Introduction
Despite the increasing illicit production and use and of fentanyl (1), decontamination strategies promoted by governing bodies and agencies across the United States, Canada, Australia and the UK are not well defined. Where federally endorsed guidelines exist, these only encourage observation of ‘proper decontamination procedures’ or defer to ‘department guidelines’.1,2 A recent summary of US state-level clean up laws continues to highlight a distinct lack of health-based standards and guidance pertaining specifically to fentanyl remediation.3
In 2017, a US-based voluntary collaborative panel known as the InterAgency Board (IAB) recommended the use of Dahlgren Decon™, a 5% solution of peracetic acid, a 10% solution of hydrogen peroxide, or a 12% solution of tri- or dichloroisocyanuric acid as effective decontaminants for fentanyl-exposed areas.4 However, these approaches remain unsanctioned by any official health or environmental authority.5 In 2018, a factsheet was released by the US. Environmental Protection Agency (EPA), to share best-practice strategies for fentanyl surface decontamination, but not as official policy.6 Pointing to literature that indicates the success of oxidants,7–9 several active chlorine and peroxy-containing products were recommended. However, the factsheet highlighted the limited data – particularly pertaining to operationally relevant reaction conditions – on which to base such recommendations.
Formulations that minimise or obviate the formation of toxic degradation products were recommended without reference to which formulations these might be. The factsheet also notes pH considerations discussed in the review by Bazley et al.10 and the likely ineffectiveness of alkaline solutions, such as unbuffered bleach. Indeed, the precipitation of fentanyl from alkaline decontamination solutions has been reported.8,11,12
A key element to managing the threat posed by uncontained fentanyl and fentanyl analogues lies in effective decontamination. The high potency of these compounds and poorly defined health-based standards to describe acceptable residual fentanyl concentrations means detection equipment with high specificity and low limits of detection are required. This has been reflected in literature to date, with many studies employing mass spectrometry (MS)- or even MS/MS-based techniques to quantify the decrease in the parent compound, equating this with decontamination efficacy.13–16
Until recently, little effort has been applied to identifying or quantifying degradation products, with some studies assessing the presence/absence of fentanyl only, and others finding no products yet assuming their presence. It is for this reason the work of both Oudejans et al.11 and Linden et al.12 is of outstanding significance within the fentanyl decontamination literature to date. Linden et al.12 has suggested N-oxides to be the near quantitative product of the reaction between peroxygenated decontaminants and fentanyl/carfentanil and this aligns with our own findings presented herein.
Critically, authors investigated the potential residual opioid activity of fentanyl- and carfentanil-N-oxides, with cell-based receptor binding assays suggesting these to be approximately 120× less potent than the respective parent compounds.12 Conversion to the respective N-oxides significantly reduces the risk of adverse effects following accidental exposure, however, authors observe that other potential (biological) targets and longer elimination half-lives of the N-oxides ought to be considered due to the long-acting properties observed.12
Though acidified bleach was also shown by Oudejans et al.11 and Linden et al.12 to be efficacious against both fentanyl and carfentanil, the liberation of chlorine gas when preparing such a solution (particularly in large volumes in field settings) without adequate safety controls complicates its use.
The effectiveness of trichloroisocyanuric acid in fentanyl decontamination has been demonstrated by Qi et al.8 However, the use of its more water-soluble counterpart, sodium dichloroisocyanurate (NaDCC), for this purpose remains unexplored in literature. NaDCC offers several advantages to decontamination over liquid (e.g. bleach) or gaseous chlorine sources. A solid with high percentage chlorine by mass, NaDCC is simple to handle, efficient to transport and stable over a number of years.17 In addition to longer shelf life, the ability of cyanuric acid to bind and release chlorine provides extended activity in solution compared to bleach.18 The handling and use of a mildly acidic aqueous NaDCC solution (pH 4–6.5), as opposed to strongly alkaline bleach (pH > 12.5) prepared to the same percentage active chlorine may be ideal or necessary depending on the contamination scenario.8,19,20 A decontamination solution of this pH range (pH 4–6) is prescribed for fentanyl decontamination for reasons previously discussed.10 At this pH range, hypochlorous acid is the predominant active form of chlorine (HOCl pKa = 7.54).21
Although other decontaminants were tested, this paper focusses on the comparison of sodium dichloroisocyanurate (NaDCC) and peracetic acid (PAA) based decontaminants. As stated in our previous review,10 the translation of experimental data to effective field-based approaches involves many considerations and relies heavily on relevant methodology and experimental design. Towards this end, we propose a new, alternative methodology for the assessment of decontamination efficacy.
