Glass or Plastic? An HPLC Vial Compatibility Study for Dilute Benzalkonium Chloride (BAK) Solutions

Benzalkonium chloride (BAK) is a widely used preservative in pharmaceutical formulations. Structurally, BAK is a mixture of homologs, each containing the quaternary ammonium with an alkyl substitution of 8–18 carbons. The C12, C14, and C16 homologs represent the major components in commercial sources. Because of potential BAK surface adsorption, this study was initiated to identify suitable vials for high performance liquid chromatography (HPLC) analysis. The plastic polypropylene vial type was more compatible with the 25 ppm BAK formulation than the glass vial types tested. With many potential factors contributing to BAK adsorption, this study highlights the challenges of predicting adsorption behavior and the importance of conducting a vial compatibility assessment early in HPLC method development for dilute solutions of BAK and analytes with similar physicochemical properties.

Benzalkonium chloride (BAK) is a commonly used antimicrobial preservative in pharmaceutical dosage forms, particularly in ophthalmic drug products (1–3). Approximately 70% of the topical eye drops commercially available are preserved with BAK, typically at concentrations ranging from 0.003% to 0.02% (30–200 ppm) (3). Concerns about ocular tissue safety mean that low BAK concentrations are generally preferred in eye drops indicated for chronic conditions. Formulations containing as low as 0.001% (10 ppm) BAK have been developed with satisfactory preservative efficacy (4).

Structurally, BAK is a mixture of alkylbenzyldimethylammonium chlorides with the alkyl chains ranging from C8 to C18 (Figure 1), and the compendial pharmaceutical grades consisting mainly of the C12, C14, and C16 homologs (1,5). The relative amounts of these homologs impact the antimicrobial spectrum and activity of BAK (1,6,7). The amphiphilic nature of BAK homologs is essential for their antimicrobial activities (8). However, this property also renders BAK highly surface-active, causing it to interact with and adsorb onto anionic or hydrophobic surfaces of containers and process materials (9–12).

During a routine high performance liquid chromatography (HPLC) method development effort in the authors’ laboratory for BAK formulations at concentrations below 50 ppm, it was repeatedly observed that the BAK peak areas from the same sample vial steadily decreased with consecutive injections. Loss of basic drugs (positively charged) to HPLC glass vials because of ionic interactions has been reported in the past (13–17). It was hypothesized that the observed loss of BAK in glass vials might be because of a similar mechanism, with the percentage loss becoming significant and noticeable in dilute solutions. This study was initiated to evaluate the BAK compatibility with high-quality premium glass vials. In addition, silanized glass vials and plastic vials were included for a comprehensive understanding.

Materials and Methods

Materials

All formulation ingredients were purchased from Medisca: Benzalkonium chloride (BAK) solution (USP/NF grade), 50%; glycerin (USP/NF) grade; sodium phosphate monobasic anhydrous (USP/NF grade); sodium phosphate dibasic anhydrous (USP/NF grade); acetonitrile and 85% phosphoric acid, both HPLC grades, were purchased from Thermo Fisher Scientific. A Milli-Q Direct 8 system from Millipore Sigma was used to produce water for the preparation of BAK formulation and HPLC mobile phase.

Six types of HPLC vials from three vendors were included in this compatibility study. All vial types have similar physical dimensions but with different surface properties (Table I) (18). The vendor and product names have been omitted to avoid commercial bias. Vendor-A glass vials served as a positive control, as BAK loss was first observed in these vials. The other four types of glass vials were all premium grades with high purity and/or reduced surface activity. One type of plastic (polypropylene) vials was also included for comparison. 

BAK Solution and Vial Compatibility Test

A 25 ppm (µg/mL) BAK solution in isotonic pH 6.8 buffer was prepared and used for all HPLC vial compatibility experiments. The detailed composition is provided in Table II. For each vial compatibility test, 1 mL of the BAK solution was accurately transferred into the vial and immediately loaded onto the HPLC autosampler for analysis. The same vial sample was analyzed by four more injections at 15 min, 30 min, 45 min, and 60 min.  

HPLC Analysis

An isocratic HPLC method, modified from the USP monograph (5), was used for BAK analysis. HPLC system: 2050C-3D HPLC system (Shimadzu); stationary phase: Luna, 150 × 4.6 mm, 5-µm CN, 100 Å column (Phemomonex); column temperature: 40 °C; mobile phase (isocratic): 0.05% phosphoric acid in 60:40 water–acetonitrile; flow rate: 1 mL/min; injection volume: 20 µL; run time: 15 min; UV detection: 214 nm.

