Forensic Chemistry: The Science Behind the Crime Lab

At 9:05 on the morning of Monday, 10 September 1984, a geneticist named Alec Jeffreys lifted a sheet of freshly developed X-ray film out of a wash bath in his laboratory at the University of Leicester. He was looking at an autoradiograph, a photographic record of where radioactive probes had bound to fragments of human DNA spread across a gel. What he saw was a smeared, ladder-like pattern of bands, and the pattern was different for every person on the gel. More than that, the bands a child carried could be traced back to the bands of that child's mother and father. In a few minutes of staring at a piece of film, Jeffreys realized he was looking at something genuinely new: a chemical signature unique to each individual yet inherited in a readable way. He called the technique DNA fingerprinting, and within four years it would secure the first criminal conviction in history obtained from a genetic match, free the first wrongly accused suspect, and remake an entire scientific discipline.

That discipline is forensic chemistry, the application of analytical chemistry to questions of law. It is older than the DNA era by a century and a half, and it is built on a demanding idea: that physical evidence, properly analyzed, can speak more reliably than any witness. This article traces how it works and how it grew, from a glass tube coated in metallic arsenic to a genetic profile with odds against a false match of better than one in a quintillion.

A morning in Leicester that changed criminal justice

The discovery in Leicester moved quickly from curiosity to courtroom. By 1985, Jeffreys had applied DNA fingerprinting to its first practical case, a paternity and immigration dispute, where the technique proved a boy's biological relationship to his family. Its first criminal application followed in Leicestershire in 1986 and 1987, and it arrived under the worst possible circumstances. Two teenage girls had been murdered in neighboring villages. Lynda Mann was killed in Narborough in 1983, and Dawn Ashworth in Enderby in 1986, and the cases bore the marks of a single attacker.

A local man, Richard Buckland, confessed to the Ashworth murder, and the investigation might have ended there, except that the police asked Jeffreys to confirm the confession against the biological evidence. His DNA analysis did the opposite. It showed that one man had committed both crimes, but that man was not Buckland, who was cleared in 1986. This was the first time DNA evidence had exonerated a suspect before conviction, and it carried an uncomfortable lesson: a confession can be wrong, but the chemistry was not. To find the real killer, investigators launched a mass screen, collecting blood and saliva from roughly 5,000 men across the local villages. The screen alone did not catch him, because a man named Colin Pitchfork persuaded a colleague to give a sample in his place. Only when that deception came to light was Pitchfork tested, matched, and charged. He pleaded guilty in January 1988. The case produced two firsts at once: the first criminal conviction secured by DNA evidence and the first pre-conviction exoneration by it.

Reading the chemical signature inside a cell

The original technique Jeffreys used was laborious, depending on large stretches of repetitive DNA and radioactive probes. Modern forensic DNA analysis is faster, more sensitive, and almost entirely automated, but the logic is the same. An analyst begins with a biological sample, which might be blood, saliva, semen, or a few skin cells, and extracts the DNA from it. Because a crime-scene trace may contain only a vanishing amount of material, the next step is amplification: the polymerase chain reaction, or PCR, copies specific target regions millions of times until there is enough to measure.

Those target regions are the key. Scattered through the human genome are short, repeated sequences called short tandem repeats (STRs), where a brief motif of DNA repeats head to tail a variable number of times. The number of repeats at any given location differs widely from person to person, and that variation is what makes a profile distinctive. Forensic labs amplify a fixed, standardized set of these locations, called loci, so that results from different laboratories can be compared. After amplification, the fragments are sorted by size using capillary electrophoresis, which pulls the DNA pieces through a thin tube under an electric field so that shorter fragments travel faster and longer ones lag behind. The output is a profile, a set of numbers describing how many repeats sit at each locus, and that profile is compared against a suspect's sample or against a database. In the United States, the CODIS system uses a core panel of twenty STR loci. Because the loci are chosen to be statistically independent, the chance that two unrelated people share a full profile by accident is better than one in a quintillion, a number with eighteen zeros.

When a single test was enough to hang a poisoner

Long before DNA, the central problem in forensic chemistry was poison, and the central poison was arsenic. It was cheap, widely available as a rat killer, tasteless in food, and it produced symptoms that mimicked natural illnesses such as cholera. For much of history a suspected arsenic murder was nearly unprovable, because the available tests were unreliable and easily dismissed in court. That changed in 1836, when the British chemist James Marsh published a method sensitive enough to convince a jury.

The Marsh test works by reduction. A sample suspected of containing arsenic is treated with zinc and acid, and any arsenic present is converted into a gas called arsine. When that gas is led through a heated glass tube, it decomposes and deposits a shiny black film of metallic arsenic on the glass, a so-called arsenic mirror. The amount deposited could be compared against standards, making the result both visible and quantifiable. The Marsh test belongs to a broader family of methods that forensic chemists call presumptive tests, meaning quick, inexpensive procedures that strongly suggest a substance is present without conclusively proving it. Another famous example is the Kastle-Meyer test for blood, developed by Erich Kastle and Erich Meyer in 1903. It relies on the chemistry of hemoglobin, whose iron-bearing heme group mimics an enzyme called peroxidase; in the presence of hydrogen peroxide, it turns a colorless phenolphthalein reagent bright pink. A positive result tells an investigator to do further testing, not to draw a conclusion, and that distinction between a hint and a proof runs through the whole field.