We utilise this methodology to demonstrate the successful degradation of several fentanyl analogues not previously explored in literature. Furthermore, we define reaction stoichiometry useful to first responders whose use of peroxygenated decontaminants is limited by time, cost or logistics.
Results and discussion
Fentanyl hydrochloride/Dahlgren Decon™ quantitative studies by GCMS/MS
A set of reaction conditions, including the number of molar equivalents of the active species, is useful to first responders who may be required to neutralise a certain quantity of fentanyl within a given timeframe. With this in mind, quantitative work was conducted to accurately define peracetic acid:fentanyl stoichiometry. Though multiple oxidative solutions were investigated, the current work focusses on Dahlgren Decon™ to complement recent reports of its effectiveness against fentanyls.11,12 Dahlgren Decon™ is a three-component commercial decontaminant which contains peracetic acid as part of a pH-modified microemulsion formulation.
Preparation of Dahlgren Decon™ requires extensive mixing of Part A (a proprietary microemulsion), Part B (sodium hydroxide) and Part C (a peracetyl borate solid responsible for liberation of peracetic acid in solution). Dahlgren Decon™ formulation is designed to provide a solution at neutral pH (∼7). The pH of Dahlgren Decon™ solutions used in this study were monitored by pH test strips and were consistently between 6.8 and 7 which is consistent with that reported by Durnal.16 All degradation experiments in this study were conducted at room temperature (∼25 °C).
A fentanyl:decontaminant mass : volume ratio of 1 : 1.5 was employed with decreasing equivalents of peracetic acid. Fentanyl hydrochloride films were treated with serial dilutions (½) of Dahlgren Decon™ applied at a constant 1 : 1.5 m : v ratio (initial fentanyl reaction concentration = 1.79 mM) for 15 minutes (Fig. 1).
Dahlgren Decon™ was shown to have a high level of efficacy against films of fentanyl hydrochloride even when diluted (1/16) to deliver 27.0 molar equivalents of peracetic acid (99.5% reduction of the fentanyl peak area). All samples treated with this amount or greater (up to 431.5 equivalents peracetic acid in undiluted Dahlgren Decon™) showed ≥99.95% decrease in the fentanyl peak area. Even treatment with 6.7 molar equivalents of peracetic acid afforded a 92% decrease in fentanyl after 15 minutes.
An increased rate of reaction observed for Dahlgren Decon™ over a peracetic acid purum (Fig. S-2) was attributed to catalysis afforded by the surfactants or other components present in Dahlgren Decon™. Using analytical techniques of GCMS/MS and UPLC-MS/MS, complete disappearance of the fentanyl peak following treatment with Dahlgren Decon™ or a commercial peracetic acid purum was demonstrated.
This is consistent with previous literature.11,12 Since such a hypothesis posits that corresponding decontamination reaction products would be produced, qualitative screening was undertaken to determine degradation products formed by the reaction between fentanyl and decontaminant solutions.
Attempted identification of Dahlgren Decon™ reaction products by GCMS, HRMS and UPLC-MS/MS
Qualitative experiments with Dahlgren Decon™ used two stoichiometric ratios of PAA : fentanyl (400 : 1 and 25 : 1) to investigate any difference in the degradation product profile caused by differing amounts of PAA. By GCMS, residual fentanyl was observed in both samples, indicating incomplete degradation of fentanyl by the PAA-based decontaminant under these diluted conditions. No structurally probable degradation products were observed in the 25 : 1 PAA : fentanyl sample.
Only one possible product, norfentanyl (2), was observed by GCMS as a very small signal following the treatment of fentanyl hydrochloride with 400 : 1 molar equivalents PAA : fentanyl. All other compounds detected in the decontamination samples were present in similar amounts in the fentanyl-only control sample and are biproducts of the commercial synthetic process.
Analysis of Dahlgren Decon™-treated fentanyl samples by fentanyl-specific MRM methods developed for UPLC-MS/MS produced two new, later-eluting peaks which were posited to arise from fentanyl N-oxide diastereomers (3.1, 3.2). Extensive interrogation of these peaks by various MS techniques is presented in SI.
