He Named My Balustrade Fracture Analysis After Himself — Then the Personal Injury Lawyer Asked Him to Explain the Load Path Diagram

The fracture origin fragment was in the acrylic case on her desk — a 6cm × 4cm shard of the failed tempered balustrade panel, mounted flat in the case with the fracture surface facing up. Dr. Ingrid Holm had placed a red felt-tip dot at the NiS inclusion site when she had mounted the specimen, to mark the fracture origin for teaching demonstrations. The dot was 3mm in diameter. The inclusion itself was 0.3mm — invisible to the naked eye, visible under the scanning electron microscope as a bright particle against the surrounding glass matrix.

Sven was at the SEM workstation, loading the imaging session parameters for the new case. He had been with her for two years. He had been present for every imaging session of the balustrade specimen — had photographed the Wallner lines at six magnifications under her direction, had run the energy-dispersive X-ray analysis that confirmed the particle’s nickel sulfide composition.

She opened the acrylic case. She lifted the fragment out and held it under the lab light — the Wallner lines radiating from the red dot, visible as fine ripple marks on the fracture surface, each wave front propagating outward from the instant the fracture had initiated. Beyond the Wallner lines, the mist region: a band of finely roughened glass where the crack had begun to accelerate. Beyond the mist, the hackle — the coarser, directional ridges where the crack had reached its terminal velocity.

“The NiS inclusion is under the red dot,” she said to Sven. “At 0.3mm it’s not visible here, but the Wallner line geometry points to it. Each of these lines is a wave front — the crack was still moving subsonically when the first waves formed, and the wave fronts are circular because the crack hadn’t yet developed a preferred direction. That’s the diagnostic signature of a spontaneous inclusion failure, not an impact event.”

Sven said: “How do you distinguish the mist radius from the hackle boundary?”

She said: “Roughness. Under the SEM, the mist region has a finer surface texture — the crack was moving fast but not yet chaotic. The hackle is where it became chaotic. The boundary is measurable. I use the mist radius to estimate the stress intensity factor at fracture — which tells you the residual stress in the panel when it failed.” She put the fragment back in the acrylic case and closed the lid. She set it on the desk. The red dot was visible through the acrylic.

“That panel,” she said, “had passed heat-soak testing at installation. The NiS inclusion was present but the particle hadn’t yet undergone the phase transformation that caused the volume expansion. It transformed in service — 14 months after installation, at ambient temperature, without any external load. The panel broke spontaneously. The person who was leaning against the balustrade at the moment of failure had nothing to do with it.”

She had gone back to the SEM session parameters after the teaching session. She had not yet read the PI claim submission. She had read it the following morning.

She had read “Whitfield Glass Failure Analysis — structural glass assessment conducted by Principal Engineer T. Whitfield.” She had read “glass analysis support: Dr. Ingrid Holm, IStructE-GE-IH-4413.”

She had looked at the red dot through the acrylic case. She had not opened the case. The fragment was inside. The dot faced up. She had opened the SEM image file on her computer — the 15,000× magnification image of the NiS inclusion, bright against the glass matrix, the inclusion boundaries clearly defined. She had looked at it. She had closed the file. She had logged the next project.

The “before” had been eight months earlier, when Whitfield had come to her office after she had submitted the fractographic report.

He had said: “The insurer needs a clear narrative — spontaneous failure, manufacturer liability. Your analysis gives us that.”

She had said: “The Wallner line geometry is unambiguous. The inclusion is at the fracture origin. The panel failed without external loading — the stress intensity factor at fracture is consistent with the residual stress of the NiS phase transformation, not with impact loading.”

He had said: “This is exactly what the PI claim needs.” She had said: “The report is signed under IStructE-GE-IH-4413.” He had said: “Good. This is the kind of technical depth that sets this firm apart.” She had said: “Thank you.” She had gone back to the SEM images.

She had noted: “sets this firm apart.” The firm. Not her analysis. Not her methodology. Not the seven years of glass fractography experience that had let her identify the mist radius boundary on the SEM image and calculate the stress intensity factor from the Wallner line spacing. The firm.

