Chapter 23: Instrumental Analysis: The Microscope, GC-MS, Spectroscopy, and SEM-EDX — How the Lab Actually Sees

"A presumptive test asks a question. An instrument is the place you go for the answer — provided you ask it honestly, and provided you trust the sample you fed it more than the result you wanted." — constructed teaching line, in the voice of this book [constructed]

Overview

For three chapters now you have watched the chemistry bench suggest. A color test turned the expected shade. A microcrystalline test threw the expected crystal. A fire investigator stood in a burned room and said the pattern looked like an ignitable liquid had been poured. Each of those is a hypothesis wearing a lab coat — useful, often right, and not, by itself, something you should be willing to send a person to prison on. This chapter is about the moment the bench stops guessing. It is about the instruments that turn looks like into is, and about the single most important sentence a competent analyst ever says on the stand: I confirmed it.

That word — confirmed — is doing real work, and the first job of this chapter is to make you respect it. A confirmatory result is not just a more expensive presumptive test. It rests on a different kind of evidence: not "this substance reacted the way the target substance reacts," but "this substance's molecular structure matches the target's, measured directly, by a method that would have given a different answer if the substance were different." The flagship of that idea is gas chromatography–mass spectrometry (GC-MS), which forensic chemists only half-jokingly call the gold standard of chemical identification. Alongside it sit the spectroscopies — FTIR, Raman, UV-Vis — that read a molecule's vibrations and its color, and the electron microscope with SEM-EDX, which can photograph a particle a thousand times too small to see and then tell you which elements it is made of.

This is the lab's confirmatory backbone. When the toxicology chapter said "screen, then confirm by GC-MS," this is the confirm. When the chemistry chapter found gasoline chemistry presumptively, this is where that finding becomes instrumentally certain. When the arson chapter established an incendiary fire on valid grounds, this chapter is what carries the accelerant from "the debris smells of it and the pattern fits" to "the laboratory identified gasoline in the fire debris by GC-MS." Earlier chapters pointed here. We have arrived.

But the chapter's other half is a warning, and it is the warning that separates a scientist from a technician. An instrument is only as honest as the sample you put into it. Garbage in, garbage out — the oldest law in the laboratory. A perfect GC-MS run on a contaminated extract gives you a perfect, confident, wrong answer. The instruments in this chapter sit at the high-validity end of the spectrum we have been building since Chapter 1 — but high validity is a property of the method and the practice, never of the method alone.

In this chapter, you will learn to:

• Explain the move from presumptive screening to confirmatory instrumental analysis, and say precisely what makes a result confirmatory.
• Describe how chromatography separates a mixture, and read a chromatogram as retention times and peak areas.
• Explain how a mass spectrum is produced and why GC-MS functions as a near-definitive identification.
• Tell FTIR, Raman, and UV-Vis apart by what they measure and what each cannot do.
• Explain SEM-EDX: how it images a particle and reads its elements, and why that is class evidence.
• Read a labeled chromatogram and a labeled mass spectrum, and state honestly where instrumental analysis sits on the validity spectrum — and where "garbage in" defeats it.

Learning Paths

This chapter is the technical heart of Part IV, and it serves the four readers differently. None of you should skip it — even the juror needs to know what "confirmed by GC-MS" actually means — but here is how the paths run.

🔎 Investigator/CSI: Your job is upstream of this chapter, and that is exactly why it matters to you. The instruments here can only confirm what you collected cleanly. Weight §23.1 (the chain from your evidence bag to the bench) and the "garbage in, garbage out" thread throughout — your packaging is the instrument's first input. 🧪 Lab analyst: This is your chapter. Sections 23.2 through 23.5 are the instruments you will run, calibrate, and defend. Note especially what each method cannot do and the controls (blanks, standards, calibration) that make a result defensible rather than merely produced. ⚖️ Law/courtroom: Weight §23.1, §23.6, and every ⚖️ In the Courtroom box. "Confirmed by GC-MS" is among the strongest sentences a forensic witness can offer — and the cross-examination lives entirely in the sample's history, the controls, and the analyst's interpretation, not in the chemistry. 👥 General reader/juror: You have heard "the lab confirmed it" a hundred times on television. This chapter tells you what is true behind that phrase and what to ask. Weight §23.1, §23.6, and the Conclusion; the rest you can read for the pleasure of seeing how the machine actually works.

23.1 Why instruments: from presumptive to confirmatory

Begin with the question every chemical analysis is really being asked: what is this substance, and are you sure enough to say so under oath? The whole architecture of the modern lab — two stages, screen then confirm — exists to answer that question honestly, because a single test almost never can.

Recall the distinction this book drew back in Chapter 10 and sharpened in Chapter 21: a presumptive test narrows the possibilities cheaply and fast, while a confirmatory test establishes identity to a standard a court should accept. A presumptive test is sensitive — it rarely misses the target — but it is not specific: many substances can trigger the same response. The roadside drug field test that turns blue for cocaine will also, notoriously, turn blue for a list of innocent substances; the Kastle-Meyer test that signals blood will also signal some horseradish. These are not broken tests. They are screening tests, and screening is the first half of a two-part sentence. The second half — the confirmation — is what this chapter is about.

