Residual Lactide Control Through Polymer Drying Before PLGA Microsphere Extrusion

Residual lactide in poly(D,L-lactide-co-glycolide) (PLGA) is a cyclic diester that remains in the polymer matrix from the ring-opening copolymerization equilibrium between lactide and glycolide monomers. In extrusion-grade PLGA with a 50:50 lactide-to-glycolide molar ratio, an acid-terminated chain architecture, and an inherent viscosity of 0.15–0.45 dL/g measured in chloroform at 25 °C at a concentration of 0.5 g/dL, the free monomer acts simultaneously as a melt plasticizer, a volatile contaminant, and a hydrolysis product reservoir. Residual lactide is not classified as a residual solvent under ICH Q3C, but parenteral and ophthalmic PLGA feedstocks are nevertheless controlled for free lactide because the cyclic dimer undergoes rapid hydrolysis in aqueous microsphere release environments to form lactic acid, lowers the local pH, and changes the autocatalytic degradation rate of the final dosage form. During extrusion-based microsphere manufacturing, in which the melt is forced through a multi-orifice die and quenched before downstream spheronization, a residual lactide level above approximately 0.5% w/w is commonly encountered as strand instability, surface porosity, die-lip fouling, and high vent condenser loading. Pre-extrusion polymer drying is the unit operation most often used to reduce free lactide and water simultaneously before the melt enters the extruder, but the two objectives impose competing thermal requirements: monomer stripping is diffusion-limited and accelerates with increasing temperature, whereas water removal must be carried out below the sintering temperature of the amorphous particle. The operating window is therefore defined by the glass transition onset, the residual monomer diffusion coefficient, the dryer vacuum level, and the acceptable loss of molecular weight due to hydrolysis.

Table 1. Comparative operating envelopes for pre-extrusion drying unit operations used with PLGA powders and pellets.

Values are representative equipment operating envelopes derived from batch dryer technical bulletins and polymer drying practice; site-specific qualification with gas chromatography and Karl Fischer titration is required because published data for individual PLGA grades in each dryer configuration is limited.

Residual Lactide Thresholds Are Not Harmonized Across Pharmacopeias

The absence of a harmonized monograph for PLGA means that residual lactide control is established through supplier-user analytical method transfer rather than a single compendial limit. Certificates of analysis for extrusion-grade PLGA typically report total monomer content by an in-house gas chromatographic method with flame ionization detection, using a non-polar capillary column and an external certified lactide reference standard. The method is usually validated over the range 0.01–1.0% w/w with a limit of quantification near 0.01–0.05% w/w, although published limits for a specific PLGA grade are limited. Residual moisture is measured by Karl Fischer titration according to USP <921> Method Ic or an equivalent validated internal procedure, and the accepted moisture ceiling for melt extrusion is generally <0.1% w/w. The control of free lactide is stricter for extrusion-based microsphere formation than for solvent-based emulsion processes because the melt residence time in a twin-screw extruder is short, and the vent surface area is small relative to the polymer volume; a starting lactide concentration that exceeds the devolatilization capacity of the vent will not be corrected downstream. In practical terms, if the pre-dried powder entering the extruder feed throat contains free lactide above 0.3–0.5% w/w, the vacuum vent is unlikely to reduce the final melt concentration below a typical certificate-of-analysis specification of <0.2% w/w without operating the screw at excessive speed or barrel temperature. The quantitative relationship between vent efficiency, screw speed, and residual lactide has not been published in a standardized compendial format; therefore, process qualification is performed with design-of-experiment studies on the specific extruder configuration. Analytical release for microsphere intermediates is commonly supplemented with gel permeation chromatography to track molecular weight loss and with differential scanning calorimetry according to ASTM D3418-21 to confirm that the glass transition temperature has not dropped because of residual lactide or water plasticization. For comparative melt viscosity measurements, ISO 1133-1:2022 may be used at low temperature and short preheat, but the values are highly sensitive to residual lactide and water and are not used as a release test for PLGA. Final microsphere suspensions, where applicable, must also comply with USP <788> limits for particulate matter.

