He Named My Turbine Blade Fracture Analysis After Himself Then the NTSB Investigator Asked Him to Read the Metallurgical Grain Pattern

The SEM specimen stub was in the pocket of her lab coat — the small aluminum cylinder, 12.7mm in diameter, with the carbon coating smear on the left side of the mounting surface. Dr. Marie-Claire Fontaine had acquired the smear in year one of her practice, when she had carbon-coated a biological sample directly on the stub instead of on the standard stub cover, and the surplus carbon had spread to the mounting surface. She had cleaned the sample off but not the smear. The smear had no effect on subsequent samples, which were mounted on the standard stub cover or directly gold-sputter coated. She kept the stub in her lab coat pocket as a reference when explaining SEM preparation to clients.

Bertrand was at the sputter coater, running the gold coating cycle on the fracture specimen — the turbine blade root fragment, 15mm × 8mm, the fracture surface facing up in the sample holder. The gold coating took four minutes. While the cycle ran, Marie-Claire was at the SEM console, reviewing the preliminary imaging parameters for the session.

The fracture surface had come from the airline’s engineering team two weeks ago — a recovered turbine blade from an engine that had experienced an uncontained blade failure during ground testing. The maintenance team’s dye penetrant inspection had identified fatigue cracking as the probable cause. They had sent the blade to her for independent fractographic confirmation.

She had placed the first sample under the SEM and looked at the fracture surface for seven minutes before she had understood what she was seeing.

Not fatigue striations. Not the banded, regularly-spaced surface topography characteristic of cyclic loading failure. The fracture origin was a smooth-walled cavity, circular in cross-section, 1.2mm in depth from the blade’s root radius. The cavity walls were clean — no oxidation, no secondary cracking, no evidence of crack growth from the surface. A void. A manufacturing void: a gas bubble or inclusion entrapment from the casting or forging process, present in the blade from its production date.

The void had been at 1.2mm depth, within the machining zone of the root radius. The maintenance team’s dye penetrant inspection — which detected surface-breaking cracks — would have found nothing because the void was subsurface. It was not visible without cross-sectional examination.

She had taken the stub from her pocket and held it while she looked at the SEM image. She held it for a long time.

Then she had called Bertrand. “Get the cross-section mount ready. I need a polished section through the root radius at the fracture origin.”

The cross-section had confirmed it. The void at 1.2mm was visible in the polished section — the circular morphology intact, the clean walls, no fatigue striations anywhere in the surrounding material. The fracture had initiated at the void under the operational stress of turbine rotation, not from cyclic fatigue loading.

Manufacturing defect. Not maintenance team error.

Breton had called her when she had sent the report.

He had said: “The void finding changes everything. The manufacturer is liable, not the maintenance team.”

She had said: “The SEM shows a clear manufacturing void — the morphology is inconsistent with fatigue initiation. The void is at 1.2mm depth, within the machining zone.”

He had said: “This is the report we needed. This is what the NTSB investigation requires.”

She had said: “The report is stamped ASNT-III-MF-6621.”

He had said: “Good. Good work, Marie-Claire. This is exactly what we needed.”

She had hung up. She had noted: “what we needed.” Twice. The void finding and the report. Both of them framed as things he had needed. He had not been at the SEM when she had understood what she was seeing.

She had put the SEM back on the specimen and continued the fractographic survey — the secondary imaging, the additional cross-sections, the energy-dispersive X-ray analysis to confirm the void’s composition profile.

She was at the SEM console when she checked the NTSB website on her second monitor — the docket listing for the blade failure investigation had been updated. She read the entry.

“The Breton Fracture Assessment — submitted by Chief of Maintenance Engineering R. Breton. SEM analysis subcontractor: Dr. Marie-Claire Fontaine, ASNT-III-MF-6621.”

She took the stub from her pocket. She examined the carbon smear on the left side of the mounting surface. The smear had been there for seven years. It had no functional significance. She put the stub back in her pocket.

She opened the ASNT certification registry on her computer. ASNT-III-MF-6621. Active. Certification class: Level III, Radiographic Testing, Ultrasonic Testing, Penetrant Testing, Visual Testing. Current renewal: eighteen months remaining.

