MENU

JinkoSolar | Company Profile | Perovskite Solar Cells

Back to Technology, Institution & Company Guides

COMPANY PROFILE / PEROVSKITE SOLAR CELLS

JinkoSolar
The TOPCon leader stacking perovskite on top, with a disclosed research budget of about CNY 211 million

JinkoSolar is one of the leading makers of N-type TOPCon, the technology that has become the mainstream in silicon solar cells. In perovskites it keeps announcing efficiencies for tandem cells that use its own TOPCon cell as the bottom cell, and in June 2026 it reported reaching 34.82%. What sets the company apart is that the annual report of its listed subsidiary shows the budget and spending for its perovskite research project as line items. It is one of the few established makers whose commitment can be checked in money terms. This profile sets out those amounts and what the company has not yet disclosed.

Sources: the 2025 annual report and 2026 interim report of Jinko Solar Co., Ltd. (Shanghai Stock Exchange), the Form 20-F of JinkoSolar Holding (NYSE) for the year ended December 2025, official press releases, and an announcement by its partner XtalPi, used as primary material. Last updated September 2026.

What this article covers
  1. JinkoSolar in 30 seconds, and where it stands in perovskite solar cells
  2. The logo and how the name is written
  3. Where the company sits in the perovskite solar cell field
  4. Scale (the silicon business in brief)
  5. Resources committed to perovskites
  6. What it has done so far
  7. What comes next
  8. What the company is aiming for in perovskite solar cells
  9. Map of perovskite partnerships
  10. Technology, products and services
  11. The risks this company carries
  12. Glossary, references and claim-to-source audit

1. JinkoSolar in 30 seconds, and where it stands in perovskite solar cells

LEGAL NAMEJinko Solar Co., Ltd.The English name of the Shanghai-listed operating company. Its parent is JinkoSolar Holding Co., Ltd.
HEADQUARTERSShangrao, Jiangxi Province, ChinaRegistered address. Offices in Minhang District, Shanghai
LISTINGSShanghai STAR Market 688223
Parent: NYSE: JKS
The parent holds about 54.16% (20-F)
CATEGORY IN THIS FIELDEstablished PV maker
(entering from silicon mass production)
Not a perovskite specialist
PRODUCT FORMATTandem with
TOPCon
Its own N-type TOPCon cell serves as the bottom cell
BEST CELL EFFICIENCY34.82%Company-reported as certified by the Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences (June 2026). Area not disclosed
RESEARCH PROJECT BUDGETAbout CNY 211 millionPlanned total investment in "high-efficiency, stable, large-area perovskite cells" (2026 interim report)
PRODUCTION OUTLOOKNo timing given by the company itselfIts partner XtalPi quotes Jinko as expecting mass production "within about three years"
The one thing to grasp first

Companies working on perovskite solar cells fall into three broad types: established PV makers, perovskite start-ups, and materials and electronics firms coming in from other industries. JinkoSolar is an established PV maker, and what stands out is that the silicon bottom cell it uses is its own mainstay, the mass-produced N-type TOPCon cell Sourced. In release after release the company says it has "demonstrated the compatibility of TOPCon with perovskite tandems". For JinkoSolar, in other words, perovskite is the next add-on to TOPCon, a technology in which it already has enormous capacity, not a technology to switch to by abandoning TOPCon.

2. The logo and how the name is written

The JinkoSolar logo is the company's trademark. Rather than recreate it with an image generator, this article uses the logo image from the official website, saved as img/jinkosolar_logo.png (the image on the website is small, at 146 × 59 pixels, so it may look coarse).

A note on names: there are two listed companies. One is the holding company listed on the New York Stock Exchange, JinkoSolar Holding Co., Ltd. (incorporated in the Cayman Islands); the other is its subsidiary, the operating company listed on Shanghai's STAR Market, Jinko Solar Co., Ltd., whose legal name is registered in Chinese. The 20-F calls the latter "Jiangxi Jinko" and puts the parent's stake at about 54.16%. Perovskite R&D is described in most detail in the operating company's annual report, so the figures in this article come mainly from its disclosures.