What, precisely, makes a result confirmatory? Three properties, and you should be able to recite them:

1. Specificity. The method responds to the target substance in a way that other substances do not reproduce. Ideally it interrogates the molecule's structure — its mass, its bonds, its arrangement of atoms — rather than a single bulk property like color, because structure is what makes one compound different from another.
2. Two orthogonal dimensions of information. The forensic consensus, codified by standards bodies for drug identification, is that confirmation should rest on at least two independent kinds of measurement that could each fail differently. GC-MS is the canonical example precisely because it combines two unrelated separations of information: how fast a molecule travels through a column (chromatography) and how it shatters into fragments (mass spectrometry). For two unrelated substances to agree on both is vanishingly unlikely.
3. Comparison to a known standard, run the same day on the same instrument. A result means nothing in the abstract. It means something only against a reference standard of the suspected substance, analyzed under identical conditions, plus a blank that proves the instrument and the glassware were clean. We will return to these controls relentlessly, because they are where confirmation is won or lost.

🔬 At the Bench Here is the shape of a real drug-identification workflow, and notice how few steps involve the dramatic instrument. A submission arrives; the analyst documents and weighs it; performs a presumptive color test to form a hypothesis; perhaps a microcrystalline test; then prepares an extract and runs it on GC-MS against a reference standard and a blank. The instrument run is minutes. The defensibility comes from everything around it: the chain of custody (Chapter 2), the clean blank, the matching standard, the calibrated instrument, the analyst's training, and a second analyst's technical review. The machine does not "identify" the drug. The analyst identifies it, using the machine, and stands behind the identification. That distinction will matter enormously in court.

The reason the field insists on this architecture is written in blood — or rather, in exonerations and lab scandals. Recall from Chapter 4 the Massachusetts disasters: analysts who "dry-labbed," reporting confirmatory results for tests they never ran, and whose presumptive-only or fabricated work tainted tens of thousands of convictions. The lesson the field drew was not "trust instruments more." It was "the confirmatory step is not optional, it must actually be performed, and it must be documented well enough that a stranger could check it." An instrument used as theater — switched on to impress, its output never truly examined — is worse than no instrument, because it lends the authority of confirmation to work that was never confirmed.

So the move from presumptive to confirmatory is not a move from "good test" to "better test." It is a move from hypothesis to conclusion, and the entire credibility of the chemistry section rests on taking that second step seriously every time. The rest of this chapter opens up the instruments that make the step possible — and keeps returning to the uncomfortable truth that the instrument is the easy part.

🔍 Check Your Understanding 1. A field test turns the "right" color for heroin. Your colleague writes "heroin identified" in the report. What is wrong with that sentence, in one word, and what step is missing? 2. Why does confirmation by GC-MS rest on two kinds of information rather than one very precise kind? What failure does the redundancy guard against?

23.2 Chromatography: separating the mixture

Most forensic samples are not one substance. They are mixtures — a street drug cut with three adulterants, a fire-debris extract holding dozens of compounds, a blood specimen carrying a parent drug and its metabolites. Before you can identify the components, you have to separate them, and the family of methods that does the separating is chromatography.

Chromatography is a technique that separates the components of a mixture by exploiting how differently each component distributes itself between two phases: a moving "mobile" phase that carries the sample along, and a stationary phase that holds it back. The word comes from the Greek for "color writing" — the technique was born around 1900 separating plant pigments into colored bands — but modern chromatography rarely involves color at all. The principle is simpler than the name. Imagine a crowd walking down a long corridor whose walls are coated in something mildly sticky. People who barely interact with the walls reach the far end quickly; people who keep pausing to touch the walls arrive later. If each kind of person interacts with the walls to a characteristic degree, then the crowd that entered the corridor as a clump will exit it sorted into groups, separated in time. That is chromatography. The "corridor" is the column; the "stickiness" is the stationary phase; the crowd's pace is set by the mobile phase; and the time each component takes to traverse the column is its retention time — the single most important number a chromatogram reports.

Gas chromatography (GC)

In gas chromatography, the mobile phase is an inert gas (helium or hydrogen) and the sample must be something that can be vaporized without destroying it. The sample is injected, flash-heated into vapor, and swept by the carrier gas through a long, thin capillary column — often thirty meters of fused silica, coiled inside a temperature-controlled oven, its inner wall coated with the stationary phase. As the oven temperature is ramped up over the run, compounds that are more volatile and interact less with the coating come off the end first; heavier, stickier compounds come off later. Each emerges at its characteristic retention time and strikes a detector that registers a signal. Plot signal against time and you have a chromatogram: a flat baseline interrupted by peaks, one per separated component, each peak's position telling you which (by retention time) and each peak's area telling you roughly how much.

GC is the workhorse of forensic chemistry and toxicology. It is superb for the volatile and the semi-volatile: drugs, accelerants, many poisons, alcohols. Its great forensic virtue for this chapter is that it separates the staggeringly complex mixture of a fire-debris extract — gasoline alone is a soup of hundreds of hydrocarbons — into a reproducible pattern that an analyst can recognize. Its limits are equally important: a substance that will not vaporize, or that decomposes when heated, is invisible to GC. That is a real gap, and it is exactly the gap the next method fills.

