The efficiency of photovoltaic cells is approaching the 30% mark: Heterojunction reduces costs and speeds up, and the industrialization of stacked cells has dawned
After decades of development, crystalline silicon solar cells are very close to their theoretical conversion efficiency limit: 29.4%. To achieve another jump in photoelectric conversion efficiency on this basis, laminated cells are considered the most promising technical direction. Recently, the development of tandem technology has been progressing rapidly: In May, King Abdullah University of Science and Technology (KAUST) announced that the certified efficiency of perovskite/crystalline silicon tandem solar cells reached 33.7%, which once again refreshed the world record of 33.2% created in April. record. A few days ago, LONGi Green Energy, a domestic leader in photovoltaics, announced that it has achieved a conversion efficiency of 33.5% for crystalline silicon-perovskite stacked cells on commercial-grade textured CZ silicon wafers.
Although the above-mentioned efficiency record is still in the laboratory stage, and there is a long distance from commercial mass production, it is enough to show the technical potential and imagination of stacking.
Compared with the imagery of stacking, many people in the industry prefer to call it a multi-junction battery. "The photovoltaic industry is showing two major characteristics, from P to N, from single to multiple. The N-type era has spawned TOPCon, heterojunction, and XBC popular technology routes. The theoretical efficiency of single-crystal silicon cells is 29.4%, and it needs to break through 30% , must be multi-junction. According to our communication with many perovskite research institutions, they also believe that perovskite technology may 'borrow the chicken and lay eggs'. In other words, crystalline silicon stacked perovskite, not perovskite Mineral-stacked perovskite is expected to be the first path to achieve mass production of double-junction cells." Zhao Xuewen, who has been involved in the photovoltaic industry for more than 20 years, told The Paper.
In 2020, LONGi Green Energy spent 1.78 billion yuan to acquire Ningbo Jiangbei Yize New Energy Technology Co., Ltd., thereby realizing the expansion of battery and module production capacity in Southeast Asia. The core of the transaction is Vietnam's photovoltaic assets - China's largest overseas photovoltaic cell module factory that has provided OEMs for domestic first-tier manufacturers for a long time. Zhao Xuewen is one of the founders of Ningbo Yize. After fulfilling the performance promise, he returned from Vietnam to start his own business, keenly smelling the next new business opportunity: the large-scale industrialization of heterojunction is just a step away, and the timetable for the mass production of heterojunction-perovskite laminated cells is also short. It is becoming clearer day by day.
The solar potential energy 10.8GW heterojunction cell and module intelligent manufacturing project with a total investment of 7.5 billion yuan has recently started construction in Lujiang, Anhui. This is the starting point for photovoltaic veteran Zhao Xuewen to enter the new photovoltaic track. According to his prediction, around 2027, the heterojunction-perovskite stack technology is expected to achieve mass production.
The cost reduction of heterojunction has reached the critical verification period
Regardless of whether the battery technology is from P to N or from single-junction to multi-junction, it is essentially an iteration of photovoltaic technology for the purpose of reducing costs and increasing efficiency.
In the past two or three years, the battle over the battery technology route has never stopped. The mainstream P-type monocrystalline PERC cell in the market is still the most economical technology in the photovoltaic power generation industry at present, but the mass production efficiency has approached its theoretical limit of 24.5%, and it is difficult to greatly improve it. The successor is N-type cells, and TOPCon, heterojunction, and XBC have become the most popular options. Among the three, the current cost-effectiveness, compatibility of equipment, and relatively controllable operational risks make TOPCon's expansion wave the most radical.
"TOPCon is really aggressive." Zhao Xuewen, the current chairman of Anhui Solar Potential Energy New Energy Co., Ltd. (hereinafter referred to as Solar Potential Energy), said.
Judging from its experience, the largest overseas foundry that once operated the PERC golden age should have a greater inclination towards TOPCon. Zhao Xuewen admitted that at the beginning of the solar potential energy preparation, he also considered the TOPCon option, but after hesitating for one night, he still chose the heterojunction and heterojunction stacking route. "Since Japan's Sanyo Corporation successfully developed the heterojunction in 1990, its reliability and stability have been proven over time. At the same time, as a platform technology, the heterojunction itself has long enough vitality and can be superimposed with innovation. The combination of mines also has natural advantages.”
TOPCon wins in the present, while HJT wins in the future. This buzzword in the industry can be described as a high-level summary of the strengths and weaknesses of the two technologies. The composition of the two camps is also quite different: new entrants are more willing to invest in heterojunction with sufficiently disruptive technology, while veteran players in the industry, especially those with huge PERC stock capacity, are more willing to choose TOPCon.
The biggest disadvantage of heterojunction is still too expensive. According to Peng Mei News, although the production process is simple, the manufacturing difficulty of heterojunction in each link is much higher than that of PERC cells, and the requirements for cleanliness, fineness, uniformity and continuity are very strict. Equipment and silver paste - these cost "two hurdles" are the crux of the expansion of heterojunction production capacity and camp expansion.
Under the clear logic of cost reduction, the new technology must run smoothly, and the stability of the equipment must be verified, and the timing is uncertain. As a result, in the past year, compared with the attack on TOPCon's production capacity, the cost reduction progress of heterojunction seems to be lower than industry expectations.