Official site: https://www.jinkosolar.com/en

3. Where the company sits in the perovskite solar cell field

Three types of company working on perovskite solar cells All of them "work on perovskites", yet their origins and aims could hardly differ more Established PV makers Perovskite start-ups Materials & electronics firms Build on silicon mass production Compete on perovskites alone Enter with skills from other fields LONGi / JinkoSolar Trina Solar / Hanwha Qcells, etc. Oxford PV / Swift Solar EneCoat / Renshine Solar, etc. Sekisui Chemical / Panasonic Toshiba / Kaneka, etc. Already own sales and factories Can lead on sharp technology Film, glass and sealing know-how Silicon remains the core business High hurdles: scale-up, funding Thin sales channels for solar JinkoSolar sits at the far left, developing tandems on top of its mainstay TOPCon cells It discloses its research project budget, but has not itself given a production date, site or capacity
Fig. 1 The three types of company working on perovskite solar cells, and where JinkoSolar sits. The classification is our own, and the companies shown are examples drawn from those covered in this series Our calculation. The use of TOPCon as the bottom cell is based on the official releases; the research project budget is based on Jinko Solar Co., Ltd.'s 2025 annual report and 2026 interim report.

The product format is a tandem with silicon (TOPCon) Sourced. The releases describe an N-type TOPCon bottom cell with perovskite stacked on top, and they also mention work on an intermediate recombination layer, so the device can be read as a monolithic two-terminal tandem (the number of terminals is not stated explicitly). The company has not announced any plan to commercialise a film-type or glass-type single-junction perovskite product.

4. Scale (the silicon business in brief)

Figures are from the disclosures of the Shanghai-listed operating company, Jinko Solar Co., Ltd., in CNY million Sourced.

CNY million (years to December)202320242025
Operating revenue118,68292,47165,492
Net profit attributable to shareholders of the parent7,44099−6,882
R&D input (company definition, including pilot-scale costs)–4,4072,535 (3.87% of revenue)

Figures rounded to the nearest CNY million Our calculation. In the first half of 2026, operating revenue was CNY 24,727 million and the net loss CNY 3,076 million.

ItemFigureAs of, and source
Employees (parent company and main subsidiaries)26,388End of December 2025 (Jinko Solar Co., Ltd. annual report)
R&D staff2,158 (8.18%)Same source. A year earlier: 1,981 (5.86%)
R&D staff (group)2,184End of December 2025 (20-F). Based in Haining, Shangrao and Xining, and in Vietnam
Module shipment target for 202660 to 70 GWCut from 75 to 85 GW in the 2026 interim report
A caution

The figures above are for the company as a whole, almost all of which is silicon wafers, cells and modules, plus energy storage. The company swung to a net loss in 2025, and its R&D input fell 42.49% from the year before (the company cites lower prices for research materials) Sourced. To gauge the scale of the perovskite effort, you need the project-level figures in the next section.

5. Resources committed to perovskites

Jinko Solar Co., Ltd.'s annual report includes a table of "research projects in progress", and fifth on the list is "Research on key technologies for high-efficiency, stable, large-area perovskite solar cells". Few established makers disclose their perovskite research spending project by project.

Item (CNY)2025 annual report2026 interim report
Planned total investment231,031,924.15210,961,124.15
Spending in the period76,776,889.17 (2025)54,462,832.41 (first half of 2026)
Cumulative spending125,668,742.58146,859,354.44
Progress as describedDeveloped a set of technologies including high-quality fabrication of large-area perovskite layers. Best laboratory tandem: 34.76%As before. Best laboratory result: 34.82%
GoalTo take perovskite cell technology from laboratory research to the development of process plans for future volume production
Our calculation: is CNY 200 million a lot or a little?

The CNY 76.78 million spent on this project in 2025 is about 3.0% of that year's total R&D input of CNY 2,535 million Our calculation. In the same table, the "all back-contact (BC)" project received CNY 325 million in 2025 and the "high-efficiency N-type bifacial module" project CNY 540 million, so perovskites run at about a quarter of the BC project and about a seventh of the bifacial module project. That is small next to TOPCon, the mainstay, and its surrounding technologies, yet it is still a single in-house project measured in hundreds of millions of yuan.

In the 2026 interim report, the planned total investment fell by about CNY 20.07 million, and the rise in cumulative spending (about CNY 21.19 million) does not match the first-half spending (about CNY 54.46 million). The annual report notes that "adjustments to sub-projects mean investment amounts differ from earlier disclosures", which suggests the scope was reorganised Not yet confirmed. This article does not adjust the numbers and shows them as disclosed.

