nm, plus phosphorus diffusion optimization, open-circuit voltage increased 5mV, boosting efficiency.

NREL 2024 report shows mainstream PERC 400W module efficiency mostly 20.5%-21%, TOPCon 1-1.5 percentage points higher than PERC, HJT (Heterojunction) even higher, e.g., Japan Kaneka's HJT module mass production efficiency 23.2%, 400W model efficiency 21.8%, due to intrinsic amorphous silicon layer passivation, surface recombination velocity reduced below 50cm/s.

For example, using 182mm or 210mm large-size wafers (M10/G12), more cells can fit in the same module area, SunPower's G12 wafer 400W module efficiency is 0.3% higher than the M6 wafer same power model.

Half-cell technology cuts cells in half, reducing current loss, REC's half-cell TOPCon module efficiency 0.2% higher than full-cell.

Shingling technology overlaps cells, saves frame space, LG NeON 6 shingled 400W module efficiency 22.1%, 0.4% higher than conventional version.

Why is thin-film mass production efficiency always lagging?

Thin-film 400W module mass production efficiency currently 17.5%-19.7%, 2-3 percentage points lower than crystalline silicon, stuck on materials and processes.

First Solar's Series 7 represents Cadmium Telluride (CdTe), mass production efficiency 19.7%, 400W model efficiency 18.1%.

They use glass substrate, the deposition process is close-spaced sublimation (CSS), vapor-depositing CdTe and Cadmium Sulfide (CdS), film thickness controlled to 3-5 micrometers.

After the 2023 upgrade, the deposition rate increased from 2μm/min to 3μm/min, yield from 92% to 95%, efficiency only increased 0.5 percentage points.

But CdTe's bandgap is fixed (1.45eV), unlike crystalline silicon which can be tuned, absorption spectrum narrower, efficiency lower.

Solar Frontier's CIGS (Copper Indium Gallium Selenide) module mass production efficiency 18.9%, 400W efficiency 17.5%.

They use stainless steel flexible substrate, roll-to-roll deposition, first create molybdenum back electrode, then sequentially deposit CIGS absorption layer, ZnO window layer.

The problem lies in large-area uniformity – on a 1.1m wide substrate, CIGS composition difference between edge and center 0.1%, efficiency drops 0.3%.

In 2024, their factory in Fukushima, Japan added online optical inspection, real-time sputtering power adjustment, yield improved from 88% to 91%, efficiency barely reached 19%.

First Solar's 100MW production line equipment investment 120 million, crystalline silicon PERC line only 80 million, allocated per module, thin-film equipment depreciation 30% higher than crystalline silicon, so even if efficiency is similar, thin-film system costs 15%-20% more expensive.

2% efficiency difference on a 400W module, how much less electricity generated per year?

Efficiency difference is not just numbers on paper; calculated to power generation, it's clear. Examples: Phoenix, USA (annual sunshine 3,800 hours) and Munich, Germany (annual sunshine 1600 hours).

Phoenix installs two 400W modules, one crystalline silicon efficiency 21.5% (actual generation 86kWh/Wp/year), one thin-film efficiency 18.5% (actual generation 74kWh/Wp/year).

After a year, crystalline silicon generates (86-74)×400=4,800 kWh more than thin-film. At local electricity price 0.12/kWh, earns 576 more per year.

Crystalline silicon efficiency 21.5%, maintains 70% efficiency under low light, actual generation 52kWh/Wp/year;

Thin-film efficiency 18.5%, maintains 85% under low light, actual generation 59kWh/Wp/year.

400W module generates (59-52)×400=2,800 kWh more per year, at €0.30/kWh, earns €840 more.

NREL 2023 actual measurement data: same 400W module, 2% efficiency difference, annual power generation difference 5%-7% in tropical regions, 8%-10% in high latitudes.

This is not trivial; a large ground-mounted power plant of 100MW, 2% efficiency difference means 5 million kWh less per year.

New developments in mass production efficiency improvement

REC's 2024 Alpha Pure-RX, uses TOPCon+SE (Selective Emitter), 400W efficiency increased to 21.5%, achieved by laser doping to locally increase emitter concentration, contact resistance reduced 20%.

