MENU
Advanced Materials Intelligence
  1. Home
  2. Charge Transport Layers and Cell Architecture Explained | Perovskite Solar Cells

Charge Transport Layers and Cell Architecture Explained | Perovskite Solar Cells

Back to Technology, Institution & Company Guides

TECHNOLOGY EXPLAINER

Charge Transport Layers and Cell Architecture
— sandwiching the absorber between two selective filters

The performance and lifetime of a perovskite solar cell depend as much on the thin layers above and below it as on the layer that absorbs the light. One layer lets only electrons through; the other lets only holes through. Which one goes on the bottom splits cells into n-i-p and p-i-n, and changes which materials and process temperatures can be used.

Built from primary sources: Solar cell efficiency tables (Version 68) and peer-reviewed papers (JACS, Science, Nature, Nature Energy, Scientific Reports, Chemical Science, Nature Communications) / Last updated September 2026

Conceptual image of a plain glass plate against a dark background, soft light shining through it from behind, evenly covered by a very thin translucent film
AI-generated concept image. An impression of the idea of a thin film that lets light through while sorting the charges. It does not represent the real colour or thickness of any film, a device structure, or a product.
What this article covers
  1. What charge transport layers are, in three points
  2. The basic structure — an absorber between two selective layers
  3. n-i-p and p-i-n — which one goes on the bottom
  4. History — a structure that started from the dye-sensitised solar cell
  5. Electron transport layer materials
  6. Hole transport layer materials
  7. A materials engineer's view (1): the dopant turned out to be a hygroscopic salt
  8. A materials engineer's view (2): stacking order decides which materials you can use
  9. Where efficiency stands, with our calculation
  10. Strengths and open problems
  11. Glossary / References / Claim-to-source audit
How claims are labelled in this article

Sourced = stated in published material (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. Every efficiency figure is given with whether it is certified, whether it is a cell or a module, and its area.

1. What charge transport layers are, in three points

When light strikes the perovskite layer, it creates electrons and holes. Carrying them to separate electrodes is the job of the charge transport layers (the electron transport layer, ETL, and the hole transport layer, HTL).

  • What they do: a review in Chemical Science (Schloemer et al., 2019) explains that the hole transport layer acts as a path that passes holes and blocks electrons, reducing recombination and raising the fill factor and efficiencySourced
  • Why they matter: the same review calls the hole transport layer critical to stability, because it is in direct contact with the perovskite, often contains hygroscopic, mobile ions such as Li⁺, and is often the least thermally stable layer in the stackSourced
  • Where materials come in: transport layers use both inorganic and organic materials: oxides (TiO2, SnO2, NiOx), fullerenes (C60, PCBM), small organic molecules (spiro-OMeTAD), polymers (PTAA, PEDOT:PSS) and monolayers (SAMs) (this article's summary)
The single most important line in this article

The perovskite dictates how the layer on top of it can be made. Schloemer et al. note that because perovskites cannot withstand harsh processing of the layers deposited afterwards, oxide hole transport layers that need high temperatures are practical mainly in configurations that place them beneath the perovskite (p-i-n)Sourced. Choosing transport-layer materials is also a question of stacking order and process temperature.

2. The basic structure — an absorber between two selective layers

Electrons go right, holes go left: transport layers as selective filters (concept) Vertical position shows energy schematically; no values are drawn (schematic by this article) Hole transport layer Perovskite (absorber) Electron transport layer Conduction band (path for electrons) Valence band (path for holes) High wall stops electrons Holes pass Electrons pass Deep wall stops holes Note: the HTL's role of passing holes and blocking electrons follows the explanation by Schloemer et al. [Ref. 6]. Note: line heights are schematic and do not show energy-level values for any particular material. Note: the left-to-right layout is for explanation; in a real cell the layers are stacked vertically.
Fig. 1 Concept diagram (vector drawing). The account of what transport layers do follows the review by Schloemer et al. [Ref. 6]. Line heights and spacing are schematic and do not show energy-level values for any particular material. The side-by-side layout is this article's choice for the sake of explanation.