She had gone back to the images. She had noted it and gone back.

She opened the IStructE registration record. IStructE-GE-IH-4413. She closed it. She opened the acrylic case. She looked at the red dot. She closed the case. She opened the PI claim submission again. She read “glass analysis support.” She closed the submission. She did not annotate it. She picked up the case and held it for a moment. She set it back on the desk. She opened the next project file.

The IStructE registration — IStructE-GE-IH-4413 — had taken her five years after her PhD. The Institution of Structural Engineers’ specialist structural glass pathway required demonstrated competence in glass material science, fracture analysis, and structural performance assessment, verified through a portfolio review and an examination by a panel of chartered engineers. It was not a general CEng credential. It was a specialist registration for engineers who worked specifically with structural glass — its failure modes, its testing standards, its interface with Building Regulations and the PI insurance framework.

She had been IStructE-GE-IH-4413 for seven years. Every fractographic report she had produced since the registration had carried the number. It was her professional seal on the analysis — the statement that the work had been conducted by an engineer whose competence in structural glass fractography had been independently verified.

The SEM itself had taken practice — not the instrument, which Sven operated, but the interpretation of what the instrument showed. A scanning electron microscope returned an image of a fracture surface. The image showed features. The question was which features were diagnostic. She had spent three years learning to read glass fracture surfaces — had studied documented cases, had practiced on reference specimens, had developed the visual vocabulary for the distinction between Wallner lines and Wallner line mimics, between genuine mist regions and surface contamination artifacts, between NiS inclusion morphology and other types of glass inclusions.

She had identified the NiS inclusion in the balustrade specimen in the third imaging session. The first two sessions had been orientation — establishing the fracture origin quadrant, mapping the Wallner line field, narrowing the search area. In the third session, she had increased the magnification to 15,000× over the origin zone and the particle had resolved: the NiS characteristic morphology, the clear phase boundary, the crystalline particle structure. She had confirmed it with EDS in the same session.

She had filed the final fractographic report that evening.

The structural glass conference was in Edinburgh — two days, 180 delegates, sessions on glass failure investigation, facade engineering, and fire performance. Whitfield presented in the second session: “Identifying Nickel Sulfide Spontaneous Breakage in Post-Heat-Soak Failures.”

His slide 6 was her SEM image — the 15,000× magnification fractograph of the NiS inclusion, the bright particle against the glass matrix, the boundary clearly defined. Her SEM image. She had set the magnification, the accelerating voltage, the working distance. She had identified the optimal imaging conditions for the nickel sulfide particle against the borosilicate glass matrix.

He said: “Our fractographic analysis capability identified the NiS inclusion at the fracture origin of a post-heat-soak balustrade panel. The inclusion’s phase transformation in service produced the characteristic Wallner line pattern radiating from a spontaneous fracture origin.”

He said “our fractographic analysis capability.”

He did not describe the Wallner line measurement methodology. He did not explain the mist radius calculation. He did not name the IStructE registration. He did not name Ingrid.

She had read the conference presentation abstract in an industry newsletter two weeks after the event. She had noted the abstract. She had gone back to the SEM.

The contact from the plaintiff’s expert had arrived in her professional email from Dr. Eleanor Price, structural glass expert, instructed by the plaintiff’s legal team.

“Dr. Holm — I am instructed by the plaintiff in the personal injury action arising from the balustrade failure at [building address]. The plaintiff’s expert panel requires the following from the IStructE-registered structural glass engineer who conducted the fractographic analysis: (1) the original fractographic analysis report, as signed; (2) IStructE-GE-IH-4413 registration documentation; and (3) your availability to provide expert testimony on the NiS inclusion identification methodology — specifically whether the inclusion was detectable before failure and whether the panel should have been quarantined after heat-soak testing. Your IStructE registration number appears on the signed report. Please confirm your availability.”

She read “IStructE-registered structural glass engineer who conducted the fractographic analysis.”

