UDCA Melting Range: Why 200–204 °C Is a Narrow Window
Share
The melting range is the quietest line on a certificate of analysis. It sits near the top, it is written as an interval rather than a single figure, and nothing about it invites a second look. Buyers spend their attention on assay and on the named impurities, because that is where arguments usually happen. The ordering is backwards. On a bile-acid solid, the melting range is the line that tells you whether the material in the drum is the same solid the specification was written around.
This note covers what the line measures, why the window is deliberately narrow rather than generous, what a result sitting at the top of that window does and does not mean, and the three questions worth sending when a number lands at an edge.
What the melting range measures, and what it does not
A melting range is a lattice measurement. When a crystalline solid is heated, the ordered packing of the molecules collapses over a temperature interval rather than at a single point. A well-formed lattice of a single molecular species gives a narrow interval; a disturbed lattice gives a wider one, and the widening usually happens at the upper end first.
Two consequences follow. The interval is an identity check with a purity signal attached, because a narrow interval close to the expected temperature is consistent with a well-formed lattice of the expected molecule. And it is not a substitute for content or impurity testing. A batch can melt inside the correct window and still carry a related substance that the melting range never notices.
The measurement is also instrument-dependent in a way that buyers rarely account for. A capillary apparatus read visually reports the interval between the first sign of collapse and the disappearance of the last crystal; a differential scanning calorimeter reports an onset and a peak, which are different numbers derived from the same event. Comparing a calorimeter peak against a capillary interval will produce an apparent discrepancy that has nothing to do with the material. This is why we report the instrument alongside the interval, and why an incoming laboratory should be told which method the released number came from before it repeats the test.
Temperature calibration matters for the same reason. A capillary apparatus is normally calibrated against a small set of reference substances with known melting points, and the residual bias of that calibration can be half a degree in either direction. Across a four-degree window, half a degree is not a rounding detail. If the interval reported by a releasing laboratory and the interval reported by an incoming laboratory differ by one degree, the first question is not which laboratory is right but whether the two instruments were calibrated against the same references.
Why a four-degree window is narrow, not generous
Our specification sets the window at 200–204 °C. Four degrees sounds like a wide allowance until it is compared with what a disturbed lattice actually does. A partially amorphous fraction, a solvate that has not fully converted, or a mixture of two crystalline habits tends to spread the interval well past four degrees, and the upper end moves before the lower one does. The window is therefore not a tolerance a drifting process can hide inside. It is tight enough that a genuine solid-state problem shows up as a number outside it.
There is a second reason the window is drawn tightly. The melting range is one of the few identity-adjacent properties that a buyer can check cheaply and independently. Content determination needs a reference standard and either a chromatographic system or a carefully standardised titration; optical rotation needs a polarimeter with a controlled cell temperature. A melting apparatus needs a few milligrams and a calibrated thermometer. A narrow window makes that cheap check meaningful, which is worth more than a wide window that never fails.
Against that, the specific optical rotation answers a different question. Rotation reports on the stereochemistry of the molecule in solution; the melting range reports on how the molecules pack in the solid. A batch can hold the correct rotation and still show a broadened melting interval, and the reverse combination also occurs. Reading the two together separates a chemical problem from a physical one, which is why we keep both on the certificate rather than one.
Reading a result that sits at the top of the window
Suppose a batch reports at the upper end of the window. The instinct is to treat that as a pass and move on. The more useful question is which end moved, and whether the movement is a property of this batch or a drift across several. The table below sets out the patterns we see and what each one is usually consistent with.
| Result pattern | Usually consistent with | Whether to ask |
|---|---|---|
| Inside the window, upper half | A well-formed lattice and a thorough crystallisation | Record the trend; no single-batch question |
| Inside the window, lower half | A slightly relaxed lattice; often accompanied by marginally higher loss on drying | Read together with loss on drying and residual solvents |
| Narrow but shifted upward | A higher-melting component or a mixed crystalline habit | Ask for solid-state data |
| Interval broadened at both ends | A partially amorphous fraction or retained solvent | Ask, and consider a heavier first-article check |
| Two events or a shoulder | Two habits, or material from more than one crystallisation | Ask, and request the retained sample |
Notice how few of these rows end in rejection. The melting range is a signal generator, not a verdict machine. Its value lies in the questions it triggers and in the trend it builds when a buyer keeps the numbers from successive batches side by side. A single interval at the upper end of the window means little. Four consecutive intervals creeping upward means something is changing.
What the melting range cannot see
Knowing the limits of the measurement is as useful as knowing the measurement. Four blind spots matter in this trade.
The amorphous fraction is the largest of them. There is no routine release test for it, on any certificate, from any specification. A batch can be entirely conforming on chemistry and still carry a surface amorphous layer introduced by milling. Crystal form and crystallinity covers this in detail; the practical point here is that the melting range is a coarse indicator of it and nothing more.
Particle size is the second. The melting range says nothing about how the material was reduced, and two batches with the same interval can behave differently under compaction because the particles are different sizes and shapes. That belongs to the particle-size specification, not to this line.
