TECHNOLOGY EXPLAINER
Encapsulation and Barrier Layers
— keeping out moisture at levels near the limit of measurement, for 20 years
Perovskites are vulnerable to moisture, oxygen and heat. So a product's lifetime depends as much on how the cell is packaged as on the absorber itself. The water vapour transmission rate required is of the same order as, or lower than, the detection limit of commercial instruments. This article sorts out water vapour transmission rate (WVTR), glass and film encapsulation, edge seals and the constraints of lamination temperature, from a materials engineer's point of view.
- What encapsulation is, in three points
- The basic formats: glass encapsulation and film encapsulation
- Where water gets in: through the face and through the edge
- WVTR requirements, and what each material can deliver
- A materials engineer's view (1): the requirement sits at the same order as the limit of measurement
- Our calculation: water entering 1 m² over 20 years, and the time for water to arrive from the edge
- What an encapsulant must do: moisture barrier alone is not enough to choose one
- A materials engineer's view (2): by-products and elastic modulus
- The lamination temperature constraint: between 150 °C and 100 °C
- A materials engineer's view (3): solving the temperature problem inside the cell
- Open problems, and what this article could not confirm
- Glossary / References / Claim-to-source audit
Sourced = stated in published material or a peer-reviewed paper (link given)
Our calculation = a value this article derived from assumptions it states
Not yet confirmed = a plan, target or outlook with no confirmed track record
Structural readings and materials-design interpretations are marked separately as Commentary.
1. What encapsulation is, in three points
- What it protects against: the NREL review notes that methylammonium lead iodide (MAPbI3) readily decomposes to PbI2 under heat and humidity, and describes encapsulation as an additional line of defence on top of stabilising the material itselfSourced
- How it is packaged: commercial silicon solar modules are typically vacuum-laminated with an encapsulant between two sheets of glass, or between glass and a backsheet, with an edge seal closing the perimeterSourced
- What makes it hard: perovskites need low water vapour and oxygen transmission rates, and they are also sensitive to the heat of lamination. Commercial lamination is carried out at about 150 °C, whereas perovskite solar cells are said to degrade thermally, particularly above 100 °CSourced
Encapsulation is not a problem solved by choosing one film that stops water. A group including the University of Rome Tor Vergata (Mariani et al., 2024) lists the requirements for a perovskite encapsulant as a low water vapour transmission rate (10⁻⁴ g/m²/day or less) and oxygen transmission rate, together with no release of degradation products, UV resistance, thermal stability up to 85 °C, processability at 120 °C or below, transparency, a low elastic modulus and sufficient adhesion, all at onceSourced. It is a materials design problem that has to satisfy barrier performance, chemical inertness, thermal and mechanical demands together (our commentary).
2. The basic formats: glass encapsulation and film encapsulation
A review by Wang and colleagues at the University of Hong Kong and elsewhere (ACS Materials Au, 2022) divides lab-scale encapsulation into thin-film encapsulation and cover-glass (or flexible cover) encapsulationSourced.
- Cover-glass encapsulation: because the cell is sandwiched between two surfaces that are practically impermeable to water, water can enter only through the encapsulant between the substrate and the cover glass, so it generally performs better than thin-film or plastic encapsulationSourced
- Polymer front and back sheets: panels laminated with glass are said to have considerably longer lifetimes than those laminated with polymers such as PET or ETFESourced
- For flexible devices: polymer covers are used mainly for flexible cells, and the review points out that they have not yet been adequately subjected to harsh tests such as damp heatSourced
For a film-type product, Sekisui Chemical describes the structure of its perovskite solar cell as a front sheet (transparent barrier, moisture sealing), a cell (power-generating element) and a backsheet (moisture sealing), and lists among its differentiating technologies the encapsulant layer and barrier materials, and a proprietary material composition optimised for perovskites (patented)Sourced. Film-type cells themselves are covered in our explainer on flexible perovskite solar cells.