On a production line consisting of a loss-in-weight gravimetric feeder, a co-rotating twin-screw extruder with a length-to-diameter ratio between 25:1 and 40:1, and a vacuum vent connected to a dry-screw or rotary-vane pump, residual lactide removal is partitioned between the upstream dryer and the extruder vent. The feed throat is maintained at 40–60 °C, the compression section is ramped from 60 °C to 120 °C, and the metering section is held at 120–160 °C for PLGA 50:50; higher-lactide grades such as 75:25 may require a metering temperature of 140–170 °C to reduce melt viscosity. The vacuum vent is located in the melt seal zone and operates at an absolute pressure of 10–30 kPa. Because the residence time under vacuum is typically less than 10 seconds, only free lactide that has migrated to the melt surface can be stripped. In practice, the vent port is heated above the lactide melting point of approximately 95–97 °C to prevent solid deposition, and the vacuum line is protected by a cold trap at -40 °C to condense sublimed monomer before the pump. If the upstream dryer leaves a high free lactide concentration, the vent condenser fills rapidly, the vacuum pressure rises, and melt strands begin to foam at the die. Foam expansion at the die exit creates microsphere diameter variability and occasionally clogs the multi-orifice die, producing a pressure spike in the extruder barrel. Production-scale observations indicate that batch-to-batch differences in free lactide are a more frequent cause of vent blocking than differences in copolymer molecular weight, but the literature does not report a unified threshold for this failure mode. The operator response is typically to lower screw speed and increase barrel temperature, but both actions can increase thermal degradation and reduce the molecular weight of the PLGA.

How Does Pre-Extrusion Drying Shift Monomer Equilibrium in PLGA Melts?

In a ring-opening copolymerization, the final polymer melt contains a temperature-dependent equilibrium concentration of cyclic monomers, and free lactide can be regenerated during melt processing by backbiting and transesterification reactions. A drying step performed in the solid state removes free lactide by diffusion to the particle surface followed by evaporation or sublimation under vacuum. The rate-limiting step is Fickian diffusion through the amorphous PLGA matrix. Below the glass transition temperature, segmental mobility is restricted, and the apparent diffusion coefficient for lactide is low; above the glass transition, particle sintering destroys the bed porosity and reduces the surface area available for monomer evaporation. For PLGA 50:50 with a DSC glass transition onset of 42–50 °C by ASTM D3418-21, the vacuum tray drying temperature is therefore set at 25–40 °C, and the bed depth is kept below 2 cm in static dryers. Solid-state removal cannot completely eliminate lactide because the polymerization equilibrium is re-established when the polymer melts in the extruder; however, lowering the starting free lactide concentration shifts the mass balance so that the extruder vent is capable of handling the regenerated monomer. Water is critical in this balance because hydrolytic chain scission produces additional carboxylic acid end groups that are known to catalyze both esterification and transesterification reactions in PLA and PLGA systems. The presence of residual tin-based catalyst from stannous octoate polymerization, when present above typical pharmaceutical limits of <100 ppm, can further accelerate monomer regeneration. Published data for the specific equilibrium lactide concentration as a function of PLGA melt temperature and catalyst residue is limited; industrial practice relies on coupling a conservative solid-state drying step with vacuum devolatilization and then confirming the lactide content before and after extrusion by gas chromatography.

An alternative or complementary unit operation is solid-state post-polymerization carried out on amorphous PLGA particles under vacuum or dry nitrogen at temperatures below the melting point. The objective is to reduce free lactide while increasing molecular weight through limited chain extension, but the low glass transition temperature of PLGA makes this operation difficult at production scale. Static trays in a vacuum oven are normally limited to 35–45 °C for acid-terminated PLGA 50:50 to avoid sintered cake formation; the bed depth must remain below 2 cm to keep the diffusion path short. Agitated vacuum dryers can operate at bed depths of 10–15 cm with intermittent rotation, but mechanical shear in a rotary dryer can generate fines that later segregate in the extruder hopper and create flow instabilities. The time required to reduce free lactide from 0.4% w/w to 0.2% w/w may exceed 48 h in a static dryer, and the same cycle can reduce intrinsic viscosity if residual water is not removed early in the heating profile. Published data for this specific configuration is limited because most dryer suppliers validate water content only and do not report free lactide removal separately. A production site must therefore qualify the drying endpoint with gas chromatography for monomer content and with Karl Fischer titration for water content, using sampling from the top, middle, and bottom of the bed to detect stratification.