On the HJT side, Switzerland's Meyer Burger 400W module efficiency 23.1%, uses low-temperature silver paste + busbar-less design, saves 20% silver paste, cost reduced $0.03/W.

UK's Oxford PV perovskite-silicon tandem module, 2023 100MW production line sample efficiency 28.3%, 400W model efficiency 26.5%, but yield only 80%, not yet meeting mass production standards.

USA's Swift Solar uses roll-to-roll printed perovskite, 2024 lab efficiency 25.8%, claims 2026 launch of 400W mass production module, efficiency 24%.

Current mass production efficiency table (mid-2024):

l Crystalline silicon PERC: 20.5%-21% (400W)

l Crystalline silicon TOPCon: 22%-22.8% (400W)

l Crystalline silicon HJT: 23%-23.5% (400W)

l Thin-film CdTe: 18.1%-19.7% (400W)

l Thin-film CIGS: 17.5%-18.9% (400W)

l Thin-film perovskite (early mass production): 24%-25% (400W sample)

Temperature and Low Light

How does temperature affect efficiency?

Crystalline silicon coefficients generally -0.3% to -0.4%/℃. For example, SunPower Maxeon 6 (IBC technology) is -0.29%/℃, REC Alpha Pure-R (TOPCon) is -0.32%/℃, Japan Kaneka HJT module slightly better, -0.28%/℃.

Thin-film coefficients significantly lower, -0.2% to -0.25%/℃.

First Solar Series 7 (CdTe) is -0.25%/℃, Solar Frontier (CIGS) is -0.22%/℃, Oxford PV perovskite tandem sample even -0.18%/℃.

Thin-film materials have wider bandgap (e.g., CdTe bandgap 1.45eV, crystalline silicon 1.1eV), electron-hole pair recombination slower at high temperatures, efficiency holds steady.

Who performs better under low light?

Crystalline silicon maintains 70%-75% efficiency, thin-film 85%-90%, difference 10-15 percentage points, significant in rainy regions.

Test conditions: Using solar simulator, irradiance reduced from 1,000W/m² to 200W/m², measure module output power proportion. Crystalline silicon e.g., LG NeON 6 (PERC) maintains 72%, REC TOPCon 74%; thin-film First Solar CdTe maintains 88%, Solar Frontier CIGS 90%, Oxford perovskite sample 92%.

Actual case: Bergen, Norway (annual sunshine 1,200 hours, overcast/rainy days 60%), install 400W modules. Crystalline silicon efficiency 21.5%, low-light maintenance 72%, actual annual generation: 400W×21.5%×1,200h×72%=74,304Wh=74.3kWh; Thin-film efficiency 18.5%, maintenance 88%, annual generation: 400W×18.5%×1,200h×88%=78,144Wh=78.1kWh. Thin-film generates 3.8kWh more per year, at €0.30/kWh, earns €1.14 more per module. 100,000 modules earn €114,000 more per year.

Munich, Germany (annual sunshine 1600 hours, overcast/rainy days 40%): Crystalline silicon annual generation: 400W×21.5%×1600h×75%=103,200Wh=103.2kWh; Thin-film: 400W×18.5%×1600h×87%=103,008Wh≈103kWh, gap narrows because more sunshine, low-light impact proportion decreases.

Technical principles:

CdTe, CIGS have larger bandgap than crystalline silicon, thermally excited carriers fewer, recombination rate lower at high temperatures, thus smaller temperature coefficient.

Simultaneously, their absorption edge is more red-shifted (can absorb more infrared light), low-energy photons can be utilized under low light, and the maintenance rate is higher.

Crystalline silicon wanting to improve these points needs to change material or structure. E.g., HJT adds intrinsic amorphous silicon layer passivation, temperature coefficient drops to -0.28%/℃, but still higher than thin-film.

For low light, PERC adds back surface field reflection, maintenance rate increases to 75%, but thin-film still leads by 10 percentage points.

Data table: Temperature and Low Light Performance Comparison

Data source: NREL outdoor measurement report (2024), Fraunhofer ISE low-light test database

What to consider in actual selection?