She read “IStructE-GE-IH-4413 registration documentation.”

She opened the IStructE registration record. IStructE-GE-IH-4413. She took the acrylic case from her desk. She opened it. She looked at the red dot — the 3mm circle of red felt-tip over the 0.3mm NiS inclusion site, the Wallner lines radiating from beneath the dot if you tilted the case in the right direction. She closed the case.

She did not call Whitfield.

She opened a reply to Dr. Price. She confirmed her availability. She attached the original signed fractographic report, the IStructE-GE-IH-4413 certificate, the certification scope document, and the renewal record. She wrote a one-paragraph attestation confirming the analysis had been conducted in compliance with the IStructE’s structural glass examination standards.

She sent the email.

She went back to the SEM. Sven had the new case specimen ready — a different glass panel, a different building, a surface impact claim rather than spontaneous breakage. She mounted it. She ran the initial scan.

Whitfield had been in his office when the PI lawsuit notification arrived — the formal letter from the plaintiff’s solicitors, served to the firm. He had read it. He had been unconcerned. The analysis was in the file. He would provide the engineering evidence the litigation required.

He had told the firm’s legal team to prepare the expert evidence bundle.

The legal team had come back within two hours. “The plaintiff’s expert has contacted Dr. Holm directly. She’s confirmed her availability and sent the signed report and her IStructE registration documentation. The deposition will require the IStructE-registered structural glass engineer — that’s Dr. Holm. Your CEng MIStructE is structural engineering. You cannot be examined under cross-examination on Wallner line geometry and NiS inclusion mechanics.”

He had looked at the PI claim submission on his screen. “Whitfield Glass Failure Analysis.”

He had been still for a long time.

He had said, finally: “Dr. Holm will be the expert witness. I’ll prepare a statement to that effect for the litigation file.”

The legal team had said: “She’s already confirmed with the plaintiff’s expert. The statement will need to go to both sides’ solicitors.”

He had said: “Yes.” He had sat at his desk. He had not said anything else for several minutes.

He had been a principal structural engineer for seventeen years. He had managed glass facade assessment programs, structural appraisals, PI insurance investigations. He had submitted fractographic reports to insurers before — always signed under his CEng MIStructE, always as the assessment program lead, always with the analysis produced by specialists under his direction.

The PI claim submission protocol at the firm had never required a distinction between the submitting engineer and the conducting specialist. He had always submitted. He was the PI insurance correspondent. He had the client relationship, the signatory authority for the firm’s report submissions. When the insurer received the Whitfield Glass Failure Analysis, they understood it as the firm’s technical product — not specifically his personal technical work, but the technical output of the assessment program he managed.

He had not thought it was wrong. He had thought it was normal. The firm’s PI correspondent submits the analysis. The analysis is part of the firm’s service. The PI claim names the firm’s correspondent.

But the litigation was not asking about the PI submission protocol. The litigation was asking about the expert witness — the person who would be placed under oath and cross-examined on the methodology. The Wallner line measurement. The mist radius calculation. The NiS phase transformation mechanism. The heat-soak test standard and its limitations.

He could not answer those questions under cross-examination. He knew the conclusions — the panel failed spontaneously due to a NiS inclusion, the manufacturer was liable, not the building owner — because he had read the report and understood it. He had reviewed the SEM images with Ingrid. He had understood what the bright particle in the glass matrix signified.

But understanding a conclusion and being able to defend the methodology that produced it were not the same thing in a deposition. The cross-examining barrister would go directly to the methodology. They always went to the methodology. And the methodology — the Wallner line geometry, the mist radius measurement, the stress intensity factor calculation, the NiS inclusion characterization — was Ingrid’s.

He had said “sets this firm apart.” He had meant it as recognition — of the quality of work that the firm produced, of the capability that had identified the failure cause when the dye penetrant inspection had found nothing. He had not examined whether “the firm’s capability” and “Dr. Holm’s IStructE registration and seven years of glass fractography experience” were equivalent descriptions of the same thing.