3. Where water gets in: through the face and through the edge
There are two ways in: through the face, permeating the cover through its thickness, and through the edge, diffusing sideways along the encapsulant layer. Sandwich the cell between glass and permeation through the face all but stops, so the problem concentrates at the edge. Sandwich it between films and permeation through the face is also set by the WVTR of the cover (our commentary).
The Wang review says that because typical encapsulants are relatively permeable to moisture, edge seals are needed to protect moisture-sensitive materials. As an example it gives breakthrough times for moisture across 12.5 mm of encapsulation of 2 days with EVA and 1.15 hours with PDMS, and says edge seals need to have low moisture diffusion and contain a desiccant (desiccant-filled polyisobutylene, PIB)Sourced. It adds that a PIB edge seal adheres well to glass but loses adhesive strength sharply at high temperature, so a secondary sealant acting as an adhesive is used alongside it, and notes a proposal to combine PIB with siliconeSourced.
Mariani et al. say that to avoid adding to cell-to-module losses from encapsulation, the edge-seal width on square-metre panels should be kept below 1 cmSourced. The edge-seal width is both the distance that keeps water out and area that generates no power (our commentary; inactive area is discussed in our explainer on roll-to-roll processing and scale-up).
4. WVTR requirements, and what each material can deliver
| Material or application | WVTR or moisture-ingress index | How the source treats it |
|---|---|---|
| Requirement for organic LEDs (OLEDs) | of order 10⁻⁶ g/m²/day | Below the sensitivity of typical commercial WVTR instruments (of order 10⁻⁴ g/m²/day), and well below what liquid crystal displays or solar cells need |
| Requirement for perovskite encapsulants | 10⁻⁴ g/m²/day or less | The condition given by Mariani et al. (oxygen transmission rate 10⁻³ cm³/m²/day/atm or less) |
| EVA | WVTR 2.61; 10 to over 1000 at 85 °C | Moisture ingress depth K = 0.38 cm/h^1/2. May release acetic acid |
| Ionomer (Surlyn) | about 1 at 20 °C | May release methacrylic acid. Elastic modulus 394 MPa |
| POE (polyolefin elastomer) | about 0.8 at 38 °C (g·mm/m²/day) | Elastic modulus 9.1 MPa (an example) |
| PIB (polyisobutylene) | K = 0.018 to 0.024 cm/h^1/2 | The smallest ingress depth in the table. Elastic modulus about 0.6 MPa |
| PIB-based encapsulant (Mariani et al.) | about 2 × 10⁻⁵ g/m²/day | Measured by calcium corrosion test (Ca test) |
| Polysilazane-derived SiNx film (Song et al.) | average 1.6 × 10⁻⁵ g/m²/day | Solution-coated, then densified with vacuum ultraviolet light. Even a film made in under 10 seconds per layer gave 3.8 × 10⁻⁵ |
All Sourced (OLED, EVA, Surlyn, POE and PIB from the text and tables of the Wang review [Reference 2]; the Mariani values from [Reference 1]; the polysilazane film from [Reference 6]). WVTR measurement temperature and humidity differ from material to material (the "20 °C", "38 °C" and "85 °C" in the table are subscripts in the original), so the rows are not values compared under identical conditions. WVTR values without units are in g/m²/day.
5. A materials engineer's view (1): the requirement sits at the same order as the limit of measurement
The Wang review writes that the WVTR needed for organic LEDs (OLEDs) is of order 10⁻⁶ g/m²/day, below the sensitivity of typical commercial WVTR instruments (of order 10⁻⁴ g/m²/day)Sourced. And the requirement Mariani et al. set for perovskite encapsulants is 10⁻⁴ g/m²/day or less, the same order as instrument sensitivitySourced.
- Mariani et al. determined the WVTR of their PIB-based encapsulant (about 2 × 10⁻⁵ g/m²/day) by the corrosion of a calcium film (Ca test)Sourced
- The polysilazane-derived film of Song et al. (1.6 × 10⁻⁵ g/m²/day) also belongs to this "10⁻⁵ range"Sourced
In other words, perovskite encapsulation is an area where whether you can measure it is as much of a barrier as whether you can make it. Measurement methods that use a material which changes when it reacts with water, such as calcium, as the "sensor"; long accelerated tests; how to guarantee lot-to-lot variation: for companies that have worked with barrier films and gas-barrier materials, evaluation technology itself is likely to be a differentiator (our commentary). This article has not checked the text of WVTR measurement standards (ISO or ASTM test methods).
