(M1030-02-11) Nitrite Detection at Trace Levels: Insights from a Lactose Case Study on Ion Chromatography and Sonication Effects

Purpose: To mitigate the risk of nitrosamine formation in drug products containing vulnerable amines, it is crucial to control the nitrite levels in excipients. Most excipients contain nitrite, often in the ppm or ppb range (1). Therefore, it is essential to establish reliable methods for nitrite quantification at trace levels. Many methods struggle with low amounts in product matrices. This study addresses challenges in determining trace nitrite levels in lactose, a common excipient in tablets, aiming to quantify nitrite levels from 20-100 µg/kg.

Methods: In this study, several different Ion Chromatography (IC) method setups were tested for their ability to quantify nitrite at 20 µg/kg. This corresponds to a standard solution containing 1 µg/L nitrite and a sample solution containing 50 g/L lactose in ultrapure water. The IC system was tested with an isocratic carbonate system (AS14A) and hydroxide gradient systems with five different columns: AS11-HC, AS15, AS18, AS19, and AS30, Detection methods included suppressed conductivity and UV detection at 210 nm. Additionally, an IC method combined with post-column Griess derivatization and UV detection at 525 nm was tested. Chromatograms of the standard with 1 µg/L nitrite were compared against those of lactose solutions (50 g/L). The influence of sample preparation on nitrite levels was examined using the optimized post-column Griess derivatization method. Studies were conducted with pharmaceutical-grade lactose dissolved in ultrapure water. References included nitrite standard solutions and mixed anion standards containing various ions. For post-column derivatization experiments, Griess reagent (1% sulfanilamide and 0.1% N-1-naphthylethylenediamine in 5% phosphoric acid) was used. Sonication experiments employed an ultrasonic cleaner (45 kHz, 80 W), with ultrapure water dispensed on the analysis day (0.055-0.057 µS/cm, TOC ≤ 5 ppb).

Results: Nitrite peaks at 1 µg/L were detectable for all tested columns, with clear separation from other ions in the standard solution. However, in lactose solutions (50 g/L), low nitrite levels could not be quantified due to peak overlap (co-elution). This was observed for all columns and detection methods. Changes in gradient or flow rate (0.25 vs 0.38 ml/min) did not resolve the issue. Chromatograms for column AS19 with conductivity detection are illustrated in figure 1, showing standard solution results and overlays with lactose solution around expected nitrite retention times. Using an IC method with post-column Griess derivatization and UV-Vis detection, very low nitrite levels in lactose were detectable (figure 2). Estimated nitrite amount of 2 µg/kg in the analyzed lactose sample. Further research tested the impact of sample preparation using the sensitive post-column Griess derivatization method. It was found that sonication during sample preparation generated high nitrite levels (table 1). Sonicated lactose samples showing nitrite content of 960 ppm after 15 minutes and 369 ppb after 30 minutes. In contrast, standard mixing resulted in 3 ppb nitrite levels, calculated based on lactose (µg/kg). The same effect was observed for ultrapure water. Without sonication, nitrite could not be detected, whereas after 15 minutes, the content increased to 332 µg/L and 178 µg/L. Nitrite content for ultrapure water refers to the solution (µg/L). These effects are supported by literature on sonication chemistry (2-7), confirming nitrite and nitrate ion formation in aqueous solutions under sonication in an air atmosphere. Sonication generates cavitation microbubbles, responsible for ultrasound's radical-mediated chemical effects, with extreme temperatures and pressures of approx. 5000 K/500 bar in collapsing bubbles (7). Nitrite levels decreasing after extended sonication may be due to nitrite formation within gas bubbles, exiting the solution after bubble collapse (5).

Conclusion: Determining trace nitrite levels, even in simple products like lactose, is challenging. At ppb levels, results can be impacted by signal/peak co-elution, background nitrite, and sample preparation. Co-elution can be avoided with optimized IC methods combined with post-column Griess derivatization and UV-Vis detection. A significant insight from this work is the impact of sonication, commonly used in sample preparation, which may lead to erroneous results due to nitrite and nitrate generation through sonochemistry. Although the generation of nitrite and nitrate through sonochemistry is well established, so far, to the best of our knowledge, no data on this effect have been published in the context of nitrosamine analysis and nitrite testing. This is an important insight since sonication is commonly used for sample preparation and could lead to erroneous results.

References: 1) Boetzel, Ruth; Schlingemann, Joerg; Hickert, Sebastian; Korn, Christian; Kocks, Grace; Luck, Bert et al. (2023): A Nitrite Excipient Database: A Useful Tool to Support N-Nitrosamine Risk Assessments for Drug Products. In Journal of pharmaceutical sciences 112 (6), pp. 1615–1624. DOI: 10.1016/j.xphs.2022.04.016.
2) Ferkous, Hamza; Hamdaoui, Oualid; Pétrier, Christian (2023): Sonochemical formation of peroxynitrite in water: Impact of ultrasonic frequency and power. In Ultrasonics sonochemistry 98, p. 106488. DOI: 10.1016/j.ultsonch.2023.106488.
3) Kwedi-Nsah, Louis-Marly; Kobayashi, Takaomi (2020): Sonochemical nitrogen fixation for the generation of NO2- and NO3- ions under high-powered ultrasound in aqueous medium. In Ultrasonics sonochemistry 66, p. 105051. DOI: 10.1016/j.ultsonch.2020.105051.
4) Merouani, Slimane; Hamdaoui, Oualid; Kerabchi, Nassim (2021): On the sonochemical production of nitrite and nitrate in water: A computational study. In: Water Engineering Modeling and Mathematic Tools: Elsevier, pp. 429–452.
5) Okitsu, Kenji; Kunichika, Riki; Asada, Shota (2023): Quantitation and evaluation of NO2-, NO3-, and H2O2 in the sonolysis of aqueous NaOH solution under air and air-Ar mixture: Effects of solution temperature, ultrasonic power, and ratio of gas mixture. In Ultrasonics sonochemistry 100, p. 106612. DOI: 10.1016/j.ultsonch.2023.106612.
6) Supeno; Kruus, P. (2000): Sonochemical formation of nitrate and nitrite in water. In Ultrasonics sonochemistry 7 (3), pp. 109–113. DOI: 10.1016/s1350-4177(99)00043-7.
7) Suslick, Kenneth S.; Hammerton, David A.; Cline, Raymond E. (1986): Sonochemical hot spot. In Journal of the American Chemical Society 108 (18), pp. 5641–5642. DOI: 10.1021/ja00278a055.

Figure 1: IC-chromatogram (overlay) standard solution (1 µg/L nitrite) and lactose solution (50 g/L), AS19, conductivity detection

Figure 2: IC-chromatogram (overlay) standard solution (1 µg/L nitrite) and lactose solution (50 g/L), with UV detection at 525 nm after Griess derivatization

Table 1: Impact of sonication on measured nitrite content