Destruction of Per-And Polyfluoroalkyl Substances (Pfas) With Advanced Reduction Processes (Arps) : A Critical Review
Destruction of Per-And Polyfluoroalkyl Substances (Pfas) With Advanced Reduction Processes (Arps) : A Critical Review
Critical Review
Destruction of Per- and Polyfluoroalkyl Substances (PFAS)
with Advanced Reduction Processes (ARPs): A Critical Review
Junkui Cui, Panpan Gao, and Yang Deng
Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.9b05565 • Publication Date (Web): 12 Mar 2020
Downloaded from pubs.acs.org on March 13, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
14
15 ABSTRACT.
18 generation of short-lived and highly reductive hydrated electrons (eaq-). This study provides a
19 critical review on the mechanisms and performance of reductive destruction of PFAS with eaq-.
2
20 Unique properties of eaq- and its generation in different ARP systems, particularly UV/sulfite and
22 perfluorooctanoic acid (PFOA), perfluorooctane sulfonate (PFOS), and others (e.g. short chain
23 perfluorocarboxylic acids (PFCAs) and perfluorosulfonic acids (PFSAs), per- and polyfluoro
24 dicarboxylic acids, and fluorotelomer carboxylic acids), are reviewed, discussed, and compared.
25 The degradation pathways of these PFAS chemicals rely heavily upon their head groups. For
26 specific PFAS types, fluoroalkyl chain lengths may also affect their reductive degradation
28 solution chemistry parameters and operating factors, such as pH, dose of chemical solute (i.e.
29 sulfite or iodide) for eaq- photo-production, dissolved oxygen, humic acid, nitrate, and
31 control actions in water industries are discussed and the priority research needs are identified.
32
33 INTRODUCTION
34 Per-and poly-fluoroalkyl substances (PFAS) have been globally incorporated into various
35 industrial and consumer products since the 1940s.1 The anthropogenic chemicals, comprising
37 aliphatic substances containing one or more carbon (C) atoms on which all the hydrogen (H)
38 substituents present in the nonfluorinated analogues, from which they are notionally derived, are
39 replaced with fluoride (F) atoms.2 Therefore, PFAS possess at least one perfluoroalkyl moiety
40 (CnF2n+1-). For polyfluoroalkyl substances, a more general concept is also proposed to embrace
41 the aliphatic substances on which at least one, but not all, C atoms are bonded to fluorine, so that
42 more partially fluorinated compounds are comprised,2, 3 such as those having “scattered” or
43 “grouped” multiple F atoms. Because of their unique resistance to heat, water, oil, and stains, in
44 addition to the reduction of friction,1 PFAS have been historically and presently used in a broad
45 range of products (e.g. food packaging and cookware, paper, textiles, leathers, carpet, mattress,
46 and firefighting materials) and in many industries (e.g. metal plating and etching, building and
48 However, concerns on PFAS have gradually been growing due to their prevalence,
49 mobility, persistence, bioaccumulation, and adverse health effects.2, 5-9 Many PFAS have been
50 demonstrated to bioaccumulate, can bind to blood proteins, and have long half-lives in humans.
51 Human PFAS exposure is linked to cancer, obesity, elevated cholesterol, immune suppression,
52 and endocrine disruption.10-12 Various PFAS, particularly perfluorooctanoic acid (PFOA) and
54 groundwater,7, 16-19 drinking water,20-22 and landfill leachate.23-27 The emerging anthropogenic
55 chemicals are challenging water and wastewater treatment, water reclamation, site remediation,
56 and landfill leachate disposal, because of a lack of effective, efficient, and practical treatment
59 Advisory (LHA) for PFOA and PFOS at 70 ng/L.21 Presently, US EPA is moving forward with
60 the maximum contaminant level (MCL) process for PFOA, PFOS, and probably more PFAS
61 chemicals in drinking water.28 The pursuit of public health and potential regulatory pressure
62 require the water industry to stay current and proactive for advancing innovative PFAS treatment
63 technologies. Among very few technically effective PFAS treatment methods, advanced
65 The term of ARPs was early used in the late 1990s to explain unintended degradation of
66 highly oxidized orgnanic compounds (e.g. carbon tetrachloride) by reducing radical species (e.g.
67 hydrated electrons (eaq-) and H atoms 7@ produced during electron beam irridation of polluted
68 groundwater.29 As a purposeful water treatment option, ARPs were proposed for addressing
69 water contaminants more than one decade later.30, 31 ARPs represent a chemical degradation
70 process producing sufficient and highly reductive radicals for destruction of contaminants in
71 water. The reducing radicals involved typically include eaq- and 7@2 in addition to others such as
72 sulfite radical anions (SO3@ -) and sulfur dioxide radical anions (SO2@ -), depending on the used
73 activation methods and chemical solutes.31 Although the term ARPs for water and wastewater
74 treatment has been very recently adopted, the investivigations on reactions of these reudcing
75 radicals with different chemicals in water began several decades ago.32-36 For environmental
76 applications, ARPs have proven very effective for the removal of various oxidized contaminants,
77 such as vinyl chloride,37, 38 perchlorate,39 bromate40, 41, nitrate42, chromium (VI)43, 44, and 2, 4, 6-
78 trichlorophenol.45
79 Special attention has been recently directed to ARPs due to their capability of effectively
80 destructing different PFAS chemicals in water,31, 46-59 advantageous over hydroxyl radical @57
81 or sulfate radical (SO4@ -) – based advanced oxidation processes (AOPs).47, 50, 58 These
82 encouraging findings demonstrate that ARPs are a promising approach to mitigating the PFAS
85 the evaluation of the effects of key factors affecting the PFAS degradation and defluorination
86 efficiencies. Furthermore, environmental implications of the knowledge for water industries are
89 PFAS
90 Although different activation methods (e.g. photolysis, radiolysis, and sonolysis) with or
91 without the assistance of chemical solutes can produce different free reducing radicals in water,
93 certain organic compounds (e.g. aminopolycarboxylic acids (APCAs) and indole derivatives)
94 was reported for effective decomposition of PFAS chemicals in water in literature.31, 56, 60-64
95 Among these solutes, SO32- and I- are the most frequently used. The effective destruction of
96 PFAS molecules during the ARP treatment is principally ascribed to the generation of eaq-, rather
98 Excess electrons can be generated in electrically neutral water directly by pulse radiolysis or
99 using ionization or detachment of a specific solute that enables the existence of electrons in
100 water rather than association with their parent cations.65, 66 Negative charges of the secondary
101 electrons polarize neighboring neutral water molecules, so that the electrons are bound to these
102 water molecules to create a metastable localized species called hydrated electrons.65, 67
103 Experimental evidence for detection of eaq- was first found in the late 1950s.67 Afterwards,
104 substantial debates have remained over the different models proposed to describe molecular
105 structure of eaq-.68 In a traditionally accepted cavity model, a hydrated electron excludes water to
106 create and occupy a small quasi-spherical region, which is surrounded by water molecules with
107 O-H bonds pointing toward the cavity.69 However, the existence of a cavity structure has been
109 Hydrated electrons are a potent reducing agent with an extremely negative standard
110 reduction potential of -2.9 V.71, 72 The short-lived radicals can have a half-life time of more than
111 300 microseconds.67 It is conventionally believed that they tend to react with chemical species
112 via a one-electron transfer mechanism.73 Whereas rate constants of these reactions broadly vary
113 from ~10 M-1@ -1 to a diffusion controlled mode, the activation energies range narrowly within 6 -
114 30 E@. -1, suggesting that the kinetics is limited by the availability of a vacant orbital on the
116 UV/sulfite. Occurrence of a chain reaction during photoionization of sulfite in water was
117 early reported in the 1920s.74 Efforts were later made to elucidate the underlying reaction
118 mechanisms.75-79 Because the UV/sulfite system for reductive destruction of PFAS is mostly
119 operated at an alkaline condition (the effect of pH would be discussed later) and bisulfite (HSO3-
120 ) has a pKa of 7.21,80 the major sulfite species in the PFAS treatment system is SO32- rather than
