Spectral Characterisation and Optical Simulation of Innovative Liquid Encapsulants for Design-for-Recycling Silicon Solar Modules

Abstract

The
transition to renewable energy, led by photovoltaic (PV) energy, is critical
for a sustainable future. However, the current linear ”take-make-waste”
economy, also used for PV modules, presents a significant challenge to
long-term sustainability, creating a massive future waste stream and reliance
on virgin materials. The primary obstacle to achieving a circular economy for
solar modules is the use of permanently cross-linked solid encapsulants like
Ethylene Vinyl Acetate (EVA), which bond all components together and severely
hinder repair, reuse, and high-value recycling. This thesis examines liquid
encapsulation as a promising alternative design-for-recycling concept, enabling
simple disassembly and material recovery. The primary objective was to
determine if liquid encapsulation is a viable alternative from an optical
perceptive, compared to conventional solid (EVA) and gas-filled (air) module
designs. To achieve this, three sub-goals were formulated and completed. First,
a comprehensive selection strategy was developed based on critical optical,
electrical, chemical, and thermal criteria, identifying seven candidate liquids
for analysis: deionised water, glycerol, SolaPro Ennogreen glycol,
Polydimethylsiloxane (PDMS), and three industrial dielectric oils (Shell Diala
S4 XZI, Midel 7131, Mivolt). Material compatibility tests with polyisobutylene
(PIB) immersed in Midel 7131 and Shell Diala, confirmed that Shell, and
therefore mineral oils in general, are incompatible with the use of PIB as a
sealant. Second, the complex refractive index (n and k) of these liquids was
experimentally measured over the 300–1200 nm solar spectrum using a novel
hybrid method. This involved a specialised ellipsometer and dual cuvette
measurements with spectrophotometry, generating the nk data needed for detailed
optical modelling. Measurements made by Semilab’s liquid-specialised
ellipsometer using a 2-term Cauchy model yielded excellent fits (R2 > 0.97)
for all liquids, in good agreement with literature values. The dual cuvette
method revealed extremely high transparency for all liquids, even slightly
yellowish fluids Midel and Shell Diala. A SNR analysis established a confidence
threshold at k ≈ 5 ∗ 10−8. Finally, these optical properties were used as inputs for detailed
optical modelling and simulation, in GenPro4, to quantify and compare the
photocurrent generated by liquid encapsulation compared to standard EVA and
air, in various solar module configurations. The optical model and its
simulated output are validated through experiments and datasheets, showing
excellent agreement around the Δ Jph = 0.2 mA/cm2. The results revealed a
fundamental trade-off: while the liquids have lower refractive indices, leading
to higher initial reflection losses compared to conventional EVA with UV
blockers, their superior transparency resulted in lower parasitic absorption.
This balance allowed top-performing liquids, notably Mivolt (highest
photocurrent and lowest reflector, Jph = 40.90 41.81 mA/cm2) and PDMS (lowest
absorber, Jph = 40.85 41.75 mA/cm2), to generate a photocurrent comparable to,
and in some cases higher than, UV-transparent EVA (Jph = 40.92 41.81 mA/cm2).
However, the optical performance of the other liquids was very close
(especially Midel and Shell, but also glycerol and glycol, followed within 0.17
mA/cm2), meaning that based on optics alone, they should not be excluded, as
other design benefits might justify their selection. A critical insight emerged
from annual energy yield simulations, which demonstrated that PDMS delivered
the highest yearly energy output (0.07% higher than EVA-UVT, 0.086% to Mivolt),
due to its extremely low absorption. The simulations showed that mitigating
reflection losses through tailored Anti-Reflection Coatings (ARCs) or glass
texturing can further unlock the full potential of liquid-encapsulated modules.
In conclusion, this thesis provides the first comprehensive optical validation
for liquid encapsulation in standard silicon PV modules. It successfully
demonstrates that, from an optical standpoint, liquids are a highly viable
technology, worthy of further research. Their primary drawback of higher
reflection is compensated by superior transparency and can also be further
mitigated with existing ARC and glass texturing technologies. This work
establishes a robust foundation for future research, confirming that liquid
encapsulation is not only a promising pathway toward a circular economy for PV
but is also an optically competitive alternative to established solid
encapsulants.