Separating a mixture, then naming every part of it

A presumptive color change can flag the possible presence of blood or a poison, but confirming which compound is present, and how much, requires a more powerful instrument. The workhorse of the modern crime lab is gas chromatography-mass spectrometry, usually abbreviated GC-MS, which combines two complementary techniques into a single analysis.

The first half, gas chromatography, solves the problem of separation. Real evidence is almost never a pure substance; an arson debris sample, for instance, is a chaotic mixture of burned material and whatever accelerant might have been poured on it. Gas chromatography vaporizes the mixture and carries it in a stream of inert gas through a very long, thin capillary column coated on the inside with a chemical film. Different compounds cling to that coating to different degrees, so they travel through the column at different speeds and emerge one at a time, neatly sorted. The second half, mass spectrometry, solves the problem of identification. As each separated compound exits the column, it is ionized and broken into charged fragments, and the instrument measures the mass-to-charge ratio of those fragments. Every molecule shatters in a characteristic, reproducible way, producing a fragmentation pattern that functions as a molecular fingerprint and can be matched against reference libraries. Together the two stages let an analyst take a single messy sample, pull it apart compound by compound, and name each one, which is why GC-MS is the standard confirmatory technique for drugs of abuse, for arson accelerants, and for the poisons of toxicology.

The evidence that never speaks but always testifies

DNA and pure chemistry are only part of the laboratory's work. A great deal of forensic science consists of reading physical traces the perpetrator could not help leaving behind. When a bullet is fired, the spiral grooves machined into a gun barrel, called rifling, scrape fine parallel scratches, or striations, into the soft metal of the bullet. Those striations are effectively unique to one barrel, and the practice of comparing them side by side under a microscope was systematized by Calvin Goddard, who established a Bureau of Forensic Ballistics and refined bullet-comparison microscopy in 1925. Questioned-document examination applies similar reasoning to paper and ink, analyzing the chemistry of a disputed signature, check, or note to determine whether it is genuine, altered, or forged.

A particularly elegant example is gunshot residue. When a firearm's primer ignites, it sprays out microscopic particles whose composition reflects the primer's chemistry, classically a fused combination of lead, antimony, and barium. The modern way to find them is scanning electron microscopy with energy-dispersive X-ray spectroscopy, abbreviated SEM-EDX. The electron microscope locates particles far too small to see by eye and reveals their characteristic rounded, melted shape, while the X-ray detector reads the elements inside each one, confirming the lead-antimony-barium signature that marks them as residue rather than ordinary dust. Different forms of evidence, then, call for different instruments matched to different analytes: capillary electrophoresis for DNA sequence variation, GC-MS for volatile organic compounds, inductively coupled plasma-mass spectrometry for trace metals, and microscopy with elemental imaging for particles. Each technique has its target, its machine, and its limit of detection.

A hundred and fifty years from arsenic to the genome

Stand back from the individual methods and a long arc comes into view. Forensic chemistry stretches across roughly a century and a half, from Marsh's arsenic test of 1836, through the Kastle-Meyer blood test of 1903, Goddard's bullet-comparison microscopy of 1925, and the founding of the FBI Crime Laboratory in 1932, to Jeffreys's DNA fingerprinting in 1984, the Pitchfork conviction of 1988, and the standardization of the CODIS STR panel through the 2000s. Each step added not just a new tool but a higher standard of certainty.

The growth of that certainty cuts both ways, and the field has had to confront its own fallibility. In 1992, the attorneys Barry Scheck and Peter Neufeld founded the Innocence Project at the Cardozo School of Law, using the very same DNA chemistry to reexamine old convictions; the organization has since used post-conviction DNA testing to exonerate more than 250 wrongfully convicted people in the United States. A few years later, the 1995 trial of O. J. Simpson in Los Angeles brought DNA evidence into the center of American public awareness, and the defense did not attack the chemistry of DNA typing itself. Instead it attacked how the evidence had been collected, stored, and handled, which underlined a lesson the discipline has had to learn repeatedly. The chemistry can be sound, but a result is only as trustworthy as the chain of human handling that surrounds it, from the crime scene to the courtroom.

Key Takeaways

Forensic chemistry is the application of analytical chemistry to legal questions, and its history runs roughly 150 years from James Marsh's 1836 arsenic test, which deposited a metallic mirror of reduced arsenic to convict poisoners, to Alec Jeffreys's discovery of DNA fingerprinting at the University of Leicester on 10 September 1984 and the modern twenty-locus CODIS profile, whose random-match probability is better than one in a quintillion because it counts the variable repeats at standardized short tandem repeat loci amplified by PCR and sorted by capillary electrophoresis. The 1986 to 1988 Pitchfork case produced both the first criminal conviction and the first pre-conviction exoneration by DNA, a pairing that captures the field's dual power to convict and to clear. The discipline distinguishes quick presumptive tests, such as the Kastle-Meyer phenolphthalein test for blood, from confirmatory instruments, above all GC-MS, which separates a mixture by gas chromatography and then identifies each component by its mass-spectral fingerprint, while specialized methods such as SEM-EDX read the lead-antimony-barium particles of gunshot residue. Running through all of it is one principle, reinforced by the O. J. Simpson trial and the Innocence Project, founded in 1992 and responsible for more than 250 exonerations: the strength of an analysis depends not only on the chemistry but on the integrity of the evidence and the people who handle it.