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

After scheduling this breakthrough meeting, Jiang Ming put down his pencil and pulled out the factory's official letterhead.

The meeting of the three parties will inevitably involve a large number of complicated experimental details, and the most incisive agenda must be used to get to the heart of the matter.

He unscrewed his pen and meticulously reviewed and compressed the issues that needed to be brought to the table, ultimately leaving only three key topics.

The first item concerns the reasons for the systematically low cutoff frequency of germanium double-diffused triodes, requiring the 502 and semiconductor research groups to submit data on material batches, junction widths, diffusion conditions, and test links, respectively.

The second item was assigned to Xie Xide for discussion, focusing on evaluating the quantitative impact of surface states on minority carrier lifetime, junction leakage current, and high-frequency response, allowing the photoconductivity spectrum to be cross-validated with the measured results of the device.

When Jiang Ming wrote down the third item, his pen stopped on the paper. The first two items were internal problems of the germanium process, but the third item would put the entire pre-research route back on the table.

He eventually wrote a discussion on the device roadmap for the 24 gigahertz pre-research target.

Wu Hanzhang picked up the invitation and read it from the first line to the last, then looked at the third item on the agenda. The empty cigarette pack in his hand was pressed into a crease.

"The first two items are to investigate the cause, and the third item is to prepare to replace the materials?"

"First, clarify the boundaries of the materials; whether to replace or retain them should be determined by the data."

Jiang Ming pushed over the frequency distribution diagrams of the twenty-seven tubes. The first batch was concentrated in the range of 120 to 150 MHz, and the second batch, after surface treatment, was raised to the range of 160 to 180 MHz. The shapes of the two curves were still similar.

"Surface treatment brings a 20% improvement, and it can be further modified later. However, the requirements of a 24 GHz preamplifier are to be increased by 1 GHz. The gap is beyond what can be explained by a single process optimization."

Wu Hanzhang stared at the vast blank space, his mind flashing through the furnace temperature curves, single crystal numbers, and device tapes from the past few months. He was reluctant to give up germanium easily, but he was even more afraid that the pre-research team would spend a year on a path that would never lead to the end.

"Send it out. We'll have a meeting in three days."

The invitation was delivered to the Institute of Physics that day. After reading the three topics, Huang Kun wrote down the words "device," "interface," and "material" in his hardcover notebook, and circled the third word with a pencil.

He picked up the phone and connected to Xie Xide's office, reading the meeting agenda from beginning to end. When he got to the material route, he could hear the sound of pages turning on the other end of the phone.

Do you have surface state comparison data for silicon?

"I just finished compiling a set recently. Both germanium and silicon have undergone mechanical polishing, acid pickling, and vacuum annealing, so the measurement conditions can be standardized."

Xie Xide spread the recording paper on the table. The defect peak in the middle of the band gap of the germanium sample was still obvious, while the curve of the silicon sample was lowered by a whole section after surface oxidation.

She originally only wanted to study the effects of different surface treatments on the same material, but the third topic listed by Jiang Ming gave her a set of entry points for cross-material verification.

"I'll be participating, and I'll be bringing the original photoconductivity spectrum curves and surface state density estimates. How far we can discuss the device roadmap depends on their experimental results."

Huang Kun wrote Xie Xide's name in the conference participant column and contacted Lin Lanying to prepare data on the properties of germanium and silicon, as well as the existing temperature range data of Furnace No. 3.

The semiconductor research group's reply was delivered to 502 the following afternoon. Wang Shouwu confirmed his attendance and attached a summary of recent point-contact germanium transistor tests.

They used higher purity germanium materials and readjusted the alloy junction formation conditions. The cutoff frequency of the highest-quality tube reached 190 MHz, while the remaining sample tubes were concentrated in the range of 150 to 180 MHz.

In his letter, Wang Shouwu explained that the material resistivity, point contact position, and test bench calibration had all been rechecked, and the remaining frequency differences could not be entirely attributed to process deviations.

On a new line at the end of the letter, the third topic seems too broad, but I agree to discuss it first.

Upon reading this sentence, Wu Hanzhang handed the letter to Jiang Ming and snorted.