She closed the registry. She put the stub back in her pocket. She went back to the SEM.

The ASNT Level III certification had taken her three years to complete after her PhD. The examination covered the theory and application of the specific NDT methods — the physics of each technique, the standards that governed its application, the interpretation protocols for the results. She had passed the examination in Radiographic Testing, Ultrasonic Testing, Penetrant Testing, and Visual Testing. Each examination was separate. Each had required a minimum of documented hours in the method — hours she had accumulated through her doctoral research, her post-doctoral work, and her first two years of independent practice.

ASNT-III-MF-6621 had been her certification number since the first examination passed. The number stayed constant across renewal cycles. She renewed every five years. She had two renewals behind her and eighteen months remaining on the current certification period.

She took the stub from her pocket and set it on the SEM console while she checked the imaging parameters. It was easier to check parameters with both hands on the keyboard. The stub sat there — 12.7mm aluminum, the carbon smear facing up. She adjusted the accelerating voltage and the working distance for the current specimen’s surface coating. She picked up the stub and put it back in her pocket.

“Ready for initial imaging,” she said to Bertrand.

He said: “First quadrant mounted.”

She centered the electron beam on the first quadrant and initiated the scan.

The airline’s engineering review board met monthly. Breton had presented the turbine blade failure analysis at the previous three sessions — using Marie-Claire’s SEM images, the void morphology photograph, the cross-section showing the 1.2mm depth.

He said: “Our SEM methodology identified the manufacturing void at 1.2mm depth. Our failure analysis reclassified the accident cause from fatigue to manufacturing defect, redirecting the liability chain to the manufacturer.”

He said “our SEM methodology.”

He said “our failure analysis.”

The SEM image on the slide was her fractograph — the void’s smooth-walled circular morphology, the fatigue-free surrounding material, the 1.2mm scale bar in the lower right corner. She had captured that image. She had identified what it showed.

He did not show her ASNT Level III certification. He did not name her. He had described the reclassification as the outcome of “our” investigation — as if the airline’s maintenance engineering department had operated the SEM, identified the void morphology, and applied the ASNT-standard fractographic interpretation methodology.

Marie-Claire was not at the engineering review board meetings. She was in her lab.

The FAA enforcement contact came from Sandra McCleod, FAA Enforcement Officer, through the regulated entity notification system.

“Dr. Fontaine — I am contacting you regarding FAA enforcement action against the turbine blade manufacturer in the matter of the [aircraft registration] failure. The manufacturer’s legal team has filed a challenge to the fractographic methodology used to reclassify the fracture cause. The challenge requires: (1) the original SEM analysis report, (2) ASNT Level III certification documentation for the certifying inspector, and (3) the inspector’s professional attestation to the methodology. ASNT-III-MF-6621 is identified on the report. Please confirm your availability to provide the above documentation and to provide professional attestation in the enforcement proceeding.”

She read “ASNT Level III certification documentation for the certifying inspector.”

She read “professional attestation to the methodology.”

She opened the ASNT certification registry. ASNT-III-MF-6621. Active. She took the specimen stub from her pocket. She looked at the carbon smear. She put it back.

She did not call Breton.

She opened a new email to Sandra McCleod. She confirmed her availability. She attached the original SEM analysis report — stamped ASNT-III-MF-6621 — the ASNT Level III certification certificate, the certification scope, and the renewal record. She wrote a one-paragraph attestation statement affirming the methodology’s compliance with ASNT standards.

She sent the email.

She went to the SEM. She had a new specimen ready — a pressure vessel fragment from a different client. She mounted it. She began the analysis.

Breton’s legal team had received the FAA enforcement notification the same day. They had reviewed the enforcement documents. The manufacturer’s challenge was directed at the SEM methodology — and the challenge required the ASNT Level III inspector to respond.

The team lead had gone to Breton.

“The FAA enforcement challenge requires the ASNT Level III inspector to provide professional attestation. ASNT-III-MF-6621 is Dr. Fontaine’s certification. You don’t hold ASNT Level III. You cannot attest to an SEM fractographic methodology you did not conduct.”

Breton had looked at the NTSB docket entry on his screen. “Breton Fracture Assessment.”

