Email for more information

Company's other products

Email to a colleague

Printer friendly format

Figure 1: Transmission and Reflectance Mode

By Frederick J. Murray, President, KORSCH America Inc. & Mark J. Sullivan, Ph.D., Bruker Optics Inc.

Introduction

In the course of pharmaceutical tablet production, it is necessary to sample tablets periodically to obtain physical measurements of the tablet weight, thickness, and hardness. In most tablet production environments, the sampling of tablets is performed manually, and recorded by hand on a batch run sheet. In more advanced production operations, the use of automated tablet sampling systems has become more prevalent. These systems will automatically measure tablet samples, by passing them through a test sequence that includes a balance, thickness tester, and hardness tester. In additional to displaying and recording these physical measurements over the course of the batch, the actual values are often used for closed loop feedback, to maintain the process at the specified setpoints, and to alarm the operator and interrupt the process if the measurements are deemed to be out of specification.

Due to the use of off-line, end product testing, such as HPLC to measure composite API assay and assess content uniformity, it is necessary to delay batch release until the batch composite assay and content uniformity can be verified. Based on the speed in which the off-line HPLC testing is conducted, this delay in batch release can be expensive, both in terms of testing, and in terms of the inventory levels that must be maintained as a result of this testing delay.

In response to the current PAT initiative for innovation in pharmaceutical production, new technology has been developed which permits the on-line measurement of API assay directly at the tablet press. This sampling technology provides the in-process measurement of tablet weight, tablet thickness, tablet hardness, and tablet content uniformity, with real-time trending and electronic data storage over the course of the batch. This technology has the potential to eliminate the necessity for time consuming and expensive HPLC analysis subsequent to batch production. In fact, the higher sampling frequency of automated testing yields statistical results with higher certainty along with expedited batch release. In addition, more comprehensive and timely CU testing during batch production will provide a valuable feedback mechanism to support efforts to reduce batch-to-batch variability, and to assess the impact of process optimization programs. This paper examines the current technology for content uniformity measurement in tablets, and offers a commercial solution for the combined on-line measurement of tablet physical properties and API assay in real-time, during tablet production.

Figure 2: Transmission and Reflectance Spectra

Current Methods of Content Uniformity Measurement

The most common methods for measuring API assay and content uniformity in tablets are HPLC and uv analysis. These conventional methods are generally slow and labor intensive, requiring sample preparation to dissolve and extract the API, in addition to the actual analysis itself. Turnaround time for transporting the samples, logging them into the QC laboratory, performing the tests, preparing and reviewing the report may be on the order of hours for a single sample. In contrast, near-infrared spectroscopy (NIR) is a non-destructive method that requires no sample preparation. The sample is preserved so that it can be used for other tests and no solvent waste is generated. The total analysis time for NIR may range from 10 sec to 2 minutes, depending upon factors such as the analyte concentration.

NIR provides a rapid alternative to the conventional test methods, but it does not eliminate the need to maintain them. NIR is a secondary method that must be calibrated and periodically verified on the basis of validated primary methods. In addition, PAT methods such as NIR may not gain regulatory approval in all markets and conventional release testing may be required in some locations.

NIR Technology

Figure 3: KORSCH XL400 Tablet Press with 3-Channel Discharge Chute

Near-infrared spectroscopy is one of the most widely used analytical techniques in pharmaceutical development because it is fast, non-destructive, and methods are transferable from the laboratory to the production environment¹. Two major advantages of working in the near-infrared region (in wavelength units 780-2526 nm and in energy units 12800-4000 cm^-1) are representative sampling due to extensive penetration of the radiation through solids, and the availability of detectors with high signal-to-noise yielding excellent measurement precision. Two different NIR sampling modes, diffuse reflectance and diffuse transmission, are commonly used for a variety of tablets. To a large extent, the two modes are complementary and selection of the preferred mode depends upon the information required and the characteristics of the tablet core such as shape, thickness, density, etc. Generally, the transmission mode that probes the entire cross section of the tablet yields more representative sampling than reflectance that is restricted to the surface layer (Figure 1). Despite this advantage, the acquisition times for transmission are generally longer because the amount of light reaching the detector is lower. In addition, transmission spectra are limited to the higher frequency 2nd and 3rd overtone regions whose peaks are generally weaker and less well resolved for specific components than the combination and 1st overtone regions observed in reflectance (Figure 2).