Schloemer et al. explain that the energy levels and mobilities of the layers must be closely matched and balanced, and that this is what prevents charge from piling up and recombiningSourced. The key point of the structure is that a transport layer is not a film that conducts electricity but a film that conducts only one kind of charge (our commentary).

3. n-i-p and p-i-n — which one goes on the bottom

Perovskite solar cells are usually made by building up layer by layer from a substrate coated with a transparent electrode. Depending on whether the electron transport layer (n) or the hole transport layer (p) sits on the substrate side, there are two structures. Schloemer et al. define n-i-p as substrate / electron transport layer / perovskite / hole transport layer, and p-i-n as substrate / hole transport layer / perovskite / electron transport layerSourced. For historical reasons p-i-n is also called “inverted”.

Typical layer stacks of n-i-p (regular) and p-i-n (inverted) cells In both, light enters through the glass at the bottom. Materials are typical examples (this article's framing) n-i-p (regular) p-i-n (inverted) Metal electrode (Au etc.) Hole transport layer: spiro-OMeTAD etc. Perovskite Electron transport layer: TiO2, SnO2 etc. Transparent electrode (FTO, ITO) Glass substrate Metal electrode (Ag, Cu etc.) Electron transport layer: C60, PCBM + BCP etc. Perovskite Hole transport layer: NiOx, SAM, PTAA etc. Transparent electrode (FTO, ITO) Glass substrate The HTL on top must be made at low temperature High-temperature oxides can go underneath Note: n-i-p/p-i-n definitions and the oxide HTL constraint follow Schloemer et al. [Ref. 6]; p-i-n layers follow Chen et al. [Ref. 9]. Note: thicknesses and colours are schematic. Real stacks vary by study; interface-treatment layers are omitted.
Fig. 2 Concept diagram (vector drawing). The definitions of the structures and the constraint on oxide hole transport layers follow the review by Schloemer et al. [Ref. 6]; the example p-i-n stack follows Chen et al. [Ref. 9]; the example n-i-p electron transport layers follow Kim et al. [Ref. 2] and Jiang et al. [Ref. 4]. Layer thicknesses and colours are schematic; this is not a cross-section of any particular record cell or product.

Which one leads today

A 2025 Nature Communications paper (Chen et al.) says the inverted (p-i-n) structure has attracted attention because it is simple to make, shows little hysteresis, is highly stable, and suits flexible and tandem devices, and that advances in self-assembled monolayers (SAMs) have taken it past the regular structure in efficiency tooSourced. For tandems, Hooijer et al. note that most high-efficiency demonstrations use thin-film recombination junctions in the p-i-n configurationSourced.

Which structure eventually dominates, however, will be decided not only by efficiency but by durability and production processes. Within the scope of this article, we could not confirm across primary sources which structure is used in mass-produced productsNot yet confirmed.

4. History — a structure that started from the dye-sensitised solar cell

How the cell structure evolved (this article's framing) Values as reported in each paper; certification status in the bottom row. Areas all differ 2009 2012 2012 2016 2025 2026 3.8% 9.7% 10.9% 19.9% 26.0% 28.0% Liquid electrolyte Mesoporous TiO2 Solid hole transport spiro-MeOTAD Insulating Al2O3 as the scaffold Planar, SnO2 made at 150 °C Inverted, SAM low-temp sequential Efficiency tables 0.051 cm² Paper's value Paper's value Paper's value Certified Certified Independent test Note: Kojima et al. 2009 [Ref. 1], Kim et al. 2012 [Ref. 2], Lee et al. 2012 [Ref. 3], Jiang et al. 2016 [Ref. 4], Chen et al. 2025 [Ref. 9], and for 2026 the efficiency tables Version 68 [Ref. 10]. Note: the 2009 value is as reported; its abstract does not mention independent certification, nor do the two 2012 abstracts. Note: areas and conditions differ, so values cannot be compared directly. Point spacing is not proportional to years.
Fig. 3 Concept diagram (vector drawing). The structure and efficiency at each point follow the respective papers [Refs. 1 to 4 and 9] and Version 68 of the efficiency tables [Ref. 10]. The choice of these six milestones is this article's own; it is not a comprehensive timeline. Area, measurement conditions and certification status differ, so the efficiencies cannot be compared directly.
YearReportStructural point
2009Kojima et al. (JACS)A photoelectrochemical cell using perovskite nanocrystals as a “sensitiser” on mesoporous TiO2. 3.8% efficiency with CH3NH3PbI3
2012Kim et al. (Scientific Reports)Replaced the liquid electrolyte with the solid hole transport material spiro-MeOTAD, filled into the pores of mesoporous TiO2. 9.7% efficiency, with much better stability than liquid-junction cells
2012Lee et al. (Science)A “meso-superstructured” cell using insulating mesoporous alumina as a scaffold, so that electrons travel through the perovskite itself. 10.9% efficiency and an open-circuit voltage above 1.1 V
2016Jiang et al. (Nature Energy)A planar cell with an electron transport layer of SnO2 nanoparticles made by low-temperature solution processing. Almost no hysteresis, certified efficiency of 19.9 ± 0.6%, and can be made at 150 °C