121 HSO3-. The photoionization of SO32- produces eaq- together with equimolar SO3@ - as shown in
122 Eq.(1).
124 When vacuum UV (VUV) irradiation (185 nm) is adopted in a UV/sulfite system, additional eaq-
125 can be directly created from photoionization of water, but the phenomena cannot be observed
127 At an anoxic condition, SO3@ - is subsequently recombined through two parallel pathways
128 to generate S2O62- (Eq. (2)) and sulfate (SO42-) (Eq. (3)), respectively.
129 SO3@ - + SO3@ - H S2O62- 2k2 = 1.1 × 109 M-1@ -1, 76, 82 (2)
131 Meanwhile, eaq- may be consumed through three mechanisms in the absence of target
132 chemical species. Firstly, eaq- can react with H+ to produce the conjugate acid, 7@ (Eq. (4)).
133 Subsequently, the produced 7@ can further promptly react with eaq- to form hydrogen gas (H2)
138 eaq- + eaq- + 2H2O H H2 + 2OH- 2k6 = 1.1 × 1010 M-1@ -1, 60 (6)
139 Thirdly, eaq- may react with S2O62- generated from Eq. (2) (Eq. (7)).
141 Of note, SO32- can be partially recycled from Eq. (2) and (7) for generation of additional eaq-. The
142 final photo-decomposition products of SO32- are sulfate and dithionite with a reported
144 When dissolved oxygen (DO) is present, eaq- generated can be substantially consumed by
145 O2 to produce superoxide (O2@ -) (Eq.(8)), which may further react with eaq- as Eq. (9).60
8
152 I@H2O* can decay to its ground state (I-) though Eq. (11). It can also be thermally degraded to an
153 intermediate state, i.e. a caged complex <@2 e-), in which electrons are less confined and undergo
154 a quasi-random walk diffusion process in the proximity of I atom <@ over water molecules (Eq.
155 (12)).84 Finally, eaq- can be generated through dissociation of the caged complex, accompanied
160 Here ( ) represents the solvent cage in which <@ and eaq- are formed in pairs.
161 More detailed mechanistic information regarding the eaq- generation from photo-
162 irradiation of I- in water was further explored. Initial photoexcitation involves a state above the
163 charge-transfer-to-solvent (CTTS) absorption, which is an absorption band for detachment of the
164 electron from I-. The photo-detached electron is an electron (etrapped-) trapped in a polarization
165 well created by water molecules oriented around I-.86, 87 88 The trapped electron would
166 subsequently experience electron solvation, which is viewed as an excited state relaxation
167 process between different electronic energy levels of the solvated electrons.89-91 Consequently,
168 two other electron species are sequentially produced primarily through a nonadiabatic electron
169 transfer process,86 including wet electrons (ewet-) and eaq-. ewet- is a pre-solvated electron with the
170 lowest excited state, while eaq- is the equilibrium solvated electron with the ground state
171 (Eq.(14)).90 Transitions among the three electron species are extremely rapid with a time scale in
174 <@ generated from Eq. (13) can react with the added I- to generate various I species, which
176 <@ + I- M I2@ - k15f = 8.8 × 109 M-1@ -1, K = 1.28 × 105 M-1, 93 (15)
178 <@ + I2@ - H I3- k17 = 4.6 × 109 M-1@ -1, 93 (17)
179 I2@ - + I2@ - H I3- + I- 2k18 = 4.6 × 109 M-1@ -1, 93 (18)
180 I- + I2 M I3- k19f = 5.6 × 109 M-1@ -1, K = 747 M-1, 93 (19)
181 Here, the subscript of f in the above rate constants means that the rate constant is for the forward
182 reaction. In the absence of target pollutants, eaq- can be consumed by several I species (Eq.(20)
183 and (21)),47 besides reactions with H+ (Eq. (4) and (5)) and self-recombination (Eq.(6)).
10
186 UV/Others. Other solutes were also used under the UV irradiation to generate eaq- for
187 destruction of PFAS in water, such as Fe(CN)64-, which can be photolyzed to generate eaq- (Eq.
