Safety assessment of poly(2‐ethyl‐2‐oxazoline) for use in food contact materials

Abstract

The EFSA Panel on Food Contact Materials (FCM) assessed the safety of poly(2‐ethyl‐2‐oxazoline) intended to be used as an additive to manufacture repeated use filter membranes based on ■■■■■. The filter membranes are used to process aqueous and alcoholic foods, for up to 24 h per day up to ■■■■■ at temperatures up to 35°C. The substance is a polymer with ≤ ■■■■■% w/w low molecular weight fraction (LMWF, < 1000 Da). Migration tests simulated the actual use conditions by repeatedly flushing ■■■■■‐based filter membranes with the food simulant 20% ethanol in consecutive contact steps. The specific migration of the sum of the LMW oligomers was below 50 μg/kg food. The specific migration of the monomer and of some impurities was not detected at the respective limits of detection (≤ 4 μg/kg food). The specific migration of the remaining impurities was not measured but calculated and was generally below 0.15 μg/kg food. Genotoxicity studies on 2‐hydroxy‐N‐(hydroxyethyl)propanamide, which is structurally related to the LMW oligomers, demonstrated that the LMW oligomers do not raise a concern for genotoxicity. The in silico assessment of the impurities did not show the presence of structural alerts for genotoxicity. The Panel concluded that the substance is not of safety concern for the consumers if (i) it is used in repeated use ■■■■■‐based filter membranes at temperatures up to 35°C for the processing of foodstuffs for which simulants A, B, C are laid down in Regulation (EU) 10/2011, except human milk, liquid infant formula and water that could be used to reconstitute infant formula; and (ii) its content in the filter membrane does not exceed 40.7% w/w, calculated based on the levels of the substance and ■■■■■ in the initial wet membrane solution and considering a residual level of other components of 2% w/w.

1. INTRODUCTION

1.1. Background and Terms of Reference as provided by the requestor

Before a substance is authorised to be used in food contact materials (FCM) and is included in a positive list, EFSA's opinion on its safety is required. This procedure has been established in Articles 8, 9 and 10 of Regulation (EC) No 1935/2004 ¹ of the European Parliament and of the Council of 27 October 2004 on materials and articles intended to come into contact with food.

According to this procedure, the industry submits applications to the Member States' competent authorities which transmit the applications to the European Food Safety Authority (EFSA) for their evaluation.

In this case, EFSA received an application from the Belgian Federal Public Service (FPS) Health, Food Chain Safety and Environment, requesting the evaluation of the substance poly(2‐ethyl‐2‐oxazoline) with the CAS number 25805‐17‐8. The dossier was submitted on behalf of Solventum Purification Inc.

According to Regulation (EC) No 1935/2004 of the European Parliament and of the Council on materials and articles intended to come into contact with food, EFSA is asked to carry out an assessment of the risks related to the intended use of the substance and to deliver a scientific opinion.

2. DATA AND METHODOLOGIES

2.1. Data

The applicant has submitted a confidential and a non‐confidential version of the dossier in support of their application for the authorisation of ‘poly(2‐ethyl‐2‐oxazoline)’ to be used in plastic FCM.

Additional information was provided by the applicant during the assessment process in response to requests from EFSA sent on 9 October 2024 (see Section 5).

In accordance with Art. 38 of the Commission Regulation (EC) No 178/2002 ² and taking into account the protection of confidential information and of personal data in accordance with Articles 39 to 39e of the same Regulation and of the Decision of the EFSA's Executive Director laying down practical arrangements concerning transparency and confidentiality, ³ the non‐confidential version of the dossier is published on Open.EFSA. ⁴

According to Art. 32c(2) of Regulation (EC) No 178/2002 and to the Decision of EFSA's Executive Director laying down the practical arrangements on pre‐submission phase and public consultations, ⁵ EFSA carried out a public consultation on the non‐confidential version of the application from 7 March to 28 March 2025 for which no comments were received.

Data submitted and used for the evaluation are:

Non‐toxicological data and information

• Chemical identity;

• Description of manufacturing process of the substance/FCM;

• Physical and chemical properties;

• Intended use;

• Existing authorisation(s);

• Estimated migration of the substance;

• Identification, quantification and migration of oligomers, reaction products and impurities.