All Sourced (Kojima et al. [Ref. 1], Kim et al. [Ref. 2], Lee et al. [Ref. 3], Jiang et al. [Ref. 4]).

Seen from the materials side, three changes overlap in this history: replacing the liquid electrolyte with a solid, thinning the porous scaffold and eventually going planar, and swapping high-temperature-fired oxides for materials that can be made at low temperature (our commentary). Schloemer et al. also write that introducing a solid hole transport layer was the first breakthrough, and that without the spiro-OMeTAD layer the technology might not have attracted as much attention as it hasSourced.

5. Electron transport layer materials

MaterialMain structureWhat the primary sources say
TiO2 (mesoporous and compact layers)n-i-pIn early cells, mesoporous TiO2 served as the scaffold that accepted electrons (Kojima et al., Kim et al.). Jiang et al. point out that planar TiO2-based n-i-p cells tend to show hysteresis and lower stabilised efficiencies
SnO2n-i-pA deeper conduction band and higher electron mobility than TiO2. Improves charge transfer from the perovskite to the electron transport layer and reduces charge accumulation at the interface (Jiang et al.)
C60, PCBM (fullerenes)p-i-n, tandemsUsed on the electron side of p-i-n cells. Chen et al. stack them as PCBM / BCP / Cu. In tandems, ALD SnO2 on top of C60 is standard (Hooijer et al.)

All Sourced (Kojima et al. [Ref. 1], Kim et al. [Ref. 2], Jiang et al. [Ref. 4], Chen et al. [Ref. 9], Hooijer et al. [Ref. 8]).

6. Hole transport layer materials

MaterialClassWhat the primary sources say
spiro-OMeTADSmall organic molecule (top side of n-i-p)The most widely used small-molecule hole transport material, originally developed for solid-state dye-sensitised solar cells. Because of its synthesis cost, low thermal stability and low intrinsic conductivity and mobility, Schloemer et al. consider it unsuited to commercialisation
PTAA, P3HT, PEDOT:PSSPolymersRelatively high intrinsic conductivity and mobility. Solution-processable, with reasonable thermal stability and hydrophobicity (PEDOT:PSS, which is hydrophilic, is the exception). However, batch-to-batch variation in molecular weight changes their properties and performance, and keeping molecular weight consistent raises cost
NiOx, CuSCN and othersInorganicPromise better stability but need high process temperatures and the like, so oxide hole transport layers are mainly used in p-i-n cells (beneath the perovskite). CuSCN can also be used in n-i-p cells, but coating it without dissolving the perovskite is difficult
SAMs (self-assembled monolayers)Monolayer (bottom side of p-i-n)Me-4PACz and others. Credited with lifting inverted cells above regular ones in efficiency (Chen et al.). Prone to detaching when annealed at high temperature (same source)

All Sourced (Schloemer et al. [Ref. 6], Chen et al. [Ref. 9]). SAMs are covered in detail in our explainer on self-assembled monolayers.