190 It should be noted that a fraction of eaq- generated from the UV/sulfite or UV/iodide
191 system can be scavenged by oxidizing species, leading to a low utilization efficiency of and a
192 short lifetime of eaq-. To address this issue, organic compounds were proposed for the generation
193 of eaq-. One such example is nitrilotriacetic acid (NTA), which plays a dual role in water.56 It is
194 firstly fully hydrated with hydration spheres and then serves as a photosensitizer to facilitate
195 water photo-dissociation and photo-ionization for generation of eaq- and @57,94
197 Afterwards, NTA with a high reactivity toward @57 (rate constant = 4.2 × 109 M-1@ -1 at pH 10.0)
198 minimizes the germinate recombination between eaq- and @57295 promoting the eaq- utilization for
199 PFAS. Therefore, under the identical conditions (0.01 mM PFOS; 2 mM NTA or SO32-; pH
200 10.0), UV/NTA was reported to achieve better degradation and defluorination of PFOS than
201 UV/sulfite.56
11
202 The other organic solutes used are indole (Eq. (24)) and its derivatives (e.g. 3-indole-
203 acetic-acid (IAA)) to generate organic radical cations and eaq- for destruction of PFAS in water.96,
204 97
206 The reductive degradation can be dramatically enhanced in the presence of organomodified
207 montmorillonite (a naturally occurring clay).63, 64 The organic modification materials are cationic
210 and trimethylphenylammonium (TMPA) bromide. The modified clay possesses an excellent
211 adsorption capability for both PFAS and indole solutes, creating a constrained interlayer space
212 for ensuing reactions. When eaq- and organic radical cations are simultaneously generated there
213 upon photolysis, the cationic radicals can be rapidly stabilized by the negatively charged planar
214 clay structure to minimize their recombination reactions with eaq-. Consequently, eaq- can more
215 efficiently attack the co-sorbed PFAS on the clay. Degradation performance of this approach is
218 Fundamental information regarding ARP degradation of PFAS in water has been
219 explored using different approaches, including identification of intermediate and final
220 degradation products, closing mass balance of F and C over the reaction time, measurements of
221 PFAS degradation and defluorination rates, and calculation with density functional theory
222 (DFT). Of note, the extent of defluorination has been quantified using three parameters with
12
223 different meanings, including the F index (Eq.(25)),46 overall defluorination ratio (overall deF%)
[F ]released
225 F index = [PFAS]degraded (25)
[F ]released
226 Overall deF% = [PFAS]0 × NC F
× 100% (26)
228 Here, [F-]released is the molar concentration of F- released due to degradation; [PFAS]0 is the initial
229 molar concentration of a specific parent PFAS chemical; [PFAS]degraded is the molar concertation
230 of the PFAS that has been degraded; NC-F is the number of C-F bonds in the parent PFAS
[PFAS]degraded
231 molecule; and DP is the degradation portion of the parent PFAS (i.e. [PFAS]0 ). F index
232 indicates the average number of F- produced from each decomposed parent PFAS molecule at a
233 specific reaction time. Overall and molecular deF% are the ratios of F- released at a specific
234 reaction time to F present in the total (i.e. degraded + residual) and degraded parent PFAS,
235 respectively. Overall deF% indicates an overall conversion efficiency from organic F in the
236 original PFAS to inorganic F anions, while molecular deF% shows the average conversion
237 efficiency of organic F to inorganic F on each decomposed parent PFAS molecule. Among the
238 three defluorination parameters, overall deF% has been the most commonly used. In the
239 following discussion, the defluorination efficiency used is an overall defluorination ratio.
240 PFOA. Reductive degradation of PFOA with eaq- follows a second order kinetic reaction
241 pattern with a rate constant of (1.7 ± 0.5) ×107 M-1@ -1 (0.01 M NaClO4; pH 10.0).62 Rate of
242 PFOA defluorination is lower than its degradation rate,48, 59 typically at the initial reaction phase,
13
243 implying that part of PFOA degradation reactions, such as the cleavage of C-C bonds on PFOA
244 for production of F-containing intermediate products, cannot synchronously translate into the
245 release of F-. PFOA degradation efficiencies can reach up to 100% in an ARP system,47, 48, 51, 56,
246 63, 64, 98 while the maximum defluorination efficiencies are observed to broadly vary between
247 ~55% and 96% in the most literature,47, 48, 51, 56, 59, 63, 64, 98, except for two studies that reported
249 likely ascribed to the different experimental conditions used in these studies. However, a
250 complete defluorination is argued to be hardly achieved from a perspective of the reductive
252 Two major intermediate product types are identified from the PFOA reactions with eaq-,
253 including less fluorinated carboxylic acids and shorter-chain PFCAs.47-49, 59 The presence of the
254 former products indicates the cleavage of C-F bonds as well as ensuing H/F exchange, while
255 formation of the latter ones implies the scission of C-C bonds.47 Concentrations of the
256 aforementioned degradation intermediate products are typically increased with time as a result of
257 accumulation due to the degradation of parent compounds at the initial reaction phase, followed
258 by a gradual decrease because the intermediates are increasingly degraded due to their own
259 reactivity toward eaq- and with the decrease of the parent compound concentrations.47, 48, 59
260 Two major parallel reaction pathways have been proposed for the PFOA degradation
261 with eaq- (Figure 1) including H/F exchange and chain shortening. Among all the – CF2 – in a
262 PFOA molecule, the T - one adjacent to the carboxyl group has exhibited a high activity
263 likely due to the inductive effect of the head group, thereby providing the preferential reaction
264 center.47, 59 Furthermore, DFT calculations reveal that for the PFCA radical anion produced after
14
265 the attachment of an extra electron, T - C-F bonds are spontaneously stretched,
266 facilitating the bond cleavage.59 The ensuing cleavage of C-F bonds leads to the formation of less
267 fluorinated organic radical anions and the elimination of F- (Eq.(28)) followed by H addition
271 The most probable structure after the two sequential H/F exchanges on T - C-F bonds is
272 Cn-1F2n-1 –CH2-COOH.59 Specifically, for PFOA (n = 7), the intermediate is C6F13 –CH2-
273 COOH.59 Of note, the presence of –CH2- is believed to increase chemical recalcitrance of the
274 degradation product.59 Meanwhile, other hydrofluorinated degradation products generated from
275 more H/F exchanges are identified (e.g. C7F12H3-COO-, C7F11H4-COO-, and C7F9H6-COO-),
276 suggesting that the C-F bonds on the middle positions of carbon chain can also be cleaved.59
277 Furthermore, the H/F exchange can occur in a similar fashion on shorter-chain PFCAs that are
279 Chain shortening is acknowledged as the other primary PFOA degradation pathway with
280 the formation of shorter-chain PFCAs as a consequence of the scission of C-C bonds. However,
281 controversial pathways have been proposed to explain the mechanism. In an early study,47
282 photo-reductive destruction of PFOA in a UV/iodide system was investigated in 18O-water with
283 the approach used previously to explore the mechanisms governing direct UV irradiation of
285 [C6F13C(16O)(18O)]-, and [C6F13C(18O)(18O)]-. Therefore, two C-C scission routes were
286 presumed to simultaneously occur during reductive degradation of PFOA with eaq-,47-49 which
15
288 respectively.