The only abundant fentanyl-related product detected by HRMS appeared at a mass consistent with fentanyl N-oxide ([M+16] (353.2227 m/z)). Inspection of MS2 data for the 353.2227 m/z precursor reveals a fragmentation pattern indistinguishable from that of fentanyl itself. Major ions at 245 m/z, 189 m/z, 146 m/z, 132 m/z and 105 m/z indicate no change to either norfentanyl or phenethyl regions of the molecule.22,23 Since none of these fragments appear to be oxidised, it is most likely that oxidation occurs at the piperidine nitrogen (3), though no fragment ion to support this hypothesis was detected.
Thus, GCMS data indicated fentanyl was not broken-down into degradation products following treatment with Dahlgren Decon™ but N-oxides were detected by UPLC-MS/MS MRM methods and by HRMS.
Attempts to isolate and quantify degradation products
To more fully investigate decontamination success and resulting reaction products, a new approach to assessing fentanyl decontamination efficacy was adopted. This approach (Fig. 2) involved upscaling the reaction between fentanyl (∼15 mg) and the decontaminant, and fractionation of the reaction mixture by medium pressure liquid chromatography (MPLC).
An upscaled approach whereby the reaction mixture is separated by MPLC is preferable to analytical methods alone for several reasons. Reaction products can be isolated in measurable quantities and characterised by non-destructive techniques such as NMR. The use of NMR as a primary characterisation technique arguably affords otherwise unattainable structural information and allows analysis of products in esse without relying on their conversion to ions or using heat or other processes capable of producing artefacts.
Yet, the method can be used in combination with analytical techniques such as LCMS which enable monitoring in real-time of reaction progress.
Attempted isolation of degradation products from Dahlgren Decon™ treatment
To demonstrate expected recovery of fentanyl in the absence of the PAA active ingredient, fentanyl hydrochloride was dissolved in Dahlgren Decon™ Part A (‘proprietary microemulsion’, DDPtA) (15.09 mg/1.05 mL) and the solution fractionated by MPLC (Fig. 3) following a period of sonication and standing (1 h 15 min). Only two peaks (Rt 3.9 min and Rt 5.6 min) were observed that did not also appear in a blank run analysed immediately prior.
An overall positive mass return of 19.57 mg for both peaks was obtained. Since these peaks overlapped, fractions were taken from extreme ends of each peak for analysis.
Peak 1 (Rt 3.9 min) yielded a white solid determined by LCMS and 1H-NMR (Fig. S-8 and S-9) to be fentanyl with a small amount of surfactant while Peak 2 (Rt 5.6 min), a yellow oily solid, also contained fentanyl with a larger quantity of surfactant (Fig. S-10 and S-11).
The co-elution of a large amount of surfactant and fentanyl likely explains the peak shape observed. 1H-NMR spectra acquired for Peak 2 differed from Peak 1 only in the level of surfactant contamination and a minor downfield chemical shift (0.02–0.04 ppm) at positions α-to the piperidine nitrogen. MPLC fractions containing non-peaks were concentrated and found to contain 370.8 mg of a yellow oil comprising surfactant.
Fentanyl hydrochloride was reacted with Dahlgren Decon™ at an initial fentanyl reaction concentration of 38.6 mM (15.13 mg/1.05 mL, 20 molar equivalents peracetic acid) for 34 minutes prior to fractionation by MPLC. LCMS analysis of the crude reaction mixture was greatly hampered by the presence of large amounts of surfactants. Within the broad signals detected at 15 and 30 minutes (Fig.
S-13), one major signal exhibiting an M+ of 353 amu, consistent with fentanyl N-oxide, was observed. No signal corresponding to fentanyl (337 m/z) was observed. RP-MPLC analysis of the reaction mixture provided three sample specific peaks (Fig.
4).
Removal of the solvent from Peak 1 (Rt = 0.7 min) produced 312.4 mg of a white foamy solid that contained insoluble inorganic material and surfactant. Peaks 2 (Rt = 3.8 min) and 3 (Rt = 5.6 min) were unresolved and therefore fractions were taken from extreme ends of each peak for analysis. Peaks 2 and 3 produced a combined mass return of 69.53 mg due to their coelution with surfactant.
