Determination of nonylphenol ethoxylates and octylphenol ethoxylates in environmental samples using [.sup.13.C]-labeled surrogate compounds.

ABSTRACT

Alkylphenol polyethoxylates (APEOs) have been widely used as nonionic surfactants in a variety of industrial and commercial products. Typical compounds are nonylphenol polyethoxylates (NPEOs) and octylphenol polyethoxylates (OPEOs), which serve as precursors to nonylphenol (NP) and octylphenol (OP), respectively. NP and 4-t-OP are known to have endocrine disrupting effects on fish (medaka, Oryzias latipes), so it is important to know the concentrations of APEOs in the environment. Because the analytical characteristics of these compounds depend on the length of the ethoxy chain, it is necessary to use appropriate compounds as internal standards or surrogates. We synthesized two [.sup.13.C]-labeled surrogate compounds and used these compounds as internal standards to determine NPEOs and OPEOs by high-performance liquid chromatography (LC)-mass spectrometry. Method detection limits were 0.015 [micro]g/L for NP (2)EO to 0.037 [micro]g/L for NP(12)EO, and 0.011 [micro]g/L for OP(3,6)EO to 0.024 [micro]g/L for OP (4)EO. NPEO concentrations in water from a sewage treatment plant were less than 0.05-0.52 [micro]g/L for final effluent and 1.2-15 [micro]g/L for influent. OPEO concentrations were less than 0.05-0.15 [micro]g/L for the final effluent and less than 0.05-1.1 [micro]g/L for influent.

INTRODUCTION

Alkylphenol polyethoxylates (APEOs) have been widely used as nonionic surfactants in a variety of industrial and commercial products and are chiefly composed of nonylphenol polyethoxylate (NPEO) and octylphenol polyethoxylate (OPEO) at a ratio of approximately 4:1. (1) Production in Japan of APEOs in 2003 comprised 18,000 t of NPEO and 1700 t of OPEO. (2) Nonylphenol (NP) and octylphenol (OP), which are used to manufacture these surfactants, have been shown to have endocrine disrupting effects on fish (medaka, Oryzias latipes). (3-6) OP did not affect the fish below a concentration of 0.992 [micro]g/L (predicted no effect concentration; PNEC), and because the concentration of OP in domestic environmental water is only approximately 0.03 [micro]g/L (predicted environmental concentration; PEC), there appeared to be little risk of fish experiencing endocrine disruption from this compound. However, the PNEC of NP is 0.608 [micro]g/L, and this concentration has been exceeded in actual environmental samples (less than approximately 0.0321 [micro]g/L). (7-9) Therefore, there remains the possibility of NP detrimentally influencing ecosystems.

Two degradation pathways are known for APEOs in aquatic environments. The degradation pathways for NPEO are shown in Figure 1 as an example. (10,11) The first (Pathway A) involves the gradual degradation of NP(n)EO to NP(n-1)EO (n: length of ethoxy [EO] chains), and so on to NP. The second pathway (Pathway B) proceeds through the formation of nonylphenol ethoxy acetic acid (NPEC). From recent findings, it appears that some NPEOs are not only precursors of NP but are, themselves, endocrine disrupting chemicals; NP(2)EO, and NP(1)elemental carbon are known to be estrogenic, although the higher ethoxymers of NP(n)EO lack estrogenic activity. (12-15) Therefore, investigations of the distribution and concentration of these compounds in the environment are necessary for predicting endocrine disrupting effects. (16-19) In an initial investigation, a high-performance liquid chromatography (LC)/fluorescence detection method was used to measure NP(n)EO(n = 1-4, n [greater than or equal to] 5). (20) In recent years, separation and quantitative determination of the compounds with each ethoxy chain length has become possible through high-performance LC-mass spectrometry, and thus the distribution of NP(n)EOs can be investigated in detail. The analytical characteristics (e.g. recovery during cleanup, chromatographic behavior, ionization efficiency) of the NP(n)EOs differ with the length of the ethoxy chain. We therefore synthesized two [.sup.13.C]-labeled surrogate compounds and developed a method for determining NPEOs and OPEOs by high-performance LC-mass spectrometry using these surrogate compounds. The generic structural formula of [.sup.13.C]-labeled NP ethoxylates is shown in Figure 2. We aimed to measure NPEOs with 1-15 ethoxyl units in the ethoxy chains, and OPEOs with 1-10 ethoxyl units.