6. Our calculation: water entering 1 m² over 20 years, and the time for water to arrive from the edge
Assumptions (set by this article)Our calculation
- The WVTR stays constant for 20 years (7,305 days), and all water that permeates stays inside
- Real WVTR varies greatly with temperature and humidity, but that is ignored here
| WVTR (g/m²/day) | Permeation over 20 years (per m²) | Value used as a guide |
|---|---|---|
| 1 | about 7,300 g | Surlyn's approximately 1 (20 °C) |
| 10⁻² | about 73 g | — |
| 10⁻⁴ | about 0.73 g | Upper limit of the Mariani requirement |
| 2 × 10⁻⁵ | about 0.15 g | The Mariani PIB-based encapsulant |
| 10⁻⁶ | about 0.0073 g (7.3 mg) | Order of the OLED requirement |
Lower the WVTR by four orders of magnitude, from 1 to 10⁻⁴, and the water entering 1 m² over 20 years falls from about 7 kg to under 1 g.
Assumptions and limits: in reality water is absorbed and diffuses within the encapsulant before reaching the cell, and temperature and humidity keep changing. How much water it takes to degrade a perovskite depends on the structure, and this article has not confirmed any primary source giving a tolerable amount. The calculation is only meant to give a feel for the orders of magnitude.
AssumptionsOur calculation
- Use the moisture ingress depth K (cm/h^1/2) from the table in the Wang review, taking the ingress depth x to advance as x = K√t
- Take the edge-seal width as 1 cm, the upper limit Mariani et al. give for square-metre panels. Then t = (x ÷ K)²
| Material | K (cm/h^1/2) | Time to advance 1 cm |
|---|---|---|
| EVA | 0.38 | about 7 hours |
| TPU | 0.23 | about 19 hours |
| Ionomer | 0.067 | about 220 hours (about 9 days) |
| PIB | 0.018 to 0.024 | about 1,700 to 3,100 hours (about 72 to 129 days) |
For the same 1 cm, PIB takes about 250 to 450 times as long as EVA ((0.38 ÷ 0.024)² to (0.38 ÷ 0.018)²).
Assumptions and limits: this article has not been able to confirm the temperature and humidity at which the K values in the table apply. Also, the EVA breakthrough time given in the same review (2 days for 12.5 mm) does not match this simple K√t calculation (about 11 hours for 12.5 mm), probably because the definitions or conditions differ. The table is meant to show the differences in order of magnitude between materials; it does not indicate the lifetime of a real product. Since even PIB alone could reach 1 cm in about three months, real edge seals contain a desiccantSourced.
7. What an encapsulant must do: moisture barrier alone is not enough to choose one
Mariani et al. set out the properties required of a perovskite encapsulant as followsSourced.
| Required property | What the paper says |
|---|---|
| Chemically inert | Must not release chemicals that cause degradation (acetic acid from EVA and methacrylic acid from Surlyn are given as examples) |
| Barrier | WVTR of 10⁻⁴ g/m²/day or less and oxygen transmission rate of 10⁻³ cm³/m²/day/atm or less. Outgassing of volatile components must also be suppressed |
| UV resistance | Must not yellow or release decomposition products under UV light |
| Thermal | Thermally stable up to 85 °C and processable at 120 °C or below (to match the thermal stability of perovskites and common charge transport layers) |
| Optical | Must be transparent |
| Mechanical | A low elastic modulus (preferably under 20 MPa at 25 °C) and sufficient adhesion, to withstand thermomechanical stress from daily temperature swings |
All Sourced (Mariani et al. 2024 [Reference 1]). The paper lists as candidate encapsulants EVA, ionomers (Surlyn, Bynel, Jurasol), PIB, POE, polyurethane and TPU, and as edge seals PIB-based butyl rubber tape, UV-curable resins, epoxies, silicones and glass frit.