Type of resourceWhat it isCategory
People (dedicated organisation)The size of any dedicated perovskite team is not disclosed. R&D staff total 2,158, and in 2025 the company says it "added experienced engineers to support the upgrade of the pilot-line platform"Sourced (dedicated headcount not disclosed)
People (leadership)The person quoted in releases is the CTO, Dr Jin HaoSourced
Money (research)The project above has a planned total investment of about CNY 211 million, with about CNY 147 million spent cumulatively (end of June 2026)Sourced
Money (production line)No investment in a perovskite tandem production line has been disclosedNot confirmed
Joint venturesA strategic cooperation agreement with XtalPi and a joint venture for AI- and robot-driven high-throughput research (January 2026; 20-F and the partner's announcement)Sourced
Universities and research institutesEfficiencies were certified by the Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, and by the National Photovoltaic Industry Metrology and Testing Center (NPVM). No joint research agreement has been disclosedSourced (joint research not confirmed)
Public fundingNo government project funding for perovskites is disclosed. "High-efficiency perovskite/TOPCon tandem solar cells" was included in the Chinese Renewable Energy Society's list of major scientific and technological advances in photovoltaics for 2024 (a recognition, not funding)Sourced (funding not confirmed)

6. What it has done so far

None of JinkoSolar's releases gives the cell area. The cells are most likely small research devices of around 1 cm², but this article treats the area as "not disclosed".

WhenWhat happenedCertification and area
Before May 202432.33% on a TOPCon-based tandemRecapped in the May 2024 release (company-reported)
31 May 202433.24%, using ultra-thin polysilicon passivating contacts, light trapping, and an intermediate recombination layer with high transmittance and high mobility, among other featuresMeasured by the Shanghai Institute of Microsystem and Information Technology, CAS (company-reported) / area not disclosed
6 January 202533.84%, with full-area passivating contacts and passivation of interface and bulk defectsSame body (company-reported) / area not disclosed
2025Reached 34.22% (cited as the "previous record" in the November 2025 release)Company-reported
27 November 202534.76%. The CTO said tandems are "the next disruptive direction after TOPCon" and that "the next stage is moving from the laboratory to large-scale application"NPVM (company-reported) / area not disclosed
January 2026Strategic cooperation with XtalPi, and a joint venture for AI high-throughput research20-F and the partner's announcement
June 202634.82%, with double-layer composite passivating contacts, multi-dimensional interface passivation, control of gradient crystallisation and stronger optical couplingShanghai Institute of Microsystem and Information Technology, CAS (company-reported) / area not disclosed
Reading the record numbers

When this article checked, JinkoSolar's perovskite tandems did not appear in NLR's (formerly NREL's) cell efficiency data table or its champion module efficiency chart Sourced. The same table does list LONGi's perovskite/silicon tandem at 35.2% (JRC-ESTI, about 1 cm²). JinkoSolar's 34.82% is a value the company says was measured by a Chinese research institute, with no area given, so it is best read within the scope of the company's own claim: a record for tandems built on TOPCon Not yet confirmed.

7. What comes next

2026 RESEARCH POLICYUse an AI research platform to drive development and industrialisationFrom the "development strategy" section of the 2025 annual report
RESEARCH PROJECT GOALTowards process plans for volume productionThe "goal" column of the research project table
AI EXPERIMENTAL LINE1,000 m², 1,000 samples a dayXtalPi's announcement. No completion date given
PRODUCTION TIMINGWithin about three years (per the partner)XtalPi writes that "Jinko expects mass production within about three years". Not in Jinko's own disclosures

For 2026, Jinko Solar Co., Ltd.'s 2025 annual report says it will "use its AI research platform to advance the R&D and industrialisation process of perovskite tandem cells and seek further breakthroughs in conversion efficiency" Sourced. The business plan section likewise speaks of "accelerating perovskite tandem R&D and laying the foundation for future industrial application", which shows the company itself placing the work before mass production rather than at it.

What could not be confirmed

None of the following appears in Jinko Solar Co., Ltd.'s reports, the 20-F or the official releases Not yet confirmed.

When tandem mass production will start ("within about three years" appears only in the announcement by its partner, XtalPi); the location, capacity and investment of a production line; the area of the record cells; efficiencies of large-area cells or modules; the ownership split and registered capital of the joint venture with XtalPi; and outdoor reliability data.