High-performance liquid chromatography (HPLC)

When the target will not survive vaporization — large molecules, heat-sensitive compounds, many modern pharmaceuticals and their metabolites — the lab reaches for high-performance liquid chromatography (HPLC). The principle is identical; only the mobile phase changes. Instead of a carrier gas, a liquid solvent is pumped at high pressure through a packed column, and compounds separate by how they partition between that liquid and the stationary packing. Because nothing has to be vaporized, HPLC handles the heavy and the fragile that GC cannot touch. It is the method of choice for many toxicology confirmations, for substances like certain benzodiazepines and opioids, and increasingly it is coupled to a mass spectrometer (LC-MS) for the same reason GC is.

🔬 At the Bench A subtlety worth internalizing: chromatography separates; it does not, by itself, identify. A peak at a given retention time tells you a component is present and roughly how much — but retention time alone is weak identification, because different compounds can coincidentally share one. (This is the chromatographic version of the class-vs-individual problem from Chapter 1: retention time is a class characteristic. Many molecules can wear it.) That is why retention time is treated as one line of evidence, confirmed against a same-day standard, and why the truly confident identification comes only when chromatography is married to a detector that reads structure. Hold that thought; it is the whole reason the next section exists.

🔍 Check Your Understanding 1. Why does a fire-debris extract need to be separated before any component can be identified? What would happen if you tried to identify the soup all at once? 2. A toxicologist needs to confirm a large, heat-sensitive drug molecule. Should they reach for GC or HPLC, and why?

23.3 Mass spectrometry: the molecular fingerprint

If chromatography sorts the crowd, mass spectrometry interrogates each person as they leave. It is the instrument that reads structure, and when it is bolted onto the end of a gas chromatograph the combination — GC-MS — is the closest thing forensic chemistry has to a definitive answer.

GC-MS is a combined instrument in which a gas chromatograph first separates a mixture into its individual components, and a mass spectrometer then identifies each component by breaking its molecules into charged fragments and measuring the masses of those fragments. The output for each separated peak is a mass spectrum — and here is the term to fix in your mind: a mass spectrum is a plot of the relative abundance of the charged fragments a molecule shatters into, arranged by their mass-to-charge ratio (written m/z); the pattern of fragments is characteristic of the molecule's structure and serves, for a great many compounds, as a near-definitive chemical identification.

How does the machine produce that pattern? As each compound emerges from the GC column, it enters the mass spectrometer's vacuum and is bombarded — classically by a beam of electrons in a process called electron ionization. The collision does two things: it knocks an electron off the molecule, giving it a charge, and it dumps in enough energy to break the molecule apart along its weakest bonds in a reproducible way. The intact, charged molecule (the molecular ion) and all of its charged fragments are then accelerated through the instrument, sorted by their mass-to-charge ratio, and counted. The result is a bar-graph "fingerprint": a forest of peaks at specific m/z values, with specific relative heights, that reflects exactly how this kind of molecule prefers to break.

Two features make that fingerprint so powerful:

• The molecular ion gives molecular weight. The heaviest significant peak (often, though not always, present) corresponds to the whole molecule and tells you its mass — a strong constraint on identity.
• The fragmentation pattern is reproducible and structure-specific. A given compound, ionized the same way, always shatters into the same set of fragments in the same proportions. Two different compounds rarely produce the same full pattern. So the spectrum can be compared — peak by peak — against a vast library of reference spectra (for example, the well-known mass-spectral databases maintained by the U.S. National Institute of Standards and Technology), and a match across the whole pattern is a very specific finding.

Now combine the two instruments and feel why the pairing is so strong. The GC supplies retention time — a class characteristic, as we said. The MS supplies the fragmentation pattern — far closer to an individual characteristic of the molecule. A substance that matches a reference standard on both retention time and full mass spectrum has agreed on two independent dimensions that fail differently. The probability of an innocent coincidence on both, for a properly run analysis against a same-day standard, is so small that GC-MS is accepted across the field as confirmatory. This is the "two orthogonal dimensions" requirement of §23.1, satisfied in a single instrument.

⚖️ In the Courtroom When an analyst testifies "I identified the substance as methamphetamine by GC-MS," they are claiming something genuinely strong — and the honest analyst will say how strong and why: a retention-time match to a reference standard and a full mass-spectral match, with a clean blank and a documented chain of custody. This is one of the few places in the whole book where the verb identified is defensible rather than overstated, because the underlying method has the specificity to earn it. But note where the cross-examination still lives, because a good attorney will go straight there: not "is the chemistry valid?" (it is) but "was this sample what you say it was?" — the chain of custody, the blank, the reference standard's provenance, whether two submissions could have been swapped, whether the extract was contaminated. The science is sound; the sample's history is the battlefield. We are back, as always, to garbage in, garbage out.

A note on honesty even here, at the high-validity end. GC-MS confirms the identity of a compound. It does not, by itself, tell you whose sample it was, how it got there, or what it means. "Gasoline is present in this fire debris" is a confident chemical fact. "Therefore the defendant set the fire" is an enormous, unearned leap that the chemistry does not license. The instrument answers what, with great authority. It is silent on who and why — and the silence is not a defect to be talked over. It is the boundary of the evidence, and a competent witness respects it.

🔍 Check Your Understanding 1. What two pieces of information does GC-MS combine, and why is agreement on both so much stronger than agreement on either alone? 2. A mass spectrum matches a library entry for cocaine across its whole fragmentation pattern. Name one important question this confirmation leaves completely unanswered.