The Wang review likewise says that a material with a low WVTR but a high elastic modulus is unsuitable as an encapsulant, and that a low modulus at all temperatures is needed to absorb stressSourced.
8. A materials engineer's view (2): by-products and elastic modulus
The table in the Wang review records harmful by-products for two standard materials of silicon PV: acetic acid for EVA and methacrylic acid for the ionomer (Surlyn). Elastic moduli are 10 MPa for EVA, 394 MPa for Surlyn, 9.1 MPa for POE (an example) and about 0.6 MPa for PIBSourced.
- By-products: that Mariani et al. put "must not release chemicals that cause degradation" first among the encapsulant requirements, with EVA's acetic acid and Surlyn's methacrylic acid as examples, can be read as a sign that they see traces of acid released by an encapsulant over many years as a potential cause of degradation (our commentary)
- Elastic modulus: Mariani et al. say low-modulus encapsulants are generally recommended to avoid delamination caused by mismatched thermal expansion coefficientsSourced. The group therefore chose a PIB whose molecular weight makes it change from a semi-solid to a highly viscous liquid between −40 and 85 °C, aiming for strain-free laminationSourced
- Electrical resistance: they state that PIB's resistivity is around 10¹⁶ Ω cm, higher than that of EVA (10¹³ to 10¹⁵ Ω cm), which helps suppress potential-induced degradation (PID)Sourced
- Additive design: adding two-dimensional hexagonal boron nitride (h-BN) flakes raised adhesive strength (pull-off test) by 25% and improved heat dissipation (the time to cool to 30 °C was 11.2% shorter)Sourced
In addition, Shi et al. (2017), from tests of cells glass-glass encapsulated with PIB, argued that heat rather than moisture was the main cause of degradation, and proposed that preventing the volatile decomposition products of the perovskite from escaping is key to stabilitySourced.
So an encapsulant has four jobs, not just keeping water out: releasing nothing from within (not producing acid itself), keeping what is inside from escaping, and not tearing the cell apart as it expands and contracts with heat. For designers of pressure-sensitive adhesives, hot melts and sealants, techniques such as tuning viscoelasticity through molecular weight, or using fillers to shift moisture permeability and thermal conductivity together apply directly here (our commentary).
9. The lamination temperature constraint: between 150 °C and 100 °C
A group from NREL and Dow (Witteck et al., 2024) tackled this problem head on. Their starting point: commercial vacuum lamination is generally performed at 150 °C to ensure crosslinking of the encapsulant and adhesion to the glass, while perovskite solar cells are heat-sensitive and prone to degradation, particularly above 100 °CSourced.
| Lamination condition | Results of Witteck et al. |
|---|---|
| 150 °C, 23 min | The harshest condition, close to conventional conditions for silicon and thin-film modules. Reference cells lost 27% of their efficiency on average (relative), with Jsc down 10%, FF down 15% and Voc down 7% |
| 125 °C, 14 min | Even the mildest condition gave an average loss of 6% (relative). Cells with a PTAA hole transport layer lost 27% (relative) |
| Lower temperature limit | Below 125 °C, adhesion to glass was insufficient and the stack became mechanically unstable, so this was taken as the minimum condition |
| Encapsulants | Compared peroxide-crosslinked EVA and POE used in silicon modules, POE grafted with silane or anhydride, and others |
All Sourced (Witteck et al. 2024 [Reference 3]).
Mariani et al., by contrast, used multi-step, differential-pressure low-temperature lamination at 90 °C (under 45 minutes in total), keeping the efficiency loss from lamination to under 1% absolute. They report that even cells using heat-sensitive spiro-OMeTAD withstood lamination at 90 °C for 10 minutesSourced. Nikbakht et al. (the 0.73 m² panel) made standard 120 °C lamination possible by using structurally robust PTAA-based modules, hot vacuum laminating them with a commercial thermoplastic filmSourced.
10. A materials engineer's view (3): solving the temperature problem inside the cell
Witteck et al. initially thought the encapsulant's viscosity or bonding chemistry caused mechanical and chemical degradation during lamination. But after swapping combinations of materials, they proposed that the main degradation mechanism during lamination is diffusion from the silver (Ag) electrodeSourced.