  Toxicological data

• Bacterial reverse mutation test on 2‐ethyl‐2‐oxazoline;

• Bacterial reverse mutation test and in vitro micronucleus tests on 2‐hydroxy‐N‐(2‐hydroxyethyl)propanamide;

• In vivo micronucleus test on p‐toluene sulfonic acid;

• In silico data and published literature on the genotoxicity potential of impurities.

2.2. Methodologies

The assessment was conducted in line with the principles laid down in Regulation (EC) No 1935/2004 on materials and articles intended to come into contact with food. This Regulation requires applicants to submit an application accompanied by a technical dossier containing the information specified in the guidelines for the safety assessment of a substance to be published by EFSA. In practice, the technical dossier should contain the information required in the EFSA Note for Guidance for the preparation of an application for the safety assessment of a substance to be used in food contact materials (EFSA AFC Panel, 2008) and relevant cross‐cutting documents from the EFSA Scientific Committee. The dossier submitted by the applicant was in line with these guidance documents.

The methodology is based on the characterisation of the substance that is the subject of the request for safety assessment prior to authorisation, its impurities and reaction and degradation products, the evaluation of the exposure to those substances through migration and the definition of minimum sets of toxicity data required for safety assessment.

To establish the safety from ingestion of migrating substances, the toxicological data indicating the potential hazard and the likely human exposure data need to be combined. Exposure is estimated from studies on migration into food or food simulants and considering that a person may consume daily up to 1 kg of food in contact with the relevant FCM.

As a general rule, the greater the exposure through migration, the more toxicological data are required for the safety assessment of a substance. Currently, there are three tiers with different thresholds triggering the need for more toxicological information as follows:

1. In case of high migration (i.e. 5–60 mg/kg food), an extensive data set is needed.

2. In case of migration between 0.05 and 5 mg/kg food, a reduced data set may suffice.

3. In case of low migration (i.e. < 0.05 mg/kg food), only a limited data set is needed.

More detailed information on the required data is available in the EFSA Note for Guidance (EFSA AFC Panel, 2008).

The assessment was conducted in line with the principles described in the EFSA Guidance on transparency in the scientific aspects of risk assessment (EFSA, 2009) and considering the relevant cross‐cutting guidance documents from the EFSA Scientific Committee.

3. ASSESSMENT

According to the applicant, the substance poly(2‐ethyl‐2‐oxazoline) (PEtOx) is a polymeric additive intended to be used as a wetting agent in the manufacture of repeated use filter membranes based on ■■■■■ The substance is intended to be used at up to ■■■■■% w/w in the wet formulation used to produce the filter membrane, resulting in a calculated maximum content of 40% w/w (see Section 3.1.2).

According to the applicant, the ■■■■■‐based filter membranes are a component of a cartridge filter unit in an inline bottling system made of ‘a single layer of microfiber nonwoven polypropylene and a single layer of ■■■■■’.

The contact conditions during the filtration process were reported to be up to 24 h per day and up to ■■■■■ per year at temperatures up to 35°C. According to the applicant, membranes may have a surface area as large as ■■■■■ m² and work at a flow rate of up to ■■■■■ L/min.

The membranes are used to filter liquid aqueous and alcoholic foods, such as beer, wine, hard seltzer, sodas and soft drinks. The applicant excluded the use for filtering bottled water and the contact with infant formulae and human milk.

The substance has not been evaluated in the past by the Scientific Committee on Food (SCF) or EFSA.

3.1. Non‐toxicological data

3.1.1. Identity of the substance ⁶

Chemical formula: [C₅H₉NO]_n

CAS no. 25805‐17‐8

Chemical structure:

The substance is a polymeric additive with a weight‐ and number‐averaged molecular mass of ■■■■■ and ■■■■■ Da, respectively. The low molecular weight fraction (LMWF, < 1000 Da) was up to ■■■■■ w/w, as measured by gel permeation chromatography (GPC). The LMW oligomers were up to ■■■■■ w/w, as measured by liquid chromatography‐mass spectrometry (LC‐MS).

PEtOx is manufactured by polymerisation of the monomer 2‐ethyl‐2‐oxazoline (EtOx) ■■■■■, using methyl p‐toluene sulfonate (MPTS) as an initiator. During the manufacturing process, MPTS is converted into p‐toluene sulfonic acid (PTSA).