If installing in deserts (e.g., Riyadh, Saudi Arabia, summer 45℃), inland Australia, thin-film power generation at high temperatures can surpass same power crystalline silicon by 3-5%;

If installing in Northern Europe (e.g., Stockholm, Sweden, 200 overcast days per year), thin-film low-light advantage increases annual power generation by 8-10%.

Conversely, in cooler regions (e.g., Vancouver, Canada, average temperature 10℃), crystalline silicon's absolute efficiency higher, more cost-effective.

Low-Light Performance

Tests show: At 200W/m² irradiance, thin-film modules output 60%-70% of rated power, crystalline silicon only 40%-50%;

Thin-film temperature coefficient 0.25%/℃ (crystalline silicon 0.4%/℃), retains 15% more power under low light in summer.

In cloudy cities (e.g., Seattle, USA), thin-film power generation loss ≤15%, crystalline silicon ≥25%, directly impacting annual generation.

Technical Differences

Spectral absorption:

l Crystalline silicon weakness: Monocrystalline silicon absorption peak in 400-700nm visible light region, almost no response above 800nm infrared (quantum efficiency <30%). NREL 2022 spectral response test shows, at 300W/m² low light, crystalline silicon utilization of 900nm infrared only 28%, while CdTe thin-film reaches 62%.

l Thin-film advantage: CdTe absorption edge at 850nm, CIGS extends to 1100nm, perovskite even to 1200nm. Fraunhofer ISE using a spectrophotometer measured, same 300W/m² light, thin-film quantum efficiency 85% at 600nm (yellow light), 60% at 900nm (near-infrared), crystalline silicon corresponding 78% and 30%.

l Actual impact: On cloudy days, the infrared light proportion is 15%-20% higher than on sunny days (NOAA atmospheric radiation data), and the thin-film thus captures more of this "escaped" light.

No PN junction:

l Crystalline silicon PN junction loss: PN junction depletion region width narrows with decreasing light, low carrier concentration under low light, electron-hole pairs easily recombine outside depletion region. EPFL (Swiss Federal Institute of Technology) using fluorescence imaging observed, at 100W/m² low light, crystalline silicon recombination rate 2.1 times that of thin-film.

l Thin-film PIN alternative: Amorphous silicon (a-Si) uses PIN structure, intrinsic layer (I-layer) thickness 0.3-0.5μm, internal electric field uniform, carrier drift distance long (>1μm). Fraunhofer ISE 2022 carrier lifetime test: thin-film lifetime 12ns at 100W/m², crystalline silicon only 5ns.

l Data evidence: TÜV Rheinland low-light IV curve analysis, thin-film short-circuit current (Isc) at 200W/m² is 1.4 times that of crystalline silicon, mainly due to less recombination.

Temperature impact:

l Crystalline silicon temperature sensitivity: Silicon's bandgap decreases with temperature increase (0.00045 eV/℃), causing open-circuit voltage (Voc) to drop quickly. NREL 2023 test at different temperatures: 25℃ crystalline silicon efficiency 19.8%, 60℃ drops to 16.2% (3.6 percentage points drop); thin-film CdTe same conditions from 18.5% to 17.1%.

l Temperature coefficient difference: Thin-film average 0.25%/℃, crystalline silicon 0.35%-0.45%/℃ (monocrystalline PERC 0.38%, polycrystalline 0.40%). Calculating with summer module surface temperature often reaching 55-60℃, thin-film loses 10%-15% less power than crystalline silicon.

l Field case: Arizona State University tracking 10 systems found, summer cloudy day (average temperature 32℃), thin-film modules daily generation 18% higher than crystalline silicon.

Surface reflection:

l Crystalline silicon reflectivity: Conventional monocrystalline silicon uses pyramid texture + silicon nitride anti-reflection coating, reflectivity ~5% (AM1.5 standard light). Under low light when intensity is low, reflection loss proportion higher.

l Thin-film anti-reflection design: CIGS uses Mo back electrode + ZnO transparent conductive film, surface smoother, paired with nanoscale texture, reflectivity 3%-4%. CdTe directly uses glass substrate + CdS buffer layer, reflectivity 2.8%.

l Effect comparison: Under the same low light, thin-film absorbs 7% more photons due to lower reflectivity, equivalent to 5% more output of rated power.

