I started with a ship full of max-level veterans, and I'm just a complete noob.

Following the force of Wu Hanzhang's fingertips, Jiang Ming did not hesitate to pull out the page of the preprint abstract and move it under the glaring halo of the work light to examine it line by line.

Compared to the title, the data presented in the abstract is extremely impactful: the dangling bond density on the surface of germanium [111] is as high as 10 to the power of 13 per square centimeter, which is about five times higher than the theoretical value calculated from the number of broken bonds in an ideal lattice.

The extra portion originates from surface reconstruction, adsorption of impurities, and the subsurface damage layer left by mechanical polishing.

This means that each dangling bond is highly susceptible to becoming a charge trap, thus forming a fixed charge region on the surface of the germanium sheet that is extremely difficult to control.

Jiang Ming opened his kraft paper notebook, first drawing the straight energy bands inside the body, then bending the conduction band and valence band near the surface to one side, and filling in the dense surface state energy levels in the middle of the band gap.

Once the Fermi level is pinned to the surface by these levels, the surface barrier will be freed from the control of the bulk doping concentration, and the depletion layer width, which was originally calculated based on the diffusion concentration, will also shift accordingly.

He picked up the two batches of peeling test pieces again and compared the oxide film cracks, leakage at the window edge, and the resistivity jump of the four probes one by one. Finally, the two problems on the paper came together.

The poor adhesion of the oxide film originates from the defect interface between the germanium substrate and the film layer. The high leakage current at the interface is also inseparable from this high-density surface state layer.

For a double-diffused triode, this effect is concentrated near the junction of the base and emitter. Minority carriers continuously fall into the trap as they pass through, affecting both current gain and noise performance.

Wu Hanzhang stared at the image on the mask, holding his cigarette in his hand for a long time without lighting it. Then he pushed the two comparison boxes in front of Jiang Ming.

"Based on your assessment, the peeling is just a visible problem right now. What if there's a hidden electrical leak inside the pipe after it's made?"

"If the interface state density is kept at this level, even if the window is protected, the electrical boundary cannot be protected. We must first treat surface treatment and masking film as the same process."

Jiang Ming copied the measurement conditions from Xie Xide's abstract to the side. The frustration of being blocked by the oxide film gradually turned into clear pressure. The root of the problem had been found, but the tools to solve it were scattered across several units.

"Please, Mr. Huang, make the arrangement. I would like to see her photoconductivity spectrum in person and also bring the data from these two batches of diffusion test pieces with me."

Wu Hanzhang picked up the secure phone and first explained the device interface problem encountered by 502 to the Institute of Physics, and then asked Huang Kun to convey his request for discussion.

That afternoon, Huang Kun called back to arrange a meeting time and reminded them to bring photos of the membrane before and after diffusion, leakage data, and substrate processing records.

When Jiang Ming arrived at the Institute of Physics, Xie Xide was organizing experimental curves for submission. Several photoconductivity spectra were spread out on the table, with wavelength scales and sample numbers written in pencil on the corners.

She first looked at the letter of introduction that Huang Kun handed her, then scanned the process briefing of 502, her gaze lingering on the words "factory research lab" for a while.

"What you're discussing right now is diffusion masking. Surface state measurement and device fabrication are separated by several layers of conditions. Let's talk about actual samples first."

Jiang Ming took out two batches of test film photos, one at 620 degrees and the other at 580 degrees, and connected the measurement locations of the oxide film cracking from the window edge, spreading laterally by about ten micrometers, and the leakage current rising on the outside to the same image.

He then turned to the diagram of band bending to explain his judgment on the interface defect, only clarifying the phenomena that 502 had already measured, without treating the inference as an experimental conclusion.

When Xie Xide heard that both batches of test pieces had peeled off from the window opening, he stopped tidying up the curves and pulled out a set of photoconductivity spectra that had not yet been copied from the bottom of the folder.

"Look at this section. After mechanical polishing, the defect energy levels of the [111] sample are most densely packed in the middle of the band gap. After acid washing, the peak value will drop, but it will return after being placed in the atmosphere."

She pressed the leakage current variation curve of 502 below the photoconductivity spectrum and adjusted it to a similar position along the horizontal axis, gradually aligning the high value regions of the two sets of data.