According to the applicant, the purity was > 99.9%, calculated as 100% minus the fraction of unreacted EtOx and other impurities determined by LC‐MS. The level of EtOx in the substance was up to ■■■■■ mg/kg. MPTS was not detected at ■■■■■ mg/kg. The impurities N‐(2‐hydroxyethyl)propanamide (NHP) and PTSA were up to ■■■■■ and ■■■■■ mg/kg, respectively. Five other tentatively identified compounds (‘Compounds 1–5’) were four propanamides (■■■■■) and 2‐aminoethyl propanoate. They were found at up to ■■■■■ mg/kg, respectively.

The Panel noted that, based on the manufacturing process, MPTS will be fully converted into its hydrolysis product PTSA, which is expected to be largely removed by degradation/evaporation at the high temperatures used in the manufacturing process of PEtOx.

LMW oligomers were tentatively identified and quantified by LC‐MS. The applicant classified them into five groups based on the repeating unit and structure (Figure 1). The group ‘oligomer series 4’ consisted of eight oligomers with an MW between 200 and 900 Da and constituted ■■■■■% w/w of all LMW oligomers (i.e. up to ■■■■■% w/w in PEtOx). The Panel noted that the backbone of the identified oligomer groups is structurally related to the impurity NHP.

FIGURE 1.

The tentative structures of oligomer series 1–5 as identified by the applicant.

3.1.2. Composition of the filter membrane ⁷

According to the applicant, the filter membrane is manufactured starting from a wet membrane solution consisting of PEtOx, ■■■■■, ■■■■■ ■■■■■ at ■■■■■% w/w, respectively. This solution is cast onto a supporting material, forming a thin layer. Then, the cast material is intensively washed with water (to remove solvents and components that are at least partly water soluble) and dried to obtain the final membrane. The users of the filter membrane are recommended to flush it with water for 30 min before its first use.

The applicant did not measure the content of PEtOx in the filter membrane but reported a maximum content of 40% w/w. It was calculated by assuming that the preparation process would leave all PEtOx in the ■■■■■ membrane while almost completely removing the other components (residual level of 2% w/w). The Panel noted that the calculation results in a content of 40.7% w/w instead of 40% w/w. However, the applicant noted that, as PEtOx is soluble in water (Section 3.1.3), the actual levels of PEtOx in the filter membrane would be lower as a result of the various washing and flushing steps applied during manufacturing and before the first use. The applicant did not support this reasoning with data, but the Panel considered it plausible.

The applicant analysed the content of impurities in the filter membrane only for NHP and PTSA, i.e. the impurities with the highest levels in PEtOx (Section 3.1.1). These were not detected with an LC‐MS method having limits of quantification (LOQs) of ■■■■■ and ■■■■■ mg/kg membrane, respectively, defined as the lowest concentration of the calibration standards. The Panel expected that the production process of the membrane greatly reduced the content of water‐soluble impurities (by washing steps) and of volatile impurities (via the drying steps).

3.1.3. Physical and chemical properties

The substance melts above 300°C and decomposes at around 380–390°C in air and 400°C under N₂, based on thermogravimetric analysis (TGA).

The substance is reported by the applicant to be soluble in water and to dissolve in organic solvents, such as methanol and dioxane. It starts being hydrolysed above 60°C.

Therefore, the Panel concluded that the substance is thermally stable and resistant to hydrolysis under the intended application conditions (i.e. up to 35°C).

3.1.4. Specific migration ⁸

The applicant determined the specific migration simulating the actual contact conditions by carrying out repeated use migration tests. A cartridge filter unit containing filter membrane specimens cut from a commercial filter membrane (surface area = 1.40 dm²) was used. The applicant did not measure the actual PEtOx content in the filter membranes used for migration testing.

The filter membranes were flushed repeatedly (3 or 6 sequential flushing steps) with fresh 20% ethanol ⁹ at 40°C and a flow rate of 133 mL/min. For each step, the liquid filtered over 5 min (665 mL) was collected and analysed with LC‐MS to determine the specific migration of specific migrants. The filter membranes were tested as such or after a pretreatment step consisting of flushing the membranes with water at room temperature at 200 mL/min for 5, 20 or 30 min. The Panel considered this approach appropriate.

The Panel noted that the flow rate and the surface/volume ratio applied in the migration tests (0.133 L/min and 2.1 dm²/L, while > ■■■■■ L/min and < ■■■■■ dm²/L in real use) were worst case with respect to the migration potential.