As a result of the BAK adsorption loss to HPLC vials, including standards, routine calibration or system suitability tasks were not performed in this study. It was established from other analytes that this HPLC system routinely met the system suitability requirement with < 2% relative standard deviation (RSD) for multiple injections of the same sample vial.

Results

The BAK homologs were well separately by the isocratic method, and a representative chromatogram is shown in Figure 2. As a result of the potential BAK adsorption loss to vials, the routine analysis of the calibration standards was not performed. The peak area was used directly to calculate the percent remaining of each homolog in the vial sample. Special efforts were made to ensure the first injection of each vial sample was made immediately following sample preparation. As shown in Figure 3, the initial (time 0) peak areas of the three BAK homologs, as well as the total, were consistent across all six vial types. This suggests that there was minimal BAK adsorption loss at time 0. In addition, the homolog distribution of this BAK source material was calculated as 57% for C12, 30% for C14, and 13% for C16, assuming the same UV response factor for all three homologs.

After the time 0 injection, each HPLC vial sample was monitored over 60 min with four consecutive injections. The percent remaining results of the BAK homologs are illustrated in Figure 4. As expected for the positive control (Vendor-A glass vials), the percent remaining of all BAK homologs decreased steadily over 60 min, reaching 96%, 81%, and 68% for C12, C14, and C16 at the end of the period. The other four glass vial types provided some improvement, but there was still 5–10% and 22–32% loss for C14 and C16 after 60 min. In contrast, the plastic polypropylene vials showed no loss of C12 or C14 and only 6% loss of C16 after 60 min.

Because C14 and C16 were present in smaller proportions than C12, the percent remaining of all homologs combined was calculated to assess the overall BAK loss. As shown in Table III, the formulation suffered more than 5% total BAK loss in all vial types except polypropylene after 60 min.

Discussion

Loss of basic drugs to HPLC vials has been reported in several studies and whitepapers in the past (13–17). This phenomenon has been attributed to ionic interactions between the positively charged drug molecules and the negatively charged silanol groups on the glass surface. Silanol groups are weak acids with pK_a values that vary depending on their substituents. On the surface of glass vials, the silanols are generally considered deprotonated and negatively charged at pH above 2.5 (19,20). Because the silanol adsorption sites on the vial surface can become saturated in each vial, the drug loss may not pose a significant issue for solutions of high drug concentrations. However, for dilute solutions, the same mass of drug loss represents a much higher percentage loss. The preliminary laboratory data described in the introduction suggested that the BAK adsorption to glass vials became problematic at sample concentrations below 50 ppm. The full data set from this current study confirmed the issue at 25 ppm BAK.

Most HPLC glass vials, including those examined in this study, are manufactured from USP Type I borosilicate glass (21). However, the requirements in the USP general chapter focus mainly on the tests for hydrolytic resistance and not on surface acidity from the silanol groups. It is known that surface properties of HPLC glass vials, including the amount of free silanol groups, vary substantially across different vendors and technical grades (15–17). The premium grades selected for this study were marketed as vials exhibiting low impurities and/or low surface activity. Silanization is a chemical treatment to deactivate the silanol group and introduce a hydrophobic barrier (18). Silanized HPLC vials have been shown to effectively minimize the adsorption of basic drugs and impurities (15).

BAK carries a permanent cationic charge on the quaternary ammonium group, independent of the formulation pH. As such, it was unsurprising that BAK exhibited adsorption loss to glass vials as a result of the ionic interactions with the silanol groups on the glass surface. This phenomenon was most pronounced with the low-cost glass vials from Vendor-A, and it did not seem to have reached saturation after 60 min. The premium glass vials (Vendor-B, Vendor-C L1, and Vendor-C L2) demonstrated some improvement, but they did not eliminate the BAK adsorption issue, suggesting the presence of residual silanols on the surface and/or additional adsorption mechanisms besides ionic interactions. The preferential loss of the C16 homolog over C14 and C12 also cannot be fully rationalized by the ionic interactions alone. Likewise, the silanized vials (Vendor-C) failed to eliminate adsorption loss, further indicating that ionic interactions are not the sole mechanism responsible for BAK loss. There might be hydrophobic interactions between the BAK alkyl groups and the siloxane groups on the glass surface (17). Another possible explanation is that the adsorbed BAK molecules may have formed a new hydrophobic layer on the vial surface for additional adsorption via hydrophobic interactions (22).

