August 1, 2014 (Vol. 34, No. 14)
Mary Stellmack Senior Research Chemist McCrone Associates
Description of a Simple Sampling Technique
After routine cleaning of metal or glass surfaces of pharmaceutical process tanks, residues may remain and appear as faint whitish or dark stains that cannot be easily wiped off. To clean the residue without damaging the tank surface, it is necessary to identify the residue so that the correct solvent or cleaning method can be used.
Due to the small amount of residue on the tank surface, it can be difficult to obtain a large enough sample volume for analysis. For unambiguous identification of the residue, it is desirable to remove a pure sample from the sampling tool for analysis, without any attached material from the sampling tool.
White cotton or polyester swabs are common sampling tools, sometimes soaked in water or alcohol prior to wiping the tank surface. Swabs are poor sampling tools because only a small amount of residue is transferred to the swab, the material of interest is difficult to remove from the swab for analysis, and light-colored residues cannot be seen on the white swab.
Our objective was to develop an alternative sampling method to obtain a larger volume of residue that could be easily isolated for micro-analysis to identify organic components by FTIR (Fourier Transform Infrared Spectroscopy), or EDS (Energy Dispersive X-Ray Spectrometry) to identify inorganic components.
We investigated several traditional and non-traditional scraping tools for removing residues from a metal or glass surface:
- Cotton swab
- Polyester swab
- Double-edge razor blade
- Tape lift with Scotch® tape
- Polypropylene spatula
- Plastic polystyrene knife
Our “tank surfaces” consisted of a flat stainless steel sheet having a thin spotty coating of rust, and a stainless steel vessel having a faint coating of white-colored hard water deposits. The stained surfaces were vigorously rubbed with the dry swabs, or moderately to lightly scraped with the other sampling devices while holding the device at about 45 degrees to the surface, except for the tape lift, which was obtained with a single firm compression of the tape against the stained surface.
The sampling devices were examined under a stereomicroscope and photographed at magnifications up to 25×. A fine tungsten needle was used to remove some of the residue from the sampling device, then the residue was transferred to a glass slide. The residue on the slide was examined under a polarizing light microscope at 500× magnification to see whether particles from the sampling tool had contaminated the residue.
When cotton or polyester swabs were used to sample a rust-colored surface, the residue was readily visible, but consisted of fine particles trapped between and below the swab fibers. It was difficult to remove the brown material from the swab for FTIR analysis without removing some cotton as well, but the entire swab head could be analyzed by EDS to identify the metallic residue (Figure 1).
When swabs were used to sample light-colored residue, no residue was visible on the swab with the naked eye. At 25× magnification, a faint residue was seen on the swab, but the residue could not be manually removed from the fibers if it was not clearly visible in the microscope. If the residue had been pure white, it would not have been possible to see the residue in the microscope, even at higher magnification.
It was determined that swabs are most useful for dark-colored residues that consist of metal corrosion, or thick clumps of dark-colored organic materials, but are poor sampling tools for light-colored residues.
The blade was held at about a 45 degree angle, and the tank surface was gently scraped. Light-colored and dark-colored residues were easily seen on the blade, and could be easily removed for analysis.
The blade may scratch some tank surfaces, and fine metal wear particles can be dislodged from the blade and mixed with the sample, making it difficult to identify residues that consist of metal corrosion. The razor blade is most useful for hard or soft nonmetallic residues.
The edge of a plastic spatula was used to scrape the tank surface to remove the residue. The spatula used in this test was dark blue in color, and light-colored residues were readily visible. For dark-colored residues, a light-colored spatula should be used.
However, polypropylene is a somewhat soft material, and rolls of plastic are generated along the edge of the spatula with the particles of interest entrained within. The residue cannot be separated from the loose plastic filaments for analysis. Some loose particles of residue (not mixed with polypropylene) were also spattered on the surface of the spatula trough, not immediately in the area of contact of the spatula with the tank surface, and these particles can be removed for analysis. Soft plastic scraping tools are best suited for soft waxy residues that do not require hard scraping of the surfaces.
The tape picked up a good quantity of residue, but the particles needed to be removed from the adhesive and cleaned with a solvent to remove any attached adhesive. This can be difficult if the particles are very small or crumbly, and some particles may be soluble in common solvents. The tape lift cannot be used for soft residues; it is only useful if the residue consists of hard material that is not firmly adhered to the tank surface.
Plastic Polystyrene Knife
The plastic knife that was used was a standard disposable plastic knife available at any grocery store. We removed the handle from the knife, and the blade end was held at about a 45 degree angle while firm pressure was applied as the teeth were scraped over the tank surface. The teeth of the knife collected a good amount of residue, and a large amount of loose residue clung to the surface of the blade adjacent to the teeth (Figure 2).
The loose residue on the blade was easily removed for analysis. The residue on the teeth contained a few scattered fine polystyrene particles that were abraded during the scraping process; polystyrene is readily soluble in many common solvents and if contamination of the residue with knife debris is suspected, it can be easily dissolved.
The infrared spectrum of polystyrene is unique, and it is substantially different from most process tank residues, thus it is easy to check for the presence of polystyrene in the sample, and in most cases the dissolution process should not be necessary.
The polystyrene knives’ hardness and solubility characteristics made them the best tools for sampling hard residues (the polystyrene knives are good sampling tools for soft residues as well). These tools are stiff enough to remove the residue without generating a large amount of abraded plastic, but do not damage the metal surfaces.
The knives are convenient because very little sample preparation is needed, and they can be obtained in many different colors; a knife color can be chosen that contrasts with the color of the tank residue, enhancing the visibility of small amounts of sample, and making it possible to isolate the sample from the sampling tool for analysis.
The selection of a scraping tool depends somewhat on the type of residue to be sampled. This brief study focused on hard residues on metal surfaces, as these are the most challenging residues for sampling. If the residue consists only of metal corrosion, any of these sampling methods, except the razor blade, are suitable.
However, if the residue contains some organic material, is present in a very small amount, or if one wishes to perform solvent extractions on the residue to look for the presence of oils or other soluble materials, the residue should be removed from the sampling tool for analysis by IR or other methods. The polystyrene knife provides a good combination of clean sample removal and little contamination from the sampling tool.
The knife, or any of the other sampling tools, can be attached to a glass slide with tape, and inserted into a standard 50 mL plastic tube with a threaded cap for shipping or safe transport to the lab for analysis. The standard glass slide fits firmly into such a tube, and will not move during shipping.