Using a SAM as the hole-selective layer was also the approach in the 2020 paper in which a group including HZB reported a silicon tandem with a certified efficiency of 29.15%, where it was shown to speed up hole extractionSourced.

7. A materials engineer's view (1): the dopant turned out to be a hygroscopic salt

Why this matters for materials engineers: the additive that raises conductivity cuts lifetime

Small organic molecules such as spiro-OMeTAD conduct poorly on their own, so they are used with dopants (additives). According to Schloemer et al., the dopants used almost without exception have been a lithium salt (LiTFSI) and a cobalt complex (FK209), which have been shown to raise hole mobility and conductivity while damaging the stability and lifetime of perovskite solar cellsSourced. The review adds that dopants often contain hygroscopic, mobile ions (such as Li⁺), which severely reduce stability whether or not the cell is encapsulatedSourced.

Materials engineers will recognise the pattern. Add an ionic additive to get conductivity, and the ions attract water and migrate. It is the same kind of problem as an antistatic agent compounded into a plastic blooming to the surface, or a resistance value that drifts with humidity (our commentary).

Polymer hole transport materials bring a different materials problem. Schloemer et al. point out that batch-to-batch variation in molecular weight changes thermal, morphological and optoelectronic properties and performance, and that keeping molecular weight consistent raises cost considerablySourced. Controlling molecular-weight distribution is exactly the kind of quality control polymer makers deal with every day. Reproducing in production the efficiency obtained from one lab batch depends on batch-to-batch reproducibility of this layer (our commentary).

8. A materials engineer's view (2): stacking order decides which materials you can use

Taking the 2025 p-i-n cell of Chen et al. as an example, here is the temperature at which each layer is made.

Process temperature falls as the stack grows: a p-i-n example (Chen et al., 2025) Bar length = deposition or annealing temperature (°C) of each layer, in stacking order from the left 500 °C 100 °C 110 °C 100 °C 70 °C 70 °C Evaporated NiO SAM Perovskite Surface treatment PCBM BCP Cu electrode Spray pyrolysis Previously 150 °C Note: temperatures and processes follow the fabrication procedure in Chen et al. [Ref. 9]. Bars drawn at 1 °C = 0.34 px. Note: an example from a single paper; it does not represent p-i-n process temperatures in general.
Fig. 4 Concept diagram (vector drawing). The temperature and process for each layer follow the fabrication procedure in Chen et al. (Nature Communications, 2025) [Ref. 9]. This article has drawn one paper's example, and it does not represent the process temperatures of p-i-n cells in general. The perovskite layer is shown at the paper's “low-temperature sequential deposition” condition (110 °C); the conventional condition for comparison is 150 °C.
Why this matters for materials engineers: the thermal budget is used up from the bottom layer upwards

In the cell of Chen et al., only the first NiO layer is made by spray pyrolysis at 500 °C; after that the SAM is processed at 100 °C, the perovskite at 110 °C, and PCBM and BCP at 70 °C, so the temperature falls as the stack growsSourced. Being able to put the oxide that needs high temperature at the very bottom is the advantage of p-i-n, and this is a concrete example of the point by Schloemer et al. that oxide hole transport layers are practical mainly in p-i-n cells (our commentary).

The interesting part is that the lower layers have temperatures they cannot tolerate either. Chen et al. explain that the two-step sequential deposition of the perovskite needs a high-temperature anneal at 150 °C for the phase transition, which causes the SAM to detach and increases non-radiative recombination at the buried interfaceSourced. So they lowered the energy barrier of the phase transition with an additive, achieving a complete transition at 110 °C, and obtained an open-circuit voltage of 1.21 V and a certified efficiency of 26.0%Sourced.

In other words, a film one molecule thick forced the process temperature of the layer above it down by 40 °C. In a stacked device, a material's heat resistance acts not as a standalone property but as a constraint on the process design of the whole stack (our commentary).

