I Have a Research Support System

In the simulation laboratory, Xu Qiu found that combining the two strategies of vacuum storage for a period of time and top battery ternaryization can indeed achieve the effect of 1+1\u003e1.

However, the highest device efficiency at this stage still has not exceeded 17%, only 16.94%, compared with the previous values ​​of 16.66% and 16.83%, the increase rate is not high.

Xu Qiu estimated that it was because the exploration time was relatively short.

This kind of pioneering exploration work can only be carried out by Simulation Lab III.

In Simulation Lab III, there are only two senior simulation experimenters. Even if they work 24 hours a day, the combined work efficiency is only about ten times that of reality.

For the system of ordinary devices, it takes one or two days for the simulation laboratory III to go out, which is equivalent to more than ten or twenty days of continuous work in reality, which is enough to explore the conditions more perfectly.

However, the groping work of stacked devices is very cumbersome, and it is obviously not enough to give the simulation laboratory a few days.

Having said that, the simulation laboratory has actually done a lot of work, and initially obtained a two-dimensional map of the device efficiency as a function of the thickness of the top cell and the bottom cell.

It's just that the precision of this two-dimensional map is not enough, and further experiments are needed to find out the best conditions.

Xu Qiu found two main isoefficiency lines from the two-dimensional map.

14% iso-efficiency line: Under the condition that the thickness of the top cell is in the range of 90-180 nm and the thickness of the bottom cell is in the range of 120-300 nm, the efficiency of the obtained laminated device can reach more than 14% in most cases.

16% iso-efficiency line: when the thickness of the top cell is in the range of 120-150 nm and the thickness of the bottom cell is in the range of 180-210 nm, the efficiency of the device can reach more than 16% in most cases.

The reason why it is said to be in most cases is that the obtained iso-efficiency line is not a rectangle, but a shape similar to a triangle.

It is also easy to understand. For example, under the condition of the first 14% iso-efficiency line, two boundary conditions are selected, the thickness of the top cell is 90 nanometers, and the thickness of the bottom cell is 300 nanometers. The performance of the device obtained in this way will definitely not be very high.

Because the bottom cell is very thick, it will absorb more light, and the short-circuit current density is higher, while the thickness of the top cell is relatively thin, and the obtained short-circuit current density is small, which is difficult to match with the bottom cell, which will cause the device performance loss.

Therefore, what Xu Qiu has to do now is to find out the highest efficiency point in the 16% iso-efficiency line.

He intends to go into battle himself and conduct experiments.

When the simulation experimenters fumbled before, they fumbled with a thickness of 30 nanometers as the interval.

Xu Qiu plans to use 10 nanometers as the precision, so there are 4 grades in the thickness of the top battery within the range of 120-150 nanometers, and there are also 4 grades in the thickness of the bottom battery within the range of 180-210 nanometers, that is, a total of 4 * 4 = 16 conditions.

If 3 batches of devices are produced for each condition, and each batch of devices is repeated 3 pieces, a total of 16*3*3=144 pieces, this is too much, and it will definitely not be completed in one and a half hours.

If 2 batches of devices are made for each condition, and each batch of devices is repeated 2 pieces, a total of 64 pieces seems to be quite a lot.

After thinking for a moment, Xu Qiu decided to continue to lower the standard:

One batch was made for each condition, and one device was replicated for each batch, so the total number of devices was reduced to only 16 devices.

Generally, photovoltaic devices have to be repeated for more than 10 batches, but now that Xu Qiu doesn't have so much time, he can only hope that he will be more European, and he can break through 17% in one shot!

Considering that the next experimental work might consume a lot of time, Xu Qiu returned to reality first.

He looked around and found nothing unusual, then adjusted a more comfortable sitting position and returned to the simulation laboratory.

Then, Xu Qiu opened the simulation laboratory I that had been dusty for a long time.

With the continuous upgrading of the system, the simulation laboratory I can now enable the acceleration function of up to 64 times.

