Polymorph and Solid-State Characterisation with XRD, DSC and TGA

Solid-state characterisation tells you what physical form an API or excipient is in: which crystalline polymorph, whether it is a hydrate or solvate, and whether any amorphous content is present. Three techniques do most of the work. X-ray powder diffraction (XRPD, often just called XRD) identifies the crystal form from its diffraction pattern. Differential scanning calorimetry (DSC) measures melting, recrystallisation and glass transitions. Thermogravimetric analysis (TGA) measures weight loss from water, solvent or decomposition. Used together, they let you identify a form, show that batches are consistent and detect conversion during processing or stability. This guide explains what each technique measures, how to interpret the results together, what regulators expect, and where solid-form studies most often go wrong.

Why solid form matters

The same molecule can pack into different crystal lattices. These polymorphs differ in solubility, dissolution rate, melting point, hygroscopicity, chemical stability, flow and compressibility. A change of form can therefore change bioavailability, cause tablets to fail dissolution or make a suspension cake. The main solid forms you will meet are:

• Polymorphs: different crystal packings of the same molecule. Usually one form is the most stable at a given temperature.
• Hydrates and solvates: crystals that include water or solvent in the lattice, in either stoichiometric or non-stoichiometric amounts.
• Amorphous forms: no long-range order. They are often more soluble, but they are physically less stable and tend to crystallise over time.
• Salts and co-crystals: multi-component solids designed to tune solubility or stability. FDA published final guidance on the regulatory classification of pharmaceutical co-crystals in 2018.

In its July 2007 guidance on solid polymorphism in ANDAs, FDA treats crystalline, amorphous, solvate and hydrate forms all as "polymorphic forms". FDA states that a difference in polymorphic form does not make a different active ingredient for ANDA purposes. A generic product can therefore use a different form from the reference product, provided it meets bioequivalence and stability requirements. This is why controlling the form you actually manufacture matters more than copying the innovator's form.

Regulatory expectations: ICH Q6A and FDA

ICH Q6A (Step 4, 6 October 1999) covers polymorphism in Decision Tree #4. In summary:

1. Drug substance: Can different polymorphs form? If so, do they have different properties, for example in solubility, stability or melting point? If those differences could affect the safety, performance or efficacy of the drug product, set an acceptance criterion for polymorph content in the drug substance. The guideline names XRPD, DSC or thermal analysis, microscopy and spectroscopy as characterisation tools.
2. Drug product: For solid dosage forms, or liquids that contain undissolved drug, ask whether a performance test such as dissolution adequately controls a change in polymorph ratio. If it does, a performance acceptance criterion may be enough.
3. Monitoring: If it does not, monitor the polymorphic form during drug product stability and set acceptance criteria if a change could affect safety or efficacy.

The 2007 FDA ANDA guidance follows similar logic. Polymorph specifications may be unnecessary when all forms have similar apparent solubility or are highly soluble under BCS criteria. Where they are needed, a USP monograph specification may be adopted if it is adequate. That guidance describes single-crystal X-ray diffraction as definitive evidence of polymorphism and XRPD as also able to give unequivocal proof. Thermal analysis, microscopy and IR, Raman or solid-state NMR spectroscopy add further characterisation.

X-ray powder diffraction: the fingerprint of the crystal form

In XRPD, a monochromatic X-ray beam strikes a powder sample. Each crystalline form diffracts at characteristic angles (2θ) that depend on its lattice spacings, so the pattern acts as a fingerprint. Our XRD page outlines the principle.

Interpreting a pattern

• Peak positions identify the form. USP <941> states that, for the same crystal form, the 2θ angles of the specimen and the reference agree within 0.2°.
• Relative intensities can vary considerably, mainly because of preferred orientation of plate- or needle-shaped crystals. Do not reject a match on intensity alone.
• A broad "halo" with no sharp peaks indicates amorphous material. Sharp peaks sitting on a halo indicate a partly crystalline sample.
• Extra peaks may come from a second polymorph, a hydrate, an excipient or a crystalline impurity. Compare against the patterns of every known form and every excipient.

Sample preparation and limits

Gentle grinding reduces preferred orientation, and USP <941> notes that in some cases particle sizes as large as 50 µm will give satisfactory results for phase identification. Excessive milling can broaden peaks, and it can also convert the form or make it amorphous, so compare a ground sample with an unground one. Detecting a minor phase is the hardest task. USP <941> describes identification as difficult below about 10% m/m in general. Lower limits are achievable only with a method developed for that purpose, using diagnostic peaks that do not overlap and spiked calibration mixtures. Never claim a detection limit you have not demonstrated in the actual matrix.