Interface treatments have side effects too

A 2022 Nature paper (Tan et al.) showed that defect passivation treatments applied to the interface between the perovskite top surface and the transport material can shift the interface work function in the negative direction (more n-type), activating halide ion migration and aggravating instabilitySourced. The paper says that the trade-off between the good and bad effects of surface treatment sets the ceiling on the stability gains achievable this waySourced. Filling defects does not automatically make things better. Ion migration is covered in our explainer on ion migration and degradation, and surface treatments in our explainer on passivation.

9. Where efficiency stands, with our calculation

CategoryEfficiencyAreaTest centre (date)Made by
Cell (very small area)28.0 ± 0.6%0.051 cm² (da)NPVM (February 2026)Hainan University
Cell (about 1 cm²)26.9 ± 0.8%1.017 cm² (da)NPVM (March 2025)Soochow University / UNSW / BaimaLake
Minimodule23.9 ± 0.5%19.48 cm² (da)NPVM (March 2025)Microquanta, 9 cells
Submodule22.9 ± 0.5%756.0 cm² (da)NPVM (September 2025)Mellow / Jinan University, 45 cells
Module22.1 ± 0.6%813.8 cm² (da)JET (November 2025)Renshine / Nanjing University
Module (large)19.3 ± 0.6%7,200 cm² (t)Fraunhofer ISE (December 2025)Renshine Solar

All Sourced (efficiency tables Version 68 [Ref. 10]). These are single-junction perovskite solar cells. Because of its small area, the 28.0% result is listed among the “Notable exceptions” and carries the table note that stability was not investigated. The other values carry notes such as initial performance. The efficiency tables do not say which transport layers or structure these cells use.

The authors of the efficiency tables note that with the 28.0% result, although the cell area is more than 3,000 times smaller, the best single-junction perovskite result has matched the best silicon result for the first timeSourced.

Our calculation: seventeen years, and the area barrier
  • Checking the efficiency: the 28.0% cell gives Voc 1.199 V × Jsc 26.76 mA/cm² × FF 0.871 = 27.95 mW/cm² → 28.0% (matches the table)Our calculation
  • Compared with 2009: 28.0 ÷ 3.8 = about 7.4 times. But the 2009 value was for a liquid-electrolyte cell, with a different area and different measurement conditionsOur calculation
  • Gap between a cell of about 1 cm² and modules: 26.9 − 22.1 = 4.8 points (813.8 cm²) and 26.9 − 19.3 = 7.6 points (7,200 cm²)Our calculation
  • Area ratio: 7,200 ÷ 1.017 = about 7,080 timesOur calculation

Assumptions and limits: the module values use different area definitions (da and t) and include losses from interconnection and scribing. The measurement dates also differ, so not all of the gap is an area effect.

10. Strengths and open problems

Conceptual image of the edge of a plain glass plate against a dark background, where faint coloured bands catching the light suggest very thin stacked films
Fig. 5 AI-generated concept image. An impression of a structure in which many thin films with different properties are stacked on glass. It does not show the real number, thickness or colour of layers, a cross-sectional image, or a product.
Strengths and challenges seen from transport-layer materials (this article's framing) Strengths Challenges Wide choice of inorganic and organic materials SnO2 can be made at a low 150 °C p-i-n lets high-temperature oxides go below SAMs lifted inverted cells past regular ones Inverted cells also suit tandems well 26.9% on about 1 cm² (independently measured) Hygroscopic dopants and mobile ions Low thermal stability of hole transport layers Polymers: molecular-weight spread and cost SAMs detach easily at high temperature Interface treatment can trigger ion migration Gap to large modules exceeds 7 points Note: items follow the papers [Refs. 1 to 9], the efficiency tables [Ref. 10] and our calculations (Section 9). Note: the split into two groups is this article's own and is not an established industry assessment.
Fig. 6 Concept diagram (vector drawing). Each item follows the papers [Refs. 1 to 9], Version 68 of the efficiency tables [Ref. 10] and this article's calculations (Section 9). Sorting them into strengths and challenges is this article's own framing, not an established industry assessment.