289 In the first one, UV photolysis is believed to excite the PFCA degradation products after
290 two H/F exchanges (i.e. CnF2n-1H2-COO-) to form @ Cn-1F2n-1, @'55-, and CH carbenes (i.e. :CH2)
291 (Eq.(30)). The produced @ Cn-1F2n-1 and @'55- can be recombined to form Cn-1F2n-1-COO- with
292 one fewer CF2 unit than its parent compound (Eq.(31)).47-49, 56
295 The stepwise –CH2– removal through direct UV irradiation and aforementioned hydrolysis
296 explains further C-C scission of PFOA. However, the pathway has recently been questioned due
297 to two reasons.59 Firstly, the degradation of Cn-1F2n-1 –CH2–COO- is challenging due to the
298 presence of -CH2-, which is supported by the sluggish degradation of CF3-CH2-COO- with eaq-. If
299 the above pathway was the case, all the PFOA molecules would finally be shortened to
300 trifluoroacetic acid (TFA). Bentel et al.59 found that TFA could be completely defluorinated in an
301 advanced reduction system, but a complete defluorination of PFOA was barely achieved.59
302 Secondly, even though Eq.(30) truly occurs, the possibility of recombination of @ Cn-1F2n-1 and
304 The second route is involved with four sequential steps, including decarboxylation
305 initiated by direct UV irradiation (Eq.(32)), hydroxylation (Eq.(33)), F elimination (Eq.(34)), and
16
312 Among the reactions, speculative Eq.(32) and (33) were proposed in a prior study on the direct
313 UV irradiation of PFOA.99 HF elimination from CnF2n+1OH in Eq.(34) was previously proposed
314 for chlorine atom – initiated photo-oxidation of methyl perfluoroalkyl ethers.100 Along the
315 DHEH pathway, the produced Cn-1F2n-1COOH can be further degraded repeatedly in a similar
316 manner, so that multiple CF2 units can be stepwise lost. However, the direct photolysis
317 hypothesis for scission of C-C bonds on PFOA is not supported by the reported evidence that a
318 very limited defluorination efficiency is achieved during direct UV photolysis of PFOA in
319 water.47, 48 In another study supporting the DHEH for chain shortening,59 how the
320 decarboxylation is initiated is not clearly stated. Particularly, whether or how hydrated electrons
321 are involved in the C-C cleavage of PFOA via the DHEH mechanism was not clarified in the
323 Of note, low energy electrons are capable of inducing decarboxylation of formic acid
324 (HCOOH) and TFA in a condensed phase.101, 102 Therefore, it is plausible that photo-induced
325 hydrated electrons can initiate decarboxylation of PFOA in water in a similar manner, which
326 may serve as the first step of the DHEH mechanism. The hypothesis was proposed to explain the
327 scission of C-C bonds on PFOA during TiO2 photocatalytic decomposition of PFOA in the
17
328 presence of oxalic acid.103 The hydrated electron-driven decarboxylation of PFCAs in water is as
329 follows.
331 Following the decarboxylation involved with e-aq, the PFOA degradation product would be
332 further degraded via Eq.(33)-(35) to accomplish the cleavage of C-C bonds via a new probable
333 DHEH pathway. Between the two aforementioned chain shortening mechanisms (i.e. Eq.(30)-
334 (31) and Eq.(32)-(36)), the DHEH route seems more plausible. In-depth mechanistic studies are
336 Quantitative information regarding relative contributions of the H/F exchange and chain
337 shortening pathways to PFOA degradation is not reported. However, the measurement of overall
338 defluorination ratio may be informative to determine relative importance of the two competitive
339 mechanisms. Theoretically, a complete defluorination is barely accomplished via the H/F
340 exchange due to chemical persistence of the degradation products, while 100% defluorination
341 can be achieved by chain shortening via the DHEH pathway with the final products of CO2, H2O,
343 Besides the two aforementioned reaction pathways, other side reactions may
345 on the activation modes of eaq-. In the UV/iodide system, various gaseous intermediates produced
346 are identified, such as CFHI2, C2F4HI, C5F6HI, C6F13I, and C6F10HI. Their formation may be a
347 result of the incorporation of <@ (Eq. (13)) into fluoridated intermediates (RF@ as follows.46
18
349 Similarly, in the UV/sulfite system, shorter chain-length fluorinated alkyl sulfonates are also
350 observed, such as C7F15SO3-, C6F13SO3-, C5F11SO3-, C4F9SO3-, and C3F7SO3-,48 likely due to the
352 PFOS. PFOS degradation with eaq- occurs through mechanisms different from the
353 decomposition of PFOA owing to the presence of a different head group (i.e. sulfonic group).46,
354 52, 55-57, 59 Two types of intermediate PFOS degradation products are identified, including 1)
355 short-chain fully and partially fluorinated PFSAs; and 2) PFOA and short-chain fully and
356 partially fluorinated PFCAs (e.g. perfluoroheptanoic acid (PFHpA), perfluorohexanoic acid
357 (PFHxA), perfluoropentanoic acid (PFPeA), and perfluorobutanoic acid (PFBA)). 52, 56, 57 Other
358 degradation products are also observed due to the incorporation of <@ or SO3-@, Particularly in a
359 UV/iodide system, more diverse iodinated gaseous products are produced from the degradation
360 of PFOS than the decomposition of PFOA.46 Similar to the intermediate products of PFOA
361 degradation, these daughter chemicals gradually build up and subsequently diminish in an ARP
362 system as the reactions proceed.52, 55, 59 Formation of the different intermediate degradation
363 products implies that the PFOS decomposition with eaq- occurs through multiple competitive
365 Three reaction pathways have been proposed to explain the reductive destruction of
366 PFOS with e-aq, including desulfonation, H/F exchange, and chain shortening via direct C-C
367 cleavage (Figure 2). 52, 55, 57, 59 During desulfonation, the attachment of eaq- breaks down the C-S
368 bond between the head group and perfluoroalkyl chain. The C-S scission relatively readily
369 occurs (Eq. (38)-(40)) because C-S (272 kJ/mol) in PFOS has a lower bond energy than C-C
370 (346 kJ/mol). Moreover, the distance between C and S atoms (4.463 Å) is longer than that of C-
19
371 C (1.529-1.627 Å) or S-O (1.651Å) on PFOS.104 DFT calculations also indicate the C-S bond
372 stretching after the formation of @ CnF2n+1SO32- (n=8) in Eq. (38), suggesting dissociation of the
375 However, two slightly different dissociation routes have been proposed to describe the following
376 C-S scission. In the first one, the desulfonation leads to the formation of CnF2n+1- (Eq.(39))
377 followed by the production of C8F17OH (Eq. (40)), which would be subsequently transformed to
381 In the other pathway, the dissociation produces @ CnF2n+1 (Eq.(41)), which can be transformed to
384 The second desulfonation pathway may be more plausible because the alkaline condition for the
385 ARP treatment disfavors the occurrence of Eq.(40). Although desulfonation does not directly
386 eliminate F or shorten the C-C chain, H/F exchange and chain shortening can take place during