He had said nothing for a moment. Then: “I’ll call Dr. Fontaine.”

The team lead had said: “She’s already responded to the FAA. McCleod confirmed receipt this morning.”

He had looked at the docket. He had said: “I’ll amend the docket entry.”

She had not considered calling Breton before responding to the FAA contact. The FAA enforcement officer had addressed the request to the ASNT Level III of record — the inspector whose certification number appeared on the fractographic report. That was her. She was the inspector of record. She had answered.

The question of whether Breton, as the submitting party for the NTSB docket, should have been notified before the ASNT Level III responded directly to the FAA — this question had not presented itself while she was composing the email to McCleod. It had not been a consideration because it had not been her responsibility. The FAA had asked a technical question of a certified technical professional. She had answered it.

She had been clear in her attestation: the analysis was conducted in compliance with the ASNT SNT-TC-1A standard. The methodology was sound. The void identification was supported by the physical evidence. She had signed it with her certification number and her professional seal.

The pressure vessel specimen was a carbon steel fragment from a pressure relief valve body — a crack that had been reported during an ultrasonic inspection, sent to her for SEM confirmation and origin identification. She had the fragment in the sample holder. She ran the initial scan.

The crack was visible at 200x — a surface-originating crack, the fracture surface showing fatigue striations from cyclic pressure loading. She noted the finding. This was fatigue. She moved to higher magnification for the origin identification.

He had been the airline’s engineering chief for eleven years. He had managed 22 maintenance investigation reports, seven NTSB submissions, and three previous enforcement proceedings. In each of those proceedings, the airline’s engineering credibility had rested on his maintenance engineering credentials and his team’s documentation.

But an ASNT Level III was a different credential. It was not a management credential. It was a technical certification — issued by the American Society for Nondestructive Testing to inspectors who had passed a certification examination in a specific NDT method, met the minimum hours of documented experience, and maintained their certification through continuing education and renewal. ASNT Level III was what gave a fractographic analysis legal standing in a regulatory proceeding: it identified the inspector as a certified expert whose methodology could be defended under technical challenge.

He did not hold ASNT Level III. He had never taken the examination. He had never had occasion to — his role was engineering management, not inspection. He had commissioned fractographic analysis from certified contractors for ten years. He had always been the one who submitted the results to regulators, because he was the regulatory interface.

He had submitted the NTSB docket entry as “The Breton Fracture Assessment” because he was the Chief of Maintenance Engineering — the airline’s technical authority on the failure. The fractographic analysis was a component of his investigation. He had commissioned it. He had reviewed the results. He had made the decision to reclassify the accident cause and redirect the liability finding.

He had made the decision. That had been his.

But the FAA enforcement proceeding was not about the decision to reclassify. It was about the methodology that had identified the void. The SEM analysis. The ASNT certification. The specific technical expertise that had looked at a smooth-walled circular cavity at 1.2mm depth and understood what it was — not fatigue, not maintenance error, but a manufacturing defect from the casting or forging process.

He had reviewed her SEM images. He had understood what “manufacturing void” meant in terms of liability. He had understood the outcome. He had never examined what understanding the void’s morphology in the SEM image — what being able to look at that surface and know what caused it — required.

He had said “what we needed.” Twice. He had said it as shorthand for the result. The result was hers. The shorthand was his.

He opened the NTSB docket amendment form.

The SEM specimen stub was in her pocket. She was at her desk preparing the FAA attestation documentation — the formal statement she would submit to the enforcement proceeding, confirming the methodology’s compliance with ASNT SNT-TC-1A and the ASNT Level III examination standards. The stub was in her pocket. It was always in her pocket.

She had reached in and touched it twice during the preparation — not deliberately, not consciously. The way she always reached in when she was at her desk working. The stub was there. The carbon smear was on the left side. She had confirmed it was there and continued typing.

She had been clear about the attestation: the fractographic analysis had been conducted in compliance with ASNT-E2001 Standard Guide for Examination of Fracture Surfaces. The methodology had followed the standard protocol for SEM fractographic examination. The analysis was sound. She was the Level III of record. ASNT-III-MF-6621.

She sent the attestation. She returned to the pressure vessel specimen.