23.4 Spectroscopy: FTIR, Raman, and UV-Vis

Mass spectrometry breaks a molecule to read it. The spectroscopies read a molecule without destroying it, by watching how it interacts with light. The umbrella term is worth defining once, cleanly: spectroscopy is the family of analytical methods that identify and characterize a substance by measuring how it absorbs, emits, or scatters electromagnetic radiation across a range of wavelengths; the resulting spectrum reflects the substance's molecular structure. Three spectroscopies earn their place in the forensic lab, and each answers a different question.

FTIR — the molecular structure

The most important for general material identification is Fourier-transform infrared spectroscopy (FTIR). FTIR measures how a sample absorbs infrared light across a band of wavelengths; because each type of chemical bond absorbs at characteristic frequencies (it vibrates like a tiny spring tuned to a particular note), the resulting absorption spectrum is a structural fingerprint of the molecule. Shine infrared light through (or off) the sample, and the bonds within it absorb the specific frequencies that match their natural vibrations — a carbonyl bond here, a hydroxyl there, an aromatic ring's signature elsewhere. The spectrum that results, plotted as absorption against frequency, contains a "fingerprint region" so specific that two different compounds essentially never share it.

FTIR's forensic virtues are real and distinct from GC-MS. It is non-destructive — the sample survives for further testing, which matters when the sample is tiny or legally precious. It is fast, needs little preparation, and excels at identifying bulk materials where the question is "what is this stuff?": plastics, fibers, paints, polymers, many pure drugs, unknown powders. Its complement to GC-MS is exact: GC-MS shines on volatile compounds and complex mixtures separated into parts; FTIR shines on solid bulk materials taken whole. Many labs use them together. FTIR's principal limit is also exact: it struggles with mixtures, because the overlapping spectra of several components blur into a composite that can be hard to disentangle — the very situation where chromatography's separate-first approach wins.

Raman — the complement that sees through glass

Raman spectroscopy answers nearly the same question as FTIR — what bonds, what structure — but by a different physics: instead of measuring absorbed light, it measures the tiny fraction of light that a sample scatters at shifted wavelengths, a shift that again encodes the molecular vibrations. Why keep a second method that answers the same question? Because their blind spots are opposite. Raman often works through transparent containers — glass, some plastics, packaging — so a suspected substance can sometimes be screened without opening the evidence, a genuine advantage for hazardous or sealed material. Raman also handles water far better than FTIR (water is nearly opaque to the infrared FTIR relies on but barely registers in Raman), which matters for aqueous samples. Modern handheld Raman units let field officers screen an unknown powder at the scene — useful, and a place to remember §23.1's discipline, because a field screen is still a presumptive result that the lab confirms.

UV-Vis — the simplest, and the most limited

Ultraviolet-visible (UV-Vis) spectroscopy measures how a sample absorbs ultraviolet and visible light, producing a far simpler spectrum than the others — often just one or two broad humps. It is genuinely useful for quantification (how much of a known substance is present, since absorbance scales with concentration) and for narrowing down classes of compounds. But you should hold it at arm's length as an identification tool: its spectra are broad and low in detail, and many different substances share similar UV-Vis profiles. UV-Vis can screen and quantify; it should not be asked to confirm an identity on its own. Treated as what it is — a screening and quantification tool — it is valuable; treated as a confirmatory identification, it overreaches.

⚠️ Junk-Science Alert The danger across all three spectroscopies is not the physics — it is the library match score presented as if it were a verdict. Modern instruments compare an unknown's spectrum to a reference library and report a percentage "match." That number is a starting point for an analyst's judgment, not a conclusion. A high match to the wrong library, a match driven by a dominant component while a mixture's minor components hide, a spectrum from a contaminated or degraded sample — each can yield a confident, high-scoring, wrong identification. The instrument's confidence is not the analyst's confidence, and the analyst's job is to interrogate the match (Is the whole spectrum explained? Is there an unexplained peak? Was the blank clean? Does it make sense in context?), not to read the percentage aloud. A spectrum is evidence; a match score is a hypothesis. Confusing the two is how good instruments produce bad conclusions.

🔍 Check Your Understanding 1. You have a small, intact pill in a sealed transparent vial and you want to screen it without opening the evidence. Which spectroscopy is best suited, and why? 2. Why is UV-Vis a poor choice for identifying an unknown, even though it is excellent for measuring how much of a known substance is present?

23.5 Electron microscopy and SEM-EDX: elemental analysis

Everything so far has identified molecules. The last instrument identifies elements, and it does so on particles far too small for any optical microscope to resolve. To reach it, start with the microscope you already half-know.

The light microscope — the criminalist's oldest instrument — magnifies by bending visible light through lenses, and physics caps its useful magnification at roughly a thousand times, because you cannot resolve detail finer than the wavelength of the light you are using. For a great deal of forensic work that is plenty: comparing fibers and hairs (Chapter 19), examining a questioned document, reading the rifling marks under a comparison microscope (Chapter 15). Call this analytical microscopy when it is used not just to look but to measure and compare — the use of magnifying optical (and, below, electron) instruments to observe, characterize, and compare the physical and structural features of trace evidence that are invisible or ambiguous to the naked eye. The optical microscope is where instrumental "seeing" begins, and it remains, quietly, one of the most-used instruments in the building.