However, because Simulation Lab II and III can automatically hang up, Xu Qiu rarely uses Simulation Lab I.

But in fact, 64 times the acceleration, this function is still very powerful.

For the evaporation operation, it would take an hour to pump a vacuum, but under the condition of 64 times speed, it only takes one minute.

It needs to be left for 12 hours after evaporation, and it only takes 10 minutes if converted at 64 times speed.

Of course, the point consumption is also very exaggerated, 6400 points need to be consumed for one hour at 64x speed.

Fortunately, Xu Qiu now has a lot of remaining points, more than 20 W, which is enough for him to squander.

In addition to the speedup in Simulation Lab I, there are many other additional benefits.

In one aspect, the substrate does not need to be cleaned.

Moreover, because stacked devices are being made recently, the current substrates have been spin-coated with two layers of transmission layers for stacked devices, zinc oxide and PFN-Br, according to the best conditions.

As a result, spin coating can be started directly from the active layer of the bottom cell, saving a considerable amount of time.

On the other hand, simulation experimenters have obtained the correspondence between spin coating speed and film thickness through several spin coating experiments, combined with optical absorption spectrometer, scanning electron microscope and other means.

Xu Qiu can directly spin coat according to the specified speed, and then get the corresponding film thickness without repeated exploration, which also saves a lot of time.

Ready to start experimenting.

Xu Qiu took the substrate that had been spin-coated with two layers of transport layers, and first started to spin-coat J4:IDIC-M with different thicknesses as the effective layer of the bottom battery.

In the process of spin coating, it cannot be accelerated by 64 times in the whole process.

For example, processes such as clamping substrates must be accelerated at normal speeds or at low speeds.

However, it is still possible to accelerate while the substrate is rotating.

In conversion, it usually takes 2 minutes to spin coat one piece, but now at 64 times speed, it takes about 10-15 seconds.

After painting 16 pieces, it took a total of less than 5 minutes of real time.

Next, Xu Qiu continued to spin-coat M-PEDOT as the third transfer layer, and spin-coat zinc oxide as the fourth transfer layer.

These two layers need to be wiped and annealed after spin coating.

Although the annealing takes ten to fifteen minutes, because all the substrates can be annealed together, the actual time consumption is negligible under the 64 times acceleration.

Two-step spin coating, plus wiping and annealing, the total time is about 30 seconds per substrate, and the total consumption is less than 10 minutes.

Xu Qiu continued to spin coat the active layer of the top cell, PCE10:PCBM:COi8DFIC.

The time consumption of this step is similar to the spin coating of the effective layer of the bottom battery, which is also less than 5 minutes.

The final step is to vapor-deposit molybdenum trioxide and silver electrodes.

Before the evaporation, Xu Qiu suddenly had an idea. He copied the 16 substrates that were spin-coated into 10 copies, and prepared to evaporate 10 times at the same time, so that 10 batches of devices could be obtained.

Although the effective layers of these 10 batches of devices are all the same, because of the difference in the evaporation operation, the influence of the evaporation operation on the device performance can be approximately excluded.

Including the subsequent vacuum placement operation, Xu Qiu also plans to replicate multiple devices at the same time, and then explore the impact of different placement times on device performance.

In this way, although he only made a batch of devices, he has actually completed repeated experiments including evaporation and vacuum placement at the same time, which has a greater probability of making his device efficiency fluctuate higher.

This can be regarded as an advantage as a human being. If the simulation personnel in the simulation laboratory are operating, they will not be able to play Xu Qiu's routine. They will only prepare batches of devices step by step.

I just don't know if Xu Qiu has developed this method, it remains to be seen whether they will learn it simultaneously.

Next, Xu Qiu evaporated 10 batches of devices and placed them in a vacuum. Under the acceleration of 64 times, the total time was less than 30 minutes.

All 10 batches of devices were prepared, which took about 45 minutes in total.

It seems that Xu Qiu only made 16 devices under the initial conditions, but after two replications, the total number of devices has expanded to more than 1,000 devices.