Material from Peak 2 (Rt = 3.8 min) was isolated as a transparent oil. Further analysis by LCMS showed a single major ion of 353 m/z (Rt = 10.1 min). This retention time is consistent with that of a synthesised fentanyl N-oxides standard.
The presence of surfactant in the recovered product hampered identification by NMR. However, the 1H-NMR spectrum (Fig. 5) confirms the fentanyl backbone to remain entirely intact, with key resonances conserved at H16 (t, δ0.99), H15 (q, δ1.99), H3 (m, δ4.80), H4/2 eq (ddd, δ2.13), H4/2 ax (bd, δ1.86, COSY spectra indicates co-resonance with surfactant peak), and all 10 aromatic protons (m, δ7.1–7.6).
A COSY spectrum (Fig. S-14) allowed the identification of four additional signals between 3.0–3.7 ppm, including H6 (m, δ3.47–3.51), H7 (m, δ3.11–3.15) and two signals assigned as H5/1 ax and H5/1 eq (∼4H) based on COSY correlations with H4/2.
With the exception of H4/2 eq, all proton signals arising from the piperidine ring appear shifted downfield with respect to the corresponding signals in fentanyl (Fig. S-14). Methylene protons H6 and H7, which appear closely grouped in the fentanyl spectrum, are here separated with H6 shifted significantly downfield by ∼δ0.6 ppm and H7 to a lesser extent (∼δ0.29 ppm).
These shifts are consistent with fentanyl N-oxide (3) formation and the 1H NMR spectra (Fig. 5) showed evidence of the presence of both isomers. The HRMS ((ESI) m/z calculated for C22H29O2N2 [M + H+] 353.2229 m/z; observed 353.2218 m/z (Fig.
S-15) was also consistent with fentanyl N-oxide (3) production. The slightly earlier elution time, relative to fentanyl, in the MPLC chromatography (Peak 2 Rt = 3.8 min, fentanyl Rt = 3.9 min) is also consistent with fentanyl N-oxide.
LCMS and 1H-NMR analysis of Peak 3 (Rt = 5.6 min) revealed it to contain a small amount of fentanyl N-oxide and predominately surfactants. No evidence of fentanyl or any other decontamination product was detected. These results indicate that treatment of fentanyl with Dahlgren Decon™ results in rapid (within 15 min) quantitative conversion to fentanyl N-oxide with no detectable degradation of the molecule.
Attempted degradation of fentanyl by sodium dichloroisocyanurate (NaDCC)
Fentanyl hydrochloride was treated with both a 12% w/v and 0.12% w/v solution of NaDCC.2H2O for 10 minutes. The pH of NaDCC solutions used in this study were regularly monitored and found to be between pH 5 and 6. Fentanyl could not be detected in the 12% w/v solution by GCMS.
In the 0.12% w/v NaDCC·2H2O sample, fentanyl was detected in the baseline of the GCMS chromatogram. In both samples, numerous potential degradation products were detected by GCMS.
Suggested products with structures plausibly derived from the oxidation of fentanyl and whose mass spectra showed high similarity to that of the library spectra are shown in Fig. 6. Common degradation pathways inferred from this analysis (Fig.
7) included N-dealkylation at the piperidine ring (F1), cleavage of the phenethyl chain between the α- and β-carbon (F2), cleavage of the phenethyl aromatic group (F3), oxidative cleavage of the piperidine ring (e.g. F4), cleavage of the amide nitrogen from the piperidine ring (F5), amide bond cleavage (F6), and cleavage of the ethyl group (F7). A reasonable prediction for the initial mechanism of degradation of fentanyl by hypochlorous acid is that the tertiary piperidine nitrogen would be chlorinated to form an unstable N-chloroammonium intermediate.
This intermediate would rapidly undergo hydrolysis or further oxidation to give reactive iminium-type intermediates which would react with water resulting in dealkylation and loss of the phenethyl substituent. Similar chlorination and oxidative cleavage processes may also occur at the amide nitrogen though this is a less reactive centre. Extensive aromatic chlorination and hydroxylation of degradation fragments is likely a secondary process occurring only after significant oxidation of the amine centre.
Common oxidation patterns of these fragments included chlorination and hydroxylation of the phenethyl chain (α and/or β position), carboxylation of the phenethyl β-position, chlorination and hydroxylation of the phenethyl aromatic ring, and chlorination of the aniline ring. Additionally, chlorophenyl isocyanate compounds were observed in the 12% w/v treated sample with close mass spectral similarity to library spectra. Considering the known instability of isocyanates and the immediacy with which samples were analysed following the 10 minutes reaction period, it is likely that these represent reactive intermediates rather than persistent products.