"He was also stuck on it, but he wanted to push it one more time."

"He has more process data than we do, and he's willing to sit down and discuss material routes, which means he's already crossed the first hurdle."

Fang Xudong posted the test summaries of the two companies on both sides of the blackboard, and then brought over the device cards, film books and curve rolls accumulated over the past few months, and reclassified them according to material batch, process parameters, crystal orientation and test conditions.

The fifth batch of crystals is numbered separately and bound separately. The device dispersion before and after Laue orientation is bound separately. The two batches of double-diffusion test pieces are divided into three groups according to the surface treatment method: original polishing, acid pickling and annealing, and silicon oxide sealing.

Jiang Ming pulled out six tables from a thick stack of documents. The first set was used to prove that crystal orientation control reduced the diode dispersion from over 40% to 6%, and the second set was used to prove that surface treatment raised the transistor cutoff frequency by about 20%.

The last table only retains three rows of numbers: the current typical value is 170 MHz, the short-term process estimate is 250 to 300 MHz, and the frequency of the devices required for pre-research is at least 1 GHz.

"At the meeting, we should first discuss the variables that have been resolved, and then discuss the remaining gaps. The contribution of crystal orientation, surface, and base region width must be separated."

Fang Xudong retraced the three sets of curves with red and blue pencils. When he saw the gap between 170 MHz and 1 GHz, his hand movements slowed down.

Over the past few months, he personally recorded every temperature fluctuation, every film, and every device. Every advancement came from repeated verification, but the more complete the curve, the clearer the material boundary became.

Jiang Ming opened the safe, took out the memo comparing the high-frequency performance of germanium and silicon, left the parameter pages involving psychic deductions in their original places, and only took out the parts that could be supported by publicly available information.

The band structure diagram is from existing solid-state physics textbooks, the mobility is based on publicly available measurements from Bell Labs, the band gap and intrinsic carrier concentration can be derived from statistical physics formulas, and the frequency determination relies on measured curves from the 502 and the Semiconductor Research Group.

He marked the sources line by line on the edge of the presentation slides, so that any attendee who asked about a particular number could find the entry point by following the literature, formulas, and original records.

Old Sun carried the cutting turntable into the room, placed the newly repaired double-sided locking screws on the table, and saw the documents all over the room. He then wiped the metal shavings off the back of his hand with his sleeve.

"How about we burn another batch, raise the diffusion temperature, shorten the time, and use the new data to make us more assertive at the meeting?"

Jiang Ming shook his head and pushed the 180 MHz column in front of Lao Sun.

"The existing data is sufficient. If we conduct another round, we can probably only push it to 210 or 220. It is worthwhile to make progress in the process, but the material conclusions will remain unchanged."

He closed the bound volume, pressed his palm against the cover, and spoke more slowly than usual.

"The purpose of the meeting is to decide where to go next. The things we have accumulated in germanium materials, single crystals, orientation, and surface treatment will all move forward with us."

The door to the next office was pushed open halfway. Wu Hanzhang, with an unlit cigarette dangling from his lips, peeked at Jiang Ming through the crack in the door for a moment before turning back to his desk.

He turned to the page on the target frequency of the device in the 24 GHz preliminary research plan. The word "germanium" on the paper had been traced several times with a pencil. He wrote "silicon" above it and added a question mark.

Before the meeting materials were packaged, an officer from the International Exchange Office brought a Pasadena postal receipt. The signature column had the date, address, and number written on it, but the response column was still blank.

Jiang Ming checked the sealing wax seal number against the mailing record, confirmed that the second letter had been delivered to Theodore von Kármán's institution, and then wrote the verification date in the upper right corner of the receipt page.

Jiang Ming could only judge how the old man in Pasadena would view temperature potential changes, nonlinear heat transfer, and variable cross-section coupling based on the next letter.

He tucked the reply slip into the old hardcover memo pad, wrote "wait window, two to four weeks" next to the outline of the third letter, and then sealed the cover.

Another document on the table had been put into a canvas bag. The cover read "Joint Technical Symposium," and on the last page, below the third topic, there was still a large blank space for someone to write.