"Last year, the double-sided microcrystalline technology did not fully run through, and the related hardware equipment is not yet mature. In addition, cost reduction technologies such as 0BB and silver-clad copper have to go through a market verification cycle." Zhao Xuewen did not evade this. He said that by the end of May this year, the four technologies of double-sided microcrystalline, double-sided silver-clad copper, 110μm thin film, and 0BB have been applied to the mass production line at the same time. In particular, the slurry consumption per watt has been reduced to a level of 16mg/W. "The next three months will be the time for the heterojunction industry to verify the data. By the end of August, we will be able to see the efficiency, cost, and yield data of the above four important cost-reducing and efficiency-enhancing technologies under full production conditions. "
Heterojunction manufacturer Risen Energy (300118) also disclosed related progress recently. The company's heterojunction mass production line in Jintan has been simultaneously imported into the pilot test line and has verified mature related technologies, including double-sided microcrystalline technology, thin silicidation technology, and low-silver paste technology. The average efficiency of the first batch of 210-size half-cell ultra-thin heterojunction solar cells rolled off the production line reached 25.4%. As of the beginning of June, the yield rate of this mass production line has reached over 90% in less than a month of commissioning.
Zhao Xuewen believes that the above progress means that heterojunction is very close to the state of "significant cost-effective mass production advantage", and the fastest wave of cost reduction will appear in the second half of this year.
Solar Potential Energy’s average module shipment power target after stable production in the first half of next year is 730W. “In 3-4 years, the average module power in the industry will reach about 770W. This is already the limit of heterojunction single-junction cells. .Then to components above 800W, it is necessary to rely on heterojunction plus perovskite stacking technology to achieve.”
"The large-scale expansion of TOPCon is not particularly worrying from the perspective of the heterojunction industry." He told The Paper that there is a lot of room for improvement in the former's real efficiency and yield, and the problem of overcapacity is also serious. "Although the overall scale of the heterojunction industry is an order of magnitude smaller than that of TOPCon, and the shipment next year will be 30-40GW, the profit per watt of heterojunction next year is worth looking forward to."
The above-mentioned yield rate is not a family statement. Some people in the industry told The Paper that the competition between different N-type technologies should not be based solely on the volume of sound. "Many TOPCon companies are not as good as they claim in terms of yield and cost. The overall progress of volume expansion is actually lower than expected."
Lamination technology that cannot be bypassed by perovskite
From aluminum back field to PERC, the industrialization success of every photovoltaic cell technology is inseparable from "the right time, place and people". Zhao Xuewen and his team have been observing the progress of heterojunction technology for many years, waiting for an opportunity to make a move.
"Until the end of last year, we judged that the problems of insufficient economics of heterojunction and immature industrial chain supporting facilities would be basically resolved by the end of 2023. Therefore, referring to this timetable, we started construction three quarters in advance and waited until the heterogeneous junction When the cost-effectiveness of mass production of mass-junction cells comes out, it will just go into production and reach production.” It is reported that the 10.8GW heterojunction project invested this time will be implemented in batches: the first phase of 3.6GW will start in June 2023, and the first phase of 1.8GW will be planned It will be put into production at the end of this year and early next year, and will reach production capacity in the second quarter of 2024; the second phase of the first phase of 1.8GW will be half a year later than the first phase; the second phase of 7.2GW will be expanded in the second quarter of 2024, and the Lujiang project will be fully operational by the end of 2025 . After the project is fully completed, it is expected to realize an annual output value of about 16.5 billion yuan.
As for the mass production node of heterojunction-perovskite laminated cells in the industry, the time point given by Zhao Xuewen is around 2027. "At present, the optimal process route and equipment route are still in the process of demonstration. After these are finalized, and then undergo pilot tests, small-scale mass production, and cost reduction, and truly enter the state of profitable large-scale mass production, the industry is still in the process of It will take some time." From the perspective of the company, Solar Potential Energy has mainly focused on heterojunction single-junction technology in the past three years. At present, it has stacked technology reserves, and the average conversion efficiency of long-term mass production of photovoltaic cells will exceed 30%.
Huasheng New Energy, which has the world's largest heterojunction production capacity and shipments, previously disclosed that in the research and development of heterojunction-perovskite stacked cells, the goal is to realize G12 heterojunction perovskite crystalline silicon stacked cells in 2025. 30% efficiency. To achieve the above goals, it is necessary to work with equipment manufacturers and material manufacturers to promote the improvement of the cost performance of heterojunction as soon as possible.
Lamination technology is in full bloom. The current mainstream view is that the stacking of crystalline silicon and perovskite can complement each other, and the heterojunction is most suitable for stacking in terms of process flow.
"For the technical direction of more than 30% efficiency, the most likely mass production is the heterojunction-perovskite stack. The laboratory record has now reached 33.7%, while the laboratory record of TOPCon-perovskite stack is not yet to 30%. The difference in mass production prospects between the two is greater, due to issues such as the complementarity of the spectrum and the conductivity of the anti-reflection layer.” Zhao Xuewen said.
The speed of large-scale commercial mass production of heterogeneous single-junction and multi-junction products also depends on the strategic changes of leading companies.
The Paper noticed that there are many new players in the heterojunction track, but all the records are currently reflected in leading manufacturers. For example, the owner of the world's highest conversion efficiency record of 26.81% for heterojunction cells is LONGi, with mass production of 210* The 66-module power record of 732.6 watts was set by Tongwei's heterojunction team.
Due to economic considerations, although leading manufacturers have not expanded their heterojunction production capacity on a large scale, they have maintained technical reserves and closely tracked the technology. Leading companies with tens of billions of cash in hand have flexibility in market strategy adjustments. Once heterojunctions show enough attractiveness in reducing costs and increasing efficiency, it is not difficult for major manufacturers to expand their production capacity by 50GW in one fell swoop.
"When the new forces of heterojunction are rushing forward, the giants are looking down in the air like eagles. When the time is right, they will bend down, rapidly expand the huge production capacity, and directly change the industry structure." Zhao Xuewen told The Paper. said the news.