Materials and methods
Polytetrafluoroethylene (PTFE)-lined polypropylene (PP) caps were extracted with 25 mL of each of the following surfactants:
1% nonionic, octylphenol ethoxylate surfactant (Triton X-100, Dow Chemical); 0.1% polysorbate 80 (PS 80); and 0.1% polysorbate
20 (PS 20). The same caps were also extracted with the following two alternative solvents: 60% IPA and 15% ethylene glycol
monobutyl ether (EGMB). The caps were submerged in each solvent at 40 °C at ambient relative humidity for seven days. The
resulting extracts were tested by gradient HPLC using a time-of-flight (TOF) LC–MS (Agilent 6500 series) equipped with a multimode
source (electrospray and atmospheric pressure chemical ionization) using positive ionization. Data were acquired using scan
mode with a range of 80 to 1500 m/z and then by extracting ions that corresponded to the compounds of interest.
The PP caps were chosen for this experiment due to the presence of known additives that could be easily tracked during extractables
screening. Compounds previously observed through extractables studies that were targeted in this experiment include a di-tert-butyl(phenyl)phosphite (Irgafos 168, BASF); a phosphate oxidative degradant of Irgafos 168; ethylene bis(heptadecanamide);
pentaerythritol tetrakis 3-(3,5-di-tert-butyl-4-hydroxyphenyl) propionate (Irganox 1010, BASF); and erucamide. Determination of extraction efficiency equivalence
was made by comparing the responses for each of the targeted extractables observed in each solvent.
Results and discussion
Figure 1: Liquid chromatography–mass spectometry (LC–MS) time-of-flight (TOF) multimode positive total ion chromatograms (visual
representation). (a): 60% isopropanol (IPA) and 15% ethylene glycol monobutyl ether (EGMB) (gray/red), 0.1% polysorbate (PS)
20 (green), 0.1% PS 80 (blue), and 1% octylphenol ethoxylate surfactant (Triton X-100, Dow Chemical) (purple). (b): 0.1% PS
80 (blue) and 60% IPA (gray). (ALL FIGURES ARE COURTESY OF THE AUTHORS)
Figure 1a presents a visual representation overlay of LC–MS total ion chromatograms (TIC) of all the solvents used for the extraction
study. The total ion chromatograms of the 60% IPA and 15% EGMB are similar; therefore, they cannot be distinguished in the
figure but are shown as the gray/red line. Figure 1b presents a visual representation of just the overlay of 0.1% PS 80 and 60% IPA to show that any potential extractables would
be masked by the PS 80 interference. Interferences are also observed with 0.1% PS 20 and 1% Triton X-100 as shown in Figure 1.
Figure 2 presents the concentration results in µg/mL of each extractable compound detected versus the type of solvent. Concentrations
were estimated based upon the average of the responses of the reserpine system suitability standards. As Figure 2 indicates, not all of the compounds of interest were extracted in each of the solvents. Irgafos 168 and Irganox 1010 were
extracted in both 60% IPA and 15% EGMB while ethylene bis(heptadecanamide) was only extracted in the 60% IPA solvent. Irgafos
168, Irganox 1010, and ethylene bis(heptadecanamide) were not extracted in the 0.1% PS 20, 0.1% PS 80, and 1% Triton X-100
solvents. Erucamide was extracted in all solvents. Irgafos 168 phosphate results were not presented because concentrations
were similar to the blank concentrations.
Figure 2: Results for the seven-day extraction study on polypropylene (PP) caps showing extractables of common PP additives
using five different types of extraction solvents; IPA is isopropanol, EGMB is ethylene glycol monobutyl ether, Triton X-100
(Dow Chemical) is a nonionic octlyphenol ethoxylate surfactant, Irgafos 168 (BASF) is di-tert-butyl(phenyl) phosphite, Irganox
1010 (BASF) is 3-(3,5-di-tert-butyl-4-hydroxyphenyl) propionate.
Based on the study results, 60% IPA was shown to be the worst-case model solvent for the extraction study because it extracted
all of the compounds of interest except for Irgafos 168 phosphate. Not only did the 60% IPA extract the same compounds as
the surfactants, it also extracted additional compounds that the surfactants did not extract. These results were comparable
to findings from a related study (5) performed in 2011 in which two types of bioprocessing bag films were submerged for seven
days at 40 °C/ambient relative humidity in various types and strengths of solvents, as shown in Figure 3. Compounds that were targeted in the related study included bis(2,4-di-tert-butyl)hydrogen phosphate, erucamide, palmitamide, stearamide, and Irgafos 168 phosphate. The study also showed that 60% IPA
had a greater extraction efficiency compared to the other solvents evaluated including 1% PS 20. Due to carryover issues associated
with the 1% PS solutions, 0.1% PS solutions were used to perform the study using PP caps.
Figure 3: Comparison of extractables of common polymer additives using 60% isopropanol (IPA) and 1% polysorbate 20 extraction
solvents in two types of polymeric bioprocessing bags; DTBHP is bis(2,4-di-tert-butyl) hydrogen phosphate, Irgafos 168 phosphate
is a common degradant of di-tert-butyl(phenyl) phosphite (Irgafos 168, BASF).
In addition to having greater extraction efficiency, as demonstrated in two separate studies using two different types of
material, 60% IPA was also shown to eliminate interferences observed in the sample chromatography of surfactants. The use
of extraction solvents that do not pose significant chromatographic interferences is critical so that potential extractable
compounds are not missed during the extractables screening. In this case, erucamide was extracted in all of the solvents and
was tracked using extracted ion analysis based upon the total ion chromatogram of the 60% IPA solvent. To perform extracted
ion analysis, the ion of interest must be known. If only the total ion chromatograms of the surfactant solvents were used
to screen for potential extractable compounds, and IPA was not used as one of the extraction solvents, erucamide would have
been missed in the chromatograms of the surfactants. Erucamide elutes at approximately 7.7 minutes in Figure 1.