The most surprising result from this study was that the plastic polypropylene vials (Vendor-C) outperformed all the glass vials, exhibiting the least amount of BAK adsorption. This finding was unexpected, since BAK contains a long hydrophobic alkyl chain and was reported to adsorb to various microplastics, including polypropylene and polyethylene (11,12). There were also separate studies documenting BAK adsorption to hydrophobic filter membranes (9,10). Nonetheless, the small amount of C16 adsorption loss suggests the presence of some hydrophobic interactions between BAK and the polypropylene vial surface. It is possible that the surface area-to-BAK ratio in the HPLC vials may be much less significant than that in porous microplastic particles or filter membranes.

The homolog proportions of BAK from commercial sources can vary substantially because of the flexible ranges defined in the compendial monographs (1). For instance, the USP monograph of BAK solution specifies that the C12 and C14 content should constitute no less than 40% and 20% of the total, respectively, with their combined proportions being no less than 70% of the total (5). Consequently, the adsorption loss of C14 and C16 may or may not significantly impact the final potency results of the total. For the BAK source used in this study, the C14 and C16 homologs represented ~43% of total content initially. As a result, the total BAK potency loss after 60 min was significant (> 5%) in all vial types except polypropylene vials from Vendor-C (Table III). In a separate study using a BAK source consisting of 68% C12 and 32% C14, the total BAK potency loss under the same conditions was negligible in all vial types except the low-cost glass vials from Vendor-A (data not shown).

In addition to vial type and BAK source, formulation compositions can influence BAK adsorption loss, though it is beyond the scope of the current study. Given the numerous factors affecting BAK adsorption, accurate prediction remains challenging. Therefore, experimental testing is recommended to identify the compatible vial types for HPLC method development of new BAK formulations.

Conclusion

This study revealed that BAK is prone to adsorption loss in glass HPLC vials, including the premium grades with low surface silanol content or silanization treatment. In contrast, minimal adsorption loss was observed in the plastic polypropylene vial type evaluated. The study results highlighted the importance of performing a vial compatibility study early in the HPLC method development process for dilute solutions of BAK and analytes with similar physicochemical properties.

References

1. Handbook of Pharmaceutical Excipients, 9th ed.; Sheskey, P. J.; Hancock, B. C.; Moss, G. P.; Goldfarb, D. J. Eds.;Pharmaceutical Press: London, UK, Washington, D.C., USA, 2020; pp. 132–135.

2. https://eyewiki.org/Preservatives_in_Topical_Ophthalmic_Medications (accessed 2024-11-28).

3. Goldstein, M. H.; Silva, F. Q.; Blender, N.; Tran, T.; Vantipalli, S. Ocular Benzalkonium Chloride Exposure: Problems and Solutions. Eye 2022, 36 (2), 361–368. DOI: 10.1038/s41433-021-01668-x

4. Kumagami, T.; Wakiyama, H.; Kusano, M.; et al. Comparison of Corneal Safety and Intraocular Pressure-Lowering Effect of Tafluprost Ophthalmic Solution with Other Prostaglandin Ophthalmic Solutions. J. Ocul. Pharmacol. Ther. 2014, 30 (4), 340–345. DOI: 10.1089/jop.2013.0132

5. USP, USP Benzalkonium Chloride Solution Monograph. USP-NF. https://online.uspnf.com/uspnf (accessed 2024-11-26).

6. Euerby, M. R. High Performance Liquid Chromatography of Benzalkonium Chlorides—Variation in Commercial Preparations. J. Clin. Hosp. Pharm. 1985, 10 (1), 73–77. DOI: 10.1111/j.1365-2710.1985.tb00718.x

7. Yao, K.; Jiang, R.; Wang, P.; et al. Comparison of Aqueous and Non-Aqueous Capillary Electrophoresis for the Determination of Four Benzalkonium Chloride Homologues in Compound Chemical Disinfectants. Heliyon2024, 10 (11), e31797. DOI: 10.1016/j.heliyon.2024.e31797

8. McDonnell, G.; Russell, A. D. Antiseptics and Disinfectants: Activity, Action, and Resistance. Clin. Microbiol. Rev. 1999, 12 (1), 147–179. DOI: 10.1128/CMR.12.1.147

9. Prince, S. J.; Allen, L. V., Jr. Sorption of Benzalkonium Chloride to Various Filters Used in Processing Ophthalmics. Int. J. Pharm. Compd. 1998, 2 (3), 249–251.

