Trace evidence is a broad term meaning any small piece of physical evidence that links a suspect to a crime scene. It's mostly circumstantial class evidence, but depending upon the expertise and rules of admissibility, it can match individual characteristics. Fingerprints, toolmarks, and bitemarks (all forms of trace evidence), for example, use a points of comparison approach.
QDE and ballistics (also trace evidence areas) rely upon the "sufficiently certain" opinion of the expert. Other trace evidence fields (like toxicology, serology, and DNA) use probability estimates or odds-ratios. Microtrace evidence is a subtype of trace evidence involving minute, microscopic particles. The science of analyzing, identifying, and comparing microtrace evidence is called microanalysis.
It should be distinguished from the science of microscopy because the word micro in microanalysis refers to the size of the evidence, not the use of a microscope. Many other types of instrumentation are used by forensic microanalysts besides microscopes. Items traditionally examined include (and by no means are limited to): hair, fibers, paint, glass, dust, wood, soil, minerals, drugs, metals, perfumes, pollens, dyes, pigments, polymers, other materials, and atoms.
It's not uncommon to see the name of the field shortened to hairs, fiber, paint, and glass. It's difficult to gain a perspective on this field. It's one of the areas where there's the most confusion about the word "criminalist" which technically refers to laboratory personnel and "criminologist" which technically refers to an academic scientist.
To make matters worse, sometimes police crime scene technicians are referred to as "criminalists" when all they do is mainly "bag and tag" stuff that may contain microtrace evidence. In all fairness, however, police should be well-versed and better equipped in this area, and the phrase "crime scene analyst" is perhaps an improvement.
Forensic microanalysis relies upon instrumentation, which is contained in standards, operating procedures, test protocols, machine tolerances, owner manuals, dictionaries, etc., and written for industry and commercial (not law enforcement) applications. They're fancy machine operators (sorry for the rather unflattering description), but they get to work with all kinds of microscopes, spectrographometers, chromographometers, and even nuclear reactors.
They really should have Master's or Doctor's degrees (to understand why the machines work, engage in original research, and develop non-commercial applications), but unfortunately, only a handful do. In all fairness, it's not their fault for many reasons:
NLECTC, for example, has taken a lead in steering commercialization toward criminal justice, as has the FBI and SEARCH via numerous symposia. There are, however, some good constitutional reasons for moving slowly. As far as expert qualifications go, forensic law seems to be moving in the direction of requiring at least a Bachelor's of Science degree and at least a year of supervised work experience.
These are minimum standards, as there are many places where a Master's degree might be required, or years of experience, or some combination thereof, with education/experience substituting for one another to qualify as an expert witness. With testimony, courts are generally reluctant to move beyond the age-old standard of similarity "in every microscopic detail" because to say more would probably dazzle the less scientific legal system.
On the other hand, forensic microanalysis is ideal for operating under the Daubert standard since along with instrumentation and machine diagnostics, it's usually easy to calculate things like Type I and II error.
Microanalysts, for a while at least, followed Gaudette's lead with testimony in phrases like 1:4500 and 1:800 odds-ratios for scalp and pubic hair. Others didn't follow this lead. A series of famous cases drew attention to the problems with probability estimates in cases of "convicted by a hair" or "fiber".
So, most microanalysts have dropped the whole probability estimate issue, except in cases where a database index exists (like with the FBI's paint file) and when there's a good chance that there are some known parameters from the population sample statistics are drawn from.
Almost all microanalysts work in crime labs, and when microtrace evidence of any kind is sent for analysis, the job is either marked for identification, comparison, or both. Identification is the process of using analytical chemistry techniques (over 30 different techniques exist, from simple titration to nuclear magnetic resonance) to identify a substance with absolute or near-absolute certainty.
It's up to each analyst in deciding the number and type of tests to run, and identification tests must be comprehensive to exclude all other substances that the sample might be. Absolutely positive identification is often achieved after five or six tests. Comparison requires that whomever sent the sample also send a control specimen, and the purpose of testing is to compare various properties of the substances to see if they have a common origin or source.
Comparison techniques consist of physical science tests (about 10 different tests exist, from determination of mass and density to refractive index) that identify the "properties" of a substance. Physical "properties" (like weight) describe a substance without reference to any other substance, leaving the sample's composition unchanged. Chemical "properties" (like color change) describe the behavior of a substance when it reacts or combines with another substance, usually called a reagent or solvent.
Comparison test results are expressed in terms of probability estimates. The smaller the probability value, the more likely absolute or near-absolute certainty is achieved. An analyst should know about how frequently "matches" occur in nature, in the crime scene environment, or among the population of people involved.
Let's look first at some laboratory comparison tests for physical properties: Melting point - every known substance has a fixed melting point Mass - the weight of a substance after accounting for the effects of gravity Density - the mass of a substance after accounting for volume or size Refractive index - the bending of a lightwave by a substance Color - several substances have all their known colors recorded These tests are easily carried out and can fairly quickly determine a "match" or not.
