UDCA Identification Tests: What IR, TLC and Colour Prove

An identification block looks like a formality. It confirms that the material is what the label says, which sounds like the one thing a certificate should not need to argue about. The block nevertheless occupies three lines on a bile-acid certificate, and buyers regularly ask whether all three are necessary. They are, and the reason is that each of the three is blind to something the other two can see.

This note sets out what each of the three tests actually establishes, where the blind spots are, and how to read the block when one of the three behaves unexpectedly.

Identification is a set of independent probes, not one test

A single measurement can only establish as much as the property it measures. Identity is a claim about a molecular structure, and no single routine test covers a whole structure. Infrared spectroscopy reports on the vibrational fingerprint of the molecule; thin-layer chromatography reports on how the substance distributes between a stationary and a mobile phase; a colour reaction reports on the presence of a particular reactive arrangement of atoms. Three different properties, three different failure modes.

The practical consequence is that the three tests fail in unrelated ways. A mixture of the expected substance with a close structural relative will pass the colour reaction easily, because the reactive group is shared. That same mixture is what thin-layer chromatography is designed to reveal. Meanwhile a solvent or a salt that shifts the infrared spectrum slightly will not trouble the colour reaction at all. Overlapping the three is what turns three weak claims into one reasonably strong one.

Infrared comparison against a reference standard

The infrared line is a comparison, not an absolute measurement, and that distinction carries most of the practical consequences. The spectrum of the sample is compared with the spectrum of a reference standard recorded under the same conditions. Agreement across the fingerprint region is a strong statement that the two solids contain the same functional groups arranged in the same way.

Three conditions have to hold for the comparison to mean anything. The sample and the standard must be prepared in the same way, because potassium bromide disc spectra and attenuated total reflectance spectra are not interchangeable. The reference standard must be within its validity period and traceable, which is why reference standards for bile-acid testing are treated as controlled material rather than as consumables. And the comparison must be made over the region where the two spectra actually differ, not only where they agree.

The blind spot is small amounts of a closely related substance. A few per cent of a structural relative contributes its own bands, but they overlap with the bands of the main substance closely enough that a visual comparison can still be judged as conforming. Infrared is therefore excellent at excluding the wrong substance and poor at detecting contamination by a near neighbour.

Thin-layer chromatography, and the plate that answers twice

The thin-layer line in this method is shared with the determination of the named impurity. The same development produces a principal spot that serves as the identification reference and a limit evaluation for the impurity being controlled. Two results, one experiment, which is part of why the block is shorter than it looks.

What the thin-layer step adds is separation. Where infrared compares two spectra as wholes, chromatography separates the components of the sample and shows how many of them there are. That is the property infrared cannot provide, and it is why the two sit together. The limit associated with the named impurity is evaluated by comparing the size of the secondary spot against a reference spot of known concentration, which makes the step semi-quantitative rather than purely descriptive.

Two practical points matter here. The plate has to be developed and evaluated under the conditions the method specifies, because the position of a spot is a function of the mobile phase, the saturation of the chamber and the temperature. And the result is read against a reference spot, so the reliability of the limit depends on how that spot was prepared.

The colour reaction, and what a family check is worth

The third line records a colour reaction with sulphuric acid and formaldehyde, which produces a characteristic coloured suspension in this family of compounds. It is a low-technology test with a specific purpose: it confirms that the material belongs to the expected structural class rather than merely resembling it on paper.

Its value is precisely that it is unrelated to the other two. Infrared can be argued about at the margin, and chromatography depends on the behaviour of the substance in a particular system. A chemical colour reaction depends on a structural feature reacting in a way that is visible to the eye. When all three agree, the agreement is worth more than the sum of the individual statements, because the three measurements do not share a failure mode.

The three probes side by side

Test	What it establishes	What it is blind to
Infrared against a standard	The functional-group fingerprint matches the reference	Small amounts of a close structural relative
Thin-layer chromatography	How many components are present, and the limit for the named impurity	Compounds that do not respond in the detection system
Colour reaction	Membership of the expected structural class	Everything quantitative

The table also explains why the block is not reducible. Removing the chromatographic step would leave no separation information. Removing the colour reaction would leave the block dependent on two measurements that share an instrumental character. Removing the infrared comparison would leave the block unable to exclude a different substance that happens to chromatograph similarly.

How the block fails, and what to ask when it does

Identification results rarely fail outright. They more often become ambiguous, and ambiguity has its own causes. A reference standard past its validity period, a plate developed in an over-saturated chamber, a disc prepared with too much sample so that the spectrum is saturated across the fingerprint region, or a colour reaction read after the suspension has settled. Each of these produces a result that is neither a pass nor a clean fail.

When a result arrives in that state, the useful questions are narrow. Which reference standard was used, and was it valid on the day. What were the preparation conditions. And, if the infrared comparison was the ambiguous one, whether a re-recorded spectrum agrees across the region where the two spectra differ. A conforming report that cannot answer those three questions is worth re-running before it is worth filing.