The distinction between “what we needed” and “what you found” had been clear to her the moment he had said it on the phone — the first time, and then the second time. She had noted it because it was precise language. He needed the report. She produced the report. He needed the finding. She found it. Those were not the same contribution.

She had not called him to point this out. There had been no practical reason to call him. She had produced the fractographic analysis, it had been submitted to the NTSB, and she had gone back to the SEM. The question of what her contribution would be called in the docket submission — “what we produced” versus “what I found and stamped ASNT-III-MF-6621” — was Breton’s question to resolve, not hers. She had no access to the docket submission form. She had submitted her analysis report to his office, not to the NTSB directly.

She had not been surprised when the docket entry said “subcontractor.” She had expected it. She was a subcontractor in the commercial sense — an independent contractor engaged by the airline for a specific technical task. The word was accurate in one context and imprecise in another. In the context of the NTSB docket, “subcontractor” erased the methodological distinction between the engineer who had commissioned the analysis and the certified inspector who had conducted it and had legal standing to defend it.

She had noted this imprecision and continued working.

The FAA enforcement proceeding had made the distinction impossible to ignore. An FAA enforcement challenge to fractographic methodology required the ASNT Level III inspector to respond — not the engineer who had commissioned the analysis, and not the engineering chief who had submitted it to the NTSB. The certification was the legal instrument. The certification was hers.

She sent the attestation. She returned to the pressure vessel specimen and the fatigue crack origin she had identified at higher magnification.

The FAA enforcement hearing was held in the federal aviation authority’s regional office — a conference room with a long table, nine people on the government side and six on the manufacturer’s. Marie-Claire had the SEM analysis report and the original fractographic images in a folder in front of her. The specimen stub was in her lab coat pocket.

The manufacturer’s lead counsel had reviewed the SEM analysis report and had developed a challenge on two grounds: first, that the void identification methodology was not compliant with the ASTM E2001 standard for SEM fractographic examination; second, that the identification of the cavity as a manufacturing void rather than a service-induced inclusion required assumptions not supported by the available evidence.

She had spent 14 minutes on the first challenge. She explained the preparation protocol — the cross-section mounting, the polishing grade, the gold sputter coating, the imaging parameters — and referenced the ASNT-MF-6621 certification scope, which included the specific examination methods used. The challenge failed on the evidence.

She had spent 23 minutes on the second challenge. She explained the distinction between manufacturing voids and service-induced inclusions in fractographic morphology: manufacturing voids have smooth, clean walls with no secondary oxidation, no diffusion gradient, and no orientation relative to the loading direction; service-induced inclusions have irregular boundaries, chemical contamination from operating fluids, and orientation that follows the stress field. The void in the turbine blade had smooth, clean walls, no chemical contamination, and no preferred orientation. It was a manufacturing void.

The manufacturer’s counsel had asked: “Could the void have been introduced during a maintenance procedure?”

She had said: “Not at 1.2mm depth in the machining zone without evidence of tooling damage in the surrounding material. There is no tooling signature in the cross-section. The void was introduced before the final machining of the root radius.”

She had answered 31 questions in three and a half hours. Every question about the methodology, she answered from the SEM images, the cross-section photographs, and the ASNT certification scope. The stub was in her pocket for all of it.

The FAA enforcement officer had confirmed the record: “ASNT Level III Analyst: Dr. Marie-Claire Fontaine, ASNT-III-MF-6621. Fractographic analysis finding: manufacturing void at 1.2mm root radius depth. Finding unchallenged. Enforcement action against manufacturer proceeds.”

During the technical recess, she took the SEM specimen stub from her lab coat pocket and held it between her index finger and thumb, the 12.7mm aluminum disk with the carbon smear on the left side of the mounting surface. The manufacturer’s counsel had asked how the fracture surface sample was prepared for the SEM analysis. She was explaining this now, in the corridor outside the hearing room. The stub was a demonstration — this surface, this coating, this is how you mount a fractographic sample so the electron beam can read the topography. The FAA enforcement record was building: “ASNT Level III Analyst: Dr. Marie-Claire Fontaine, ASNT-III-MF-6621.” Bertrand was waiting at the end of the corridor, watching the explanation from a distance. The counsel leaned in to look at the smear. She held the stub steady. She showed him the smear. She explained the preparation.