He had looked at the SEM image in her report many times. He had looked at the Wallner lines radiating from the inclusion. He had understood them as evidence of the failure cause. He had never been able to generate them himself.

He opened the PI claim amendment form.

The acrylic case was on her desk, closed. The red dot was visible through the top panel of the acrylic. She was preparing the deposition documentation — the fractographic report, the signed original, the IStructE certification package, the SEM image set at all six magnifications. The case was on the desk in its usual position. She had not opened it since closing it after the plaintiff’s expert contact.

She had not reached for it while working. She was working through the deposition documentation checklist — she had a list of eight items from Dr. Price’s instruction letter, and she was going through them in order. The acrylic case was not on the list. It was on her desk. She was working through the list.

She finished the documentation package. She sent it to Dr. Price. She went to the SEM room. The surface impact case was ready for the second imaging session. She loaded the parameters. She began the scan.

She was clear about why she had not called Whitfield when Dr. Price’s contact arrived. The contact had been addressed to the IStructE-registered structural glass engineer who had conducted the analysis — the registered expert. That was her. It was not a communication that required her to loop in the PI correspondent before responding. It was a direct professional contact from an instructed expert to the expert of record.

Whether she had needed to think about it as a question — whether to call Whitfield or not — was itself a data point. She had not needed to think about it. She had opened the registration record, confirmed it was her, opened the reply field, and begun preparing the response. The sequence had not included a decision point about Whitfield.

This was not a choice she had made against him. It was the absence of a reason to include him. The contact had come to her. The answer was hers to give.

She sent the documentation package to Dr. Price. She checked it against the eight-item checklist. All eight items confirmed. She closed the checklist. She went to the SEM room for the second session on the surface impact case.

The personal injury deposition was held in a conference room at the plaintiff’s solicitors’ offices — a long table, the plaintiff’s barrister and Dr. Price on one side, the defense solicitor and Whitfield on the other, Ingrid at the end of the table as the expert witness. The acrylic case was in front of her, closed. The fractographic report was in a folder beside it.

Whitfield’s introduction was three sentences: “Dr. Holm is the IStructE-registered structural glass engineer who conducted the fractographic analysis of the failed panel. The fractographic methodology questions are for her. I’ll be available for questions about the firm’s engagement with the PI claim and the client relationship.”

He sat back. He did not speak again on technical topics for the remainder of the deposition.

The plaintiff’s barrister led the examination. She began with the heat-soak test: what it was designed to detect, what its limitations were, whether a NiS inclusion of 0.3mm was above or below the detection threshold of standard heat-soak testing under EN 14179-1.

Ingrid said: “Heat-soak testing per EN 14179-1 is designed to accelerate the NiS phase transformation that causes spontaneous breakage — the test cycles the glass between 280–300°C to trigger transformation in inclusions that would otherwise transform in service over years. The test has a documented false-negative rate for inclusions below approximately 0.5mm. A 0.3mm NiS inclusion, depending on its position within the glass thickness, may not transform during the test cycle. This panel passed heat-soak. The inclusion was present at installation. It transformed in service 14 months later.”

The barrister said: “So the panel passed the industry standard test and still failed spontaneously?”

She said: “Yes. The test reduces but does not eliminate the risk. EN 14179-1 acknowledges this limitation. The industry adopted the test knowing it has a false-negative rate for sub-0.5mm inclusions.”

She opened the acrylic case. She took out the fracture origin fragment and placed it beside the fractographic report on the table. She held the fragment under the room light.

“These lines radiating from the red dot are the Wallner lines — wave fronts propagating from the moment the crack initiated at the NiS inclusion site. The dot marks the fracture origin. The inclusion is 0.3mm, at approximately 4mm depth from the tension face of the panel. The Wallner lines are visible to the naked eye here because I’m showing you the physical specimen at this scale. Under the SEM, the inclusion itself is visible as a bright particle.”

The barrister said: “At the time of the failure — the moment the person was leaning against the balustrade — what was the applied external load on the panel?”