But when the question requires magnifications of tens of thousands, or the elemental composition of a speck, light is not enough — and the lab switches from light to electrons.

The scanning electron microscope (SEM)

A scanning electron microscope (SEM) illuminates a specimen not with light but with a focused beam of electrons, scanned across the surface in a raster while detectors collect the signals that bounce back. Because electrons have a far shorter effective wavelength than visible light, the SEM resolves structure orders of magnitude finer than any optical scope, producing images of extraordinary depth and detail at magnifications well beyond ten thousand times. On its own the SEM is a seeing instrument: it shows you the shape, texture, and surface structure of a particle — the cratered morphology of a gunshot-residue particle, the layer edges of a paint chip, the surface of a pollen grain or a fiber. But morphology is only half the story. The other half is composition, and that is where the second instrument rides along.

Energy-dispersive X-ray analysis (EDX) and SEM-EDX

When the SEM's electron beam strikes the specimen, it does more than bounce — it knocks inner-shell electrons out of the sample's atoms, and as the atoms relax they emit X-rays whose energies are characteristic of the specific elements present. An energy-dispersive X-ray (EDX, also written EDS) detector catches those X-rays and reads off which elements are in the spot the beam is hitting, and roughly in what proportion. Bolt the two together and you have the instrument this section is named for: SEM-EDX is the combination of a scanning electron microscope, which images a particle's shape and surface at very high magnification, with an energy-dispersive X-ray detector, which determines the elemental composition of that same particle; together they reveal both what a microscopic particle looks like and which chemical elements it is made of.

The forensic showcase for SEM-EDX is gunshot residue (GSR), which Chapter 24 will treat in full. When a firearm discharges, the primer vaporizes and condenses into microscopic particles with a characteristic morphology (tiny spheroids) and a characteristic elemental signature — classically the combination of lead, barium, and antimony fused in a single particle. SEM-EDX is uniquely suited to find such a particle on a sampling stub and confirm both its telltale shape and its elemental composition. That two-part confirmation — morphology and elements together — is what makes a particle "characteristic of" GSR rather than merely "consistent with." The same instrument characterizes paint pigments by their elements, glass by its minor constituents, and unknown particles of every kind.

Now the honesty, which for this instrument is unusually important. Elemental composition is class evidence, not individualization. SEM-EDX can tell you a particle is made of lead, barium, and antimony in proportions characteristic of primer residue. It cannot tell you which gun, which cartridge, or which person — only that the particle belongs to a class of particles produced by a class of events. A paint chip's elemental profile can say "consistent with an automotive paint of this type"; it cannot say "this car and no other." This is the class-versus-individual logic of Chapter 1, applied to the smallest things the lab can see: SEM-EDX is superb at saying what a particle is made of and weak at saying where, uniquely, it came from. An analyst who lets a jury hear an elemental match as a unique source has stepped off the validity spectrum, no matter how impressive the instrument. And — the recurring warning of this whole chapter — the GSR result is only as clean as the sampling: residue transfers easily, contaminates readily, and a positive can arrive on a sleeve by routes that have nothing to do with firing a weapon. The instrument reads the particle faithfully. It cannot vouch for how the particle got onto the stub.

🔬 At the Bench A practical reason SEM-EDX is trusted where it is trusted: it is automatable and non-destructive of the particle. Modern systems can scan a GSR stub for hours unattended, flagging every particle whose composition matches the target, and the flagged particle survives for a human to re-examine and photograph. That combination — exhaustive automated search plus a preserved, re-checkable particle plus a two-dimensional confirmation (shape and elements) — is what lifts characteristic-GSR identification above the pattern-matching disciplines. But automation also imports a quiet bias risk: a system tuned to find a particular signature will, by construction, report that signature and not others. The criterion of what counts as a "characteristic" particle is a human choice, made in advance, and it deserves the same scrutiny as any other interpretive decision in the lab.

🧠 Cognitive-Bias Watch It is tempting to believe that an instrument cannot be biased — that once the chemistry is automated, the human, with all the expectations the detective put in their head, has been engineered out. That belief is half true and dangerous for the other half. The instrument is objective: the mass spectrometer reports the fragments it finds, the EDX detector reports the elements it sees. But the analysis is studded with human choices that expectation can tilt — which marginal library match to accept; whether an unexplained minor peak is "noise" or a second component; where to draw the line on what counts as a "characteristic" GSR particle; how far to go in the wording of testimony. An analyst who knows the lab "needs" a positive can, without any dishonesty, resolve each of those judgment calls in the wanted direction — the same contextual drift this book names everywhere and dissects in Chapter 31. The safeguard is the same here as for the pattern disciplines: keep domain-irrelevant information (the suspect, the wanted answer) away from the bench until the result is fixed, then compare. The objectivity of the instrument is not a substitute for the independence of the analyst.

🔍 Check Your Understanding 1. The SEM gives you a particle's shape; EDX gives you something else. What does EDX add, and why is the combination stronger than either signal alone for identifying GSR? 2. SEM-EDX reports that a particle is composed of lead, barium, and antimony. State the single most important thing this finding does not establish.