The third is low-level volatile content. A small amount of residual solvent can sit in the lattice without moving the interval beyond the window, even though it will be visible on the loss-on-drying line and on the residual-solvent panel.
The fourth is the distinction between a genuine solid-state change and an artefact of the measurement. Over-fast heating pushes every interval upward and narrows it, because the sample does not have time to equilibrate. A result that looks unusually clean is sometimes just a fast ramp.
Three questions to send when a number sits at an edge
Suppose a result arrives at the boundary of the window, or the interval widens. Three questions resolve most cases without a re-test, and they are worth sending as a set rather than one at a time.
First, what instrument and what heating rate produced the number. This establishes whether the figure is comparable with the previous batch at all, and it is the single most common explanation for an apparent shift.
Second, what the previous five batches read. A single value at an edge is noise until it is placed in a series. When the series is flat and one point moves, the point is investigated; when the series drifts, the process is investigated.
Third, whether a solid-state measurement exists for that batch. X-ray powder diffraction and differential scanning calorimetry both answer the crystallinity question directly, and either can be requested as a first-article measurement where the history justifies it.
How we report it, and why
We report the full interval rather than a single figure, we state the instrument and the heating rate next to it, and we keep the raw trace with the batch record. A buyer comparing our certificate against EP versus USP limits can then work from the same numbers we worked from rather than from a summary that has already dropped the shape of the curve. Where a result sits at an edge of the window, we say so on the certificate and attach the trace, because an edge result that is explained is worth more to a reviewer than a clean result that happens to be unexplained.
Running the test in your own laboratory
If you repeat the melting range on receipt, three preparations decide whether your number will be comparable with ours. The apparatus has to be calibrated against reference substances on the day, because a capillary apparatus drifts and an uncalibrated instrument can sit half a degree away from the true value without anyone noticing. The sample has to be representative of the drum rather than of the scoop, which means taking a portion from more than one position where the container has been in transit. And the heating rate has to be the same as the one stated on the certificate, because a faster ramp narrows the interval and pushes it upward.
The conventions for reading the end-point are set out in the general chapters of the major pharmacopoeias. The European Pharmacopoeia and the United States Pharmacopeia both describe the apparatus and the closing temperature in similar terms, but the wording has changed between editions and the British Pharmacopoeia follows its own drafting history. Where two laboratories each work to a different general chapter, a difference of a degree between their results can be a convention rather than a material difference, and it is worth establishing which chapter each laboratory used before the results are compared.
Where a difference does appear, the cheapest way to resolve it is a split sample. The same portion is divided, one part measured in each laboratory, and the two intervals compared. That test removes every variable except the laboratory, and it settles in an afternoon what a correspondence can fail to settle in a month.
Where the interval belongs in a specification
The melting range belongs in a specification as an identity-adjacent check, not as a purity control. Writing it as an identity item keeps the limit tight and the purpose clear; writing it as a purity item invites the reader to draw conclusions the measurement cannot support.
Two practices make the line more useful than it looks. The first is to state the instrument alongside the window, so that a reader knows what kind of number to expect. The second is to keep the interval, rather than only the end-point, in the batch record. An interval of 202–203 °C and an interval of 200–204 °C both sit inside a four-degree window, and the two say different things about the lattice. Only one of them can be recovered from a record that stores a single figure.
A specification also benefits from saying what the interval is not. A line on a certificate that silently doubles as a purity claim will eventually be read as one, and the resulting argument is harder to settle than the measurement itself.
One habit separates buyers who use this line well from those who ignore it: keep the interval, not just the end-point, in the incoming records. A file of intervals shows when a lattice started to relax; a file of end-points shows only that nothing crossed a line.
Frequently asked questions
Why is the melting range written as an interval instead of a single temperature?
Because a crystalline solid does not melt at one instant. The ordered lattice breaks down over a temperature interval, and the width of that interval carries information. Reporting a single figure would throw away the part of the measurement that shows whether the lattice is well formed.
Can a batch with a correct melting range still have a purity problem?
Yes. The melting range is a lattice measurement. Small amounts of a closely related sterol can sit in the same lattice without disturbing it enough to move the interval outside the window. That is why the range never replaces the impurity panel; it complements it.
What does a result at the top of the window mean?
Usually nothing on its own. A result at the upper end is consistent with a well-formed lattice and a thorough crystallisation. It becomes worth a question only when the top end moves while the bottom end stays, or when the same material previously read lower.
Is the melting range a pharmacopoeial test for UDCA?
It appears in the monographs as an identity-related check, but the exact limits differ between monographs and editions. We therefore state the window our specification sets, and we keep the raw trace so that a buyer comparing against another edition can do so on the numbers rather than on a summary.
Does milling change the melting range?
Milling can, by introducing a partially amorphous fraction at the surface of the particles. A small amorphous fraction tends to broaden the interval and pull the onset down. This is one reason we ask for the particle-size history when a melting interval changes without any chemical change.
What should I request if a melting result looks unusual?
Three things: the instrument and the heating rate, the results of the previous five batches, and a solid-state measurement such as X-ray powder diffraction or differential scanning calorimetry for the batch in question. The first two usually settle the question without the third.