Both chlorinated and non-chlorinated small molecules were observed by GCMS in the solution of fentanyl treated with NaDCC (Fig. 6). More heavily chlorinated products were observed in the 12% w/v NaDCC·2H2O solution than in the 0.12% solution, for which only one trichlorinated product was suggested.
Although widely used in various industries, polychlorinated benzenes such as those detected in these samples are persistent organic pollutants (POPs).24,25 The level of toxicity of these chlorinated phenyl compounds increases with increasing chlorination,26 so the observation that we may be able to control the level of chlorination in the products may be important in future method development efforts. Polychlorinated phenolic compounds are of significant concern as environmental pollutants as they have been found to be toxic to aquatic organisms and to accumulate in the food chain.27 Toxicity to humans has recently been highlighted for polar phenolic biproducts from disinfectant reagents28 and halogenated biphenolic compounds27 which appear to interfere with the transport and action of thyroid hormones in humans. While the amount of halogenated aromatic chemicals likely to be produced by the NaDCC decontamination of fentanyl analogues in the field by first responders is likely to be very low and insignificant compared to the large amount of these compounds produced by industries such as pesticide, pharmaceutical and dye manufacturing, not to mention water purification processes,29 it is still important to acknowledge the production of potentially toxic compounds.
Quantitation of these products is therefore required to properly assess the risk of potential human exposure and inform environmental waste management strategies.
The detection of numerous, structurally diverse products derived from at least two major regions of the fentanyl molecule indicates degradation of fentanyl by NaDCC to be extensive and likely to proceed via multiple oxidation pathways even at low concentrations of NaDCC. This is a significant finding considering the need to evidence true degradation by the detection of new, smaller products. Though quantitation of these products remains outstanding, these are expected to represent a more acceptable and manageable risk to personnel than the parent compound or a simple oxidised derivative of the parent compound.
Products of the expended oxidant itself (i.e. cyanuric acid and/or isocyanuric acid) are non-toxic.30
Our experiments clearly show quantitative conversion of fentanyl analogues to the single N-oxide product by peracetic acid reagents while reaction with NaDCC results in a mixture of multiple small molecule degradation products, none of which are present in any significant concentration. While the opioid activity and associated toxicity of fentanyl-N-oxide is less than the parent compound,12 it is not zero, and what is not known is the remaining toxicity of N-oxide derivatives of the more potent designer analogues which continue to be introduced into the illicit drug market. The toxicity of the proposed phenolic and chlorophenolic degradation products of hypochlorous acid degradation is primarily associated with long term, cumulative exposure to these products rather than acute toxicity from a single exposure at low concentrations.27,28 Exposure to these small molecule degradation products can be managed in the field using standard chemical hazard personal protective equipment.
Furthermore, if the environmental impact of the volume of waste is a concern, this can be managed by use of absorbent material to retain the waste which can then be incinerated. These practices are common in field procedures. It is argued, therefore, that disassembly of the fentanyl molecular structure by NaDCC represents significantly less risk to first responders and bystanders when destruction of fentanyl analogues is required than quantitative conversion to an oxidised analogue of the parent opioid.
Attempted isolation of degradation products from NaDCC treatment
Attempted dissolution of fentanyl hydrochloride (15.55 mg) in a 12% w/v NaDCC·2H2O solution (1 mL) resulted in a cloudy solution and a sticky, white aggregate which was not easily dispersed manually or by sonication. This aggregate was observed to turn yellow after 20 minutes of sonication through to brown after 1 hour. Schenning et al. also described a ‘sticky, semi-solid (that) inhibited mixing’ of NaDCC and bleach reactions with various fentanyls.31 As proof of principle, Dahlgren Decon™ Part A (DDPtA) was therefore used to enhance solubility of the NaDCC reaction components.
This allowed complete dissolution of fentanyl and successfully prevented the formation of aggregates.
A fentanyl-only control sample prepared in DDPtA and analysed by LCMS after 15 minutes was shown to contain fentanyl (Fig. S-18). By contrast, no fentanyl was detected after 15 minutes when reacted with 8% w/v NaDCC·2H2O in DDPtA (reaction [fentanyl]i = 34.5 mM; 9.06 mol DCC:1 mol fentanyl).