(1) There is still no definitive transport layer

As of 2019, the review by Schloemer et al. found it hard to beat spiro-OMeTAD on performance, and summarised research as moving in two directions: materials conductive enough without doping, and new dopantsSourced. Inverted cells using SAMs have since advanced, but which combination of materials will become the production standard could not be confirmed in the primary sources within the scope of this articleNot yet confirmed.

(2) Long-term durability has to be checked structure by structure

Chen et al. assessed encapsulated cells with the ISOS-D-3 (85 °C / 85% RH in the dark) and ISOS-L-2 (maximum power point tracking under illumination) protocols, but these are accelerated tests, and how many years the cells last outdoors could not be confirmed in the primary sources within the scope of this articleNot yet confirmed. Test methods are covered in our explainer on outdoor testing and durability assessment.

The article in summary
  • A charge transport layer is a film that lets only one kind of charge through, and one sits on each side of the absorberSourced
  • n-i-p puts the electron transport layer at the bottom; p-i-n (inverted) puts the hole transport layer there. Inverted cells are said to have overtaken regular ones in efficiency too, thanks to advances in SAMsSourced
  • Hole-transport-layer dopants (such as LiTFSI) contain hygroscopic, mobile ions and cut lifetimeSourced
  • Oxides that need high temperatures can only go at the bottom, and process temperatures fall as the stack grows. In one example the perovskite deposition temperature was lowered from 150 °C to 110 °C to keep the SAM from detachingSourced
  • 26.9% for a cell of about 1 cm² and 19.3% for a 7,200 cm² module (efficiency tables). The gap is 7.6 pointsOur calculation

11. Glossary

Electron transport layer (ETL)
A layer that carries electrons generated in the absorber to the electrode and blocks holes.
Hole transport layer (HTL)
A layer that carries holes generated in the absorber to the electrode and blocks electrons.
n-i-p (regular)
A structure stacked from the substrate as electron transport layer / perovskite / hole transport layer.
p-i-n (inverted)
A structure stacked from the substrate as hole transport layer / perovskite / electron transport layer.
spiro-OMeTAD
A small organic molecule long and widely used for hole transport. Used with dopants.
Dopant
An additive used to raise conductivity or mobility, such as LiTFSI or a cobalt complex.
Fullerenes (C60, PCBM)
Cage-shaped molecules of carbon atoms and their derivatives. Used as the electron transport layer in p-i-n cells.
BCP
A thin organic layer inserted between the electron transport layer and the metal electrode.
SAM (self-assembled monolayer)
A film one molecule thick that bonds to the underlying surface. Used on the hole side of p-i-n cells.
Hysteresis
Current-voltage characteristics that differ depending on the direction of the voltage sweep.
Mesoporous structure
A film with nanometre-scale pores. Used as a scaffold for the absorber in early cells.
Sequential deposition (two-step method)
Making a lead iodide film first and then reacting it with an organic salt to form the perovskite.
Work function
The minimum energy needed to remove an electron from a solid. It governs how energies line up at an interface.

12. References (primary sources)

  1. Kojima, A. et al. “Organometal Halide Perovskites as Visible-Light Sensitizers for Photovoltaic Cells”, Journal of the American Chemical Society 131, 6050–6051 (2009) https://doi.org/10.1021/ja809598r
  2. Kim, H.-S. et al. “Lead Iodide Perovskite Sensitized All-Solid-State Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%”, Scientific Reports 2, 591 (2012, open access) https://doi.org/10.1038/srep00591
  3. Lee, M.M. et al. “Efficient Hybrid Solar Cells Based on Meso-Superstructured Organometal Halide Perovskites”, Science 338, 643–647 (2012) https://doi.org/10.1126/science.1228604
  4. Jiang, Q. et al. “Enhanced electron extraction using SnO2 for high-efficiency planar-structure HC(NH2)2PbI3-based perovskite solar cells”, Nature Energy 2, 16177 (2016) https://doi.org/10.1038/nenergy.2016.177
  5. Tan, S. et al. “Stability-limiting heterointerfaces of perovskite photovoltaics”, Nature 605, 268–273 (2022) https://doi.org/10.1038/s41586-022-04604-5
  6. Schloemer, T.H. et al. “Doping strategies for small molecule organic hole-transport materials: impacts on perovskite solar cell performance and stability”, Chemical Science 10, 1904–1935 (2019, open access) https://doi.org/10.1039/c8sc05284k
  7. Al-Ashouri, A. et al. “Monolithic perovskite/silicon tandem solar cell with >29% efficiency by enhanced hole extraction”, Science 370, 1300–1309 (2020) https://doi.org/10.1126/science.abd4016
  8. Hooijer, R. et al. “Why more junctions do not yet deliver: interconnection challenges in perovskite multijunction solar cells”, Energy & Environmental Science 19, 4582–4596 (2026, open access) https://doi.org/10.1039/d6ee01631f
  9. Chen, M. et al. “Low-temperature sequential deposition for efficient inverted perovskite solar cells”, Nature Communications 16, 5746 (2025, open access) https://doi.org/10.1038/s41467-025-61144-y
  10. Green, M.A. et al. “Solar cell efficiency tables: Version 68”, Joule 10, 102494 (2026) https://doi.org/10.1016/j.joule.2026.102494