Finally came the exciting testing session.

Because of the acceleration, the test is still relatively fast.

Basically, the connection is completed, and the result can be displayed in seconds.

It takes about 15 seconds on average to test and get a result.

Xu Qiu selected the batch of devices with the best hand feeling during evaporation, and uniformly selected the vacuum storage time condition as 12 hours.

Open the cabin and test it.

Xu Qiu's strategy is to directly discard devices with an efficiency lower than 16.5% in the initial test.

If it can reach 16.5% for the first time, then give it three scanning opportunities, and if the performance does not reach 17%, just discard it.

In this way, Xu Qiu tested 1# to 12# one after another, a total of 12 devices.

Among them, the best system is 7#, with a maximum efficiency of 16.96%, which is very close to 17%.

At that time, Xu Qiu made an extra exception and gave it a few more chances. As a result, the test got lower and lower, and finally he had no choice but to give up.

Until the 13th device, Xu Qiu finally got the data of the initial efficiency exceeding 17% for the first time, reaching 17.07%.

He changed the position of the baffle and scanned ten consecutive data, the highest efficiency was 17.11%.

Xu Qiu thought about it, first tested 14#, 15#, and 16#, and found that there were no other devices with an initial efficiency exceeding 17%, and then began to specialize in 13#.

The processing condition corresponding to 13# is the condition that the thickness of the top cell is about 130 nanometers, and the thickness of the bottom cell is about 190 nanometers.

Xu Qiu found hundreds of brothers and sisters of the 13# device for testing.

First of all, the other nine 13# devices in different evaporation batches under the same 12-hour vacuum storage time, after the test, the highest efficiency increased to 17.27%.

Next, for the 13# devices under the best batch of 17.27% efficiency, the devices under different storage times were tested. The results showed that the performance of the devices with a storage time of 16 hours was the best, up to 17.36%.

After getting the final result of 17.36%, Xu Qiu let out a long breath.

After more than an hour of tossing, he basically didn't have a moment to rest, and he was focused on performing high-intensity operations throughout the whole process.

Xu Qiu felt as if his body had been hollowed out, and he had entered a state of sage time.

Fortunately, the results are very good, and finally a breakthrough has been made.

Moreover, it is only the result of preliminary exploration. If the amount of data can be increased before, there is still room for further improvement in device performance.

Therefore, the probability of repeating more than 17% efficiency in reality is very high.

Xu Qiu glanced at the time, there were still about 20 minutes until the end of the exam, so he was not in a hurry to go out.

He first handed over the current best conditions to the simulation experimenters for batch repetition, and then began to take stock of the process of stacking devices step by step.

At the beginning, it was based on a semi-transparent device to prepare a relatively simple four-terminal method laminated device. At that time, the bottom cell used a semi-transparent device. It was found that even with a thin layer of metal electrodes, the light loss was still very high. The final device The efficiency is always less than 10%, which shows that the four-terminal method is not suitable for organic photovoltaic systems.

Therefore, Xu Qiu chose the two-terminal method and started to try again. As a result, several systems were tried and finally the efficiency reached 10%.

Later, after a period of process exploration, Xu Qiu chose two high-efficiency systems, the bottom cell J2: IDIC-4F, and the top cell PCE10: IEICO-4F. The device efficiency finally broke through 12%, breaking the current standard of stacked devices at that time. world record.

Later, Xu Qiu optimized the device structure of the two-terminal method stacked device. Instead of using the middle thin layer electrode as the charge recombination layer, it was directly replaced by two almost transparent transport layers, which can significantly reduce the charge recombination layer. The light loss of the small top cell device greatly improves the current density of the top cell, and the efficiency jumps to 14%.

Then, Xu Qiu introduced PCBM into J2:IDIC-4F to regulate the light absorption between the top cell and the bottom cell, so that the short-circuit current densities of the two can be matched more easily, successfully increasing the efficiency to 15%, a breakthrough A recognized threshold in the field of organic photovoltaics.