8. What the company is aiming for in perovskite solar cells

This section goes beyond what the company says directly. It sets out inferences that can be drawn from primary sources, each with its basis.

What can be readPrimary evidence behind itCategory
The real target is a tandem that makes use of the existing TOPCon linesEvery release stresses the "compatibility of TOPCon with perovskite tandems". The 20-F also discusses its TOPCon patent count and tandems side by sideInference
It is trying to move from research to designing production processesThe research project's goal is "from laboratory research to the development of process plans for volume production". In 2025 it added engineers to support the upgrade of the pilot-line platformSourced (the stated goal)
It is buying speed in materials discovery with AI and robotsThe joint venture with XtalPi, the 1,000-samples-a-day experimental line, and the "AI research platform" in the 2026 research policyInference
Space applications are in view as well as ground-based onesXtalPi's announcement mentions reducing the area of satellite solar wings. Jinko Solar Co., Ltd.'s annual report also refers to commercial space expanding the range of usesInference
For now, the money-maker remains TOPConThe 2026 R&D and production plans centre on raising TOPCon efficiency, cutting costs and mass-producing new modulesSourced
One inference, explained in full

This article reads the insistence on a TOPCon bottom cell as a consequence of most of JinkoSolar's equipment and patents being in TOPCon. The 20-F says the company held 1,484 registered TOPCon-related patents at the end of 2025, and Jinko Solar Co., Ltd.'s annual report says it has "the largest number of N-type TOPCon patents in the world" Sourced. Whereas LONGi's tandems use HJT (heterojunction) as the bottom cell, JinkoSolar appears to be seeking a path in which the TOPCon cells it already mass-produces become the bottom cell as they are, so that its existing cell factories can carry on as the front end of tandem production Not yet confirmed. TOPCon cells, however, involve high-temperature steps, so their surface condition and compatibility with the layers stacked on top differ from HJT. That the company keeps citing work on "double-layer composite passivating contacts" and an "intermediate recombination layer" shows that this junction is where the difficulty lies.

9. Map of perovskite partnerships

JinkoSolar's perovskite (tandem) relationships Only ties stated in the operating company's reports, the 20-F, official releases and XtalPi's announcement JV, joint development (navy) Ownership, certification (dotted); certification is not a partnership Jinko Solar Shanghai-listed XtalPi AI high-throughput research JV (January 2026) JinkoSolar Holding NYSE-listed parent Stake of about 54.16% SIMIT (CAS, Shanghai) Measured 33.24%, 33.84%, 34.82% (certification body) National PV Metrology Center NPVM: certified 34.76% (certification body)
Fig. 2 JinkoSolar's perovskite (tandem) relationships, centred on the operating company, Jinko Solar Co., Ltd. Sources: JinkoSolar Holding's Form 20-F (year ended December 2025), official releases (May 2024, January and November 2025, June 2026) and XtalPi's announcement. Solid navy lines are joint ventures and joint development; dotted lines are ownership and efficiency measurement or certification. SIMIT is the Shanghai Institute of Microsystem and Information Technology of the Chinese Academy of Sciences (CAS). Certification bodies are third parties contracted to take measurements, not partners. XtalPi's announcement names the counterparty to the agreement as "a subsidiary of Jinko Solar". Partnerships with materials or equipment makers could not be confirmed in primary sources and are not shown.

10. Technology, products and services

As of September 2026, JinkoSolar does not sell any product that uses perovskites Not yet confirmed. The technical elements it has published are as follows.

ElementWhat the primary sources say
Bottom cellA high-efficiency N-type TOPCon (passivating contact) cell developed in house, improved step by step with ultra-thin polysilicon, full-area passivating contacts and double-layer composite passivating contacts
Middle layerAn intermediate recombination layer with high optical transmittance and carrier mobility, and new intermediate composite-layer materials
Perovskite layerPassivation of defects at the bottom interface and in the bulk, a new crystallisation method (kinetic control of gradient crystallisation) and optimised vertical charge transport
OpticsNew light-trapping technology, and stronger optical coupling and light management
Scaling upA set of technologies, including high-quality fabrication of large-area perovskite layers, already developed (from the progress column of the research project table). No large-area efficiency has been published
Research methodA closed-loop experimental line using AI and robots with XtalPi (formulations, processes and test results are turned into data and fed back into the model)
A materials engineer's view: the hard part of a TOPCon base is the joint

In a two-terminal tandem, current flows because electrons generated in the lower silicon cell and holes generated in the upper perovskite cell meet and annihilate in the intermediate recombination layer. That layer has to meet conflicting demands: let light through to the cell below with as little absorption as possible, while connecting the two cells electrically with no resistance. This is why JinkoSolar's 2024 release highlighted "an intermediate recombination layer with high transmittance and high mobility" Sourced.