13. Claim-to-source audit

Claim in the textBasisLabel
A photoelectrochemical cell using perovskite nanocrystals as a sensitiser on mesoporous TiO2, with 3.8% efficiency for CH3NH3PbI3Kojima et al. (JACS, 2009). Reference 1 https://doi.org/10.1021/ja809598rSourced
The solid hole transport material spiro-MeOTAD filled into the pores of mesoporous TiO2, 9.7% efficiency, and much better stability than liquid-junction cellsKim et al. (Scientific Reports, 2012). Reference 2 https://doi.org/10.1038/srep00591Sourced
A meso-superstructured cell with insulating mesoporous alumina as the scaffold and electrons travelling through the perovskite, 10.9% efficiency and open-circuit voltage above 1.1 VLee et al. (Science, 2012). Reference 3 https://doi.org/10.1126/science.1228604Sourced
The tendency of planar TiO2-based n-i-p cells to show hysteresis and lower stabilised efficiency. That SnO2 has a deeper conduction band and higher electron mobility than TiO2. SnO2 nanoparticles made by low-temperature solution processing, almost no hysteresis, certified efficiency of 19.9 ± 0.6%, and fabrication at 150 °CJiang et al. (Nature Energy, 2016). Reference 4 https://doi.org/10.1038/nenergy.2016.177Sourced
That surface passivation treatments can shift the work function in the negative direction (more n-type), activating halide ion migration and aggravating instability, and that this trade-off sets the ceiling on stability gainsTan et al. (Nature, 2022). Reference 5 https://doi.org/10.1038/s41586-022-04604-5Sourced
That the hole transport layer passes holes and blocks electrons, reducing recombination and raising fill factor and efficiency. The need to match the energy levels and mobilities of the layers. The reasons the hole transport layer is critical to stability (direct contact with the perovskite, hygroscopic mobile ions such as Li⁺, often the lowest thermal stability). The definitions of n-i-p and p-i-n. That introducing a solid hole transport layer was the first breakthrough and that without spiro-OMeTAD the technology might not have drawn attention. That perovskites cannot withstand harsh processing of later layers, so oxide hole transport layers, which need high temperatures and the like, are practical mainly in p-i-n cells. The difficulty of coating CuSCN. The conductivity and mobility of polymer HTMs (PTAA, PEDOT:PSS, P3HT), the hydrophilicity of PEDOT:PSS, and molecular-weight variation and cost. That spiro-OMeTAD was developed for solid-state dye-sensitised cells and is unsuited to commercialisation because of synthesis cost, low thermal stability and low conductivity and mobility. That LiTFSI and FK209 are used almost universally and damage stability and lifetime. That dopants contain hygroscopic, mobile ions that reduce stability whether or not the cell is encapsulated. The two research directions of dopant-free materials and new dopantsSchloemer et al. (Chemical Science, 2019). Reference 6 https://doi.org/10.1039/c8sc05284kSourced
Use of a self-assembled monolayer as the hole-selective layer to speed hole extraction, and a certified efficiency of 29.15% for a silicon tandemAl-Ashouri et al. (Science, 2020). Reference 7 https://doi.org/10.1126/science.abd4016Sourced
That most high-efficiency tandem demonstrations use thin-film recombination junctions in the p-i-n configuration. That in tandems ALD SnO2 is deposited on top of C60Hooijer et al. (Energy & Environmental Science, 2026). Reference 8 https://doi.org/10.1039/d6ee01631fSourced