3.1.4.1. Specific migration of EtOx, NHP, PTSA and MPTS

Two filter membranes (i.e. with or without a 5‐min pretreatment step) were tested with three sequential flushing steps. EtOx, NHP and PTSA were not detected in any of the steps at limits of detection (LODs) of 2.9, 4.0 and 2.7 μg/kg food, respectively. MPTS was not detected with a method having an LOQ of 0.15 μg/kg food, defined by the lowest concentration of calibration standards.

The specific migration of the five compounds that were tentatively identified in the compositional analysis (i.e. compounds 1–5: four propanamides and 2‐aminoethyl propanoate; Section 3.1.1) was calculated by the applicant pro‐rata to the migration of NHP. This approach was considered acceptable by the panel. Compounds 1–3 and 5 were estimated to potentially migrate below 0.15 μg/kg food. The specific migration of compound 4 was further refined by the applicant by migration modelling, resulting in < 0.5 μg/kg food.

A summary of the migration results is provided in Table 1.

TABLE 1.

Migration of the substances addressed in this evaluation.

Substance	Specific migration (μg/kg food)
EtOx	< 2.9 a
MPTS	< 0.15 b
PTSA	< 2.7 a
NHP	< 4.0 a
■■■■■ (compound 1)	< 0.018 c
■■■■■ (compound 2)	< 0.043 c
■■■■■ (compound 3)	< 0.082 c
■■■■■ (compound 4)	< 0.5 d
2‐aminoethyl propanoate (compound 5)	< 0.13 c
LMW oligomers – series 4 oligomers (individual)	< 20a

^a Not detected at this limit of detection.

^b Not detected at this limit of quantification.

^c Calculated pro‐rata to the migration of NHP.

^d Calculated pro‐rata to the migration of NHP and refined by migration modelling.

3.1.4.2. Migration of LMW oligomers

The migration of the LMW oligomers was tested with two filter membranes (membrane 1 and 2, pretreated at 200 mL/min for 20 or 30 min, respectively) using six sequential flushing steps with 20% ethanol. The filtered solutions were analysed for the migration of (i) non‐volatiles, (ii) nitrogen‐containing substances and (iii) series 4 oligomers (i.e. the most abundant LMW oligomer group, ■■■■■% w/w).

1. The migration of non‐volatiles was measured gravimetrically and decreased from 4.4 mg/L in step 1 (both membranes) to 0.63 mg/L and 0.93 mg/L in step 3 (membrane 1 and 2, respectively). It remained below 1 mg/L in all three following steps. The Panel considered these values as indicative for the overall migration during repeated use.

2. The migration of nitrogen‐containing species was measured by combustion followed by chemiluminescence nitrogen detection. As the level of ■■■■■ was measured separately by LC‐MS and subtracted, the results encompassed the sum of the N‐containing impurities and the oligomers. The Panel considered the results as indicative for the oligomer migration.

   For membrane 1, in the six sequential flushing steps, the migration was 2700, 86, 27, 8, 37 and 4 μg/kg food. For membrane 2, it was 4800 μg/kg food in step 1, decreased to around 100 μg/kg food in step 2–3, followed by scattering results between 230 and 460 μg/kg food in steps 4–6. The Panel considered that the high values for step 1 (2700 and 4800 μg/kg food) were likely due to compounds that were not part of the LMWF, possibly released from the membrane surface due to the filter preparation and assembly (e.g. the cut of the membrane from a larger commercial membrane, Section 3.1.4). Therefore, only the results of steps 2–6 were considered to represent the migration of the LMWF. Taking into account the expected stability of PEtOx (Section 3.1.3) and of the ■■■■■‐based membrane under the testing conditions, the Panel considered that the trend of increasing migration observed for membrane 2 cannot be attributed to an actual increase in the migration of the LMWF substances. Hence, only results from membrane 1 (step 2–6) were considered for the evaluation of LMWF migration.

3. For every flushing step and for both membranes, none of the oligomers of the ‘oligomer series 4’ group was detected at an LOD of 20 μg/kg food by LC‐MS. Assuming that all eight series 4 oligomers migrated at the LOD, their total potential migration could be 160 μg/kg food. The Panel considered this as a very conservative estimate, as the migration of the total of the nitrogen‐containing species (ii), which include all LMW oligomers, was reported as < 50 μg/kg food in the third flushing step. Therefore, the Panel concluded that the migration of the sum of all LMW oligomers is below 50 μg/kg food.