A TOPCon cell carries a thin polysilicon film on its surface; unlike HJT, it does not come with a transparent conductive layer already in place. To put perovskite on top, the polysilicon contact has to be made thin and transparent, and a new recombination layer built on top of it. The repeated references to "ultra-thin polysilicon" and "double-layer composite passivating contacts" can be read as the process of re-engineering TOPCon for tandem use Not yet confirmed. For charge transport layers and cell structure, see also our explainer on perovskite-silicon tandems.

11. The risks this company carries

RiskSubstanceCategory
Losses in the core business and shrinking R&DA net loss of CNY 6.88 billion in 2025 and another of CNY 3.08 billion in the first half of 2026. R&D input fell 42.49% year on year in 2025Sourced
Shipment target cutThe 2026 module shipment target was lowered from 75 to 85 GW to 60 to 70 GWSourced
Limited verification of the recordsCell areas are not disclosed, and the records are not in NLR's table. No large-area or module efficiency has been publishedSourced
Unclear production timing"Within about three years" appears only in the partner's announcement; Jinko's own disclosures give no timingNot yet confirmed
Tug of war with TOPCon investmentWhile the company is losing money, improvements to existing TOPCon (new modules, cost cuts) and investment in tandem production compete for the same fundsInference
Opaque joint ventureThe ownership split, registered capital and allocation of intellectual property in the joint venture with XtalPi have not been disclosed (XtalPi says only that commercialisation will be joint and led by Jinko)Not yet confirmed

12. Glossary

TOPCon
Tunnel oxide passivated contact. An N-type silicon cell that uses a thin oxide film and a polysilicon film to suppress surface recombination. Today's mainstream in volume production.
Tandem solar cell
Two cells that absorb different parts of the light spectrum, stacked on top of each other. JinkoSolar places perovskite on top of TOPCon.
Intermediate recombination layer
The layer that joins the upper and lower cells in a two-terminal tandem. It lets light through while electrons and holes recombine there so that current can flow.
Passivation
Neutralising defects at surfaces and interfaces so that charge carriers are not needlessly lost.
NPVM
The National Photovoltaic Industry Metrology and Testing Center, one of China's bodies for measuring solar cell efficiency.
Research projects in progress
A table of ongoing R&D projects that Chinese listed companies disclose in their annual reports, showing each project's planned total investment and spending.
High-throughput experimentation
A research method in which robots automatically make and test large numbers of samples. Combined with AI, it speeds up materials discovery.
Form 20-F
The annual report a foreign company listed in the United States files with the SEC.
Concept diagram: cross-section of a tandem cell with a thin perovskite layer stacked on a production silicon cell and joined by a middle layer
Fig. 3 Concept diagram of the TOPCon-based perovskite tandem JinkoSolar is developing. This is an AI-generated concept image and does not depict actual equipment or products.