Vacuum Tray Drying Parameters for High-Lactide Copolymer Grades

PLGA grades with higher lactide molar fractions, such as 65:35, 75:25, and 85:15, present different drying boundaries because the glass transition temperature is higher and the initial free lactide concentration is typically higher from the polymerization feed. The maximum shelf temperature can be increased moderately, but the risk of lactide plasticization is also greater in the early stages of drying. Differential scanning calorimetry according to ASTM D3418-21 is used to determine the glass transition onset of the incoming lot; the dryer setpoint is then placed 5–10 °C below the measured onset to avoid sintering. For a 75:25 PLGA with a typical glass transition of 50–60 °C, a vacuum tray dryer can be operated at 45–50 °C at 1–5 kPa for 24–48 h, but the bed depth should not exceed 2 cm for static trays. High-lactide copolymers also possess lower melt viscosity at a given molecular weight, which means that residual lactide plasticization can lead to premature melting in the feed zone if the feed throat temperature is not controlled. The moisture specification remains <0.1% w/w by USP <921> Method Ic because water-induced hydrolysis is independent of the lactide-to-glycolide ratio. The analytical verification of dried powder should include residual lactide by GC-FID, residual glycolide by the same method where glycolide is detected, residual moisture by Karl Fischer titration, and intrinsic viscosity by capillary viscometry to detect molecular weight loss during drying.

Table 2. Reference drying and extrusion limits for common PLGA compositions.

Reference values are based on supplier certificates of analysis and thermal analysis practice; they are not compendial limits and require lot-specific confirmation for amorphous acid-terminated PLGA.

After drying, PLGA must be transferred to the extruder feed system under dry nitrogen or in sealed aluminium-lined packaging, because amorphous PLGA powders re-absorb atmospheric moisture rapidly. Exposure to ambient air at relative humidity above >60% for 30 min can increase the surface moisture content into the 0.1–0.3% w/w range, depending on particle size and porosity. The extruder hopper should be purged with dry air or nitrogen at a dew point below -40 °C, and the feed throat should be jacketed at 20–30 °C to prevent condensation on cold metal surfaces. A hopper purge rate of 5–10 L/min is often sufficient for small production-scale feeders, but excessive gas flow can fluidize low-bulk-density powders and cause feed rate fluctuation. The transfer line from the dryer to the feeder should be made of conductive tubing to reduce electrostatic accumulation, and the receiving container should be grounded because dried PLGA fines can adhere to charged surfaces and later dislodge as agglomerates. Moisture ingress during storage and transfer is a more frequent cause of extruder instability than residual lactide reformation, but both variables should be monitored at the feed throat because free lactide and water together reduce the effective glass transition temperature and can cause melting in the feed zone. In cleanroom environments classified under ISO 14644-1:2015, the relative humidity is often controlled between 20% and 45%; however, local enclosures around the dryer discharge and feeder hopper are still required if the product moisture specification is <0.1% w/w.

When Twin-Screw Venting Alone Fails to Meet Residual Monomer Specifications

If the pre-extrusion drying step is shortened or the incoming PLGA lot contains free lactide near 1.0% w/w, vacuum venting alone may not bring the melt concentration below the target because the devolatilization rate is controlled by diffusion through a viscous melt and by the short residence time under the vent. Increasing the vacuum level to below 10 kPa can cause the melt surface to foam into the vent port, and the resulting pressure fluctuations propagate backward into the melt seal, creating screw surging and inconsistent die flow. Raising the metering zone above 160 °C can improve monomer volatility, but PLGA undergoes chain scission, discoloration, and generation of lactic acid and carbon dioxide at elevated temperatures; the apparent residual lactide may remain high because thermal degradation regenerates monomer at the same time devolatilization removes it. For this reason, the operational boundary for PLGA extrusion is not defined by the vacuum pump capacity alone, but by the combination of pre-dried residual lactide, residual moisture, screw speed, barrel temperature, and vent port geometry. A production extruder with atmospheric venting in the feed zone and vacuum venting in the melt zone usually requires a pre-dried powder containing less than 0.3% w/w free lactide to achieve a final melt concentration below 0.2% w/w at a melt temperature of 140–150 °C. If the final product remains above specification, the common corrective actions are to lower the dryer bed depth, extend the drying time at a temperature 5–10 °C below the measured glass transition onset, or replace static drying with vacuum drying under nitrogen sweep. Additional compounding aids are generally avoided; amine-based additives in particular are incompatible with PLGA before extrusion because nucleophilic amines can promote ester aminolysis, accelerate molecular weight loss, and change the residual monomer profile.