23.6 Reading a chromatogram and a mass spectrum

Theory becomes skill when you can read the output. This section teaches the two graphs at the center of the chapter — a GC chromatogram and a mass spectrum — by walking through a labeled example of each. Both diagrams are schematic and not to scale; a real instrument's output carries exact axis values, and a real analyst reads them against same-day standards and blanks. The legend, stated once: peaks rise above a flat baseline; the horizontal axis is time (chromatogram) or mass-to-charge ratio, m/z (mass spectrum); peak height/area means relative amount (chromatogram) or relative abundance of a fragment (mass spectrum).

Reading a GC chromatogram

A chromatogram is the record of what came off the GC column, and when. Here is a schematic fire-debris chromatogram of the kind §23.2 described — gasoline separated into its many hydrocarbons, the pattern an analyst learns to recognize on sight.

FIGURE 23.1 — "A gasoline chromatogram: the pattern in the debris"     [constructed teaching example]

 Detector
 signal
   │
   │                         ▲                  the tall, evenly-spaced peaks in the
   │              ▲          ║║      ▲           mid-to-late region are aromatics
   │             ║║   ▲     ║║║║    ║║           (toluene, xylenes, the C3-alkylbenzenes) —
   │        ▲   ║║║  ║║    ║║║║║   ║║║   ▲       the "signature ridge" of weathered gasoline
   │   ▲   ║║║ ║║║║ ║║║║  ║║║║║║  ║║║║  ║║   ▲
   │  ║║  ║║║║ ║║║║ ║║║║║ ║║║║║║║ ║║║║║ ║║║  ║║
───┼──╨╨──╨╨╨╨─╨╨╨╨─╨╨╨╨╨─╨╨╨╨╨╨╨─╨╨╨╨╨─╨╨╨──╨╨───────►  time (retention time →)
   │  ↑                          ↑                  ↑
   │ solvent              individual hydrocarbons   heavier
   │  /early                 (each peak = one        compounds
   │  light ends             separated compound)     elute last
   │
   └─ baseline (flat where nothing is eluting)

 Each PEAK = one separated compound.  Its POSITION (retention time) = which compound (a class
 characteristic, confirmed against a standard).  Its HEIGHT/AREA = roughly how much.
 The whole PATTERN — the relative sizes and spacing of the aromatic peaks — is what an analyst
 recognizes as "gasoline," compared to a reference gasoline standard run the same day.
 Illustrative, not to scale; a real chromatogram carries exact retention-time and abundance values.

Walk through it. The flat stretches are baseline — the column is delivering nothing but carrier gas, and the detector reads near zero. Early on, a cluster of peaks marks the volatile "light ends" and any solvent. Then comes the body of the chromatogram: a procession of peaks, each one a single compound that the column separated out, arriving at its own retention time. Read left to right and you are reading the sample sorted by volatility, lightest first. The position of each peak tells you which compound it might be — but only might, because retention time is a class characteristic that several compounds could share; the identity is confirmed against a standard, and ultimately by the mass spectrum of each peak. The height or area of each peak tells you roughly how much of that compound is present. And the feature that matters most for fire debris is not any single peak but the whole pattern — the characteristic distribution and relative heights of the aromatic compounds, the "signature ridge" a trained analyst recognizes as gasoline, especially gasoline that has partly evaporated (weathered) in a fire. The analyst does not eyeball one peak and declare an accelerant; they compare the total pattern against a reference gasoline standard, following a published classification scheme for ignitable liquids. Pattern, against a standard, with a clean blank — that is the discipline.

What the chromatogram does not tell you, on its own, deserves a sentence of its own: it does not, by retention time alone, prove the identity of any single peak, and it says nothing about who poured the liquid or when. It tells you what is present, and in what pattern. Identity confirmation comes when each significant peak is fed to the mass spectrometer — which is the next graph.

Reading a mass spectrum

If the chromatogram is the lineup, the mass spectrum is the interrogation of one suspect pulled out of it. Select a single peak from Figure 23.1, send that one separated compound into the mass spectrometer, and you get a fragmentation fingerprint. Here is a schematic spectrum of one such aromatic compound.

FIGURE 23.2 — "A mass spectrum: the molecular fingerprint of one peak"     [constructed teaching example]

 Relative
 abundance
 (%)
 100 ┤                 █                         ← BASE PEAK (most abundant fragment;
     │                 █                            here m/z 91, the tropylium ion, the
  80 ┤                 █                            signature fragment of an alkylbenzene)
     │                 █
  60 ┤                 █
     │                 █                    █     ← MOLECULAR ION (M⁺), the whole charged
  40 ┤                 █                    █        molecule = molecular weight (here m/z 106)
     │        █        █                    █
  20 ┤   █    █    █   █        █     █     █
     │   █    █    █   █    █   █     █     █
   0 ┼───┴────┴────┴───┴────┴───┴─────┴─────┴──────────►  m/z (mass-to-charge ratio →)
        39   51   65  91   77       103   106
        └── lighter fragments ──┘            └─ molecular ion

 Each BAR = a charged fragment the molecule broke into.  Its POSITION (m/z) = that fragment's mass.
 Its HEIGHT = how abundant that fragment is.  The BASE PEAK (tallest, set to 100%) and the overall
 PATTERN are compared, peak-by-peak, against a reference library; the MOLECULAR ION (heaviest major
 peak) gives the molecule's weight.  Agreement across the WHOLE pattern + the matching retention time
 from Figure 23.1 = a confirmatory identification.
 Illustrative m/z values for teaching; a real spectrum is matched to a standard and a library.