13. References

  1. Jinko Solar Co., Ltd. Jinko Solar 2025 Annual Report (in Chinese; disclosed on the Shanghai Stock Exchange, 30 April 2026). Corporate information, operating revenue, net profit and R&D input, employees and R&D staff, budget and spending for the research project on high-efficiency, stable, large-area perovskite solar cells, description of perovskite/TOPCon tandem technology, and the 2026 development strategy and business plan. http://static.cninfo.com.cn/finalpage/2026-04-30/1225261496.PDF
  2. Jinko Solar Co., Ltd. Jinko Solar 2026 Interim Report (in Chinese; disclosed on the Shanghai Stock Exchange, 27 August 2026). First-half 2026 results, the revised shipment target, 34.82% (Shanghai Institute of Microsystem and Information Technology, CAS), and updated budget and spending for the research project. http://static.cninfo.com.cn/finalpage/2026-08-27/1225513196.PDF
  3. JinkoSolar Holding Co., Ltd. Form 20-F for the fiscal year ended December 31, 2025 (filed with the U.S. SEC on 29 April 2026). The stake of about 54.16% in Jiangxi Jinko, 2,184 R&D staff and their locations, 1,484 TOPCon patents, 34.76%, and the strategic partnership with XtalPi (January 2026). https://www.sec.gov/Archives/edgar/data/1481513/000110465926050737/jks-20251231x20f.htm
  4. JinkoSolar Press release, "33rd World Record! JinkoSolar Sets New Efficiency Benchmark for Perovskite/TOPCon Tandem Cells at 34.82%" (16 June 2026). 34.82% (Shanghai Institute of Microsystem and Information Technology, CAS) and the technical elements. https://www.jinkosolar.com/en/site/newsdetail/2952
  5. JinkoSolar Press release, "JinkoSolar Breaks World Record Again with 34.76% Efficiency for Perovskite/TOPCon Tandem Cell" (27 November 2025). 34.76% (NPVM), the previous record of 34.22%, and comments from CTO Jin Hao. https://www.jinkosolar.com/en/site/newsdetail/2789
  6. JinkoSolar Press release, "JinkoSolar's Perovskite Tandem Solar Cell Based on N-type TOPCon Sets New Record with Conversion Efficiency of 33.84%" (6 January 2025). 33.84% (Shanghai Institute of Microsystem and Information Technology, CAS) and the technical elements. https://www.jinkosolar.com/en/site/newsdetail/2502
  7. JinkoSolar Holding Co., Ltd. Press release, "JinkoSolar's Perovskite Tandem Solar Cell Based on N-type TOPCon Sets New Record with Conversion Efficiency of 33.24%" (31 May 2024). 33.24%, the previous record of 32.33%, and technical elements including the intermediate recombination layer. https://ir.jinkosolar.com/news-releases/news-release-details/jinkosolars-perovskite-tandem-solar-cell-based-n-type-topcon
  8. XtalPi Announcement, "Powerhouse Collaboration: XtalPi and JinkoSolar Form Joint Venture to Push Photovoltaic Efficiency Limits with AI" (January 2026). The strategic cooperation agreement with a subsidiary of Jinko Solar, the joint venture, the 1,000 m² experimental line handling 1,000 samples a day, the approach to commercialising intellectual property, and the statement that "Jinko expects mass production within about three years". https://en.xtalpi.com/powerhouse-collaboration-xtalpi-and-jinkosolar-form-joint-venture-to-push-photovoltaic-efficiency-limits-with-ai/
  9. NLR (U.S. national laboratory, formerly NREL) Cell efficiency data table. Absence of any JinkoSolar perovskite tandem entry when this article checked, and the entry for LONGi's 35.2%. https://www.nlr.gov/docs/libraries/pv/cell-efficiency-data-table.xlsx
  10. NLR (U.S. national laboratory, formerly NREL) Champion Module Efficiencies chart. Absence of any JinkoSolar perovskite-based module when this article checked. https://www.nlr.gov/docs/libraries/pv/champion-module-chart.pdf
  11. JinkoSolar Official website (source of the company name style and the logo image). https://www.jinkosolar.com/en

