Insights

Characterizing Amorphous Solid Dispersions: Why XRPD Alone Isn’t Enough

A single diffractogram cannot confirm miscibility, physical stability, or manufacturability.

10 November 2026 5 min read PCL Research & Manufacturing
Characterizing Amorphous Solid Dispersions: Why XRPD Alone Isn’t Enough

An amorphous halo on an XRPD pattern is often read as confirmation that an amorphous solid dispersion is ready to move to the next development stage. In reality, X-ray powder diffraction cannot distinguish a molecularly miscible dispersion from a phase-separated blend of amorphous domains, and it cannot predict recrystallization risk on storage. Teams that stop at XRPD often discover stability or bioavailability problems only after formulation lock, when the cost of rework is highest.

What XRPD Actually Tells You

X-ray powder diffraction measures long-range periodic order. The absence of Bragg peaks indicates that no crystalline fraction is detectable above the instrument’s sensitivity threshold, which depends on crystallite size, habit, and the scattering power of the API.

An amorphous halo is consistent with the absence of detectable crystallinity. It is not proof of a single, homogeneous amorphous phase, nor does it indicate molecular-level miscibility between API and polymer carrier.

Nanocrystalline domains and low-level crystalline fractions can remain below the detection limit, particularly in polymer-dispersed systems where the crystalline signal is diluted and broadened by the amorphous polymer background. A dispersion can be XRPD-negative and still contain a crystalline seed population sufficient to drive recrystallization during storage.

Thermal Analysis: DSC and TGA Beyond the Glass Transition

Differential scanning calorimetry is the standard follow-up to XRPD. A single glass transition temperature (Tg) intermediate between the API and the polymer is commonly used as circumstantial evidence of miscibility, often benchmarked against Gordon-Taylor or Fox equation predictions.

The interpretation is not always straightforward. At low API loadings the Tg signal can be weak or broad, enthalpic relaxation can overlap with the transition, and residual solvent evaporation can mask or shift the event. Modulated DSC, separating reversing and non-reversing heat flow, is frequently necessary to resolve these overlapping thermal events.

Thermogravimetric analysis is not optional in this context. Residual solvent and moisture plasticize the dispersion, lowering the apparent Tg; without TGA data, a measured Tg may reflect a wet-state artifact rather than the true dry-state value. TGA coupled with mass spectrometry or FTIR can further identify the species responsible for a given weight loss, supporting root-cause investigation.

Moisture Sorption and Physical Stability: The Role of DVS

Amorphous solid dispersions are, by definition, in a higher free-energy state than the corresponding crystalline API. Moisture uptake plasticizes the polymer matrix, can depress the Tg below storage temperature, and accelerates recrystallization kinetics.

Dynamic vapor sorption quantifies hygroscopicity and sorption/desorption hysteresis, which is itself diagnostic of structural relaxation or incipient phase change. Combining DVS humidity cycling with post-exposure XRPD or Raman analysis identifies the critical relative humidity at which a given dispersion begins to recrystallize.

This information directly informs packaging configuration, desiccant strategy, and shelf-life claims. Without it, ICH stability studies risk producing failures with no mechanistic explanation, late in the program when corrective options are limited.

Resolving Phase Separation: Solid-State NMR and Vibrational Spectroscopy

A single Tg by DSC indicates miscibility only at the length scale that calorimetry can resolve. Finer-scale phase separation, involving drug-rich or polymer-rich nanodomains, can remain undetected by thermal analysis alone and still compromise long-term stability.

Solid-state NMR relaxation measurements (T1 and T1-rho) probe spin diffusion over short length scales and offer higher sensitivity to local heterogeneity than DSC. They are a useful orthogonal check whenever miscibility is a critical quality attribute, particularly for high-potency APIs where recrystallization of even a small fraction has disproportionate safety implications.

Raman and FTIR microspectroscopic mapping add a complementary, spatially resolved view, generating chemical images that reveal surface enrichment, phase boundaries, or early crystallization fronts before they are visible by XRPD or bulk DSC. These phase-separated regions, even when below the detection limit of other techniques, can act as nucleation sites that undermine shelf-life.

Building an Orthogonal Characterization Strategy

No single technique provides a complete picture of an amorphous solid dispersion. A defensible CMC characterization package combines complementary methods, each targeting a different physical property and length scale.

The same logic extends into process development. In-line PAT tools, such as near-infrared or Raman probes during spray drying or hot-melt extrusion, allow residual solvent and phase homogeneity to be monitored in real time, reducing batch-to-batch variability before scale-up decisions are made.

Ultimately, the dissolution and supersaturation behavior of the dispersion, together with HPLC-based chemical stability and degradation profiling, must confirm the performance advantage that justified choosing an amorphous form in the first place. A physically stable dispersion that does not sustain supersaturation has not solved the original bioavailability problem.

Amorphous solid dispersions are a powerful tool for addressing poorly soluble APIs, but they are also metastable systems whose risk profile is easy to underestimate with a single analytical lens. PCL Research & Manufacturing builds orthogonal characterization packages around XRPD, DSC, TGA, DVS, solid-state NMR, and vibrational spectroscopy, tailored to the specific API, polymer system, and containment requirements of each program, including OEB4/OEB5 compounds. If your current ASD data package relies on XRPD and a single DSC run, it may be worth a technical conversation before the next stability pull.

Share

More insights

All articles