The NIR spectra of tablets can be analyzed qualitatively to monitor batch-to-batch variations or quantitatively to obtain actual assay values. Most qualitative methods can be applied directly to the raw data and require little method development. By contrast, quantitative methods require a significant investment in method calibration and validation². The calibration set should ideally consist of more than 100 tablets that span the range of API concentrations from 75% to 125% of label claim and randomly incorporates the range of weights and hardness that will be obtained in production. An independent validation set of 50-75 tablets that covers the same range of variation in API content and physical properties as the calibration set is also required. After both sets of samples are analyzed by NIR, they are submitted for determination of the reference concentration value by the primary method (e.g., HPLC). The combined spectral data and primary reference values are treated with chemometric software to generate the calibration equation to predict API content. The calibration equation is then applied to the validation set to validate the method. Additional spectra of routine production samples will subsequently be added to further improve the accuracy and robustness of the equation. The NIR method can then be used to generate results for on-line analysis.

Figure 4: Schleuniger Pharmatron StepOne Tablet Sampling System

On-Line Tablet Sampling Technology

Tablet weight is the most critical parameter that must be controlled during high-speed tablet production. Even with advances in tablet weighing technology, the tablet press is still capable of producing tablets at a faster rate than they can be measured for weight. The result is that a secondary variable, which may be measured in real-time, must be used to insure tablet weight control. For more advanced tablet compression equipment, a press force control system is used to control tablet weight. In general, when a die is filled to a preset volume, and compressed to a final tablet thickness, then a resulting compression forced is measured. Since the compression rollers are in a fixed position, and the working length of the upper and lower press tools are produced to a precise tolerance, the tablet thickness during compression remains constant. Assuming consistent particle size die to die, then the press force of each tablet will reflect the amount of material in the die, or the tablet weight. Tablet press control systems will measure the individual compression force of each and every tablet as it is produced, and use the average compression force measurement to compare to a force setpoint. The dosing cam, which controls the amount of fill in the die, is then adjusted in very small increments by a precise stepping motor, to maintain the press force, and consequently the tablet weight, at the specified weight setpoint.

Over the course of the batch, the correlation between press force and tablet weight must be re-confirmed periodically, to account for any in-batch variation in the granulation. This is achieved by measuring actual tablet weight and comparing this value to the tablet weight setpoint. The operator can then adjust the press force setpoint, as required, to maintain the process centered at the specified tablet weight.

In addition to the tablet weight verification, the operator will also periodically measure the tablet thickness and tablet hardness to insure that these parameters are within preset specifications. Manual adjustments may be made to tablet thickness to increase or decrease tablet hardness over the course of the batch. Each time that a tablet thickness adjustment is initiated, the press force setpoint must be adjusted accordingly to insure that the control system will maintain the specified tablet weight. In general, the use of a tablet sampling system eliminates the requirement for manual sampling, and eliminates the subjectivity of a manual adjustment, when the data indicates than an adjustment is required.

Figure 5: FT-NIR Measurement Station

Several methods of tablet sampling are generally available on more advanced tablet compression equipment. The first method permits the sampling of single tablets from the die table, which permits tablets to be selected from specific or random punch stations, and for the tablet identity (punch station) to be preserved through the sampling process. (Figure 3). In this case, the tablet weight, thickness, hardness and content uniformity can be matched to a specific punch station, and corresponding compression force. The second method permits the random sampling of a group of tablets, which are generally diverted from the good channel in the tablet discharge chute, to a pneumatic transfer system that transports the tablets to the tablet sampling unit, which is positioned remote from the tablet press. This system requires robust tablets that can withstand the pneumatic transfer process. With this method of sampling, the system will report on tablet weight, tablet thickness, tablet hardness and tablet content uniformity, however, the tablet may not be correlated to a specific punch station or compression force.