- Time-of-flight secondary ion mass spectrometry (ToF-SIMS) and cross-sectional Kelvin probe force microscopy (cKPFM) were used to correlate chemical and electrical changesSourced
- Adding a self-assembled monolayer (SAM) hole transport layer on the ITO side and atomic-layer-deposited (ALD) SnOx on the electron-extraction side as "internal diffusion barriers" meant no degradation even with lamination at 150 °C, the same as commercial conditionsSourced
- Cells with 2PACz (a SAM) plus SnOx showed no degradation beyond measurement error at 125 °C for 14 minutes, and only a 3% average (relative) loss at 150 °C for 14 minutesSourced
What this shows is that the temperature constraint on encapsulation is not a problem of the encapsulant alone. There is the route of developing encapsulants that bond at low temperature (Mariani et al.), and the route of putting diffusion barriers inside the cell so it can run on existing 150 °C lines (Witteck et al.). The first is a matter for encapsulant makers; the second for makers of charge transport layer and electrode materials, and of equipment. Which becomes mainstream will change the flow of materials business (our commentary). SAMs are covered in our explainer on self-assembled monolayers.
11. Open problems, and what this article could not confirm
(1) Cases that passed accelerated tests, and the problems that remain
The NREL review cites a report in which FA1-yCsyPb(I1-xBrx)3 cells, glass-glass encapsulated with commercial encapsulants (EVA and butyl rubber), passed the IEC damp heat test (85 °C, 85% RH, 1,000 hours)Sourced. Mariani et al. report that, without any edge seal, their devices kept more than 80% of their initial efficiency through ISOS-D-2 (85 °C, over 1,000 hours), ISOS-L-1 (light soaking, over 1,000 hours), an in-house thermal shock test (−40 to +85 °C, 200 cycles) and a modified humidity-freeze test (10 cycles)Sourced. The same paper, however, mentions a separate report in which cells that had passed damp heat and thermal cycling still lost about 20% of their efficiency after 1,000 hours of continuous illumination at 55 ± 5 °CSourced. Encapsulation can stop water, but it cannot on its own stop internal degradation driven by light and heat. What the tests involve is covered in our explainer on outdoor testing and durability assessment, and internal degradation in our explainer on ion migration and degradation.
(2) The long-term record of film encapsulation
As of the Wang review (2022), encapsulation with polymer covers was described as not yet adequately subjected to harsh tests such as damp heatSourced. Sekisui Chemical states for its film-type product that "durability equivalent to 10 years has been achieved" (our translation), and aims for 20 years, on a par with siliconSourced. However, the method behind "equivalent to 10 years" (test conditions and conversion method) and whether it has been confirmed by a third party are not stated in the material this article consulted. The 20 years is a targetNot yet confirmed.
(3) What this article could not confirm
The tolerable amount of moisture at which perovskites begin to degrade, measured WVTR values for the barrier films of film-type products, and the text of WVTR measurement standards could not be confirmed in primary sources within the scope of this article, so they are not stated. The role of encapsulants in limiting lead leakage is covered in our explainer on lead and iodine.
- Perovskite encapsulants are required to have a WVTR of 10⁻⁴ g/m²/day or less. That is the same order as the sensitivity of typical commercial instrumentsSourced
- Lowering the WVTR from 1 to 10⁻⁴ cuts the water entering 1 m² over 20 years from about 7,300 g to about 0.73 gOur calculation
- With glass encapsulation, water gets in at the edge. Edge seals use materials such as desiccant-filled PIBSourced
- EVA can release acetic acid and Surlyn methacrylic acid. A low modulus, inertness and low-temperature processing are all needed at onceSourced
- Commercial lamination runs at 150 °C, while perovskites degrade easily above 100 °C. One main cause is silver diffusion, and cells with internal diffusion barriers have withstood 150 °CSourced
- Encapsulation can stop water, but internal degradation from light and heat has to be solved separatelySourced
12. Glossary
- Encapsulation
- Wrapping a cell in encapsulant and cover materials to protect it from outside air, moisture and oxygen.