388 H-F exchange represents another PFOS degradation pathway, which can eliminate F but
389 cannot break down C-C bonds. Besides the occurrence of the aforementioned H/F exchange on
390 the PFCAs (the PFOS degradation products), H/F exchange may happen on relatively weak C-F
391 bonds of PFOS after the attachment of eaq-.46, 55, 59 The eaq- attachment preferentially attacks one
20
392 of the centermost C-F bonds, producing PFOS raidcals.46, 52, 105 Recent DFT calculations show
393 that the two center ones on PFOS have the lowest bond dissociation energy (BDE) (446.31
394 E@. -1) among all the C-F bonds, implying that the C-F cleavages preferentially take place in
395 the middle –CF2– chain.59 One H/F exchange plausibly proceeds successively through Eq. (42)
396 and (43),55 in a manner similar to H/F exchange for PFOA (Eq. ((28) and (29)) except for the
400 Cyclic H/F exchanges can lead to the formation of multiple H/F exchange products, depending
402 The third degradation mechanism is direct cleavage of C-C bonds on PFOS. Lowest
403 unoccupied molecular orbitals (LUMO) of PFOS are situated on the moiety of C4 - C8 atoms in
404 the perfluoroalkyl chain, which exhibits a sigma antibonding nature.52 Following the attachment
405 of eaq- on the middle – CF2 –, the binding between the C and C atoms would be attenuated,
406 facilitating the breakdown of these C-C bonds. Particularly, from a thermodynamic point of
407 view, the formation of C3F7-, C4F9-, and C5F11- is favorable because these dissociation fragments
408 of PFOS have the lowest relative energies W= with respect to 5 @,52 For example, two
409 pathways are proposed to explain plausible occurrence of reductive C-C scission on C5. The first
410 C-C scission mechanism begins with the attachment of eaq- on the middle – CF2 – to produce
411 CF3(CF2)2CF2 X –(CF2)4SO3-, which then complexes with a proton followed by C-C scission due
413 F F F F
21
414 Y Y Y Y
415 CF3(CF2)2C @ Z C (CF2)3SO3 + H H CF3(CF2)2'Z7 + @ C(CF2)3SO3-
– - + (44)
416 Y Y Y Y
417 F F F F
418
419 However, the contribution of Eq. (44) to the overall chain shortening is insignificant because of
420 limited availability of protons at an alkaline condition typically used for ARP treatment of PFAS.
421 The other reaction pathway firstly proceeds through F elimination (Eq.(42)) to produce
422 @'nF2nSO3- , which subsequently transforms to a carbanion (C8F16SO32-) after eaq- attachment
425 Thereafter, the carbanion is dissociated due to reductive cleavage of C-C to generate a shorter
426 chain carbanion CF3CF2CF2- (i.e. C3F7- fragment) and an olefin (Eq.(46)).46, 52
427
428
429 F F F F
430 Y Y Y Y
431 CF3CF2'Z'Z' -Z CF2(CF2)2SO3- H CF3CF2C - + C [ C(CF2)3SO3- (46)
432 Y Y Y Y Y Y
433 F F F F F F
434
435 Carbanion C3F7- would recombine @'557 from the PFOA degradation to generate PFBA. The
436 reaction pathway is indirectly evidenced from the abundance of PFBA observed among the
437 PFOS degradation prodcuts.52 Or the carbanion reacts with H2O+ (Eq.(40)), followed by F
438 elimination (Eq.(34)) and hydrolysis (Eq.(35)) to produce pentafluoropropionic acid (PFPrA).
22
446 Other PFAS. ARP destruction of other PFAS chemicals in water have also been
447 investigated, including shorter chain PFCAs and PFSAs, per- and polyfluoro dicarboxylic acids
449 COOH), in addition to PFOA and PFOS alternatives (e.g. GenX and F-53B).46, 58, 59, 106-108
450 Generally, degradation behaviors of the PFAS chemicals rely heavily upon their head groups.
451 For specific PFAS types, the degradation patterns may be also influenced by lengths of
452 fluoroalkyl chain.46, 59 Below are an in-depth discussion on hydrated electron reductive
453 degradation of other PFCAs and PFSAs, PFdiCAs, and FTCAs in water.
454 Degradation of other PFCAs (n ^ 2) in an ARP system proceeds through the two
455 aforementioned competitive PFOA decomposition mechanisms (i.e. H/F exchange and chain
456 shortening via DHEH). Their decomposition rates are typically greater than the rates of F-
457 release.59 Moreover, both degradation and defluorination rates of these PFCAs (n ^ 2) appear
458 independent of their chain lengths. Park et al.46 found that three PFCAs (n = 4, 6, and 8) had
459 similar pseudo-first-order rate constants (~ 1.3×10-3 min-1) in a UV/iodide system and their F
460 indexes slightly varied between 1 and 2, suggesting a chain length-independent degradation
461 mechanism. And Bentel et al.59 reported that the overall defluorination ratios of PFCAs (n = 2-
463 In contrast, different mechanisms for reductive degradation of TFA (n = 1 PFCA) with
464 eaq- have been proposed.47, 59 One viewpoint ascribes the decomposition of TFA to two parallel
465 pathways:47 1) sequential H/F exchange (Eq. (28) and (29)) that successively produces
466 difluoroacetic acid (DFA, CF2HCOOH) and monofluoroacetic acid (MFA, CFH2COOH), finally
467 generating acetic acid (CH3COOH); and 2) the direct cleavage of C-C leads to the formation of
25
468 fluoroform (CHF3) and CO2, and/or generates two fragments (i.e. @' 3 and @'557 2 which react
469 with water to form fluoroform and formic acid, respectively, or @' 3 of which recombines with
470 itself to generate hexafluoroethane (C2F6). Acetic acid, formic acid, fluoroform, and
471 hexafluoroethane are identified among the intermediate products of eaq- -initiated degradation of
473 However, the aforementioned TFA degradation pathways are not supported by
474 experimental evidence from direct UV/sulfite degradation of TFA, DFA, and MFA.59 ARP is
475 observed to achieve complete defluorination for TFA, but not for DFA or MFA, thereby ruling
476 out the principle role of the sequential H/F exchange pathway. Meanwhile, the rates of TFA
477 degradation and defluorination are nearly synchronized, indicating that F- is rapidly released
478 accompanied with TFA degradation. The observation does not support the aforementioned direct
479 C-C cleavage route, because the alteration of TFA structure cannot cause an immediate release
480 of F-. Instead, the two unique properties of the TFA degradation, i.e. 100% defluorination and
481 the synchronized degradation and defluorination patterns, suggest that the DHEH route is a
482 plausibly dominant mechanism, which finally leads to the production of inorganic F-, H2O, and
484 However, degradation and defluorination rates of different PFSAs are noticeably reduced
485 with the decreasing chain length.46, 59 For example, in a UV/iodide system, the decay rate of
486 perfluorobutane sulfonate (PFBS) was reported to be only 13% of the PFOS decomposition rate
487 under identical experimental conditions.46 At n = 1, the decay and F- release of CF3SO3- are
488 almost marginal.59 The chain-length reliance is ascribed to the increasing strengths of primary
489 and secondary C-F bonds with the decreasing chain length. During the initial defluorination
26
490 phase, F- release preferentially occurs from the middle – CF2 – where extra electrons tend to
491 attach.
