Phosphorus flame retardants: properties, production, environmental occurrence, toxicity and analysis

Since the ban on some brominated flame retardants (BFRs), phosphorus flame retardants (PFRs), which were responsible for 20% of the flame retardant (FR) consumption in 2006 in Europe, are often proposed as alternatives for BFRs. PFRs can be divided in three main groups, inorganic, organic and halogen containing PFRs. Most of the PFRs have a mechanism of action in the solid phase of burning materials (char formation), but some may also be active in the gas phase. Some PFRs are reactive FRs, which means they are chemically bound to a polymer, whereas others are additive and mixed into the polymer. The focus of this report is limited to the PFRs mentioned in the literature as potential substitutes for BFRs. The physico-chemical properties, applications and production volumes of PFRs are given. Non-halogenated PFRs are often used as plasticisers as well. Limited information is available on the occurrence of PFRs in the environment. For triphenyl phosphate (TPhP), tricresylphosphate (TCP), tris(2-chloroethyl)phosphate (TCEP), tris(chloropropyl)phosphate (TCPP), tris(1,3-dichloro-2-propyl)phosphate (TDCPP), and tetrekis(2-chlorethyl)dichloroisopentyldiphosphate (V6) a number of studies have been performed on their occurrence in air, water and sediment, but limited data were found on their occurrence in biota. Concentrations found for these PFRs in air were up to 47 μg m(-3), in sediment levels up to 24 mg kg(-1) were found, and in surface water concentrations up to 379 ng L(-1). In all these matrices TCPP was dominant. Concentrations found in dust were up to 67 mg kg(-1), with TDCPP being the dominant PFR. PFR concentrations reported were often higher than polybrominated diphenylether (PBDE) concentrations, and the human exposure due to PFR concentrations in indoor air appears to be higher than exposure due to PBDE concentrations in indoor air. Only the Cl-containing PFRs are carcinogenic. Other negative human health effects were found for Cl-containing PFRs as well as for TCP, which suggest that those PFRs would not be suitable alternatives for BFRs. TPhP, diphenylcresylphosphate (DCP) and TCP would not be suitable alternatives either, because they are considered to be toxic to (aquatic) organisms. Diethylphosphinic acid is, just like TCEP, considered to be very persistent. From an environmental perspective, resorcinol-bis(diphenylphosphate) (RDP), bisphenol-A diphenyl phosphate (BADP) and melamine polyphosphate, may be suitable good substitutes for BFRs. Information on PFR analysis in air, water and sediment is limited to TCEP, TCPP, TPhP, TCP and some other organophosphate esters. For air sampling passive samplers have been used as well as solid phase extraction (SPE) membranes, SPE cartridges, and solid phase micro-extraction (SPME). For extraction of PFRs from water SPE is recommended, because this method gives good recoveries (67-105%) and acceptable relative standard deviations (RSDs) (<20%), and offers the option of on-line coupling with a detection system. For the extraction of PFRs from sediment microwave-assisted extraction (MAE) is recommended. The recoveries (78-105%) and RSDs (3-8%) are good and the method is faster and requires less solvent compared to other methods. For the final instrumental analysis of PFRs, gas chromatography-flame photometric detection (GC-FPD), GC-nitrogen-phosphorus detection (NPD), GC-atomic emission detection (AED), GC-mass spectrometry (MS) as well as liquid chromatography (LC)-MS/MS and GC-Inductively-coupled plasma-MS (ICP-MS) are used. GC-ICP-MS is a promising method, because it provides much less complex chromatograms while offering the same recoveries and limits of detection (LOD) (instrumental LOD is 5-10 ng mL(-1)) compared to GC-NPD and GC-MS, which are frequently used methods for PFR analysis. GC-MS offers a higher selectivity than GC-NPD and the possibility of using isotopically labeled compounds for quantification.