14. Claim-to-source audit

Claim in the articleCategorySource
Legal name, registered address and offices, STAR Market code 688223SourcedReference 1 http://static.cninfo.com.cn/finalpage/2026-04-30/1225261496.PDF
The parent, JinkoSolar Holding (NYSE: JKS), holds about 54.16%SourcedReference 3 https://www.sec.gov/Archives/edgar/data/1481513/000110465926050737/jks-20251231x20f.htm
Operating revenue and net profit (2023 to 2025), R&D input and its 42.49% fall, 26,388 employees and 2,158 R&D staffSourcedReference 1 http://static.cninfo.com.cn/finalpage/2026-04-30/1225261496.PDF
First-half 2026 operating revenue and net loss, and the shipment target cut to 60 to 70 GWSourcedReference 2 http://static.cninfo.com.cn/finalpage/2026-08-27/1225513196.PDF
Planned total investment, spending in the period and cumulative spending (2025) and the goal of the research project on high-efficiency, stable, large-area perovskite solar cellsSourcedReference 1 http://static.cninfo.com.cn/finalpage/2026-04-30/1225261496.PDF
The same project's planned total investment and spending as of the first half of 2026SourcedReference 2 http://static.cninfo.com.cn/finalpage/2026-08-27/1225513196.PDF
CNY 325 million on the BC project and CNY 540 million on the bifacial module project (2025 spending)SourcedReference 1 http://static.cninfo.com.cn/finalpage/2026-04-30/1225261496.PDF
Perovskite spending at about 3.0% of R&D input, about a quarter of the BC project and about a seventh of the bifacial module projectOur calculationOur own division of the figures in reference 1
The 2026 policy (advancing R&D and industrialisation with an AI research platform) and the business plan wording, the engineers added to support the pilot-line upgrade, and inclusion among the major advances of 2024SourcedReference 1 http://static.cninfo.com.cn/finalpage/2026-04-30/1225261496.PDF
2,184 R&D staff (group), 1,484 TOPCon patents, and the partnership with XtalPi (January 2026)SourcedReference 3 https://www.sec.gov/Archives/edgar/data/1481513/000110465926050737/jks-20251231x20f.htm
34.82% (Shanghai Institute of Microsystem and Information Technology, CAS) and the technical elementsSourcedReferences 4 and 2 https://www.jinkosolar.com/en/site/newsdetail/2952
34.76% (NPVM), the previous record of 34.22%, and the CTO's commentsSourcedReference 5 https://www.jinkosolar.com/en/site/newsdetail/2789
33.84% (Shanghai Institute of Microsystem and Information Technology, CAS) and the technical elementsSourcedReference 6 https://www.jinkosolar.com/en/site/newsdetail/2502
33.24%, the previous record of 32.33%, and technical elements including the intermediate recombination layerSourcedReference 7 https://ir.jinkosolar.com/news-releases/news-release-details/jinkosolars-perovskite-tandem-solar-cell-based-n-type-topcon
The joint venture with XtalPi, the size of the experimental line, the approach to commercialising intellectual property, and the "mass production within about three years" outlook (as stated by the partner)SourcedReference 8 https://en.xtalpi.com/powerhouse-collaboration-xtalpi-and-jinkosolar-form-joint-venture-to-push-photovoltaic-efficiency-limits-with-ai/
No JinkoSolar perovskite entry in NLR's cell efficiency data table or module chartSourcedReferences 9 and 10 https://www.nlr.gov/docs/libraries/pv/cell-efficiency-data-table.xlsx
The three types of perovskite solar cell companyOur calculationOur own classification. Companies shown are examples from those covered in this series
The inconsistency in the research project figures for the first half of 2026 (a possible change of scope)Not yet confirmedThis article's observation from comparing the figures in references 1 and 2. The company has given no specific explanation http://static.cninfo.com.cn/finalpage/2026-08-27/1225513196.PDF
Area of the record cells, large-area and module efficiencies, production timing, lines and investment, the joint venture's ownership split, and outdoor reliability dataNot yet confirmedNot in references 1 to 7. Figures from press reports and research firms are not used http://static.cninfo.com.cn/finalpage/2026-04-30/1225261496.PDF
That the device can be read as two-terminal (the number of terminals is not stated)Not yet confirmedThis article's commentary based on the description of the intermediate recombination layer in reference 7
The reading that the real target is a tandem that makes use of existing TOPCon linesInferenceThis article's interpretation of references 3 to 7
The reading that it is buying speed in materials discovery with AI and robotsInferenceThis article's interpretation of references 1 and 8
The reading that space applications are also in viewInferenceThis article's interpretation of reference 8 and the industry trends described in reference 1
The reading that TOPCon is being re-engineered for tandems, and the risk of a tug of war with TOPCon investmentInferenceThis article's interpretation of the technical elements and results in references 1, 4 and 7

Last updated 25 September 2026 / Troy Technical
Every figure in this article comes from statutory filings Jinko Solar Co., Ltd. has made with the Shanghai Stock Exchange, the Form 20-F JinkoSolar Holding has filed with the U.S. SEC, the company's official press releases, the announcement by its partner XtalPi and material published by NLR (formerly NREL). No research-firm estimates or press-based figures have been used. Passages marked "Inference" are this article's interpretation of primary sources, not statements made by the company.

🌐 Japanese