492 Reductive destruction of three other PFAS including PFdiCAs, FTCAs, and GenX in
493 water has also been studied.58, 59 PFdiCAs have two – COO- on the two chain ends, different
494 from corresponding PFCAs (with the same perfluoroalkyl chain length) having a – COO- on one
495 end and a recalcitrant – CF3 on the other. One more relatively weak link between the
496 perfluoroalkyl chain and – COO- on PFdiCA favors its degradation via H/F exchange and/or
497 DHEH mechanisms. Bentel et al.59 found that PFdiCAs were more rapidly decomposed than
498 corresponding PFCAs (n=3-10) in a UV/sulfite system. Moreover, they observed very similar
499 overall deF% (~67%) of these compounds, indicating that defluorination of PFdiCAs is
501 Reductive degradation of FTCAs was investigated in detail.59 The presence of – CH2CH2
502 – considerably increases chemical persistence to eaq- attack, so that FTCAs are more recalcitrant
503 to reductive degradation or defluorination than the PFCAs with the same perfluoroalkyl chain
504 length.59 As their perfluoroalkyl chain lengths decrease, reactivity of FTCAs is lowered.
505 Consequently, significant degradation can only be observed for n ^ 6 FTCAs. Similar to the
506 degradation of PFSAs, the decay of FTCAs is believed to firstly occur on the middle C-F bonds
507 with relatively low BDEs via H/F exchange. The assumption is supported by identification of
508 H/F exchange degradation products of FTCAs. In the degradation product analysis, shorter-chain
509 PFCAs are also found in low abundance, implying that dissociation of the head group
511
27
513 Hydrated electron–driven reductive degradation has proven highly effective for chemical
514 destruction of PFAS in water. Nearly complete degradation of various PFAS can be
515 accomplished at appropriate operating conditions.47-49, 52, 53, 56, 57, 63, 64 Key factors affecting the
517 Solution pH. Initial pH can significantly influence PFAS reductive degradation and
518 defluorination with eaq-. Meanwhile, pH may evolve as the reaction proceeds, depending on the
519 degradation extent and solution buffer capacity. The pH range studied in literature broadly varies
520 within 2.4 - 11.48, 50, 53, 55, 56 Experimental evidence unanimously indicates that an increasing pH
521 favors the reductive destruction, and thus the optimal pH for reductive degradation of PFAS falls
523 The enhancement effect of an alkaline pH is principally because less H+ scavenges eaq-
524 (Eq. (4)-(5)) with the increasing pH, so that eaq- can be more efficiently utilized for the
525 degradation of PFAS.60 Based on Eq.(4), a pH increase by 1.0 means that the reaction rate of H+
526 scavenging by eaq- is decreased by one order of magnitude. Therefore, a slight pH change may
528 In a UV/NTA system, the presence of NTA can greatly eliminate @57 due to its high
529 reactivity toward @572 thereby promoting the utilization efficiency of eaq- for destruction of
530 PFAS.56 Distributions of successively deprotonated NTA species with the pKa values of 0.8, 1.9,
531 2.48, and 9.65 depend on solution pH. Because a fully deprotonated NTA is more reactive
532 toward @57 than its fully or partially protonated forms, an alkaline solution would benefit the
533 reactions of NTA and @57 to reduce the consumption of @57 for eaq-. 56
28
535 organomodified clay, the inhibiting effect of pH decrease (3.0-11.0) on the PFAS degradation
537 environment on the modified clay, where protons in the clay interlayers are appreciably replaced
538 by large modification organic cations and eaq- is generated in the vicinity of sorbed PFAS.
539 On the other hand, solution pH may change as the PFAS degradation proceeds in an ARP
540 system with a limited solution pH buffering capacity.46, 55, 57 Various pH evolution patterns have
541 been reported in literature, likely because the pH variation is simultaneously controlled by
542 multiple factors. A major reason for the pH increase is the H+ consumption by eaq- (Eq.(4)).
543 Moreover, when direct photolysis (e.g. VUV irradiation) generates @57 in the presence of sulfite,
544 @57 can react with sulfite to generate OH- and increase pH (Eq.(47)).79
545 @57 + SO32- H OH- + SO3@ - k47 = 4.5 × 109 M-1@ -1 (47)
546 For example, in a UV/sulfite system (2.4 g/L sulfite; UV irradiation: 500 W and 365 nm; and
547 400 mL solution), pH at an initial level of 7.00 was reported to consistently increase by 0.16 and
548 0.22 during degradation of PFOA (20 mg/L) and PFOS (20 mg/L), respectively.57
549 In contrast, other reasons can contribute to a pH decrease. For example, a rapid F- release
550 from PFAS defluorination can lead to form a weak acid HF (pKa = 3.45) in water, thus reducing
551 solution pH.46, 55 In a UV/iodide solution (10 mM iodide; UV irradiation: 8 W and 254 nm; and
552 30 mL solution) for the destruction of PFOS (24 µM), pH was noticed to initially increase and
554 Besides the aforementioned reasons, others such as concentrations and species of
555 intermediate degradation products generated over the reactions likely influence the activity of H+
29
556 in water. Consequently, the final pH change patterns may be very complex, depending on
557 specific operational conditions. Gu et al.55 treated PFOS with VUV/sulfite (37.2 µM PFOS; pH
558 10; UV irradiation: 10 W and 185 nm; and 800 mL solution) using a broad range of sulfite dose
559 (1-20 mM). A consistent pH decrease was observed at 1 mM sulfite, whereas the pH exhibited an
560 initial decrease followed by an increase for other doses, at which the pH increase extent was
562 Solute dose. Dose of the chemical solute for photo-production of eaq- is essential to the
563 reductive degradation patterns of PFAS because it is directly related to the yield of eaq-. When
564 SO32- or I- is used, the PFAS degradation and defluorination efficiencies are typically increased
565 with an increasing solute dose until a critical level, beyond which the PFAS destruction
566 efficiencies would decrease as the dose further increases.47, 48, 57 For example, Sun et al.57
567 reported that the degradation efficiencies of five PFAS all increased with the SO32- dose, peaked
568 almost at 100% at 2.4 g/L SO32-, and decreased, to different degrees, as the SO32- concentration
569 increased to 4.0 g/L (20 mg/L PFDA, PFNA, PFOA, PFBA, or PFOS; pH 7.0; UV irradiation:
570 500 W and 365 nm; and 400 mL solution). And Qu et al.47 found that the initial PFOA
571 decomposition rate increased with an increase in the I- concentration within [I-]:[PFOA] = 0 – 12,
572 but declined as the I- concentration increased at [I-]:[PFOA] = 12-28 (25 µM PFOA; 0.0-0.7 mM
573 I-; pH 9.0; UV irradiation: 15 W and 254 nm; and 740 mL solution).
574 The positive dependence of solute dose observed within a low solute dose range is
575 ascribed to the fact that more solutes for photo-excitation can translate into more eaq- through Eq.