[FIGURE 1 OMITTED]

EXPERIMENTAL PROCEDURES

Reagents and Chemicals

Although two surrogate reagents for APEOs, [.sup.13.C.sub.2]-NP(1)EO and [.sup.13.C.sub.2]-NP(2)EO, are commercially available, they were not suitable for those APEOs that have longer ethoxy chains, and, additionally, it is preferable that four or more [.sup.13.C] atoms are included to avoid isotopic interferences in LC-mass spectrometry analysis. So we synthesized two compounds, NP(8)EO-[.sup.13.C.sub.4] and NP(10)EO-[.sup.13.C.sub.4] for analysis of APEOs with longer ethoxy chains. The synthetic route for NP(8)EO-[.sup.13.C.sub.4] is shown in Figure 3.

[FIGURE 2 OMITTED]

A dry tetrahydrofuran solution containing NP(6)EO (1.9 g, 3.92 mmol) was added dropwise to a dry tetrahydrofuran solution containing 60% sodium hydride (221 mg, 5.52 mmol). After stirring this mixture for 1 hr in an ice bath, a dry tetrahydrofuran solution of ethyl bromoacetate-[.sup.13.C.sub.2] (729 mg, 4.32 mmol; Aldrich) was added and the reaction mixture was stirred for 6 hr at room temperature. Cooled water was then carefully added dropwise, followed by ethyl acetate, and the products were partitioned between the two phases. The ethyl acetate phase was separated, dried and concentrated, and yielded Compound A after purification by column chromatography. A solution of Compound A (957 mg, 1.67 mmol) in dry tetrahydrofuran was then added dropwise to a cooled (ice bath) solution of lithium aluminum hydride (102 mg, 2.67 mmol) in dry tetrahydrofuran (6 mL), and the resultant mixture was stirred for 2.5 hr at room temperature. Cooled water was then added dropwise to the reaction mixture; extraction with ethyl acetate and column chromatography yielded Compound B [NP(7)EO-[.sup.13.C.sub.2]]. By repeating these operations, the target compound NP(8)EO-[.sup.13.C.sub.4] was obtained. For the synthesis of NP(10)EO-[.sup.13.C.sub.4], NP(8)EO was used as the starting compound. These synthesized reagents are now available from Sumika Chemical Analysis Service, Ltd. Table 1 shows the list of standards and [.sup.13.C]-labeled surrogate compounds.

[FIGURE 3 OMITTED]

Acetonitrile and methanol were high-performance LC grade from Kanto Chemical Co. Inc. Stock solutions of NP(n)EO and OP(n)EO were prepared in acetonitrile, and working solutions were prepared from the stock solution by appropriate dilutions in acetonitrile.

[FIGURE 4 OMITTED]

Apparatus and Procedures

Each water sample was collected in a brown glass bottle with a metallic screw lid lined with polytetrafluoroethylene. The bottle was cleaned with methanol and acetone and dried before sampling. After collection, the sample was analyzed immediately, or when this was not possible, samples were stored at 4 [degrees]C until analysis. The samples comprised influent and final effluent of the city type sewage treatment plant (STP) collected as spot samples on April 10, 2003, at 11:00.

Figure 4 shows the flowchart of the analysis method of NP(n)EO and OP(n)EO, which involved solid phase extraction, partial purification by gel permeation chromatography (GPC), and quantification by LC-mass spectrometry. Thus, to 400 mL of liquid sample were added 250 ng of NP(1)EO-[.sup.13.C.sub.2] and NP(2)EO-[.sup.13.C.sub.2], and 100 ng of NP(8)EO-[.sup.13.C.sub.4] and NP(10)EO-[.sup.13.C.sub.4]. Larger amounts of NP(1)EO-[.sup.13.C.sub.2] and NP(2)EO-[.sup.13.C.sub.2] were needed to reduce the influence of the natural isotopes in the target compounds. Samples were filtered through a glass fiber filter (GF/F Whatman). In the case of water samples from the STP influent, more filters were needed than for the river water samples. For example, the influent of the wastewater treatment plant needed eight filters for a 400-mL sample, whereas only one filter was required for river water. Filters and suspended solids were extracted with acetone (3 x 5 mL) using an ultrasonic bath. The extract was concentrated by rotary evaporation to approximately 1 mL and added to the filtered sample, which was then loaded onto a solid phase disk (Empore TM Disk C18 ff 47 mm[phi], 3M) previously conditioned with methanol (10 mL) and purified water (10 mL). The target compounds were extracted from the solid phase disk with methanol (10 mL); the extract was concentrated by [N.sub.2] purging at 40 [degrees]C and then diluted to 1 mL with methanol/acetonitrile (50/50). The extracted sample was subjected to GPC under the conditions shown in Table 2, and the fraction obtained from 9.5 min to 14.5 min was concentrated by [N.sub.2] purging and measured by LC-mass spectrometry using the conditions shown in Table 3.

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]