Then, Xu Qiu saw the failure of the Y series receptors in the multilayer device, and felt that when designing the structure of the multilayer device, the efficiency of the original single-junction device should not be used as the benchmark, but more consideration should be given to the bottom layer. The adaptation of the battery and the top battery, so he replaced IDIC-4F with IDIC-M whose light absorption range is biased towards the short wavelength range, and further improved the device efficiency to 15.5%.

Then, Xu Qiu tried to find near-infrared non-fullerene acceptors developed by other research groups to replace the IEICO-4F used in the original group, and found that the COi8DFIC developed by Li Dan's research group at the National Center for Nano Science and Technology was similar to the previous one. His own system is the best match, and the final efficiency exceeds 16%.

A few days ago, the school girl made a batch of devices on a whim. As a result, the efficiency of the actual devices surpassed the results of the simulation laboratory. After research, it was found that vacuum placement can improve the performance of some system devices. Through this strategy, the efficiency was successfully improved. to more than 16.5%.

At the same time, Xu Qiu also got inspiration from the Sanyuan article of Li Dan's research group, took the PCBM out of the effective layer of the bottom cell, put it in the top cell, and finally increased the device efficiency to more than 16.5%.

Now, Xu Qiu has integrated the two major strategies of vacuum storage and top battery ternaryization, and personally operated, and finally achieved a device efficiency of 17.36%!

After taking inventory, Xu Qiu was deeply moved. It was really not easy to get the current result.

Even with the big killer of the system, it took more than two months to optimize the stacked device step by step from scratch.

This is Xu Qiu's longest task so far.

Calculated based on the fact that the work efficiency of the simulation laboratory is ten times that of reality, if it is converted into real time, the process of exploration may take as long as one or two years.

In fact, this is also a CNS-level job, which generally requires a lot of work.

Unless it's the kind of discovery that breaks new ground, like tearing graphene with tape or something like that.

But at the current stage, it is very difficult to open up this new field.

Even a Nobel Prize-winning scientific researcher cannot guarantee that he will be able to open up a new field in his lifetime. It depends on luck.

Therefore, at this stage, they are all fighting in the existing field, trying to make a breakthrough in a certain dimension.

The time required for a breakthrough is usually calculated in years.

This is also the difference between CNS and AM. Xu Qiu’s step-by-step optimization process of stacked devices, if he disassembles and publishes an article, every step of optimization can publish an AM, and some even reach the level of the big sub-journal of Nature. level.

Now Xu Qiu put them together just to hit a CNS article.

In addition to articles, Xu Qiu broke the world record of organic photovoltaics again, and was significantly ahead of other international researchers. This is a very milestone event.

Xu Qiu can be regarded as winning glory for the country.

In fact, before 2015, domestic research on photovoltaic energy was very backward.

According to the progress chart of various types of photovoltaic efficiency issued by the National Renewable Energy Laboratory NREL, including silicon, copper indium gallium selenide, gallium arsenide, cadmium telluride, organic photovoltaics, perovskite and other types of efficiency world Records, less than 10% of the domestic list.

If it is less than 10%, it is a good way of saying it. If it is changed to a bad word, it is almost 0. There are almost no domestic universities or research institutes on the list...

NREL itself accounts for a large proportion. They are beautiful countries' research institutes specializing in photovoltaics and energy. There are several departments in it, which tackle different fields. Wei Xingsi came from there before.

In addition, there are many organizations on the list, including the beautiful country of University of California, Los Angeles, UCLA; Kangaroo country, University of New South Wales, UNSW; Swiss Federal Institute of Technology in Lausanne, EPFL and so on.

Of course, the statistical time of this chart starts from 1975, and there are almost no domestic lists, which is also related to the fact that the domestic scientific research circle has only started to make efforts in the past ten years.

At this stage, in addition to organic photovoltaics, the domestic development of perovskite photovoltaics is also very good, and many breakthroughs have been made, which can be regarded as a two-pronged approach with organic photovoltaics.