Material thickness:

l Crystalline silicon thickness disadvantage: Silicon wafer thick, carrier diffusion length needs >100μm to be collected, under low light diffusion speed slow (Einstein relation: diffusion coefficient D=μkT/q, low light μ decreases), easily recombined. EPFL micro-region current mapping shows, crystalline silicon current collection efficiency at edges only 60%, center 80% at 100W/m².

l Thin-film thickness advantage: CdTe thickness 2-3μm, carrier diffusion distance 0.5μm (far less than thickness), almost all collected. CIGS thinner (1-2μm), perovskite only 0.5μm, low-light collection efficiency >95% (NREL carrier dynamics research).

l Data validation: In 400W module, thin-film due to thin thickness, effective generation area under low light 12% more than crystalline silicon (calculated by carrier collection efficiency).

Application Scenarios

Northern Europe winter

Stockholm, Sweden (latitude 59°N, winter sunshine <6 hours/day), Trondheim, Norway (latitude 63°N, December average sunshine 2.8 hours/day), these places have low sun angle, 40% annual generation relies on dawn/dusk low light (300W/m²).

l Thin-film module performance: CdTe modules maintain 55%-65% rated power output at 100-300W/m² low light (Fraunhofer ISE 2023 field monitoring). Stockholm 10kW power plant tracked 5 years, winter (Nov-Mar) thin-film generation 28% higher than crystalline silicon, because quantum efficiency is high under dawn/dusk twilight (when infrared light proportion is high, thin-film QE 20%+ higher than crystalline silicon).

l Crystalline silicon weakness: Monocrystalline silicon output only 35%-45% at 200W/m², winter generation loss 30%-35% (NREL Nordic project report). Trondheim household power plant case: crystalline silicon winter daily average generation 8kWh, thin-film reaches 11kWh, difference 3kWh enough for evening peak heating electricity.

l Key data: High latitude regions (north of 50°N) winter, thin-film annual generation 15%-20% more than crystalline silicon (calculated by NREL regional model), mainly because low-light duration proportion is high (40%+ of annual generation period).

US Northwest

Seattle, USA (annual overcast days 158), London, UK (annual overcast days 145), these regions annual sunshine <2000 hours, 70% generation days in low-light environment with >50% cloud cover (200-500W/m²).

l Thin-film advantage: CdTe modules annual generation in Seattle 1.12kWh/W (rated power), crystalline silicon only 0.92kWh/W (21% difference), because cloudy days thin-film temperature coefficient low (0.25%/℃), summer 30℃ loses 12% less power than crystalline silicon (TÜV Rheinland test). London commercial power plant data: thin-film low-light day (cloud cover >70%) generation loss ≤12%, crystalline silicon ≥25%.

l Crystalline silicon limitation: Polycrystalline silicon prone to increased reflectivity (from 5% to 7%) due to surface condensation on cloudy days, further reducing low-light efficiency (EPFL environmental simulation experiment). Seattle community power plant comparison: thin-film rainy season (Oct-Apr) total generation 18MWh/year more than crystalline silicon.

l Installation adaptation: Thin-film low tilt angle (15°-20°) installation, low-light efficiency 17% higher than crystalline silicon (Munich University of Technology rooftop experiment), suitable for Seattle's many sloped roofs.

Desert Edge

Phoenix, Arizona, USA (summer average temperature 38℃, winter 15℃), Alice Springs, inland Australia (annual sunshine 3,800 hours, day-night temperature difference >20℃), these places midday strong light (1,000W/m²) efficiency high, but dawn/dusk low light (200-400W/m²) proportion 30%.

l Thin-film supplementary light role: CdTe output reaches 60% rated power at dawn/dusk (6:00-8:00, 18:00-20:00) low light, crystalline silicon only 40%. Arizona State University tracking 10 systems: thin-film annual comprehensive generation 5% higher than crystalline silicon, because morning/evening supplementary power offsets crystalline silicon's slight midday advantage (crystalline silicon midday efficiency 8% higher, but low light 20% worse).

l Crystalline silicon midday burst capability: Monocrystalline PERC efficiency 22% at 1,000W/m², 3.5 percentage points higher than thin-film (18.5%) (NREL standard test), but only evident during 3 midday hours. Alice Springs farm power plant: crystalline silicon 2-hour midday generation 35% of daily total, thin-film 28%, but thin-film daily total generation 3% more (low-light cumulative effect).

l Temperature impact: Desert daytime module surface temperature 60℃+, thin-film temperature coefficient low (0.25%/℃), loses 13% less power than crystalline silicon (0.38%/℃) (Fraunhofer ISE desert measurement).