576 (1) or Eq. (10)-(13), provided that sufficient photons are available. For example, the relative
577 quasi-stationary concentration (RQSC) of eaq- in a UV/SO32-/N2 system was concluded to reveal
30
578 that RQSC is approximately linearly increased with the increasing sulfite dose at pH> 9.48
579 However, the potential scavenging effects for eaq- can also be gradually enhanced with the
580 increase in the solute dose. In a UV/sulfite system, eaq- can be increasingly consumed with the
581 increasing sulfite dose through the enhanced self-recombination of eaq- (Eq.(6)) and/or reaction
582 with intermediate S2O62- (Eq.(7)). And for a UV/iodide system, an increasing I- dose can enhance
583 the reactions of eaq- with different I species (Eq.(20) and (21)), besides its self-recombination.
584 Once the solute dose reaches a specific level beyond which the scavenging begins to overweigh
585 the eaq- generation, the negative dependence of the solute dose would become dominant.
586 DO. The presence of DO can substantially decrease rates and efficiencies of PFAS
587 degradation and defluorination in the most ARP systems.46-48, 50, 55 The negative impact is
588 principally attributed to the scavenging effects of O2 (Eq.(8) and (9)) .60 Moreover, DO is
589 capable of adsorbing part of UV photons to reduce the yield of eaq- and thus compromise the
590 PFAS decomposition.50, 109 Of note, the inhibiting extent of PFAS decomposition due to DO is
591 typically substantial. For example, in a UV/sulfite system for PFOA degradation (20 µM PFOA;
592 10 mM sulfite; UV irradiation: 10 W and 254 nm; 200 mL solution; and pH 10.3), an overall
593 defluorination ratio was reported to considerably decline from 88.5% in a N2 purged solution to
594 6.4% in an air-aerated solution within 24 hours.48 These findings suggest that effective
596 Efforts have been made to mitigate the negative DO impact. A strategy is the use of a
597 high photon flux UV irradiation. Gu et al.52 found that PFOS remained a rapid decomposition at
598 an initial 5 mg/L DO in an open UV/sulfite system (9.93 × 10-8 einstein/cm2@ using a high
599 pressure mercury UV lamp. The UV lamp provided a higher emission density and a broader
31
600 emission spectrum than low or medium-pressure UV lamps, thus promoting the eaq- quantum
601 yield. Another approach is adoption of appropriate chemical solutes for photo-ionization to
602 minimize the scavenging effect of DO for eaq-. Sun et al.56 noticed a slight difference of PFOS
603 degradation and defluorination at oxic and anoxic conditions in a UV/NTA treatment, because
604 NTA and UV-excited NTA (NTA*) could rapidly react with O2 and different reactive oxygen
605 species (ROS) (e.g. O2@ -) produced in the presence of O2. Consequently, the quenching of eaq- by
607 Water matrix constituents. Water matrix constituents co-existing with PFAS may
608 significantly affect reductive destruction of PFAS in water through different mechanisms. Two
609 constituents extensively studied are humic acid (HA) and nitrate (NO3-).
610 Humic acid. Dissolved organic matter (DOM) is ubiquitously present in different PFAS-
611 polluted water matrixes, such as natural organic matter (NOM) in surface freshwater,112-114
612 effluent organic matter (EfOM) in biologically treated municipal wastewater,115, 116 soil organic
613 matter (SOM) in groundwater, and leachate organic matter (LOM) in landfill leachate.117-119
614 Because DOM is typically much more abundant than PFAS in polluted water, its impact on ARP
615 destruction of PFAS is of interest. Although various organic types are comprised in aquatic
616 DOM, only the effect of HA, a primary hydrophobic DOM fraction, was investigated in
617 literature,53, 55, 98 among which in-depth studies for evaluation of the role of HA were
619 For a UV/iodide ARP treatment, enhanced and inhibiting effects of HA on the PFAS
32
621 competitive mechanisms govern the influence of HA. Experimental evidence reveals that HA
622 can accelerate the PFAS degradation with the increasing HA concentration, but the
623 decomposition is slowed down once HA is beyond a specific concentration. At a too high HA
624 concentration, the PFAS degradation kinetic rate is even below that in the absence of HA.