3.2. Toxicological data ¹⁰

The migration of the monomer, of the eight impurities and of the LMW oligomers was not detected at LODs below 0.05 mg/kg food. Therefore, in accordance with the ‘EFSA Note for Guidance’ (EFSA AFC Panel, 2008), the applicant only submitted data to assess their genotoxic potential.

3.2.1. Genotoxicity

No genotoxicity studies were provided on the LMWF of PEtOx, i.e. the oligomeric fraction and the impurities, which is the toxicologically relevant fraction for the genotoxicity assessment (EFSA Scientific Committee, 2019). Instead, the applicant submitted literature data on a PEtOx polymer, ¹¹ on EtOx and two unpublished genotoxicity studies on 2‐hydroxy‐N‐(2‐hydroxyethyl)propanamide as source chemical for read‐across to the impurity NHP and to the LMW oligomeric fraction. The applicant submitted in silico data for the other impurities having an estimated migration in food above 0.15 μg/kg food, i.e. the migration level leading to a potential exposure above the threshold of toxicological concern (TTC) of 0.0025 μg/kg body weight (bw) per day (EFSA Scientific Committee, 2012). For one of these impurities (PTSA), a published mutagenicity study was also submitted.

3.2.1.1. 2‐Ethyl‐oxazoline (EtOx)

In the framework of a study on the biocompatibility of poly(ethyleneimine), EtOx was tested in a limited bacterial reverse mutation (Ames) test with Salmonella Typhimurium TA98 and TA100 (Wiegand et al., 2013). It was tested over the dose range of 5.1–5000 μg/plate without metabolic activation and 5.1–510 μg/plate with metabolic activation, using the preincubation and plate incorporation procedures, respectively. The results, reported graphically, were evaluated as negative, based on the lack of dose‐related and/or twofold increases in revertant colonies. The Panel noted the reduced set of tester strains used in this study and that the dose range tested in the presence of metabolic activation was not selected based on the toxicity of the test compound but on that of some polyethyleneimines tested in parallel. Based on these limitations, the study was evaluated as of insufficient reliability and the results of low relevance.

However, the Panel also noted that the in silico assessment of EtOx with the OECD Quantitative Structure‐Activity Relationship (QSAR) Toolbox did not identify any relevant alerts for genotoxicity. Overall, the Panel concluded that EtOx does not raise concern for genotoxicity.

3.2.1.2. Other PEtOx impurities

The specific migration of the impurities MPTS, ■■■■■ (compound 1), ■■■■■ (compound 2), ■■■■■ (compound 3) and 2‐aminoethyl propanoate (compound 5) was estimated to be below 0.15 μg/kg food. Therefore, these five impurities do not need to be further addressed. The Panel noted that this also applies to the potentially genotoxic substance MPTS, ¹² as the risk posed by genotoxic substances is considered negligible below 0.15 μg/kg food.

The migration of the remaining three impurities, PTSA, NHP and ■■■■■ (compound 4), was estimated above 0.15 μg/kg food and below 50 μg/kg food. Therefore, their genotoxicity potential was addressed by the Panel.

The applicant submitted a genotoxicity study only for PTSA. PTSA was reported negative in an in vivo micronucleus test in ICR mice following intraperitoneal administration at a single dose (30 mg/kg body weight (bw), corresponding to half of the median lethal dose, LD50) given to groups of six male mice 36, 48 or 60 h before study termination. No increase in micronucleated reticulocytes was observed compared to controls. Although no information on polychromatic to normochromatic erythrocytes (PCE/NCE) ratio was provided, the Panel noted that PTSA is predicted to be bioavailable based on QSAR and that the test conditions (half of LD50 by intraperitoneal exposure) would plausibly lead to systemic exposure. It considered the study reliable with restrictions because of the non‐oral route of administration and limited reporting, and the negative results of limited relevance.