What the block does not tell you

Identification is not purity, and the distinction gets blurred more often than any other point in this note. A certificate can show a completely conforming identification block alongside an impurity result that fails, because the two lines measure different things. A batch can also pass identification while carrying a fraction of amorphous material, retained solvent or an unexpected particle-size distribution, none of which the identification block is designed to detect.

For the wider picture of what the impurity lines control, the EP impurities reference sets out the named substances and their limits. Read the identification block as the answer to one question only: is this the expected substance. Everything else on the certificate answers a different one. Buyers who keep those questions apart find that most apparent certificate contradictions dissolve; buyers who merge them spend their time chasing discrepancies that were never there.

What a second-source comparison looks like

Buyers who qualify more than one source for the same material tend to compare certificates line by line, and the identification block is where that comparison is least informative and most often attempted. Two laboratories can both pass, and their reports can look different, for reasons that have nothing to do with the material.

The first source of difference is the reference standard. Infrared comparison depends on the standard used, and two suppliers may compare against different standards with different histories. The spectra will agree with each standard and may differ slightly from each other, which produces an apparent discrepancy that is an artefact of the anchors rather than of the samples.

The second is the presentation of the chromatographic step. One certificate may report the identification spot and the impurity evaluation as separate lines, another may report them as one, and the underlying work can be identical. The third is the colour reaction, where the description of the observed result varies between laboratories even when the observation is the same.

The practical way to compare two sources is to compare the method references rather than the result descriptions. Where the methods match, the results are comparable; where they do not, the certificates are describing two different experiments that happen to share a heading.

Documentation that should sit behind the block

A conforming identification block on a certificate is a summary. The evidence behind it is the set of records that produced it, and a buyer who may need to defend the material in an audit should know what those records are.

At minimum they are: the reference standard's identity and validity period, the spectrum or chromatogram as recorded, the plate or system suitability records for the chromatographic step, the preparation details for the colour reaction, and the analyst's and reviewer's signatures. None of that appears on a certificate, and all of it is what a laboratory would produce if asked why a result is what it is.

The ICH quality guidelines frame the expectation at the level of the test rather than the record, which is why the record set is a matter for the laboratory's own system rather than for a prescription. The useful question for a buyer is not whether the records exist, since they must, but whether the laboratory can produce them on request within a reasonable time. A laboratory that cannot answer that question in principle will not answer it in practice.

How the three tests sit in the wider framework

Identification tests are not chosen freely. The general chapters of the European Pharmacopoeia and of the United States Pharmacopeia describe the techniques and the acceptance conventions, and the monograph for a material selects the combination that is appropriate for it. The selection is a judgement about which properties distinguish the substance from its likely companions, which is why different materials in the same family carry different identification blocks.

Where a material does not have a monograph of its own — a situation that applies to some members of this family — the combination has to be justified by the specification rather than inherited from a published text. In that case the reasoning behind the choice becomes part of the technical file, and a buyer is entitled to ask what the three tests are each there to exclude.

Where the identification block sits in your own release testing

A buyer repeating the identification block on receipt has three choices about scope. The full block can be repeated, which is thorough and rarely necessary. The chromatographic step can be repeated alone, which is cheap and covers the greatest number of failure modes. Or the block can be accepted on the certificate and re-tested only on the first batch from a new lot of material, which is the usual compromise.

Whichever scope is chosen, the specification for the material itself should be the reference for what the block contains. The product page for ursodeoxycholic acid sets out the specification as we release against it, and the identification items there are the ones a buyer will be comparing when they design an incoming test.

It is also worth deciding in advance what a failed identification means operationally. A failure means the material is not what it was ordered to be, which is a different situation from a purity shortfall and calls for a different response.

Frequently asked questions

Why are three identification tests needed instead of one?

Because each test is blind to something the others detect. Infrared spectroscopy compares the whole molecular fingerprint but is insensitive to small amounts of a closely related sterol. Thin-layer chromatography separates and shows what is present. The colour reaction is a family-level chemical probe. Together they overlap in a way that a single test cannot.

What does the colour reaction with sulphuric acid and formaldehyde indicate?

It is a group reaction used in this family of compounds to confirm that the substance under test belongs to the expected structural class, producing the characteristic colour and suspension described in the monograph. It is a family check, not a purity check.

Can the thin-layer step and the impurity-C test share one plate?

In the method used here they do, which is one of the reasons the identification block is efficient. The same chromatographic development yields both the identification spot and the limit evaluation for the named impurity, so the two results are read from one experiment.

What happens if a reference standard is past its expiry?

The comparison loses its anchor. Infrared comparison against a standard is only meaningful while the standard is valid and its spectrum is traceable, so we treat an expired standard as a failed test rather than as an approximate one.

Is a passing identification block evidence of purity?

No. Identification establishes that the material is the expected substance. Purity comes from the impurity panel, the residual-solvent panel and the content determination. Confusing the two is one of the most common reading errors on a certificate.

How often should identification be repeated on a stored batch?

Identification is normally performed at release. Where a batch is held beyond its retest period, the retest programme decides what is repeated; in most cases identity is the least likely thing to have changed and the content and impurity tests carry the weight.