Jiang Ming bent down to check the sample surface treatment records and found that the high-density area measured by Xie Xide corresponded to the area where the resistivity jump was most concentrated on the outside of their diffusion window.

This overlap is not enough to directly prove causality, but it incorporates surface states, film adhesion, and interface leakage into the same verifiable path.

Xie Xide then took out three more sets of processing data. One set of germanium sheets was first soaked in concentrated hydrochloric acid, then washed with deionized water, and then immediately sent into a low vacuum device and heated at 400 degrees Celsius for ten minutes.

"After this treatment, the density of dangling bonds can be reduced by about 30%. However, when the sample leaves the vacuum and comes into contact with air again, the effect deteriorates quickly. I judge that the surface is re-oxidized, or it may have adsorbed gas molecules."

As Jiang Ming stared at the first curve after processing, the structure of the old vacuum chamber 502 came to mind. When Yuan Lei No. 1 was doing anode copper block welding, it had undertaken the tasks of evacuation and heating, and there was still room for modification.

"After the low-vacuum heating is completed, if a protective film is deposited directly in the vacuum to seal the treated surface, can this 30% be retained?"

Xie Xide placed the pencil next to the spectrum and flipped back to the experimental conditions page before answering.

"I can do photoconductivity testing here, but the thin film deposition equipment is temporarily unavailable. This method is worth doing. You'll need to figure out the film material and interface stress yourselves."

Over the next two hours, Jiang Ming listed out the vacuum level, heating temperature, and quartz material conditions that 502 could provide, while Xie Xide indicated the control samples that needed to be retained for the retest of surface density of states.

In the latter part of the discussion, she took the initiative to write the pickling concentration, cleaning time, and exposure time into the joint test form, and also proposed that at least one blank control piece be retained for each batch of samples, which was only annealed and did not deposit.

"You can copy the raw photoconductivity spectrum data, and the subsequent processing results should be shared. The official report should also indicate that these data came from my preprint."

"Okay, the deposition record, film thickness, and post-diffusion leakage data for 502 will be sent simultaneously, and the sample will also be archived according to your numbering rules."

Huang Kun remained seated at the recording table to the side until both sides agreed on the terms, at which point he wrote down two lines in his hardcover notebook: "502 device direction and surface state, device interface, joint project".

When Jiang Ming saw those two lines of records, he realized that the collaboration between 502 and the Institute of Physics had officially extended from furnace temperature, purification, and single crystal growth into device physics.

After returning to the factory, he wrote the entire process on the blackboard, from polishing germanium wafers to soaking in hydrochloric acid, cleaning with deionized water, low-vacuum annealing, and then to vacuum evaporation of silicon oxide thin films and double diffusion.

Fang Xudong copied the items into the timetable one by one, and when he wrote about the evaporation of silicon dioxide, his pen tip hovered over the column for material sources.

"While the vacuum chamber can be modified, where will the evaporation source come from? And how can the purity of readily available silica powder be verified?"

Old Sun took out a paper package left over from the last time he processed the quartz crucible from the materials cabinet. Inside were the cut edges of the quartz tube material that Wang Shouwu had sent. The serial number and purity certificate were still affixed to the seal.

"We've been keeping these scraps; they're too small to make crucibles, but they're perfect for evaporation crucibles."

Jiang Ming took out a piece and examined it under the light. He then instructed Fang Xudong to add the cleaning, drying, and empty-load evaporation of the quartz fragments to the process. The vacuum chamber renovation was also scheduled to be among the top priorities this week.

Just as Fang Xudong posted the new timetable on the blackboard, Wu Hanzhang brought back an airmail progress notice from the factory.

The von Kármán's second reply has entered the delivery area near Pasadena. Based on the current communication cycle, if the reply is sent in a timely manner, a response signal may appear within two weeks at the earliest.

Jiang Ming tucked the notice into the old hardcover memo book, turned to the reserved page for the third communication, wrote "variable cross-section coupling" after "material interface change", and drew a connection line that was not yet closed.

On the other side of the blackboard, Lao Sun was pouring a bag of quartz fragments into a clean tray, the crisp sound of the fragments colliding echoing across the workbench.

According to the newly finalized plan, the next round of experiments must be conducted before the germanium surface re-adsorbs impurities, and then it must be firmly sealed into a new film.