No other genotoxicity studies were submitted on PTSA, NHP and compound 4. An in silico assessment was performed using knowledge‐based (OECD QSAR Toolbox) and statistical (Leadscope) models. However, no reliable information on potential genotoxicity was provided by the Leadscope models and, hence, this section only reports the results from the OECD QSAR Toolbox. Regarding ■■■■■, no relevant structural alerts were identified by the general mechanistic and end‐point‐specific profilers. Regarding compound 4, no relevant structural alerts were identified by end‐point‐specific profilers. One general mechanistic profiler (DNA binding by OECD) that was highlighted for compound 4 and the oligomers was the presence of ■■■■■ as a potentially alerting structure for the formation of an ■■■■■ following oxidation. In this respect, the Panel noted that general mechanistic profilers mainly provide supporting mechanistic information in case of positive predictions of end‐point‐specific profilers, while the significance of such predictions in isolation is limited as unspecific and only broadly associated with molecular structure. Regarding NHP, no relevant structural alert for genotoxicity was identified with the OECD QSAR Toolbox. An alert for in vivo genotoxicity (H‐acceptor‐path3‐H‐acceptor) was identified by the ISS profiler, but it was disregarded as demonstrated to be devoid of predictive value (Aljallal et al., 2024; Pradeep et al., 2021). Adequate negative genotoxicity data were available for the related substance 2‐hydroxy‐N‐(2‐hydroxyethyl)propanamide (see Section 3.2.1.3), which was proposed by the applicant as source substance for read‐across. The Panel considered that the two substances presented sufficient structural similarity to support a read‐across.

3.2.1.3. 2‐Hydroxy‐N‐(2‐hydroxyethyl)propanamide as a source substance for read‐across

The applicant submitted an Ames and an in vitro micronucleus test on this source substance to support the read across to the targets NHP and LMW oligomers of PEtOx. The studies were carried out in compliance with GLP and following the OECD TGs 471 (OECD, 2020) and 487 (OECD, 2016), respectively.

In the Ames test (Verspeek‐Rip, 2004), a sample of 2‐hydroxy‐N‐(2‐hydroxyethyl)propanamide (purity 95%) was dissolved in water and tested using the plate incorporation procedure with S. Typhimurium (TA98, TA100, TA1535 and TA1537) and E. coli WP2uvrA in two separate tests without and with S9 mix (with 5% and 10% v:v S9 in the first and second test, respectively). The following concentrations were tested using triplicate plates. First test: 0, 3, 10, 33 (TA 100 and WP2 only), 100, 333, 1000, 3333 and 5000 μg/plate (all strains), equivalent to 0, 3, 10, 33, 94, 310, 940, 3140 and 4720 μg/plate. Second test: 100, 333, 1000, 3333 and 5000 μg/plate, equivalent to 94, 310, 940, 3140 and 4720 μg/plate (all strains). No increase in revertant colonies and no precipitation of the test item or signs of toxicity were observed at any dose, with or without S9 mix. The study was evaluated as reliable without restrictions and the negative results of high relevance.

In the vitro micronucleus test (Roy, 2019), a sample of 2‐hydroxy‐N‐(2‐hydroxyethyl)propanamide (purity 98.8%) was tested in human peripheral blood lymphocytes using the cytokinesis‐block procedure. The substance was dissolved in water and tested in duplicated cultures at 250, 500, 750 and 1330 μg/mL (10 mM, maximum recommended concentration), with both short (4 h with and without S9 mix) and extended (24 h without S9 mix) treatments. Micronuclei were scored in 2000 (1000/culture) binucleated cells per concentration; the toxicity of the treatment was evaluated by the Cytokinesis Block Proliferation Index (CBPI) in 1000 cells (500/culture). No increase of binucleated cells with micronuclei and no signs of cytotoxicity (as decreased CBPI) were observed in treated cells at any dose, with or without S9 mix. The study was evaluated as reliable without restrictions, and the negative results of high relevance.

Supplementary information was reported from a previous evaluation of the substance by the EFSA CEF Panel (EFSA CEF Panel, 2010) as a food flavouring. This included a negative in vivo micronucleus test in mice given 2000 mg 2‐hydroxy‐N‐(2‐hydroxyethyl)propanamide/kg bw via gavage (limit dose for non‐toxic chemicals), with no change in the PCE/total erythrocyte ratio as an indicator of bone marrow toxicity. The Panel noted that, according to current criteria for the evaluation of in vivo micronucleus test results (OECD TG 474, 2014), the reported results cannot be considered clearly negative owing to the absence of evidence of toxicity to the bone marrow. An expert judgement should be applied for the evaluation of the results. The Panel noted that 2‐hydroxy‐N‐(2‐hydroxyethyl)propanamide did not elicit any toxicity in vitro tests up to the maximum recommended concentration (10 mM). Therefore, no toxicity in bone marrow is expected to be elicited by the chemical under in vivo test conditions as well, despite its predicted bioavailability . The Panel, therefore, considered the study reliable with restrictions, and the results of limited relevance. Overall, based on adequate negative data in the Ames and in vitro micronucleus tests, the Panel concluded that 2‐hydroxy‐N‐(2‐hydroxyethyl)propanamide does not raise a concern for genotoxicity.