DSC: thermal events and form stability

DSC measures the heat flow into or out of a sample relative to a reference as the temperature is ramped. Typical conditions are 2–5 mg of sample in a crimped aluminium pan, heated at 10 °C/min under nitrogen, with the instrument calibrated using certified reference materials. See our DSC page for an overview of the technique.

Interpreting a thermogram

• Endotherm: melting, desolvation or a solid–solid transition. Report the extrapolated onset temperature and the enthalpy (J/g), not just the peak temperature.
• Exotherm: crystallisation of amorphous material, conversion to a more stable form, or decomposition.
• Step change in the baseline: a glass transition (T_g), which indicates amorphous content.

Relative stability of polymorphs

The Burger–Ramberger heat-of-fusion rule is a quick first assessment. If the higher-melting form has the lower heat of fusion, the pair is probably enantiotropic: the stability order reverses at a transition temperature below the melting points. If the higher-melting form also has the higher heat of fusion, the pair is probably monotropic, and one form is the more stable at all temperatures. Confirm the result with slurry competition experiments, because thermal data alone can mislead.

Pitfalls

Heating rate shifts the onset temperature and can hide or reveal transitions. A metastable form may melt, recrystallise and melt again as the stable form, which looks like two forms in the sample. Volatile loss in a sealed pan raises the internal pressure and distorts the peaks, so for solvates compare sealed pans with pin-holed pans and always read the DSC alongside the TGA.

TGA: water, solvent and decomposition

TGA records the mass of a sample as it is heated. It distinguishes loosely held surface moisture, which is lost gradually from low temperature, from lattice-bound water or solvent, which is lost in a defined step, often at the same temperature as a DSC endotherm. It also shows where decomposition begins. Our TGA page describes the measurement.

Worked example: is it a hydrate?

Suppose an API has an anhydrous molecular weight of 350.4 g/mol. TGA shows a single, sharp 4.9% weight loss between 60 and 120 °C, which matches a DSC endotherm, followed by melting at a higher temperature. The theoretical water content of a monohydrate is 18.02 ÷ (350.4 + 18.02) × 100 = 4.89%, so the data fit a monohydrate. A hemihydrate would lose about 2.5%. Then confirm:

• Water content by Karl Fischer titration agrees with the TGA loss, which rules out an organic solvent.
• After drying, XRPD shows a different pattern (dehydration to an anhydrate) or a halo (collapse to amorphous material).
• If a solvate is suspected, identify the volatile by headspace GC. Our residual solvents guide covers the method.

Combining the techniques

The table below summarises what each technique contributes and where it falls short. FT-IR is a useful fourth technique, because polymorphs often show shifted bands in the fingerprint and hydrogen-bonding regions.

A practical workflow

1. Characterise a well-defined reference batch of the intended form by XRPD, DSC, TGA, FT-IR and Karl Fischer titration, and archive its data.
2. Screen for other forms by recrystallising from solvents of different polarity, by slurrying, by cooling and evaporation, and by milling. Check every solid by XRPD.
3. Assess the relative stability of the forms found, using DSC and competitive slurries.
4. Choose a release test, usually XRPD identity against the reference pattern, and justify whether a polymorph limit is needed using Decision Tree #4.
5. Check the form after granulation, compression and on stability, particularly under heat and humidity.

How SDS Labs can help

SDS Labs, an independent analytical laboratory in MIDC Mahape, Navi Mumbai, offers XRD, DSC, TGA and FT-IR analysis under ISO/IEC 17025:2017 and ISO 9001:2015 quality systems, with results generated in line with 21 CFR Part 11 and cGMP requirements. We support solid-form identification, thermal profiling and moisture or volatile assessment for APIs, intermediates and formulations.

To plan a solid-state study, share your sample details and objectives through our contact page, call +91 90295 20144 or email info@sdslabs.in.

Frequently asked questions

References

• FDA — ANDAs: Pharmaceutical Solid Polymorphism, Chemistry, Manufacturing and Controls Information (2007)
• FDA — ICH Q6A Specifications: Test Procedures and Acceptance Criteria
• EMA — ICH Q6A scientific guideline
• FDA — Regulatory Classification of Pharmaceutical Co-Crystals
• ICH — Quality guidelines