She said: “The panel failed without applied external load. The stress at failure was entirely from the residual stress of the NiS phase transformation — the particle expanded, created a stress concentration, and the residual stress in the tempered panel initiated the fracture. A person leaning against a balustrade applies a lateral load of a few tens of Newtons. The stress intensity factor at the fracture origin — calculated from the mist radius measurement — corresponds to the residual stress of the tempered glass, not to any external loading event.”

She pointed to the mist region — the band of finer surface texture around the fracture origin, visible on the physical specimen if you knew where to look.

The defense examined on three points: the EN 14179-1 standard’s detection limitations (she cited the standard directly), the stress intensity factor calculation method (she explained the mist radius measurement and the Paris Law derivation), and whether the inclusion could have been introduced during installation damage rather than during manufacture (she explained the inclusion morphology — the crystalline NiS particle surrounded by a clear phase boundary with no surrounding damage zone, characteristic of a manufacturing inclusion, not a post-installation defect).

She answered 27 questions in four hours.

The deposition record built throughout: “IStructE Chartered Structural Glass Engineer: Dr. Ingrid Holm, IStructE-GE-IH-4413. Fractographic finding: NiS inclusion at fracture origin, panel failed spontaneously. Liability: manufacturer.”

Dr. Price emailed her after the session.

“Dr. Holm — your fractographic methodology and your IStructE registration are the technical foundation of the plaintiff’s case. The NiS identification changes the liability from user-impact to manufacturer-defect. The deposition record is clear.”

Sven had been at the lab when she returned. He said: “IStructE-GE-IH-4413 in the court record.”

She said: “Yes.”

He said: “The red dot.”

She said: “The red dot is in the record.” She picked up the acrylic case. She looked at the red dot through the lid. She set the case back on the desk.

Whitfield called that evening. He said: “The deposition outcome was satisfactory. Your analysis held up under cross-examination.” She said: “The fractographic methodology was sound.” He said: “Yes. I’ve amended the PI claim attribution — your name and IStructE registration going forward. And I’m implementing a firm protocol requiring IStructE-registered engineers to be named authors on all glass failure analysis reports submitted to insurers or courts.” She said: “Yes.” He said: “This is the kind of technical depth that sets the firm apart.” She said: “Yes.” She put the acrylic case in the desk drawer. She opened the next project file.

He had said it again. “Sets the firm apart.” She had said “yes.” She had put the case in the drawer. She had opened the next project.

The court record, once the PI proceedings concluded, would carry the deposition transcript in the full case file. She had the case reference number from Dr. Price’s final instructions letter. She had noted it in her project folder for the balustrade case.

The firm protocol had arrived in her email two days after Whitfield’s call — the formal document, countersigned by the firm’s Managing Partner, requiring IStructE-registered structural glass engineer named authorship on all glass failure analysis reports submitted to insurers or courts. One page. She had read it. She had filed it.

She had thought, briefly, about whether the protocol was a change or a statement. It described a procedure that the firm’s PI submissions had not followed. It described it as a new requirement. It did not describe what the previous practice had been, or why the procedure had not been followed before, or what the firm would do differently in retrospect for the reports that had already been submitted.

She had not annotated this thought in the document. She had read the protocol. She had filed it. She had gone back to the new case.

“Sets the firm apart,” he had said. Both times. Before the PI claim and after the deposition. The same phrase. The content of the firm’s capability that set it apart had not changed between the two uses. She had said “thank you” the first time. She had said “yes” the second time.

The acrylic case was in the desk drawer. She had put it there after the deposition — not for any reason she had examined, but because the deposition was complete and the case belonged with the ongoing reference materials rather than on the desktop where she had been working with it actively. It was in the drawer. The red dot was in the drawer.

She was preparing the new case file — a different panel, a different building, a different fracture pattern. The SEM stage was loaded with the new specimen. She reached for the acrylic case on her desk — the teaching reference, the 6cm × 4cm fragment from the balustrade case, the red dot at the NiS inclusion site.