European old cities

Berlin, Germany (city roof average tilt angle 10°-15°), Brooklyn, New York, USA (apartment roofs often have chimney/antenna shading), these places fixed installation angles, local shading common, low-light anti-shading capability more important.

l Thin-film anti-shading advantage: CdTe series resistance low (0.5Ω·cm²), local shading power loss <10% (TÜV Rheinland shading test); crystalline silicon due to fewer bypass diodes, hot spot effect can cause up to 25% loss (Berlin apartment power plant case: thin-film shaded 20% area by chimney, generation only drops 8%; crystalline silicon drops 22%).

l Small space efficiency: Thin-film higher low-light output per unit area, 400W module at 200W/m² generates 12% more power than same power crystalline silicon (NREL space optimization report). Brooklyn rooftop power plant: thin-film installs 20 modules (8kW), annual generation 5% more than crystalline silicon's 22 modules (8.8kW), because low-light efficiency high + no shading loss.

l Low tilt angle adaptation: Berlin roof average tilt 12, thin-film low-light efficiency 15% higher than crystalline silicon at 10° tilt (German VDE certification test), no need for extra mount adjustment.

US Midwest

Western Texas, USA (annual dust storm days 30+), New South Wales, Australia (haze season PM2.5 >50μg/m³), dust coverage reduces light transmittance, impact more obvious under low light.

l Thin-film self-cleaning potential: CIGS surface hydrophobic (contact angle >90°), dust accumulation 15% less (Fraunhofer ISE dust experiment), low-light transmittance retention 8% higher than crystalline silicon. Texas ranch power plant: thin-film after dust accumulation, low-light efficiency only drops 5%, crystalline silicon drops 13%.

l Anti-reflection stacking advantage: Thin-film itself reflectivity 2.8%-4% (crystalline silicon 5%), after dust accumulation reflectivity increases smaller (thin-film rises 1.2%, crystalline silicon rises 2.5%), under low light retains 7% more photons (NREL optical simulation). New South Wales power plant tracking: thin-film haze day generation loss 10% less than crystalline silicon.

Installation Adaptability

Crystalline silicon heavy 22-28kg, thick 3.5-4cm, requires ≥20kg/m² load-bearing surface, efficiency 22-24%;

Thin-film only 10-15kg, thick 0.3-0.5cm, slightly bendable (bend radius >1m), efficiency 14-18%, low-light shading loss reduced 30-50%, high temperature (-0.25%/-0.3%/℃ coefficient) daily average generates 1.2kWh more per module.

Physical Properties

How heavy?

Crystalline silicon 400W modules use 3.2mm tempered glass + aluminum alloy frame, single module weight concentrated in the 22-28kg range – equivalent to two full water buckets.

Florida, USA project measurement shows, when installing on old color steel roof (load limit 15kg/m²), each square meter needs to allocate 2 modules, total weight reaches 44-56kg/m², far exceeding roof load capacity, forced to add steel support frames, extra cost $35 per square meter.

Taking CIGS technology as example, the 400W model weighs only 10-15kg (about 50% of crystalline silicon), thickness compressed to 3-5mm (similar to a mobile phone thickness).

In Bavaria, Germany, a wooden farmhouse roof (load-bearing 12 kg/m²) installation, thin-film modules laid directly without reinforcement, load per square meter only 20-30 kg/m², saving mount cost $1,200/project.

Weight difference also affects transport efficiency: Crystalline silicon truck load capacity 30% less than thin-film, but efficiency high, $0.08/W overall system cost lower.

Size and flexibility determine scenarios

Crystalline silicon module standard size approx. 2000×1000×35mm, rigid structure makes it non-bendable, only suitable for flat or <30° tilt roofs.