- WVTR (water vapour transmission rate)
- The amount of water vapour passing through 1 m² of a film or material per day, in g/m²/day.
- OTR (oxygen transmission rate)
- The amount of oxygen passing through a film or material.
- Encapsulant
- A resin that fills and bonds the space between the cell and the cover materials, such as EVA, POE, ionomers or PIB.
- Edge seal
- Material that closes off a panel's perimeter to stop moisture coming in from the side, such as desiccant-filled PIB.
- Barrier film
- A transparent film made resistant to water vapour and oxygen, used on the front and back of flexible products.
- Backsheet
- The sheet covering the back of a module.
- Vacuum lamination
- A process that bonds cover materials, encapsulant and cell together by heating and pressing under reduced pressure.
- Ca test
- A method that determines very low WVTRs from the change as a calcium film reacts with water and turns transparent.
- Moisture ingress depth K
- The coefficient (cm/h^1/2) when the depth moisture travels into a material is written as x = K√t.
- Elastic modulus
- How resistant a material is to deformation. The lower it is, the softer the material and the better it absorbs stress from thermal expansion and contraction.
- Damp heat test
- An accelerated test holding devices at 85 °C and 85% relative humidity for a long time. IEC 61215 specifies 1,000 hours.
13. References (primary sources)
- Mariani P et al. “Low-temperature strain-free encapsulation for perovskite solar cells and modules passing multifaceted accelerated ageing tests”, Nature Communications 15, 4552 (2024) (full text on PMC) https://pmc.ncbi.nlm.nih.gov/articles/PMC11137052/
- Wang Y et al. “Encapsulation and Stability Testing of Perovskite Solar Cells for Real Life Applications”, ACS Materials Au 2, 215–236 (2022) (full text on PMC) https://pmc.ncbi.nlm.nih.gov/articles/PMC9888620/
- Witteck R et al. (NREL, Dow) “Reducing Thermal Degradation of Perovskite Solar Cells during Vacuum Lamination by Internal Diffusion Barriers”, ACS Applied Energy Materials 7, 10750–10757 (2024) (full text on PMC) https://pmc.ncbi.nlm.nih.gov/articles/PMC11600410/
- Li Z, Klein TR, Kim DH, Yang M, Berry JJ, van Hest MFAM, Zhu K (NREL) “Scalable fabrication of perovskite solar cells”, Nature Reviews Materials 3, 18017 (2018) (DOE Public Access accepted manuscript) https://www.osti.gov/servlets/purl/1430821
- Nikbakht H et al. “Upscaling Perovskite Photovoltaics: from 156 cm² Modules to 0.73 m² Panels”, Advanced Science 12, 2416316 (2025) (full text on PMC) https://pmc.ncbi.nlm.nih.gov/articles/PMC12165116/
- Song L, Sun H, Suzuri Y et al. “Polysilazane-Coated Films Achieving Record-High Moisture Barrier Performance with Sub-10 Seconds Densification Using High-Power VUV Irradiation”, Advanced Science 12, 2415721 (2025) https://doi.org/10.1002/advs.202415721
- Shi L et al. “Accelerated Lifetime Testing of Organic–Inorganic Perovskite Solar Cells Encapsulated by Polyisobutylene”, ACS Applied Materials & Interfaces 9, 25073–25081 (2017) https://doi.org/10.1021/acsami.7b07625
- Cheacharoen R et al. “Encapsulating perovskite solar cells to withstand damp heat and thermal cycling”, Sustainable Energy & Fuels 2, 2398–2406 (2018) https://doi.org/10.1039/c8se00250a
- Sekisui Chemical “Perovskite solar cell business briefing”, presentation slides, 7 January 2025 (PDF, in Japanese) https://www.sekisui.co.jp/ir/event/other/__icsFiles/afieldfile/2025/01/07/20250107PVK.pdf
14. Claim-to-source audit
| Claim in the text | Basis | Label |
|---|---|---|
