(W1030-02-08) Raman Spectroscopy as a PAT Tool for Hot Melt Extrusion

Purpose: Hot Melt Extrusion is an established pharmaceutical manufacturing technology used for improving drug solubility and availability, controlled release, as well as taste masking and abuse deterrence. As part of Quality by Design (QbD), Process Analytical Technology (PAT) is employed to ensure consistent product quality and process control. Raman spectroscopy was used in this study to monitor not only the drug concentration but also quality attributes such as crystalline/amorphous form and drug-polymer interaction in real-time during HME. The data provided by inline analysis enables immediate adjustments to the process, ensuring optimal product quality. Consequently, this approach enhances the efficiency and reliability of the HME process, leading to improved pharmaceutical products.

Methods: Blends of acetaminophen in Soluplus, prepared at concentrations of 0%, 10%, 20%, 25%, 40%, or 50%, were fed into the HME under constant temperature, feed rate, and screw speed settings. No attempts were made to optimize the HME conditions. The All-In-One Raman analyzer with the Raman process probe installed at the die recorded measurements at 24 second intervals. The setup is shown in Figure 1. Extrudate samples were collected every few minutes for offline assay by HPLC or UV-Vis spectroscopy. The Raman spectra were pre-processed and paired to the assay values to develop a Partial Least Squares (PLS) regression model. The acetaminophen form and interactions were determined by spectral subtraction of the Soluplus features and comparison of the resulting spectra with reference spectra of the various forms of acetaminophen.  The Raman spectra of the amorphous and three crystalline forms of acetaminophen differ, allowing for the estimation of the amounts of amorphous form present.

Results: A PLS regression model was developed for the quantification of acetaminophen concentration. Raman spectra were collected in-line with the addition of 0%, 10%, 25%, and 50% acetaminophen in Soluplus.  The Raman spectra displayed clear changes in peak intensities corresponding to different acetaminophen concentrations (see Figure 1). Samples were collected during for High-Performance Liquid Chromatography (HPLC) analysis every two to three minutes. HPLC confirmed there was no degradation of the acetaminophen and provided calibration data for the Partial Least Squares (PLS) regression model. The PLS model showed good agreement between Raman predictions and HPLC results using two latent variables (RMSEC=0.5%, RMSECV=0.7%, R²_Cal=0.998). The model was then used to predict the concentration of acetaminophen during a second HME run after adding 20% and 40% acetaminophen. HPLC results of the collected samples were used to validate the model. The measured values were in good agreement with the Raman predictions with a standard error of prediction (SEP) of 1.4% as shown in Figure 2. Spectral subtraction confirmed that acetaminophen remained in the amorphous state in the die throughout the process. The Raman features of Soluplus were subtracted from selected spectra of the calibration run by scaled subtraction.  A comparison of the subtracted spectra to the spectra [1] of pure amorphous and crystalline forms of acetaminophen shows very good agreement with the amorphous form for all three concentrations (see Figure 3). The amide I band (C=O stretch) positions and shape near 1650 cm^-1 had significant differences when comparing the 10%, 25%, 50% and pure amorphous forms (see Figure 3).  The peak is shifted to higher energies at lower concentrations (10% and 25%). This could suggest stabilizing hydrogen bonding or diopole-dipole interactions with itself or Soluplus.  The peak at 1615 cm^-1 is attributed to the aromatic ring (C=C stretch) and shows no shifts. The amide II peak at 1555 cm^-1 has a slight shift to lower energy for lower concentrations (10% and 25%).

Conclusion: The Partial Least Squares (PLS) regression model accurately predicted acetaminophen concentration, facilitating real-time process monitoring of acetaminophen concentration during the process.  The results confirmed that acetaminophen remained in the amorphous state throughout the HME process, with no observable transition to crystalline forms, ensuring consistent product quality. Insights into API polymer interaction were also determined from the Raman spectra.  This highlights Raman spectroscopy's potential as a powerful PAT tool for the pharmaceutical industry.

References: 1. J.F. Kauffmana, L.M. Batykeferb, D.D. Tuschelb, J. Pharm. Biomed. Anal. 48 (2008) 1310-1315


Fingerprint region of Raman spectra collected during the extrusion of 0%, 10%, 25%, and 50% acetaminophen. Reference spectra (dotted) for amorphous acetaminophen [1] and Soluplus are shown for comparison.


Acetaminophen concentration prediction by Raman spectroscopy during validation run of hot melt extrusion of 20% and 40% acetaminophen in Soluplus. HPLC values are indicated by *.


In-line HME Raman spectra of 10% (b), 25% (c), and 50% (d) melts of acetaminophen in Soluplus with Soluplus peaks removed by scaled subtraction of the Raman spectrum of Soluplus. Spectra are compared with the Raman spectrum of the pure amorphous form of acetaminophen (a) [1].