3.2.1.4. LMW oligomers of PEtOx

No genotoxicity data were provided on PEtOx LMW oligomers. Five groups of PEtOx oligomers with a backbone structurally related to NHP were identified via LC‐MS (Section 3.1.1). The end groups did not bear additional alerting structures for genotoxicity, including the piperazine ring present in series 3 and the unsaturation in series 1. The Panel noted that adequate negative in vitro genotoxicity data (Ames and micronucleus test) were available for 2‐hydroxy‐N‐(2‐hydroxyethyl)propanamide. By read‐across, the structurally related NHP and hence the LMW oligomers were also considered not to raise concern for genotoxicity.

3.2.1.5. Concluding remarks on genotoxicity

No indication of potential genotoxicity was provided by the in silico assessment of the monomer EtOx and the three impurities for which the potential migration was determined to potentially exceed 0.15 μg/kg food.

All LMW oligomers were structurally related to the impurity NHP, with no additional alerting structures in their end groups. As adequate negative genotoxicity data were available for the related substance 2‐hydroxy‐N‐(2‐hydroxyethyl)propanamide, the Panel concluded by read‐across that the PEtOx LMWF does not raise concern for genotoxicity.

Overall, the FCM Panel concluded that the LMW oligomeric fraction and the impurities present in PETOx do not raise concern for genotoxicity.

3.3. Discussion

The maximum content of the substance in continuously used ■■■■■‐based filter membranes to process liquid aqueous and alcoholic foods was calculated as 40.7% w/w. According to the applicant, the use of the filter membrane is the last step of an inline bottle filling system.

Based on the description of the preparation process of the filter membrane, which involves intensive water washing steps, it can reasonably be assumed that the water‐soluble impurities (such as EtOx and PTSA) will be largely removed. The same is expected for part of the LMW oligomers. The recommended water flushing at the customer's facility before the first use may further reduce their residue levels.

The applicant investigated the potential migration simulating the actual contact conditions by a repeated sequential flushing of filter membranes cut from commercial ■■■■■‐based filter membranes. Membranes were flushed with 20% ethanol (simulant C in Reg. (EU) 10/2011) at a flow rate of 133 mL/min at 40°C, then the liquid filtered over 5 min was collected and analysed. The Panel considered this approach appropriate. It also considered that the migration from this membrane into 20% ethanol covers the migration of the substance into foodstuff for which simulants A and B ¹³ are laid down in Reg. (EU) 10/2011.

The monomer EtOx and the other impurities were not detected at LODs at or below 4 μg/kg food. The migration of the potentially genotoxic impurity MPTS (which was not detected at a level of 18 mg/kg in the substance) was not detected at an LOD below 0.15 μg/kg food. The migration of each series 4 oligomer (i.e. the most abundant LMW oligomer fraction, ■■■■■% w/w of the total LMW oligomers) was not detected at an LOD of 20 μg/kg food. Based on the results from the migration of nitrogen‐containing species, the Panel estimated that the total migration of all LMW oligomers would be below 50 μg/kg food.

The actual content of PEtOx in the filter membrane is expected to be lower due to the washing and drying steps applied during the filter membrane preparation and the flushing before use. Therefore, the Panel noted that an actual 40.7% w/w content would, in principle, trigger a migration of impurities and LMW oligomers higher than that reported in Section 3.1.4. Nevertheless, considering the LODs and LOQs in the low μg/kg food range and taking into account the preparation procedure of the membrane, the potential migration from a membrane with 40.7% w/w PEtOx is expected to remain within the same range as that reported in Section 3.1.4.

As the specific migration of the LMW oligomers and impurities was not detected at LODs below 0.05 mg/kg food in migration tests or determined to be below 0.05 mg/kg food by pro‐rata or migration modelling calculations, only genotoxicity was considered for the safety assessment. The Panel concluded that the impurity NHP does not raise concern for genotoxicity. The same conclusion applies to the LMW oligomers, as their repeating unit is structurally similar to NHP. The oligomers' end groups do not bear alerting structures for genotoxicity. The in silico assessment of the impurities carried out with the OECD QSAR toolbox did not show relevant structural alerts for genotoxicity. Overall, based on the available data, the Panel concluded that the PEtOx LMWF does not raise a genotoxicity concern.