Sandra McCleod had contacted her after the hearing was complete.

“Dr. Fontaine — your fractographic testimony and your ASNT Level III credentials are the foundation of this enforcement action. The void identification at 1.2mm depth is the central technical finding. The record is clear.”

She had read the email in her lab. Bertrand had been finishing the prep for a new specimen.

He had said: “ASNT-III-MF-6621 in the FAA record.”

She had said: “Yes.”

He had said: “The void at 1.2mm.”

She had said: “1.2mm.” She had taken the stub from her pocket. She had looked at the smear. She had put it back. She had gone to the SEM.

Breton had called the evening after the hearing. He said: “Good outcome. Your testimony was decisive.” She said: “The fractographic analysis is sound.” He said: “Yes. I’ve amended the NTSB docket and I’m updating our NDT submission protocols — the ASNT Level III will be named going forward.” She said: “Yes.” He said: “Good work, Marie-Claire.” She said: “Thank you.” She put the stub in her pocket. She went back to the SEM.

The amended NTSB docket entry arrived by email three days after the hearing: “Fractographic analysis by Dr. Marie-Claire Fontaine, ASNT Level III, ASNT-III-MF-6621. Finding: manufacturing void at 1.2mm depth, root radius machining zone. Cause: manufacturer production process.” She read it. She filed it in the blade failure project folder alongside the original NTSB docket entry with “SEM analysis subcontractor.” Both entries were in the folder. She had not annotated either of them.

The NTSB preliminary probable cause report — the one that had used the “Breton Assessment” language before the enforcement action — was in the NTSB’s public docket archive, accessible at the NTSB accident number. The report had not been reissued. She had the accident number in a note at the bottom of the project folder index. She had not looked it up since she had written it down.

The new specimen was a compressor disk fragment from a turbine engine teardown — a routine inspection case, no failure event, preventive fractography to document the disk’s condition after 8,000 operating cycles. She had the disk mounted in the preparation room. Bertrand was sectioning it.

She had removed the stub from her pocket when she had begun the SEM session preparation — not to use it, but because she was cleaning out her lab coat pockets before putting on the fresh coat for the SEM room. The stub had been in the left pocket for seven years. She had set it on the lab bench and looked at it for a moment before putting on the fresh coat.

She had put it in the fresh coat’s left pocket.

Bertrand had brought the first section. She positioned it in the SEM sample holder. She set the imaging parameters — 20kV accelerating voltage, secondary electron imaging mode, 500x initial magnification for orientation, higher magnification for feature identification if needed. She ran the initial scan.

The compressor disk surface was clean — no anomalies in the low-magnification scan, the grain structure uniform, no evidence of fatigue cracking at the exposed grain boundaries. She increased the magnification to 2,000x. Still clean. She noted the result and moved to the next section.

Bertrand was at the bench preparing the second section. He had been with her for two years. He had been present for the turbine blade analysis, had prepared the cross-section mount that confirmed the 1.2mm void. He did not mention it. He prepared the sections. She imaged them. The work continued.

The stub was in her pocket.

The updated airline NDT submission protocol arrived from Breton’s office — a policy document requiring that all ASNT-certified analysts be named in their own right in regulatory submissions, not subsumed under the submitting engineer’s authorship. She read it. She filed it in the blade failure project folder alongside the original docket entry, the amended entry, and the FAA enforcement record.

Four documents in one folder. The project was complete.

She closed the folder. She opened the compressor disk case file. She had four more sections to image. She went to the SEM room. Bertrand had the second section ready.

She positioned the second section in the sample holder. She set the parameters. She initiated the scan.

The disk surface at section two was the same as section one — clean grain structure, no anomalies, uniform fatigue marks from normal operating load cycles. She noted it. She moved to section three.

The stub was in her pocket. She had carried it for seven years through examinations, client briefings, enforcement hearings, and routine NDT work. It had acquired the carbon smear in year one and had remained unchanged since. She did not think about it while she was imaging the compressor disk. She was noting the grain structure at section three.

Section three was clean. She moved to section four.