On-Line NIR Measurement

The tablet press control system will permit key set-up parameters to be established, prior to batch production, as follows:

More advanced tablet press control systems will permit these product setpoints to be recorded and retrieved via a product recipe capability. In this way, the sampling system can be readily set-up for any number of products by simply selecting the appropriate product name from the recipe list.

Figure 6: NIR Probe for Product Hopper or Feed Pipe

Once the batch has been started, then the sampling system will request tablets at preset intervals, as defined by the sampling frequency and sample size. The Schleuniger Pharmatron StepOne (Figure 4) system will process the tablets through a sequence of test stations, and will generate the corresponding measurement for each tablet and relay this data to the tablet press control system. The first position is where the FT-NIR measurement of API assay is made (Figure 5). The second position measures the tablet weight, followed by the tablet thickness, and tablet hardness measurement. The system may be configured to bypass the tablet hardness station to permit in-process samples to be collected for additional quality control testing (friability, dissolution). Once the full sample size has been processed, then the data are evaluated by the tablet press control system to permit real-time trending on the HMI, and if required, feedback to the control loops for tablet weight, tablet thickness, or tablet hardness. As there is no feedback on the API assay measurement, the system can report only if the API assay is trending toward the preset limits established in the set-up, or stop the tablet press, if the API assay value exceeds preset limits.

The use of the StepOne tablet sampling system will permit the real-time display of tablet weight, tablet thickness, tablet hardness, and content uniformity over the course of the batch, including average, range, and standard deviation statistics. These data can easily be adapted to a SCADA system for remote monitoring or centralized batch data storage. The electronic audit trail for machine adjustments (Event Log) on the tablet press will record any machine adjustments that are initiated by the sampling system - based on the real-time sampling results, and all associated press stops are recorded in the alarm history audit trail.

In addition to the NIR measurement at the tablet sampling station, it is also possible to mount NIR probes (Figure 6) at various locations on the tablet press, to insure consistency in the manner in which material is transported from an overhead tote, through the feed pipe, and potentially, in the feed frame itself. These NIR probes can provide a continuous measurement of API assay as the material passes the product hopper, and feed frame, in the tablet press. These additional sensors can potentially elucidate any gradient in API assay that occurs during the feeding process.

Conclusion

The use of an on-line tablet sampling system with an FT-NIR measurement capability will permit batch content uniformity to be assessed in real time, with a more comprehensive and representative sample size. While the actual opportunity to affect an expedited batch release is highly dependent on the internal quality procedures of the manufacturer, the use of on-line NIR measurement technology does provide the technical basis to support this initiative. The use of on-line NIR technology also supports process optimization efforts geared to insuring maximum batch-to-batch consistency and quality.

About the Authors: Frederick J. Murray is the President of KORSCH America Inc. and is responsible for the KORSCH operation in North America. KORSCH America is a Division of KORSCH AG, of Berlin, Germany, one of the premier manufacturers of tablet compression equipment worldwide. Mr. Murray has been with KORSCH for 19 years. Mr. Murray has authored and published technical papers on tablet press automation, tablet press validation, and the application of expert systems to pharmaceutical processing equipment. He can be reached at 508-238-9080 or fred.murray@korschamerica.com.

Mark Sullivan is currently head of the Process Analytical Technology team at Bruker Optics based in Billerica, Massachusetts. He has over 20 years of experience with optical and non-optical spectroscopic techniques in the pharmaceutical and chemical industries. He was previously a Research Fellow working in PAT at Merck & Co., Inc. in West Point, PA and has served on USP Project Team 18 on PAT. His professional interests are to facilitate the integration of new technologies to develop the next generation of PAT tools and to expand the scope of methodologies used in research and manufacturing in a regulated environment. He can be reached at 978-439-9899 or mark.sullivan@brukeroptics.com.

References:
¹ Gabriele Reich, "Near-infrared spectroscopy and imaging: Basic principles and pharmaceutical applications" Advanced Drug Delivery Reviews 57, 1109-1143 (2005).
² M. Blanco, M. A. Romero, M. Alcala, "Strategies for constructing the calibration set for a near infrared spectroscopic quantitation method" Talanta 64, 597-602 (2004).

Pharmaceutical Processing
Advantage Business Media