If encountering curved silo (e.g., Iowa, USA project), need to cut modules and customize mounts, material loss rate 15%, installation period extended 5 days.

Thin-film module size more flexible, common specification 1800×900×3mm, key advantage is slight bendability (bend radius >1m).

Spain Andalusia greenhouse project laid them on arched roof (bend radius 1.5m), fit rate 98%, compared to crystalline silicon custom solution saved $5,000.

Thickness advantage also evident in irregular installation surfaces: Oslo, Norway ship top installation, thin-film modules can conform to hull curvature, while crystalline silicon needs cutting into fan shape, edge sealing failure risk increased 20%.

Material determines durability

Crystalline silicon module front side low-iron tempered glass (light transmittance >91%), back side TPT composite material, frame anodized aluminum (thickness 1.2mm).

Lab tests show, can withstand 25mm diameter hail impact at 23m/s speed (equivalent to hail impact under category 10 wind), but glass shatters leading to moisture ingress, entire module scrapped.

Thin-film modules use multilayer coating structure: stainless steel foil or polymer substrate deposited with CIGS/CdTe absorption layer, surface covered with fluorocarbon resin protective layer.

Arizona desert test shows, wind/sand abrasion resistance 40% higher than crystalline silicon, but surface easily scratched by sharp objects (e.g., branch scrape).

However, some high-end products (e.g., First Solar Series 6) use anti-scratch coating, surface hardness increased to 3H pencil hardness, minor scratches affect generation efficiency <3%.

Load-bearing and space actual measurement

Installation convenience can be compared via specific scenario data. On industrial roof with 20 kg/m² load-bearing (Ohio, USA case):

Crystalline silicon module requires 4 people to carry (single 22 kg), each installation takes 25 minutes (including mount fixing);

Thin-film module 2 people can carry (single 12 kg), no-mount laying each only 15 minutes.

Space utilization: Crystalline silicon efficiency 22.5%, 100㎡ roof can install 26 modules (total power 9.75kW);

Thin-film efficiency 17% only installs 19 modules (total power 7.6kW).

But in shaded environment (e.g., California orchard project), thin-film modules due to better low-light response, the actual generation gap narrows to 12%.

Temperature coefficient difference more obvious: Crystalline silicon -0.38%/℃, thin-film -0.29%/℃, summer high temperature thin-film daily average generates 1.1kWh more per module (calculated for 400W).

Can it withstand extreme environments?

US National Renewable Energy Laboratory (NREL) extreme environment tests on two module types show:

l Low temperature test (-40℃): Crystalline silicon frame shrinkage causes encapsulant cracking probability 8%, thin-film polymer substrate shrinkage <0.5%, no cracking;

l Salt spray corrosion (coastal environment): Crystalline silicon aluminum alloy frame annual corrosion rate 0.02mm, thin-film stainless steel substrate almost no corrosion;

l Mechanical load: Crystalline silicon static load (2400Pa) deformation 1.2mm, thin-film (3600Pa) only 0.8mm.

Space Utilization

How many modules can fit on a piece of land?

Crystalline silicon 400W modules use PERC technology, mass production efficiency 22-24% (monocrystalline highest 24.5%).

Houston, Texas, USA flat roof project, 100㎡ area (20m long × 5m wide), based on module size 2000×1000mm, each row lays 5 modules (occupies 10m width), total 5 rows (occupies 10m length), exactly lays 25 modules.

Each actual power 380-400W, total power 9.5-10kW.

Based on local sunshine 4.5 hours/day, annual generation approx. 9.5kW×4.5h×365≈15,600kWh.

Thin-film modules efficiency 14-18% (CIGS highest 18.2%).

Same 100㎡ roof, due to lower efficiency, each module power 250-320W, to reach 10kW total power, need to lay 31-40 modules (100㎡ maximum lays 25 standard size, actually can only install 25, total power 6.25-8kW).

Bavaria, Germany farm, 100㎡ roof installs thin-film modules (efficiency 17%), 25 modules total power 8kW, annual generation 8kW×4.2h×365≈12,300kWh, 21% less than crystalline silicon.