She used it to calibrate her initial fracture surface assessment before loading new specimens: she held the teaching fragment up, identified the Wallner lines radiating from the dot, confirmed her eye was calibrated for the origin geometry, then set it beside the new specimen for comparison. Sven was at the SEM workstation, adjusting the imaging parameters.

The deposition record was in the court archive — “IStructE Chartered Structural Glass Engineer: Dr. Ingrid Holm, IStructE-GE-IH-4413, NiS inclusion fractographic identification” — alongside the amended PI claim. She held the teaching fragment under the light one more time. She opened the acrylic case. She set the fragment on the bench.

The amended PI claim had arrived by email from the firm’s administrative system: “Analysis by Dr. Ingrid Holm, CEng MIStructE, IStructE-GE-IH-4413.” She had read it. She had filed it in the balustrade case folder alongside the original PI claim with “glass analysis support.” Both were in the folder. She had not separated them. She had not annotated either.

The firm protocol had arrived the same morning — the formal policy document requiring IStructE-registered structural glass engineer named authorship on all glass failure analysis reports submitted to insurers or courts. She had read it. She had filed it.

The new case was a facade panel from an office building — a laminated safety glass unit, not tempered, so NiS spontaneous breakage was not a concern. The failure pattern suggested a manufacturing defect in the interlayer — delamination at the edge zone, with crack propagation from the delamination boundary. Different mechanism, different surface features, different SEM preparation protocol.

She had already identified the fracture origin in the preliminary optical examination — a circular delamination zone at the corner of the panel, visible as a slightly clouded area before the crack had propagated. She had circled it in her notes. She would confirm it under the SEM.

Sven had the imaging parameters ready. She mounted the new specimen beside the balustrade teaching fragment on the comparison bench. She held both under the lab light — the balustrade fragment with its Wallner lines radiating from the red dot, the new specimen with its delamination boundary and the faint crack initiation site at the corner.

“The distinction between the two,” she said to Sven, “is in the fracture surface topography at the origin. Here” — she pointed to the balustrade fragment — “the Wallner lines are concentric and well-defined because the fracture initiated from a single point. Here” — she indicated the new specimen — “you’re going to see a different pattern. Delamination-initiated cracks propagate along the interlayer boundary first, then through the glass. The surface features will be elongated rather than concentric.”

She put both specimens in the SEM specimen holder. She set the parameters. She ran the first scan.

The insurer’s original PI claim file — “Whitfield Glass Failure Analysis” — was in the insurer’s archive. The court had filed the original as an exhibit before the amendment had been processed. The amended report was in the updated court bundle. The original exhibit record in the insurer’s file had not been updated. She had the claim reference number: PI-2024-GF-0347. She had not looked it up since writing it down.

She checked the initial scan. The delamination boundary was visible. She noted the surface feature and increased the magnification for the origin identification. The SEM stage hummed. The image built.

She opened the acrylic case. She set the fragment on the bench.

The new case analysis was proceeding normally. The delamination-initiated fracture had confirmed at the SEM: the interlayer boundary was visible in the secondary electron image, the crack propagation path was along the boundary for approximately 8mm before transitioning into the glass body. The surface features at the transition point were elongated, as she had predicted to Sven — a different topography from the concentric Wallner lines of the NiS inclusion case.

She noted the findings in the project log. She prepared the preliminary fractographic assessment for the client — not the final report, which would wait for the second session, but the preliminary finding that would allow the client’s legal team to begin their liability assessment. The finding was clear: manufacturing defect in the interlayer adhesion at the corner zone, not impact damage, not installation error.

Sven printed the SEM images at four magnifications. She reviewed each one, annotating the key features in pencil on the print margins — the delamination boundary at 200×, the crack propagation path at 500×, the interlayer surface texture at 2,000×, the fracture origin at 5,000×. IStructE-GE-IH-4413 in the project header.

She set the prints beside the preliminary assessment document. She reviewed the complete package. She wrote the summary conclusion.

She opened the acrylic case. She set the fragment on the bench.