10. Bin, T.; Kulshreshtha, A. K.; Al-Shakhshir, R.; Hem, S. L. Adsorption of Benzalkonium Chloride by Filter Membranes: Mechanisms and Effect of Formulation and Processing Parameters. Pharm. Dev. Technol. 1999, 4(2), 151–165. DOI: 10.1081/pdt-100101350

12. Kim, T. K.; Jang, M.; Hwang, Y. S. Adsorption of Benzalkonium Chlorides onto Polyethylene Microplastics: Mechanism and Toxicity Evaluation. J. Hazard. Mater. 2022, 426, 128076. DOI: 10.1016/j.jhazmat.2021.128076

(12) Jang, M. H.; Kim, T. K.; Park, J. W.; et al. Elucidating Adsorption Mechanisms of Benzalkonium Chlorides (BACs) on Polypropylene and Polyethylene Terephthalate Microplastics (MPs): Effects of BACs Alkyl Chain Length and MPs Characteristics. J. Hazard. Mater. 2024, 468, 133765. DOI: 10.1016/j.jhazmat.2024.133765

13. Shia, J.; Xu, J.; Murphy, B. P.; Chambers, E. E. Overcoming Glass Vial Adsorption Effects for Trace Analysis of Basic Compounds by LC-MS/MS. Poster #WP09, ASMS 2011. https://www.waters.com/webassets/cms/library/docs/720004027en.pdf (accessed 2024-12-04).

14. Nakashima, M.; Sato, Y.; Kuwahara, T.; et al. Evaluation of a Novel Glass Vial Overcoming Adsorption Effect for Pharmaceutical Drugs. 2015. https://lcms.labrulez.com/paper/9002 (accessed 2024-12-04).

15. Huang, Y.; Hu, Y.; Yuill, E. M.; et al. Circumventing Glass Vial and Diluent Effects on Solution Stability of Small Molecule Analytes during Analytical Method Development and Validation. J. Pharm. Biomed. Anal.2022, 213, 114676. DOI: 10.1016/j.jpba.2022.114676

16. Schaaf, L.; Sawazki, J.; Ingenrieth, K.; et al. Why Is Gold-Grade Glass Quality So Important When You Want to Detect Low Concentrations of Analytes? ThermoFisher Scientific White Paper 21833, 2022. https://www.fishersci.com/content/dam/fishersci/en_US/documents/programs/scientific/technical-documents/white-papers/glass-quality-white-paper.pdf (accessed 2024-12-10).

17. Murakoshi, M.; Fukuzawa, K.; Sato, Y.; Asakawa, N. Adsorption Phenomenon and Development of Low Adsorption Vials for LC and LC/MS. https://www.shimadzu.com/an/sites/shimadzu.com.an/files/pim/pim_document_file/applications/posters/11099/adsorption_phenomenon.pdf(accessed 2024-12-04).

18. ThermoFisher Scientific. Sample Handling Solutions. https://assets.thermofisher.com/TFS-Assets/CMD/brochures/br-21472-sample-handling-br21472-en.pdf (accessed 2024-12-04).

19. Duncan, M. R.; Lee, J. J.; Warchol, M. P. Influence of Surfactants upon Protein/Peptide Adsorption to Glass and Polypropylene. Int. J. Pharm. 1995, 120 (2), 179–188. DOI: 10.1016/0378-5173(94)00402-Q

20. Jednačak-Bišćan, J.; Pravdić, V.; Haller, W. Adsorption Phenomena on Glass Surfaces: IV. Surface Acidity of Silica and Glasses. J. Colloid Interface Sci. 1988, 121 (2), 345–354. DOI: 10.1016/0021-9797(88)90438-9

21. USP, USP General Chapter <660>, “Containers – Glass,” USP-NF, https://online.uspnf.com/uspnf (accessed 2024-11-26).

22. Rutland, M. W.; Parker, J. L. Surface Forces between Silica Surfaces in Cationic Surfactant Solutions: Adsorption and Bilayer Formation at Normal and High pH. Langmuir 1994, 10 (4), 1110–1121. DOI: 10.1021/la00016a024

Authors

Avery Stadler is an undergraduate student and researcher pursuing a degree in Biochemistry at Binghamton University, State University of New York. She previously interned in Dr. Fang Zhao’s laboratory in 2023 and 2024, where she conducted research in pharmaceutics.
Brett Seaman is currently an undergraduate student majoring in Pharmaceutical Chemistry at St. John Fisher University, in Rochester, New York.
Dr. Priyanka Bhatt is an Assistant Professor in the Department of Pharmaceutical Sciences at Wegmans School of Pharmacy, St. John Fisher University, in Rochester, New York.
Dr. Fang Zhao is an Emeritus Professor in the Department of Pharmaceutical Sciences at Wegmans School of Pharmacy, St. John Fisher University, in Rochester, New York.