Who loses less power when shading comes?

Napa Valley, California, USA vineyard project, modules partially shaded by grapevines (shading area 10%), measured generation loss 52%, entire string 25 modules daily average 12kWh less.

Lab using xenon lamp simulates shading (20% shading area), crystalline silicon power plummets 60%, thin-film only 25-30%.

University of Freiburg, Germany test: Same 20% shading area, thin-film module generation loss 28%, 32 percentage points less drop than crystalline silicon.

Melbourne, Australia suburban roof, chimney shadow covers 2 hours daily, thin-film modules annual generation 8% more than crystalline silicon (approx. 400kWh/year).

How to squeeze out generation from small spaces?

Summer roof hot like an oven, module temperature rises to 60-70℃, two module types' "high temperature power drop" magnitude different, actual space utilization efficiency changes.

Crystalline silicon temperature coefficient -0.35% to -0.45%/℃ (per 1℃ rise, generation drops 0.4%).

Phoenix, Arizona summer, roof temperature 65℃, crystalline silicon module actual power 12-18% lower than rated (400W becomes 328-352W).

Thin-film temperature coefficient -0.25% to -0.3%/℃, same 65℃ only drops 10-12% (400W becomes 352-360W).

Local ground-mounted power plant comparison: 100 crystalline silicon (400W) total power 40kW, summer daily average generation 220kWh;

100 thin-film (400W) total power 40kW, daily average generation 235kWh, 6.8% more.

Who can occupy irregular spaces better?

Iowa, USA silo, curved roof bend radius 2m, area 80㎡.

Crystalline silicon modules are too rigid, can only be installed on a flat top part (40㎡), install 20 modules (total power 7.6 kW);

Thin-film modules conform to curved surfaces, uses entire 80㎡, installs 25 modules (total power 8kW), 25% more modules installed.

Berlin, Germany factory curved exterior wall (tilt angle 45°), area 150㎡.

Crystalline silicon needs cutting into fan shape, edge sealing failure risk 20%, finally only installs 30 modules (11.4 kW);

Thin-film directly adheres, installs 37 modules (14 kW), utilizes 22% more space.

How to arrange ground-mounted power plants?

Open land (e.g., Nevada desert, USA): Crystalline silicon efficiency high, arranged with 2.5m row spacing, 1.8m column spacing, 1MW requires 2.5 hectares land;

Thin-film efficiency low, same 1MW requires 3 hectares, but thin-film wind/sand resistance strong (surface fluorocarbon resin layer), dust accumulation cleaning frequency 1 less per year than crystalline silicon (saves $0.02/W maintenance fee).

Mountain power plant (e.g., Italian Alps): Terrain undulating, crystalline silicon module tilt angle fixed, some areas shadow overlap;

Thin-film can adjust installation angle (flexible substrate allows ±5° fine adjustment), reduces shadow waste.

Local project measurement, thin-film utilizes 15% more mountain area than crystalline silicon, generates 9% more power.

Complex Environment Compatibility

Can it directly attach to curved roofs?

Iowa, USA silo, curved roof bend radius 2 meters, area 80㎡.

Crystalline silicon modules (2000×1000×35mm) can only be installed on the flat top part (40㎡), install 20 modules (total power 7.6kW), and the remaining 40㎡ curved area is left empty.

Installation also requires cutting module corners, edges sealed with silicone, resulting in water seepage in the second year rainy season, 2 modules scrapped, repair cost $600.

Same silo, using CIGS thin-film (1800×900×3mm), directly covers 80㎡, installs 25 modules (total power 8kW), 25% more than crystalline silicon.

Installation uses double-sided tape + battens fixing, no cutting, no mounts, saves $5,000 mount cost.

Barcelona, Spain water tower (bend radius 1.8 meters), thin-film modules after adhering, annual generation 12% more than crystalline silicon custom solution (approx. 500kWh).

How to install on vertical walls without looking ugly?

Thin-film modules can be used as "building material", directly attached to walls, called BIPV (Building-Integrated Photovoltaics).

Berlin, Germany car factory, curved exterior wall 150㎡, tilt angle 45°.