Fentanyl was also undetected by LCMS when the reaction residue was analysed after 1 hour (immediately following injection of the mixture onto RP-MPLC).
Fractionation and analysis of the control sample demonstrated 100% mass return of fentanyl from a single MPLC peak (Rt = 5.6 minutes) alongside a small amount of surfactant by 1H-NMR (Fig. 8). By contrast, no fentanyl was recovered following the reaction of fentanyl and 8% w/v NaDCC·2H2O in DDPtA (reaction [fentanyl]i = 34.46 mM; 9.06 mol DCC:1 mol fentanyl, 1 h).
For this reaction, peaks eluting from 9.0 minutes–15 minutes were poorly resolved (Fig. 9). Fractions containing these peaks were combined as accurately as possible and lyophilised to yield the products in Table 1.
Positive mass returns for all recovered material were again attributed to residual surfactant from DDPtA and potentially some column material present in Peak 1 (Rt = 0.9 minutes).
Notably, no fentanyl was detected by LCMS after 15 minutes, 60 minutes or in any recovered product following the reaction. LCMS analysis showed predominantly those compounds known to be derived from the proprietary microemulsion and ions unique to reaction samples have not provided any significant structural insight into potential degradation products. The absence of fentanyl N-oxide (353 m/z) in all spectra is notable and suggests the DCC reaction to proceed via a different route to peroxygenated oxidants.
1H-NMR spectra of all fractions showed no peaks corresponding to fentanyl or fentanyl N-oxide (Fig. S-21). The signals present were identified as corresponding to the components of the DDPtA microemulsion and weak signals appeared that could be associated with small organic degradation products but no compounds could be identified from these signals.
The absence of fentanyl by LCMS, 1H-NMR and MPLC (retention time) suggests the decontamination of fentanyl hydrochloride with 9.1 molar equivalents of NaDCC to be effective at an initial fentanyl concentration of 34.5 mM. Since these conditions were shown to be successful, similar conditions were trialled for other fentanyl analogues.
Results of decontamination of fentanyl analogues with NaDCC
The reactions performed herein provide evidence for the degradation of fentanyl, and other fentanyl analogues tested, by NaDCC. In many of these experiments (Table 2), the commercial microemulsion DDPtA was used as a surfactant system to solubilise reaction components and facilitate a complete reaction. This proprietary surfactant blend effectively solubilised the reaction between various fentanyls and NaDCC and prevented aggregate formation.
Subsequent testing was performed to identify a suitable alternative surfactant or mixture of surfactants which may be cheaper and more accessible to first responders. Though by no means exhaustive, this investigation identified an aqueous mixture of 15% v/v Hydrapol® RP40 and 5% v/v butane-1,3-diol to adequately solubilise the reaction between various fentanyl analogues and NaDCC, including the major cyanuric acid biproduct. It should be noted, however that this system does not remain a stable emulsion over long standing periods following mixing.
A series of reaction conditions employing the latter surfactant system are included in Table 3. Further experimentation is expected to reveal more suitable surfactants, or mixtures thereof, for this purpose. However, it is necessary that such a system meet a number of requirements, such as resistance to oxidation by NaDCC (at least over the reaction time) and environmental considerations.
Reaction conditions appearing in the current discussion were selected with the intent to inform minimum requirements in the field. Though the volume of an aqueous decontamination solution that can be prepared is rarely limited (especially if grey water or seawater can be used as an effective alternative), there are some scenarios in which decontamination can be limited by (a) the amount of volume which can be practically applied, and (b) quantity of active ingredient available. With this in mind, concentrated reactions were prepared and then the number of molar equivalents reduced using lower concentrations of decontamination solution.
The lowest number of molar equivalents of DCC : fentanyl found to provide successful decontamination was 9 : 1. This was applied at an initial fentanyl reaction concentration of 34.5 mM (i.e. 12.9 g fentanyl/L, or ∼13 : 1 m : v, or 78 mL decontamination solution/1 g fentanyl) for a period of 1 hour although LCMS data suggests the time frame for complete degradation to be much more rapid (<15 minutes).
The degradation of furanylfentanyl and ocfentanil was shown to proceed under similar conditions. These reactions were allowed to proceed for shorter periods of time since previous data suggested one hour was unnecessary. Data supports the ability of NaDCC to degrade both of these analogues.
