Loss on Drying vs Water Content: Why UDCA Shows Both

Two numbers on a UDCA certificate look like the same number. Loss on drying says not more than 0.5 per cent. Water content says 0.2 per cent. A buyer reads them as two views of the same thing, and puts the certificate down. They are not the same thing, and the difference has a practical consequence: you cannot substitute one for the other, and you cannot compare one supplier's water content figure against another supplier's loss on drying figure.

The short version is this. Loss on drying weighs what leaves the sample when you heat it under a defined set of conditions. Water content measures water, and only water. Everything that evaporates at 105 degrees counts in the first number and not in the second.

What each method actually removes

The two tests answer different questions, and the gap between them is not an error. It is information.

Test	What is measured	Typical pharmacopoeial conditions	What the gap tells you
Loss on drying	Everything volatile under the stated temperature and time: water, residual solvents, and any other volatile matter	Dried at 105 °C for 3 hours, weighed before and after	A large gap against water content points to residual solvent
Water content	Water only, by a titrimetric method that reacts specifically with water	Karl Fischer titration, volumetric or coulometric	A small gap means the volatiles are essentially all water

Two consequences follow from that table, and both of them matter at goods-in.

The first is that loss on drying is an upper bound on water. It can be higher than the true water content because something else evaporated too. It cannot be lower. If a certificate shows loss on drying of 0.4 per cent and water content of 0.5 per cent, one of the two figures is wrong, and that is worth a question.

The second is that residual solvent becomes visible in this comparison before it becomes visible anywhere else. Our note on residual solvents under ICH Q3C covers the limits themselves, which are set out in the ICH Q3C guideline; the moisture pair is where you first see the solvent's shadow, because a clean drying-loss figure with a matching water figure leaves very little room for solvent to hide.

Why the two pharmacopoeias print different limits

The same substance, tested against the European Pharmacopoeia and against the USP, can carry different drying-loss limits. The USP Ursodiol monograph specifies drying at 105 degrees for three hours with not more than 0.5 per cent loss. Against the European monograph limit of 1.0 per cent for the same drying-loss test, that is the tighter figure. That divergence is not an inconsistency in the science. It reflects two different judgements about what a drying step should be allowed to leave behind, arrived at independently.

For a buyer, the practical rule is simple: quote the pharmacopoeia you are releasing against, and make the supplier state the same one back. A supplier who answers "both" without saying which figure belongs to which text has not answered the question.

Where the water actually is

Bile acids are not strongly hygroscopic, and UDCA in its usual crystalline form takes up little moisture in ordinary storage. The water that appears on a certificate comes from four places, and knowing which one is in play tells you whether the number is a process observation or a storage one.

1. Retained solvent from the final crystallisation or washing step. This is the most common origin, and it is why the drying-loss figure and the residual-solvent profile should be read together rather than separately.
2. Surface moisture taken up during milling, sieving or packaging in an environment with uncontrolled humidity. Particle-size work is the usual moment this happens, and it is one reason a change in milling conditions can move the drying-loss number.
3. Water bound in the crystal lattice where a hydrate or partial hydrate forms. This fraction behaves differently under drying, and the standard conditions may not remove all of it.
4. Moisture picked up in the container between release and arrival, if the packaging chain is not tight. This is the one that makes a certificate technically true and a sample out of specification, and it is why container and storage conditions belong in the conversation rather than in the small print.

Two different chapters, two different questions

Loss on drying and water determination sit in different parts of the pharmacopoeia, and that placement is not an accident of editing. Loss on drying is a gravimetric method: the sample is weighed, dried under a stated temperature and time, and weighed again, and the result is the difference. Water determination is a chemical method built on a reaction that is specific to water.

The Karl Fischer reaction consumes water stoichiometrically, one mole of water for one mole of iodine, and a coulometric cell measures the quantity of electricity needed to keep that reaction supplied with iodine. That is why the two methods behave so differently at the bench. Gravimetry costs almost nothing and answers a question about the whole volatile fraction. Karl Fischer needs a dedicated reagent system, has to be kept dry, and answers a question about one molecule. A coulometric cell can reach water at the microgram level on a sample weighed in milligrams, which is why the technique is specified once in the general chapter rather than described inside each monograph.

Two practical consequences follow. The first is that a water figure and a drying figure can be compared only if the same test was run at both ends. The second is that the sample size is part of the result. A drying-loss figure determined on 1.000 g and a water figure determined on 200 mg are not measurements of the same thing to the same precision, and treating them as interchangeable because both carry one decimal place is a mistake that surfaces in a specification argument rather than in a laboratory.