Crystalline silicon modules need installing on metal mounts, protrude 20 cm from the wall, have high wind resistance, need painting to match color, single installation cost $45.

Finally, only installs 30 modules (11.4 kW), because mounts occupy space, actual utilized area is only 100㎡.

Thin-film modules (e.g., Solar Frontier FS-385) use polymer substrate, color optional dark gray or blue-black, directly attach to exterior wall.

Same 150㎡ wall, installs 37 modules (14kW), utilizes 22% more area.

Appearance flush with wall, wind resistance 30% less, annual maintenance saves $200 (no mount repair).

California, USA parking lot partition, thin-film modules transmittance 15%, daytime no need for lights, generates power while shading, user feedback "better looking than concrete wall".

What about agricultural greenhouses needing light transmission?

Crystalline silicon modules transmittance <10%, covering them harms crop growth.

Thin-film modules can adjust transmittance, 15-30% optional.

Andalusia, Spain greenhouse project, arched roof area 2000㎡, uses CdTe thin-film (transmittance 25%), grows tomatoes underneath.

Measured tomato light sufficient (PAR value ≥400μmol/m²/s), yield only 8% less than open field, simultaneously generates 18kW, annual electricity sales $2,500.

Crystalline silicon solution can only be installed on greenhouse sides (non-growing area), 2000㎡ greenhouse only installs 50 modules (19 kW), generation 40% less than thin-film.

Netherlands flower greenhouse tried both: thin-film transmittance 20%, rose blooming rate 92%;

Crystalline silicon installed on sides, blooming rate 85%, flower farmer said "thin-film turned greenhouse roof into 'power-generating window'".

Can it withstand coastal salt spray corrosion?

Crystalline silicon frame aluminum alloy, annual corrosion rate 0.02mm (NREL test), 10 years frame thins 0.2mm, sealing failure risk increases 30%.

Bergen, Norway coastal project, crystalline silicon modules installed 5 years, 15% frames rusted, frame replacement cost $300/module.

Thin-film modules use stainless steel foil or polymer substrate, salt spray test (ASTM B117 standard, 500 hours) later, surface no corrosion.

Same Bergen project, thin-film modules installed 5 years, only 2 modules edges slight fading, cleaned with detergent back to normal, maintenance cost $50/module (1/6 of crystalline silicon).

Florida Keys, USA project, thin-film modules salt spray resistance 50% higher than crystalline silicon, after typhoon season (wind speed 180 km/h), thin-film intact rate 98%, crystalline silicon 85% (frame loosened).

How to arrange on mountain slopes without blocking light?

Italian Alps power plant, slope 15-25, crystalline silicon installed at 20° tilt, some areas shadow overlap, 10% area wasted.

Measured 1MW power plant, annual generation 1.1 million kWh, 120,000 kWh less than theoretical value.

Thin-film modules flexible substrate allows ±5° fine adjustment.

Same mountain, thin-film adjusts angle according to terrain (15-25° gradient), shadow overlap area reduced to 3%, 1MW annual generation 1.25 million kWh, 150,000 kWh more (enough for 50 households for 1 year).

Colorado, USA mountain project, thin-film modules use "slope-following mounts" (cost 0.1/W), 0.05/W cheaper than crystalline silicon fixed mounts, simultaneously utilizes 15% more sloped land area.

Do they become brittle in cold regions?

Crystalline silicon glass thermal expansion coefficient 8×10⁻⁶/℃, -40℃ frame contracts, encapsulant (EVA) prone to cracking.

Yukon, Canada project, crystalline silicon modules installed 3 years, after -40℃ cold wave, 8% module edge encapsulant cracked, moisture entered interior, generation dropped 20%.

Thin-film modules use polymer substrate (thermal expansion coefficient 12×10⁻⁶/℃), contraction rate higher than crystalline silicon, but flexible, less prone to cracking.

NREL low temperature test (-40℃, 24 hours) shows, thin-film modules no cracking, power only drops 3% (crystalline silicon drops 8%).

Tromsø, Norway project, thin-film modules installed 5 years, -40℃ environment, annual generation more stable than crystalline silicon (fluctuation <5%), because low temperature thin-film efficiency actually increases 2-3% (carrier mobility improves).