Walk through this one too. Every bar is a charged fragment the molecule shattered into when the mass spectrometer ionized it; its horizontal position is the fragment's mass-to-charge ratio (m/z), and its height is how abundant that fragment is. The tallest bar, scaled to 100 percent, is the base peak — the fragment the molecule most readily breaks into; for the alkylbenzenes of gasoline, a fragment at m/z 91 (a famously stable seven-carbon ring fragment) is a giveaway of that whole family. Off to the right sits a smaller but crucial peak: the molecular ion, the intact molecule with its charge, whose m/z equals the compound's molecular weight — a hard constraint on what the molecule can be. Everything in between is the fragmentation pattern: the specific set of pieces, in their specific proportions, that this molecular structure prefers to break into.

The identification is made not from any one peak but from the whole pattern at once, compared bar-by-bar against a reference spectrum (from a same-day standard and from a spectral library such as the NIST database). When the unknown's full fragmentation pattern matches the reference and its retention time from the chromatogram matches the standard, you have the two orthogonal agreements of §23.1, and the identification is confirmatory. That is the moment "looks like gasoline" becomes "gasoline, identified by GC-MS." One graph found the component and timed it; the other read its structure; together they are as close to certainty about a compound's identity as forensic chemistry gets.

And the same boundary applies, stated plainly because it is the chapter's spine: this confirms what the compound is. It is silent on how the debris came to contain it, on who was involved, and on when. The mass spectrum is a triumph of specificity about identity and says nothing at all about responsibility. Reading it well means reading both what it establishes and where it stops.

🔬 Read the Evidence

text FIGURE 23.3 — "The fire-debris extract on the bench" [the cold case] THE ITEM Charred flooring and soil from the front room of the Mill Creek cabin, sealed at the scene in a clean, vapor-tight metal can; a same-day reference gasoline standard; and an instrument blank. THE CONTEXT Chapter 21 found gasoline chemistry presumptively; Chapter 22 established an incendiary fire (multiple origins + ignitable-liquid pattern) on valid grounds. The debris vapor is extracted (passive headspace onto an adsorbent) and run on GC-MS against the standard and the blank. The blank is clean. WHAT IT SHOWS The chromatogram reproduces the characteristic weathered-gasoline pattern (the aromatic "signature ridge" of Figure 23.1); the mass spectra of the diagnostic peaks match the reference and the spectral library; retention times match the standard. The classification scheme for ignitable liquids is satisfied for gasoline. WHAT IT DOESN'T It does not say who introduced the gasoline, when, or why; it does not, by itself, prove arson (the fire science of Chapter 22 does that); and it cannot date the pour. A clean result also depends entirely on clean collection — a contaminated can could mimic or mask. THE INFERENCE The accelerant is now CONFIRMED INSTRUMENTALLY: gasoline is identified in the fire debris by GC-MS, corroborating — at a higher level of validity — the presumptive and fire-science findings that preceded it. Honest status: *accelerant confirmed instrumentally.* THE LESSON Confirmation upgrades the *certainty of identity*, not the *reach of the inference*. "Gasoline is present, confirmed" is a strong, defensible fact; "therefore the defendant" is a leap the chemistry never makes.

🔍 Check Your Understanding 1. In Figure 23.1, what does the position of a peak tell you, and why is it not enough by itself to identify the compound? 2. In Figure 23.2, what is the difference between the base peak and the molecular ion, and what does each contribute to the identification?

🗂️ The Case File

The accelerant, confirmed — and a particle that only whispers. Two instrumental results come back on the Mill Creek case this chapter, and the discipline is in stating each at exactly its strength.

First, the fire debris. Chapter 21 found gasoline chemistry presumptively; Chapter 22 established, on valid fire-science grounds (multiple origins, an ignitable-liquid burn pattern, none of the debunked Willingham-era folklore), that the fire was incendiary. This chapter delivers the confirmation those chapters pointed toward. The state laboratory extracted the vapor from the sealed fire-debris can — clean blank alongside it, reference gasoline standard run the same day — and analyzed it by GC-MS. The chromatogram reproduced the characteristic weathered-gasoline pattern; the mass spectra of the diagnostic peaks matched the reference and the spectral library; the retention times matched the standard; the ignitable-liquid classification scheme was satisfied. Honest status: the accelerant is confirmed instrumentally — gasoline is identified in the fire debris by GC-MS. This is the chapter's promised beat, and it is a genuinely strong, high-validity result. Note carefully what it does not do: it does not name who poured the gasoline, when, or why. It upgrades the certainty of the accelerant's identity, not the reach of any inference about a person. "Gasoline is present, confirmed" is a fact the chemistry earns; everything beyond it must come from other evidence.

Second, a particle on a sleeve. A garment sleeve recovered in the course of the investigation was sampled and examined by SEM-EDX. The instrument imaged microscopic particles and read their elemental composition. The findings are reported here at class-level, consistent-with strength only — particle morphology and elemental signatures consistent with the kinds of particles such an environment can produce, the sort of result that cannot be excluded and may corroborate but, on its own, associates a class, not a person. Elemental composition is class evidence (§23.5); it does not individualize a source, it cannot say how a particle reached a sleeve, and a particle can arrive by transfer and contamination along routes that have nothing to do with the event in question. No name is attached to this sleeve here, and none can be on the strength of the particle work alone. It is a thread to be weighed later, with everything else, not a finger pointed now.