The applicant reported information only on the use of the substance in ■■■■■‐based filter membranes with a maximum content recalculated by the Panel of 40.7% w/w. Therefore, to cover the risk related to the exposure to oligomers and impurities, the Panel considered it necessary to restrict the uses of the substance to the reported content and membrane type.

The applicant excluded the use of the membrane in contact with water that could be used to reconstitute infant formula (such as powdered formula). Therefore, this use was not considered in this evaluation.

4. CONCLUSIONS

Based on the above‐mentioned data, the FCM Panel concluded that the substance poly(2‐ethyl‐2‐oxazoline) is not of safety concern for the consumers if:

• ‐it is used in repeated use ■■■■■‐based filter membranes at temperatures up to 35°C for the processing of aqueous and alcoholic foodstuff for which simulants A, B, C are laid down in Regulation (EU) 10/2011, except human milk, liquid infant formula and water that could be used to reconstitute infant formula; and
• ‐the content of poly(2‐ethyl‐2‐oxazoline) in the filter membrane, calculated based on the levels of the substance and ■■■■■ in the initial wet membrane solution and considering a residual level of the other components of 2% w/w, does not exceed 40.7% w/w.

5. DOCUMENTATION AS PROVIDED TO EFSA

Dossier on poly(2‐ethyl‐2‐oxazoline) as a new substance to be used in plastic for food contact uses. June 2024. Submitted by Solventum Purification Inc.

Additional data, October 2025. Submitted by Solventum Purification Inc.

ABBREVIATIONS

bw

body weight

CAS

Chemical Abstracts Service

CBPI

Cytokinesis Block Proliferation Index

CEF Panel

EFSA Panel on Food Contact Materials, Enzymes, Flavourings and Processing Aids

Da

Dalton

EtOx

2‐ethyl‐2‐oxazoline

FCM Panel

food contact materials Panel

FCM

food contact materials

GLP

good laboratory practice

GPC

gel permeation chromatography

LC‐MS

liquid chromatography – mass spectrometry

LMW

low molecular weight

LMWF

low molecular weight fraction

MPTS

methyl p‐toluene sulfonate

MW

molecular weight

NCE

normochromatic erythrocytes

NHP

N‐(2‐hydroxyethyl)propanamide

OECD

Organisation for Economic Co‐operation and Development

PCE

polychromatic erythrocytes

PEG

poly(ethylenglycol)

■■■■■

■■■■■

PEtOx

poly(2‐ethyloxazoline)

Po/w

octanol/water partition coefficient

PTSA

p‐toluene sulfonic acid

QSAR

Quantitative Structure‐Activity Relationship

SCF

Scientific Committee on Food

SML

specific migration limit

TGA

thermogravimetric analysis

TTC

Threshold of toxicological concern

REQUESTOR

Belgium competent authority

QUESTION NUMBER

EFSA‐Q‐2024‐00224

COPYRIGHT FOR NON‐EFSA CONTENT

EFSA may include images or other content for which it does not hold copyright. In such cases, EFSA indicates the copyright holder and users should seek permission to reproduce the content from the original source.

PANEL MEMBERS

Riccardo Crebelli, Maria Joao da Silva, Konrad Grob, Claude Lambré, Evgenia Lampi, Maria Rosaria Milana, Marja Pronk, Gilles Rivière, Mario Ščetar, Georgios Theodoridis, Els Van Hoeck, Nadia Waegeneers.

LEGAL NOTICE

Relevant information or parts of this scientific output have been blackened in accordance with the confidentiality requests formulated by the applicant pending a decision thereon by EFSA. The full output has been shared with the European Commission, EU Member States (if applicable) and the applicant. The blackening may be subject to review once the decision on the confidentiality requests is adopted by EFSA and in case it rejects some of the confidentiality requests.

EFSA FCM Panel (EFSA Panel on Food Contact Materials) , Lambré, C. , Crebelli, R. , da Silva, M. J. , Grob, K. , Lampi, E. , Milana, M. R. , Pronk, M. , Ščetar, M. , Theodoridis, G. , Van Hoeck, E. , Waegeneers, N. , Cariou, R. , Castle, L. , Di Consiglio, E. , Franz, R. , Barthélémy, E. , Comandella, D. , & Rivière, G. (2026). Safety assessment of poly(2‐ethyl‐2‐oxazoline) for use in food contact materials. EFSA Journal, 24(3), e9932. 10.2903/j.efsa.2026.9932