It depends, however, upon the substance. You obviously don't want to use melting point unless you've got ample quantities, but it's good for dust, wood, soil, minerals, metals, and other materials. Freezing point is always an alternative, as that tends to also be fixed. Mass is good for unknown substances. Density and refractive index tend to be good for soil and glass particles.
With either substance, the sample and specimen are floated or immersed in side-by-side test tubes containing various liquids, and "matches" float to the same line on each tube (the density-gradient technique). Soil, paint, and many other materials have all their known colors recorded. From geology, for example, we know that there are 2200 minerals, 1100 soil colors, and only 40 common rocks. It has then been mathematically determined that there is only a 1:50 chance of finding two different soils 1000 feet apart.
Laboratory analytical techniques for identification:
ELECTROPHORESIS involves putting substances in a gel or liquid, applying a positive and negative charge to each end, and forcing components in the substance to migrate across the gel. Among other things, it's a quick and easy way to separate organic from inorganic matter because all proteins carry an electrical charge (and migrate at known speeds). It's commonly used in blood and DNA samples.
SPECTROPHOTOMETRY is based on the wave (nonquantum) theory of light and the principle that substances selectively absorb certain frequencies (wavelengths) and reflect or transmit others. Light that goes through a substance will have a frequency roughly equivalent to the energy requirements (photon absorption capabilities) of a substance. The selective absorption of light by a substance is measured by a spectrophotometer. These machines essentially consist of a light source beamed through a prism, the sample, and then into a decoder (electromagnetic to electrical) which records output in the form of a line graph. The light source can be: ultraviolet; visible light; or infrared. Infrared (IR) is the most common, as practically all substances absorb IR very selectively. Ultraviolet (UV) and visible light are used for simple pieces of evidence, like drug samples. The IR spectra of thousands of substances have been collected, indexed, and cataloged. IR spectra are the "fingerprints" of microanalysis.
CHROMATOGRAPHY is based on Henry's Law of equilibrium that a volatile compound (in transition from solid, gas, or liquid states) will have a fixed ratio of molecules escaping and staying. All substances reach this point of equilibrium at different speeds, known as the time of elution (which accounts for the separation of component molecules). Machines exist that record these times of elution by applying a heat source (volatility) to an injection of sample through a tube or chamber filled with a carrier medium (gas or liquid), the recording being called a chromatograph. There are three different types of techniques: gas chromatography (GC); high-performance liquid chromatography (HPLC); and thin-layer chromatography (TLC). Gas (GC) is the most commonly used technique for hair and fibers, but it vaporizes or destroys samples. HPLC doesn't use as much heat, and is good for heat-sensitive materials like explosives and certain drugs. TLC uses a gel (much like electrophoresis) that is positioned vertically, colored dyes, and measures the distance a spot travels up a glass plate. MASS SPECTROMETRY (MS) is the combining or coupling of gas chromatography with a special kind of spectrophotometer that measures mass/charge. Components are first separated using chromatography and then enter a high-vacuum chamber where they are bombarded with high-speed electrons, creating an ionization effect. The ions decompose rapidly into smaller fragments which then travel through a magnetic field to have their masses weighed, controlling for ionic charge. No two substances have the exact same "fragmentation pattern" evident in the recorded spectra. Therefore, MS spectra are also considered the "fingerprints" of microanalysis.
ATOMIC ABSORPTION/EMISSION is a type of spectrometry that measures the dividing lines, or breaks, between different colors of a spectrum produced by a sample-activated light. This is called line spectrum, and the technique is based on analyzing the dark spaces between a color spectrum. Emission spectrum analysis is the most common technique. The sample is placed between two carbon electrodes and an electric current is applied. The emitted light is focused through a prism, displayed on a plate, and the lines recorded. Absorption spectrum analysis requires that the light or heat source be of the same element as the sample, so the analyst has to guess or use the technique for confirmatory purposes only. The word atomic with these techniques refers to absorption being the jumping of an electron to a higher orbit and emission being an electron falling back to a lower orbit.
NEUTRON ACTIVATION ANALYSIS is based on the principle of isotopes. An isotope of an atom is the same element with the same number of protons but different number of neutrons. Most elements have 2-10 known isotopes. Some are stable, others unstable. The unstable ones release radioactivity. Neutron activation analysis is bombarding a sample with neutrons and measuring the isotope radioactivity. Since at least the gamma radiation of every known isotope has been cataloged and indexed, it's a fairly simple matter to produce a table showing the percent concentration of every trace element based on the intensity of the different gamma radiations.
X-RAY DIFFRACTION only works with crystalline samples. Many rocks, minerals, and soils are crystals. Other evidence, such as blood, can be converted to crystal form with iodine and so forth. Crystals are composed of parallel atomic planes which deflect X-rays, and every crystalline substance produces a different X-ray diffraction pattern. This is another "fingerprint" method of microanalysis.