Status after this chapter: the arson finding stands, now instrumentally confirmed as to the accelerant — gasoline. The sleeve particle is, at most, a consistent-with corroborator awaiting context. The file has gained certainty about what the fire debris contains and has, deliberately, gained no certainty at all about who. That is the honest line, and we hold it.

Conclusion

This chapter was the lab's confirmatory backbone, the place earlier chapters were pointing. We drew the line between a presumptive hypothesis and a confirmatory conclusion, and named what makes a result confirmatory: specificity, two orthogonal dimensions of information, and comparison to a same-day standard with a clean blank. We watched chromatography separate a mixture into a readable pattern of peaks; mass spectrometry read each peak's molecular structure; and the marriage of the two — GC-MS — earn the rare, defensible verb identified. We sorted the spectroscopies by the questions they answer (FTIR for bulk molecular structure, Raman as its complement through glass and water, UV-Vis for quantification but not confident identification) and met the electron microscope and SEM-EDX, which see and chemically read a particle far too small for light — class evidence about elements, never a unique source. And we learned to read the two central graphs, a chromatogram and a mass spectrum, at honest strength.

Two truths sit at the center of it all. The first is that these instruments belong at the high-validity end of the spectrum this book has built since Chapter 1 — grounded in analytical chemistry and physics, quantified, reproducible, and, for GC-MS, genuinely confirmatory of a compound's identity. This is the part of forensic science that most deserves the public's trust, and we should say so plainly. The second truth is the one that keeps the first honest: garbage in, garbage out. A flawless instrument run on a contaminated, mislabeled, or poorly collected sample produces a flawless, confident, wrong answer. The validity lives in the method and the practice — the chain of custody, the blank, the standard, the analyst's judgment, the second reviewer — never in the machine alone. And even a perfect confirmation answers only what, never who or why; the silence past that boundary is not a gap to be filled with inference but a limit to be respected.

In the cold case, the instruments did exactly what they should: they confirmed the gasoline in the fire debris, upgrading the accelerant finding to instrumental certainty, and they kept the sleeve particle honestly at consistent-with. The next chapter, Chapter 24, takes up the microscopic transfer evidence those particles belong to in full — gunshot residue, paint, glass, and soil — where SEM-EDX does much of the work and where the discipline of saying "class, not individual" matters most. The instruments can see almost anything. Knowing what their seeing does and does not mean is the science.

Key Terms

• GC-MS (gas chromatography–mass spectrometry) — a combined instrument that first separates a mixture into its components by gas chromatography and then identifies each by its mass-spectral fragmentation pattern; the forensic gold standard for confirming a compound's identity.
• Chromatography — a technique that separates the components of a mixture by how differently each distributes between a moving (mobile) phase and a stationary phase; its output is a chromatogram of peaks at characteristic retention times.
• Mass spectrum — a plot of the relative abundance of the charged fragments a molecule breaks into, arranged by mass-to-charge ratio (m/z); the fragmentation pattern is structure-specific and serves as a near-definitive identification of the molecule.
• FTIR (Fourier-transform infrared spectroscopy) — an analytical method that identifies a substance by how it absorbs infrared light, since each bond type vibrates at characteristic frequencies; yields a structural "fingerprint" of the molecule, non-destructively.
• Spectroscopy — the family of methods that identify and characterize a substance by measuring how it absorbs, emits, or scatters electromagnetic radiation; the spectrum reflects molecular structure.
• SEM-EDX (scanning electron microscopy with energy-dispersive X-ray analysis) — the combination of a scanning electron microscope (imaging a particle's shape and surface at very high magnification) with an energy-dispersive X-ray detector (reading the particle's elemental composition); reveals both what a microscopic particle looks like and which elements it is made of. Elemental composition is class evidence.
• Analytical microscopy — the use of magnifying optical and electron instruments to observe, characterize, and compare structural features of trace evidence that are invisible or ambiguous to the naked eye.

Spaced Review

1. Distinguish a presumptive test from a confirmatory one, and explain why GC-MS counts as confirmatory while a color test does not. (§23.1; recall the presumptive/confirmatory pairing from Chapters 10 and 21.)
2. In Chapter 21 the lab found gasoline chemistry presumptively; in this chapter it is confirmed. Explain in one or two sentences what changed, scientifically, between those two statements — and what did not change about who or what the evidence implicates. (§23.6, the Case File)
3. Validity-spectrum question. Where does instrumental confirmation by GC-MS sit on the NAS 2009 / PCAST 2016 validity spectrum, and why — and what single practical condition can collapse its reliability despite its high standing? (§23.1, Conclusion; contrast with bite-mark comparison, Chapter 16.)
4. The autopsy found no soot in the airways (Chapter 11), establishing the victim was dead before the fire; this chapter confirms gasoline in the debris. Explain how these two findings, from different disciplines, fit together in the cold case without either one alone proving a homicide. (Chapters 11, 22; §23.6)
5. An analyst testifies that SEM-EDX found a particle of lead, barium, and antimony on a sleeve. A juror concludes the wearer fired a gun. Name two distinct overstatements in that leap, using terms from this chapter and from Chapter 1. (§23.5)