However, detection of a small (<4%) amount of ocfentanil by LCMS after 45 minutes suggests that a greater number of molar equivalents should be applied. Residual ocfentanil was not reduced by increased reaction time (up to 3 h). The one notable difference in the structure of ocfentanil compared to the other compounds tested is the presence of a fluorine substituent on the phenyl group attached to the amide nitrogen.
The presence of the electron withdrawing fluorine substituent is likely to slow oxidation by hypochlorous acid of the phenyl ring and the amide group of ocfentanil. Reaction with ocfentanil may be slowed enough that more of the reagent reacts with other components of the mixture and so is consumed. Reaction conditions under which ocfentanil cannot be detected warrants further testing and would begin with increasing the molar equivalents of NaDCC added to the reaction.
The continual emergence of ‘designer’ fentanyl analogues combined with specificity limitations of detection equipment means the exact fentanyl analogue encountered at an incident may be unknown. As a result, it is ideal that the decontaminant exhibit broad efficacy across this class of compounds. Evidence supports multiple cleavage sites on the fentanyl molecule by NaDCC and the successful degradation of fentanyl analogues possessing unique functionality.
The strategy prescribed herein may therefore provide efficacious decontamination of existing and new members of the 4-anilinopiperidine class. Further testing on as many analogues as possible should be of high priority to establish any analogue-specific modifications or exceptions.
Conclusions
The overall findings of this work hold significant implications for both first responders and laboratory practitioners responsible for evaluating decontamination efficacy. These results highlight the dangers of equating parent peak disappearance with degradation. It also highlights the need to clearly define ‘degradation’ and demonstrate it by the quantitation of new, smaller products.
The need to evidence degradation by detection and recovery of products was perhaps most clearly demonstrated in the case of Dahlgren Decon™, where elimination of the fentanyl peak – achieved reproducibly using GCMS/MS and UPLC-MS/MS – was shown to be due to the simple transformation to fentanyl N-oxide. Though fentanyl N-oxidation significantly reduces the risk of adverse effects following exposure, it in no way constitutes breakdown of the parent compound and therefore cannot reasonably be considered degradative. The logistics, cost and accessibility of commercial peroxygenated products such as Dahlgren Decon™ may impact various users and scenarios.
In contrast, treatment of fentanyl and three fentanyl analogues with sodium dichloroisocyanurate resulted in molecular disassembly of these compounds and detection of several small organic degradation products. These conditions ensure genuine, irreversible degradation and are therefore recommended herein pending toxicity and environmental risk assessments of degradation products. Satisfying many requirements of an appropriate decontamination solution, further formulation development of this agent may realise a cheap, effective and widely available solution to the problem of fentanyl decontamination.
Our research suggests that an aqueous solution containing Hydrapol® RP40 and 5% v/v butane-1,3-diol adequately solubilises the reaction between various fentanyls and NaDCC, including the major reaction biproduct cyanuric acid, allowing it to proceed to completion. The development of a NaDCC/surfactant-based decontamination formulation for use in the field by first responders is propitious for many reasons besides its efficacy. As a solid, NaDCC is easy to store, transport and handle.
NaDCC is usually manufactured in the form of granules, allowing for accurate and convenient weighing during on-site solution preparation. The compound is readily available due to its demand in swimming pool maintenance, and is inexpensive due to its economical and simple synthesis. Reaction products do not themselves require additional neutralisation procedures following decontamination, although rinsing of the surface may be desirable if cyanuric acid residues remain.
Each of these factors contributes to broad accessibility and utility of the proposed formulation.
The upscaled decontamination approach applied and developed herein may represent an important reform in decontamination methodology by providing a wider lens through which to assess decontamination efficacy. Decontamination reactions were upscaled and fractionated such that products could be isolated on a scale characterizable by elucidation techniques and quantified using a microbalance. Though here described in the context of fentanyls, further development of this methodological approach may see it find wide utility across the decontamination field.
Author contributions
J. T. Blanchfield and A.
A. B. Robertson supervised the project.
All authors designed the experiments and interpreted results. M. M.
Bazley performed all experimental work, including data processing and analysis. M. M.
Bazley wrote this manuscript, which was reviewed and edited by all authors.
Conflicts of interest
There are no conflicts of interest to declare.
Supplementary Material