| The encapsulant requirements (no degradation products, with EVA's acetic acid and Surlyn's methacrylic acid; WVTR of 10⁻⁴ g/m²/day or less, OTR of 10⁻³ or less and suppressed outgassing; UV resistance; thermal stability to 85 °C and processing at 120 °C or below; transparency; a low modulus under 20 MPa at 25 °C and adhesion). The list of candidate encapsulants and edge seals. That the edge-seal width on square-metre panels should be under 1 cm. That low-modulus encapsulants are recommended to avoid delamination from mismatched thermal expansion. Strain-free lamination with a PIB that changes from semi-solid to highly viscous liquid between −40 and 85 °C. PIB resistivity of 10¹⁶ Ω cm against 10¹³ to 10¹⁵ Ω cm for EVA. h-BN raising adhesive strength by 25% and shortening the time to 30 °C by 11.2%. About 2 × 10⁻⁵ g/m²/day by Ca test. Multi-step differential-pressure lamination at 90 °C (under 45 minutes) with a loss under 1% absolute, and spiro-OMeTAD cells withstanding 90 °C for 10 minutes. Keeping over 80% without an edge seal through ISOS-D-2 and ISOS-L-1 (each over 1,000 hours), 200 thermal shock cycles and 10 humidity-freeze cycles. The other report of about a 20% loss after 1,000 hours of illumination at 55 ± 5 °C despite passing damp heat and thermal cycling. That the authors' affiliations include the University of Rome Tor Vergata | Reference 1 https://pmc.ncbi.nlm.nih.gov/articles/PMC11137052/ | Sourced |
| That commercial silicon modules are vacuum-laminated glass-glass or glass-backsheet with an edge seal. The classification into thin-film and cover-glass encapsulation, with cover glass generally performing better. That glass laminates last considerably longer than polymer (PET, ETFE) laminates. That polymer covers have not been adequately tested under conditions such as damp heat. That the OLED requirement is of order 10⁻⁶ g/m²/day, below commercial instrument sensitivity (of order 10⁻⁴). That high-modulus materials are unsuitable even with low WVTR. The table of encapsulant properties (EVA: WVTR 2.61, 10 to over 1000 at 85 °C, K = 0.38, modulus 10 MPa, acetic acid; Surlyn: about 1 at 20 °C, 394 MPa, methacrylic acid; POE: about 0.8 g·mm/m²/day at 38 °C, 9.1 MPa; TPU K = 0.23; ionomer K = 0.067; PIB K = 0.018 to 0.024, about 0.6 MPa). The need for edge seals; breakthrough times across 12.5 mm of 2 days for EVA and 1.15 hours for PDMS; desiccant-filled PIB; PIB losing adhesion at high temperature and being used with a secondary sealant. That the corresponding author is affiliated with the University of Hong Kong | Reference 2 https://pmc.ncbi.nlm.nih.gov/articles/PMC9888620/ | Sourced |
| That commercial vacuum lamination is generally performed at 150 °C and that perovskites are prone to degradation particularly above 100 °C. An average 27% (relative) loss at 150 °C for 23 minutes, with Jsc 10%, FF 15% and Voc 7%. An average 6% even at 125 °C for 14 minutes, and 27% with PTAA. Insufficient adhesion below 125 °C. The comparison of EVA, POE and grafted POE. The proposal of silver diffusion as the main degradation mechanism. ToF-SIMS and cKPFM. No degradation at 150 °C with internal diffusion barriers of a SAM and ALD SnOx; 2PACz plus SnOx within error at 125 °C for 14 minutes and 3% at 150 °C for 14 minutes. That the authors are affiliated with NREL and Dow | Reference 3 https://pmc.ncbi.nlm.nih.gov/articles/PMC11600410/ | Sourced |
| That MAPbI3 readily decomposes to PbI2 under heat and humidity. That encapsulation is an additional line of defence. The report of FA1-yCsyPb(I1-xBrx)3 cells glass-glass encapsulated with EVA and butyl rubber passing the IEC damp heat test (85 °C, 85% RH, 1,000 hours) | Reference 4 https://www.osti.gov/servlets/purl/1430821 | Sourced |
| That the 0.73 m² panel used PTAA-based modules to allow standard 120 °C lamination, hot vacuum laminated with a commercial thermoplastic film | Reference 5 https://pmc.ncbi.nlm.nih.gov/articles/PMC12165116/ | Sourced |
| SiNx films derived from perhydropolysilazane and densified with vacuum ultraviolet light, with an average WVTR of 1.6 × 10⁻⁵ g/m²/day and 3.8 × 10⁻⁵ g/m²/day at under 10 seconds per layer. Intended for flexible devices including perovskites | Reference 6 https://doi.org/10.1002/advs.202415721 | Sourced |
