Particle Sciences - Technical Brief: 2010: Volume 8
The need for particle size control of pharmaceuticals is becoming more important as the industry attempts to formulate active pharmaceutical ingredients (API’s) with poor aqueous solubility, which constitutes up to 40% of new chemical entities1. Advances in drug delivery have been made where a nanonized or micronized API, with the concomitant increase in particle surface area, results in increased bioavailability. The use of semisolid vehicles (i.e., aqueous gels, creams and ointments) provide a good option for the formulation of hydrophobic pharmaceuticals. Semisolid formulations can also allow for the safe delivery of high potency compounds with a better patient compliance when used as an alternative to needles or when an API has significant oral dosage side effects. Semisolid dosage forms can be developed to deliver topically for local action as well as transdermally for systemic delivery. When a semisolid delivery form is combined with controlled particle sizes, it becomes necessary to measure and monitor the particle size in situ. As a result, the need for highly reproducible particle sizing techniques has grown in the past decade. This trend will continue, particularly in view of FDA recommendations that a more thorough characterization of particle size distributions is needed in submissions where a drug product claim is based in a tightly controlled particle size.
Particle sizing in dispersions can be accomplished using a variety of techniques, including laser diffraction, dynamic light scattering (DLS), disc centrifugation, and light microscopy. All of these techniques have their advantages and limitations. Laser diffraction relies on a well controlled presentation of the sample to the measurement region and is limited to samples with narrow range of particle concentrations. Dilution is often required and this may affect the particle size, particularly in compounds with high solubility. DLS also uses low particle concentrations, usually requiring significant sample dilution and suffers the same limitations. Since DLS is based on Brownian motion, the dispersion media available for use are limited by viscosity. Also, the particles, in media with viscosities similar to water, must be below 1 µm in size, otherwise settling occurs too rapidly for accurate size assessment. Disc centrifugation relies on the ability of the particles to move through the dispersion due to centrifugal force. This requires that the dispersion viscosity be low enough to allow the centrifugal force on the particles to overcome viscous resistance. Disc centrifugation is a highly accurate technique, but it is not amenable to use with semisolid formulations with high viscosities. All the methods discussed so far may require dilution and while some drug product formulations can be diluted without changing the particle size distribution, others may be affected by dilution, which makes meaningful particle size measurement challenging.
Optical microscopy, however, is a technique which lends itself quite naturally to the determination of particle size in situ. Particle sizing by optical microscopy and image analysis is a technology-intensive method requiring automated acquisition of photomicrographs. This is usually controlled by software that also handles the analysis of the images collected. The acquisition of a large number of image objects is required for statistically relevant measurements, which may include the determination of length, width, area, equivalent circle diameter, roughness, etc.
With careful selection of objectives and camera, the technique has a broad dynamic range where the upper limit is several mm at low magnification and the lower limit is close to 1 µm, which is the minimum resolution achievable with the use of normal white light illumination. An advantage of microscopy over laser diffraction is the verifiable and calibrated accuracy, as calibration of the instrument may be performed with NIST traceable stage micrometers and verification of analysis methods with the use of latex microsphere size standards. This is in contrast to laser diffraction, which, based on first principles, may only be verified with latex microspheres. However, as mentioned earlier, optical microscopy requires a very large number of particle observations. ISO 13322-1 contains a guide for the number of particles required at a variety of confidence values based on the width of the particle size distribution. The shortcoming in this is that the analyst has no knowledge of this fact a priori and the range of distribution widths described by ISO 13322-1 are too few to describe many real world particle size applications even though the largest of the listed distributions requires an extraordinarily high number of observations (~65,000).
Light microscopy was used to monitor the particle size distribution of an API in gels in stability studies. Linearity and accuracy were demonstrated and the data are presented below. A comparison of the particle size determinations performed by both light microscopy and laser diffraction are also provided.
The accuracy and precision results of the method are collated in Table 2 and Table 3, respectively. The accuracy of the method shows significant deviations. A deviation of 86% was found for placebo gels spiked with 1.9 µm latex beads. The absolute error of ~1 µm is quite small, to be expected, and not considered an issue for tracking the change in particle size, for which this method was developed.
Acceptable precision is demonstrated with a maximum RSD of 2.6% for four determinations performed on the 20.9 µm latex beads spiked into the placebo.
Method robustness was investigated for an active gel by varying the gel thickness. This was accomplished by using spacer tapes of 1 mil, 2 mil, or 5 mil in thickness (25, 50, and 125 μm, respectively) to separate the cover slip from the slide to create a gap for the sample. The results in in all three cases gave a median particle size (d50) of the API in the gel of 2.2 μm and indicate that the sample thickness had negligible effect on the measured particle size.
Ideally, the verification of analytical techniques involved in particle size analysis would be in agreement with another technique. Two techniques were used to examine samples of a polydisperse particle size standard suspended in water. Figure 2 shows an overlaid microscopic particle size distribution and laser diffraction particle size distribution of the 1-10 µm polydisperse standards (PS192). The laser diffraction data are the average of five distribution measurements. The distribution results from each method agree well with the 95% confidence intervals provided with the certified standard values (Figure 3).
We have shown that optical microscopy can be used to monitor the particle size of API in an aqueous gel. Precision and robustness are excellent. Accuracy shows significant deviations at the limits of smaller particle size, as can be anticipated from pushing the limit of detection for optical microscopy. However, depending on the application, this can be acceptable and there may be some modifications that can improve accuracy. Microscopy is well suited when only small volumes of sample are available and only minimal sample preparation is required.
1. Bo, Tang, et.al. Drug Discovery Today, Vol. 13, Number 13/14, July 2008.
Particle Sciences is a leading integrated provider of formulation and analytic services and both standard and nanotechnology approaches to drug development and delivery.
3894 Courtney Street, Suite 180
Bethlehem, PA 18017