That inverted cells have drawn attention for ease of fabrication, low hysteresis, high stability and compatibility with flexible and tandem devices, and have overtaken regular cells in efficiency through advances in SAMs. That the phase transition in sequential deposition needs 150 °C, at which the SAM detaches and non-radiative recombination increases at the buried interface. A complete phase transition at 110 °C with an additive, open-circuit voltage of 1.21 V and certified efficiency of 26.0%. The fabrication procedure (FTO / NiO: spray pyrolysis at 500 °C / Me-4PACz: 100 °C / perovskite: 110 °C (conventionally 150 °C) / PEACl: 100 °C / PCBM: 70 °C / BCP: 70 °C / Cu 90 nm, evaporated). Assessment by ISOS-D-3 and ISOS-L-2Chen et al. (Nature Communications, 2025). Reference 9 https://doi.org/10.1038/s41467-025-61144-ySourced
Single-junction perovskite entries in Version 68 of the efficiency tables: 28.0 ± 0.6% (0.051 cm², da, NPVM, February 2026, Hainan University, Voc 1.199 V, Jsc 26.76 mA/cm², FF 87.1%, with the note that stability was not investigated); 26.9 ± 0.8% (1.017 cm², NPVM, March 2025); minimodule 23.9 ± 0.5% (19.48 cm²); submodule 22.9 ± 0.5% (756.0 cm²); modules 22.1 ± 0.6% (813.8 cm², JET, November 2025) and 19.3 ± 0.6% (7,200 cm², total area, FhG-ISE, December 2025). The note on the 28.0% result that, although more than 3,000 times smaller, it matched the best silicon result for the first timeEfficiency tables Version 68 (Joule, 2026). Reference 10 https://doi.org/10.1016/j.joule.2026.102494Sourced
The check of 28.0% as Voc × Jsc × FF (27.95 mW/cm²). The ratio of about 7.4 to 2009. Cell-to-module gaps of 4.8 and 7.6 points. The area ratio of about 7,080. The proportional scale in Fig. 4Our calculation. The values compared differ in area definition, measurement conditions and dateOur calculation
Which structure and which combination of transport layers are used in mass-produced products. Long-term outdoor durabilityNot stated because they could not be confirmed in the primary sources within the scope of this article (commentary by this article)Not yet confirmed
Framing the transport layer as a film that passes only one kind of charge. Reading the history as three materials changes. Treating the ionic-dopant problem as the same kind as antistatic-agent bloom and humidity dependence. The reading that control of molecular-weight distribution affects reproducibility in production. The framing that the thermal budget is used up from the bottom layer, and the reading that a film one molecule thick forced the upper layer's process temperature down by 40 °C. The classification of materials (Fig. 2 and tables) and the sorting into strengths and challengesThis article's own framing and commentary based on published content. Not views expressed by the researchersCommentary
That Figs. 1, 2, 3, 4 and 6 are explanatory drawings, and that the hero image and Fig. 5 are AI-generated imagesA note by this article (commentary)Commentary

Last updated 25 September 2026. Sources are limited to primary material (the efficiency tables and peer-reviewed papers). Every efficiency value is given with whether it is certified, whether it is a cell or a module, its area and the test centre. Which transport layers the cells and modules in the efficiency tables use, the structure of mass-produced products, and long-term outdoor durability are not stated, because they could not be confirmed in published primary sources. All figures are for explanation. Figs. 1, 2, 3, 4 and 6 are vector drawings, and the hero image and Fig. 5 are AI-generated images; none of them shows a real cross-section, micrograph or physical product.

🌐 Japanese