| Glass-glass encapsulation with PIB, evaluated by damp heat and thermal cycling following IEC 61215:2016. The argument that heat rather than moisture was the main cause of degradation, and the proposal that preventing the escape of volatile decomposition products is key | Reference 7 https://doi.org/10.1021/acsami.7b07625 | Sourced |
| That there is research in which encapsulation with glass and an edge seal enables stability under damp heat and thermal cycling tests (as far as the abstract) | Reference 8 https://doi.org/10.1039/c8se00250a | Sourced |
| The structure of Sekisui Chemical's film-type product (front sheet = transparent barrier and moisture sealing, cell, backsheet = moisture sealing); the encapsulant layer, barrier materials and proprietary material composition as differentiating technologies; durability equivalent to 10 years achieved, aiming for 20 years | Reference 9 https://www.sekisui.co.jp/ir/event/other/__icsFiles/afieldfile/2025/01/07/20250107PVK.pdf | Sourced |
| Sekisui Chemical's target of 20 years' durability | A company target, not an actual result. The method behind "equivalent to 10 years" and whether it was confirmed by a third party are not stated in the material (commentary) | Not yet confirmed |
| The table of permeation over 20 years (about 7,300 g at WVTR 1, about 0.73 g at 10⁻⁴, about 0.15 g at 2 × 10⁻⁵, about 7.3 mg at 10⁻⁶). The times to reach 1 cm by K√t (EVA about 7 hours, TPU about 19 hours, ionomer about 9 days, PIB about 72 to 129 days) and the PIB/EVA ratio of about 250 to 450. The simple calculation of about 11 hours for EVA over 12.5 mm | Our calculation. A constant WVTR with all permeated water retained, x = K√t, and a 1 cm width are this article's assumptions. The measurement conditions for K have not been confirmed by this article | Our calculation |
| The tolerable amount of moisture at which degradation begins, measured WVTR values for the barrier films of film-type products, and the text of WVTR measurement standards | Not stated because they could not be confirmed in primary sources within this article's scope (commentary) | Commentary |
| Framing encapsulation as a materials design problem; dividing water ingress into face and edge; the reading that measurability is a barrier; the four roles of an encapsulant; the reading that traces of acid may cause degradation; and the reading of two routes, low-temperature encapsulants and in-cell diffusion barriers, and their effect on the flow of business | This article's own framing and commentary based on published content. Not views expressed by the institutions or authors | Commentary |
| That Figs. 1, 2, 3, 4 and 6 are explanatory drawings rather than real cross-sections or measured data, and that the hero image and Fig. 5 are AI-generated images | Our note (commentary) | Commentary |
Last updated 25 September 2026. Sources are limited to primary material (peer-reviewed papers, NREL accepted manuscripts published through DOE Public Access, and Sekisui Chemical's published material). Because the article includes structural readings and materials-design interpretations, those are marked as Commentary and kept separate from sourced fact. The tolerable amount of moisture at which degradation begins, measured WVTR values for the barrier films of film-type products, the text of WVTR measurement standards, and the method behind Sekisui Chemical's "equivalent to 10 years" are not stated because this article could not confirm them. WVTR values differ from source to source in measurement temperature, humidity and method, and are not compared under identical conditions. All figures are explanatory concept graphics. Figs. 1, 2, 3, 4 and 6 are vector drawings; the hero image and Fig. 5 are AI-generated images, and none of them shows a real cross-section photograph or physical product.