Lab Life: A Journey from Failure to Breakthrough

Oral: Rayleigh High Materials Application Laboratory, Engineer Zhang

I still clearly remember that Thursday afternoon.

Professor Li from the School of Chemistry and Chemical Engineering at Shandong University called, his voice tight with suppressed anger: "Xiao Zhang, one of the acrylic viewports on your last batch of high-pressure reactors developed a micro-crack at 1 2 MPa. Fortunately, our lab technician caught it early; otherwise, all the experimental data for our entire research group would have been wasted."

My palms were sweating as I held the phone.

This is the second-generation seamless viewing window developed in collaboration with Shandong University, designed for a theoretical pressure resistance of 15MPa. During our factory testing, the burst pressures of three samples were 14.8, 15.2, and 15.5MPa, all meeting specifications. So why did issues arise at only 12MPa under actual laboratory operating conditions?

The next morning, I drove to Jinan with the technician.



Act 1: Reproduction Failed

In the high-pressure laboratory of the School of Chemistry and Chemical Engineering at Shandong University, we inspected the "offending" viewport. It was installed on a horizontal high-pressure reactor used for supercritical fluid extraction experiments. The acrylic body of the viewport remained intact, but under oblique lighting, a silver-colored micro-crack approximately 8mm long was clearly visible near the flange sealing groove.

"It's not a brittle fracture," I observed through the magnifying glass for five minutes. "It's stress corrosion cracking, originating at the root radius of the seal groove."

Professor Li stood by, adjusting his glasses. "This occurred after 72 hours of continuous operation in a supercritical CO₂ environment at 1 MPa and 80°C. We previously used imported quartz viewports without issue, but they are too expensive and cannot be manufactured in such large, custom shapes."

I fell silent.

Quartz has excellent resistance to stress corrosion, but we work with acrylic. We haven't systematically evaluated acrylic's resistance to stress cracking in supercritical CO₂ environments before.

"Give me two weeks," I said.



Act II: Finding the Way in Darkness

Back in the Dongguan lab, I sliced open that micro-cracked viewport.

Under the electron microscope, the crack cross-section exhibits typical brittle fracture features, yet a ring of micro-crazing surrounds the crack origin—a classic morphology indicative of stress cracking caused by organic solvents or small molecule penetration. Due to its extremely small molecular size and high penetrability, supercritical CO₂ infiltrates the gaps between acrylic polymer chains, gradually widening micro-cracks at stress concentration points until visible crazing forms.

This is the real enemy.

Our previous testing only covered pure pressure burst tests, without simulating supercritical fluid conditions. In other words, we passed the wrong exam.

That week, my colleagues and I worked overtime continuously. We tried three approaches:

1. Switch to a thicker high-impact PC sheet: However, PC softens when used long-term at 80°C and exhibits poor dimensional stability.

2. Chemical vapor deposition on acrylic surface for barrier protection: However, the viewport is too thick, causing the coating to crack easily during flange clamping.

3. Optimize the fillet design at the seal groove root to reduce stress concentration; simultaneously switch to a modified acrylic formula with improved stress-crack resistance.

The third option is theoretically the most viable, but we don't have modified sheets in stock. I've contacted Professor Zhou at the School of Materials Science and Engineering, South China University of Technology (SCUT). We've collaborated before, and they have a small-scale twin-screw extruder that can help us trial-produce a batch of modified acrylic with different formulations.



Act III: The Breakthrough at 4 a.m.

The modification approach involves incorporating a certain proportion of copolymer monomers into the acrylic substrate to enhance chain flexibility and reduce small-molecule permeability, while maintaining light transmittance and compressive strength.

It's like trying to have your cake and eat it too: adding a copolymer improves stress crack resistance but reduces compressive strength and heat resistance. We need to find the optimal balance.

Five formulation iterations were completed, each following the full cycle: extrusion granulation → plate pressing → annealing → CNC machining → supercritical CO₂ simulation testing. Each cycle took three days.

In the first four rounds, either the pressure resistance dropped below 10 MPa, or internal stresses after annealing caused optical distortion in the sheets. In the fifth round, we added a nano-scale crosslinking agent to compensate for the strength loss resulting from introducing flexible chain segments.

It was 4:00 AM. The second sample from the fifth batch, after 72 hours of supercritical CO₂ immersion in the simulation vessel, was loaded into the burst test machine. The pressure gauge needle climbed steadily—8 MPa, 10 MPa, 12 MPa, 14 MPa—until 15.2MPa, with no leakage detected.

The pressure relief valve on the test machine is sounding.

I stared at the intact viewport. A colleague tapped my shoulder. I couldn't speak; I just stood there, grinning like an idiot.



Act 4: Delivery Is Not the End

The sixth round was stability validation. We tested ten additional samples; all passed the 15MPa burst test, and after 72 hours of supercritical CO₂ immersion, no microcracks were observed.

On day 14, I personally delivered the third batch of windows to Jinan.

Professor Li smiled, a rare sight, after seeing the test results pass completely. He said something I'll never forget: "The biggest fear in experiments isn't equipment failure—it's failing without knowing why. You found the root cause this time. That's where you outperform imported products."

This modified acrylic high-pressure viewport has been operating stably in the Shandong University research group for eight months, accumulating over 2000 hours of service with no cracks or fogging. We have filed a patent for both this formulation and the sealing groove design, titled "Stress-Corrosion-Resistant Seamless Viewport for Supercritical Fluid Systems."



Final Thoughts

In Rayleigh's lab logs, this experience is internally referred to as the "CO₂ Campaign."

It taught me three things: First, real-world user conditions are always more complex than we imagine—passing tests in the factory doesn't guarantee success on-site. Second, failure itself holds no value; what matters is the rigor of thoroughly understanding why it failed. Third, every collaboration with university labs isn't a client-vendor transaction—it's a joint effort to solve an unsolved problem.

The lab lights often stay on until dawn. But every time they shine, we are one step closer to the answer.

Rayleigh High: In the world of transparent materials, we refuse to let even a "micro-crack" become a safety hazard for our users.