Why moisture is not a cosmetic number

Three parameters on the same certificate are calculated on the dried basis, and each of them moves when the moisture figure moves.

• Assay. A content figure expressed on the dried basis is corrected by the drying-loss result. Get the moisture wrong and the assay shifts with it.
• Specific optical rotation. The rotation test is also reported on the dried basis. Our walkthrough of the +58.0 to +62.0 degree test sets out where the correction enters the calculation.
• Related substances. Impurity percentages are conventionally reported on the same basis, so a material sitting near an impurity limit can be pushed over it or pulled back under it by a change in moisture alone.

This is why a supplier's internal shift from a drying-loss test to a Karl Fischer water test is not a neutral analytical choice. It changes the basis of three other calculations on the same certificate. If the substitution is made without saying so, the figures remain internally consistent and become externally incomparable, which is the worst of both outcomes.

What to ask, and how to read the answer

1. Ask which test the certificate reports, by name. Loss on drying and water content are two different lines. If only one appears, ask whether the other was run.
2. Ask for the drying conditions if loss on drying is reported. Temperature and time are part of the result. A figure obtained at 60 degrees is not comparable with one obtained at 105 degrees.
3. Compare the two figures when both are present. Water content above loss on drying is impossible. A large gap means solvent or another volatile, and that should trigger the residual-solvent discussion.
4. Confirm the basis of the assay and rotation calculations. Both should state whether they are on the dried basis, and the drying result used should be traceable to the same batch.
5. Check the container and the closure. A double polyethylene liner inside a fibre drum is a moisture barrier as much as a contamination barrier. If the packaging changed, the drying-loss history changed with it.

Read the moisture pair on a certificate as a pair. The absolute numbers matter less than whether they agree with each other, whether the test names match the pharmacopoeia you are releasing against, and whether the basis of the other calculations is stated. That is a five-minute check, and it prevents the most common argument at goods-in, which is not about whether the material passed. It is about which test the passing figure came from.

Frequently asked questions

Is loss on drying the same as water content?

No. Loss on drying weighs everything volatile that leaves the sample under the stated temperature and time, which includes water and residual solvents. Water content, usually by Karl Fischer titration, measures water specifically. Loss on drying is therefore always greater than or equal to the true water content, never less.

Why is the USP limit lower than the EP limit?

The two pharmacopoeias were written independently and reached different judgements about how much volatile matter a drying step should be allowed to leave behind. For ursodiol the USP monograph specifies drying at 105 degrees for three hours with not more than 0.5 per cent loss, which is tighter than the European limit of 1.0 per cent for the same test. Neither is wrong; they are different texts.

Can I substitute a Karl Fischer result for a loss on drying test?

Not as a like-for-like replacement, because the two measure different fractions of the sample, and because the drying-loss result is used as the correction basis for the assay and the optical rotation. A substitution can be defensible if it is declared and if the receiving specification is written for the water test, but it must be a documented decision on both sides, not a quiet change of method.

Where does the water come from in a bile acid?

Most often from retained solvent in the final crystallisation or washing step, and from surface moisture picked up during milling, sieving or packing in humid conditions. A smaller fraction can be lattice-bound in a hydrate, and a further fraction can be picked up in the container between release and arrival if the packaging chain is not tight.

Does moisture affect the assay result?

Yes. Assay, specific optical rotation and the related-substances percentages are all conventionally calculated on the dried basis on this material, so the drying-loss result feeds directly into all three. A change in moisture moves three other numbers on the same certificate.

What should I do if the certificate shows only one of the two?

Ask which test was performed and whether the pharmacopoeial test was among them. If the certificate reports only a water content figure while your release specification is written against a loss on drying limit, you are holding two documents that do not speak to each other, and that gap needs closing before the batch is accepted.

Where this leaves you

The moisture section of a certificate is small and easy to skip. It is also the section that quietly governs three other figures on the same page. Read the test name before the number, check that the two moisture figures agree with each other, and make sure the drying basis behind the assay is the same basis you are releasing against.

What we ship against, and the tests each lot is released on, is set out on our ursodeoxycholic acid page. If you want the moisture and drying-loss lines reviewed against your own specification before an order, that can be arranged through a quotation request. For the wider question of what each certificate line proves, see our breakdown of the certificate of analysis.

This article describes analytical methods and pharmacopoeial testing for quality and procurement professionals. It states no therapeutic use, dose or indication for any substance, and no order quantity or lead time. Always work from the pharmacopoeial edition in force at the time of your release.