At this time, in reality.

Wang Jizeng on the podium has been observing Xu Qiu, so he has naturally heard of Xu Qiu's name.

Not only that, but he often went to the website of Wei Xingsi's research group to check if Xu Qiu had any new work published recently.

As a result, there are almost new articles coming out every time I read them, and they are all articles above the AM level that I dream of.

Therefore, Wang Jizeng was very curious about how this legendary figure managed to post articles like drinking water?

As a green pepper scholar, the article is a mountain on their heads.

Especially for an associate professor like Qingjiao who has no background like him, whether he can be promoted to a professor depends almost entirely on his academic achievements.

Seeing Xu Qiu on the list of invigilating teaching assistants earlier, Wang Jizeng still felt a little inexplicable.

I wondered if I could take this opportunity to get acquainted with each other and ask for the secret of posting articles, but found that Xu Qiu was very cold and sat in the back seat after the test papers were distributed.

Then, nothing happened.

It's the real kind of motionless.

Wang Jizeng sighed in his heart, sure enough, other people's success is not accidental.

This kind of determination cannot be achieved by oneself.

At this moment, Wang Jizeng suddenly saw Xu Qiu move, as if he was about to get up.

Then I got up halfway, didn't get up, fell back and sat down again, and made a slight knocking sound.

What is this doing, a special method of cultivation?

Sure enough, the boss's world is different from that of ordinary people like me.

On the other hand, when Xu Qiu returned to reality from the simulation laboratory, he found himself sitting in one motion for an hour, and his buttocks and legs were numb.

I wanted to stand up, but I found that I couldn't get up halfway, so I sat back down again, but I couldn't control my strength well and made a duang sound. Fortunately, there were no other students around, otherwise it would be embarrassing.

This reminded him of the dark history before.

At that time, Xu Qiu entered the simulation laboratory in a squatting state in the bathroom, but when he returned to reality, he found that his legs were numb and almost fell into the pit...

Therefore, if he wants to do experiments in the simulation laboratory for a long time in the future, he will basically choose to enter the simulation laboratory on the bed in the dormitory.

If you are usually outside, you usually just go in and take a look at the results before running away, and you won't stay inside for too long.

After sitting on the seat for a while, Xu Qiu finally returned to normal.

He got up and inspected it, and found that the students in this year were relatively well-behaved, and none of them cheated.

But if you think about it carefully, this is a professional elective course, and the difficulty may not be very high.

At the same time, the difficulty of the questions in the closed-book exam is generally relatively low, mainly to prevent students from failing.

On the contrary, because the information can be consulted in the open-book exam, the teacher will set the difficulty higher, otherwise it will be difficult to give a score if all the students get full marks in the end.

In order to distinguish Grade A from other grades in the open-book exam, the teacher will either ask a few differentiated questions when writing questions, for example, the teacher of Composite Materials will test out the questions that are not in the PPT that have been mentioned in the class; Or set a lot of questions in the test paper to screen out those students who write slowly.

Xu Qiu strolled around and finally stepped onto the podium.

He found that looking down from above, the angle of view is very clear, almost unobstructed, and if there is anything abnormal about the students below, he can spot it at a glance.

Xu Qiu suddenly understood that there was a high probability that the invigilators could also detect the cheating that he could detect before, but they chose to turn a blind eye and close one eye.

After a while, the bell for the end of the exam finally rang again.

There were not many people who handed in the exam ahead of time, only three, and the rest were dragged to the end.

Generally, if you want to hand in the papers in advance, you hand in very early, like 20 minutes or half an hour in advance.

When the exam is about to end, if you hand in the paper in advance, the invigilator will generally not agree.

Because it may be messy, it will affect the teacher's collection of papers, and when binding the test papers, they need to be arranged according to the student number.

If there are too many people who submit the papers in advance, the order is out of order, and it needs to be arranged manually, which will be more troublesome.

PS: Ask for a guaranteed monthly pass.