625 Therefore, an optimal HA appears to exist to maximize the rate of PFAS decomposition in a
626 UV/iodide system.53 Of note, though HA may considerably alter the kinetic patterns of PFAS
627 degradation, it may not accordingly change the final PFAS degradation efficiency. On the other
628 hand, HA may exhibit different impacts on the defluorination behaviors. In a UV/iodide system
629 (30 µM PFOA; 0.3 mM iodide; UV irradiation: 14 W and 254 nm; and pH 10.0), rapid PFOA
630 degradation and defluorination were noticed in the presence of HA (1.0 mg/L) only within the
631 initial 3 hours. Thereafter, the PFOA decomposition and defluorination efficiencies with time
632 were independent of the initial HA concentration.98 In another set of PFOS degradation
633 experiments with the nearly identical operational conditions,53 HA (0.0-30.0 mg/L) played a
634 more complicated role. HA at 1.0 and 30.0 mg/L significantly promoted and inhibited the initial
635 PFOS decomposition rate, respectively. However, the greatest defluorination efficiency (almost
637 Four mechanisms governing HA-enhanced PFAS degradation are proposed for the
638 UV/iodide system. The principal enhancement pathway is that HA reacts with oxidizing I-
639 containing intermediates (e.g. I2, HOI, IO3- and I3-) to again produce I-, which is subsequently
640 photo-activated to generate additional eaq-.49, 53 In the second mechanism, specific functional
641 groups on HA (e.g. quinone moieties) may serve as an electron shuttle,120, 121 thereby mediating
642 electron transfer between I- and PFAS. The third one is associated with the confinement effect of
33
643 HA, in which HA bridges I- and PFAS to form I- -HA-PFAS adducts.53 The adducts can enable
644 the occurrence of ensuing eaq- formation and PFAS decomposition in a local region, so that the
645 quenching of eaq- by others becomes minimal and the diffusion distance for eaq- to PFAS is
646 shortened. And the fourth mechanism is the direct formation of extra eaq- from photo-ionization
647 of HA in water.122 However, contribution of this pathway is minor due to its low quantum yield
648 of eaq-.53
649 Meanwhile, three plausible mechanisms can inhibit the reductive degradation of PFAS in
650 the presence of HA. Firstly, an inhibition can be potentially caused by the UV quenching
651 property of HA. UV blocking effects of HA fractions in NOM, EfOM, and LOM have been
652 reported elsewhere.119, 123, 124 Competitive absorption of HA with SO32- or I- for photons can
653 directly alleviate the yield of eaq-. Secondly, the HA absorption of UV can generate various
654 reactive species (e.g. @572 1O2, H2O2, and excited triplet state DOM (DOM*)), which
655 ineffectually decompose PFAS, but can oxidize the generated eaq-.50 Thirdly, the presence or
656 absence of specific moieties may inhibit the reductive decomposition of PFAS. Electron
657 withdrawing groups (EWGs) (e.g. carboxylic acids on aromatic rings) on HA preferentially react
658 with the generated eaq-.125 And the scarcity of quinone moieties would lessen electron transfer
660 A net effect of HA on the PFAS degradation in the UV/iodide system relies upon
661 competition among the aforementioned enhancement and inhibiting mechanisms. For the eaq-
662 reductive decomposition of PFAS, the high molecular weight (MW) HA faction with more
663 electron donating groups (EDGs) and electron transfer mediators exhibits an overall
34
664 enhancement effect, whereas the low MW molecules characterized with abundant EWGs and
665 saturated aliphatic moieties marginally promote or even inhibit the PFAS degradation.53
666 Nitrate. NO3- is a common surface water and groundwater solute. NO3--N concentrations
667 in the U.S. groundwater are mostly below 3 mg/L,126 while its level in surface freshwater without
668 nutrient pollution is typically less than 1 mg/L.127 Gu et al. 55 observed that PFAS degradation
669 was slowed down with the increasing NO3- level (0.0-0.5 mM) in a VUV/sulfite system (37.2
670 µM PFOS; pH 10; UV irradiation: 10 W and 185 nm; and 800 mL solution). However, a
671 significant difference among the final PFOS degradation efficiencies was not observed at 0.0-0.5
672 mM NO3-, unless the NO3- was increased to 1.0 mM. The NO3- suppression of PFAS
678 The reductive degradation of PFAS is improved with an increasing temperature,50, 51 likely due to
679 the growing collision frequency of molecules in water as the temperature increases. Pseudo-
680 first-order rate constants of PFAS degradation or defluorination are reported to fit the Arrhenius
681 equation at different temperature ranges (20-40oC for PFOA;51 and 35-100oC for PFOS50).
682 Moreover, the same intermediate degradation products are identified in the UV/iodide
683 degradation of PFOA at different temperatures, suggesting that the temperature change cannot
684 alter the pathways of PFAS degradation.51 However, the time required for a specific intermediate
35
685 product to reach its maximum concentration declines with the increasing temperature, because
686 intermediate products become more reactive at a higher temperature and thus tend to more
687 promptly react with eaq- rather than accumulate in the system. 51
689 Implication. Destruction of aqueous of PFAS through eaq- -driven chemical reduction
690 provides a technically viable treatment approach to mitigation of the PFAS pollution for water
691 and wastewater industries. This approach is capable of effectively degrading a broad range of
692 PFAS molecules as compared to many other chemical oxidative or reductive methods, offering a
693 promising ultimate solution to the PFAS impacts. Unique nature of the advanced chemical
694 reductive process has profound implications for its application in realistic treatment.
695 Firstly, effluent total dissolved solids (TDS) can influence how ARPs are applied to the
696 treatment of different PFAS-polluted water matrixes. TDS would be inevitably increased after
697 the ARP treatment due to the formation of inert salt residuals, such as sulfate and iodide. In the
698 literature, a typical dose of sulfite or iodide for the generation of eaq- for PFAS treatment ranges
699 from several to a few tens of mM, which leads to a TDS increase by several hundreds to
700 thousands of mg/L. Further TDS increase is expected when acid and/or alkaline is used for pH
701 adjustment. Consequently, the direct application mode is infeasible for drinking water treatment
702 or municipal wastewater reuse, which typically has a strict limit on the effluent salinity. For
703 example, TDS and sulfate in U.S. potable water should be a 500 and 250 mg/L, respectively,
704 according to the U.S. secondary drinking water standards. Instead, under the circumstances,
705 indirect application of ARPs to treatment of the water or wastewater may be considered. One
36
706 such example is that reverse osmosis (RO) concentrates PFAS into membrane brine, which is
707 later treated with ARPs to remove PFAS accumulated in the brine. For remediating subsurface
708 PFAS-polluted groundwater, TDS is not a limiting factor in most cases. However, in-situ
709 remediation with the ARPs is impossible because direct UV irradiation is barely applied to
710 subsurface. Rather, PFAS-polluted groundwater, after pumped out, can be readily ex-situ treated
712 Secondly, complexity of the system design, operation, and maintenance can be
713 significantly increased due to the pH adjustment and the requirement for an anoxic environment.
714 Solution pH in the most water, particularly drinking water sources and municipal wastewater, is
715 nearly neutral. However, a moderately to highly alkaline pH is preferred to ensure an effective
716 reductive destruction of PFAS, thus requiring a pH adjustment. However, addition of acid and/or
717 base increases the system complexity, effluent TDS, and treatment costs. Another factor
718 inhibiting the PFAS decomposition efficiency is DO, which is ubiquitous in various PFAS-
719 polluted water. To remove dissolved oxygen from water and subsequently maintain a low or no
720 oxygen condition in realistic water and wastewater treatment is challenging, particularly for
721 high-capacity treatment facilities. To control DO in water and wastewater typically requires
722 additional treatment devices and costs. However, DO may not be an issue for treatment of a low-
725 fluoride-containing degradation products may be produced, of which some remain toxic or are
726 greenhouse gases. Furthermore, when I- is adopted for the eaq- generation, different iodinated
727 byproducts have been identified as a result of iodide incorporation. Formation and toxicity of
37
728 iodide-containing byproducts produced from drinking water disinfection have been well
729 documented. 129-133 These compounds generally exhibit more significant mammalian cell
730 cytotoxicity and genotoxicity than their brominated and chlorinated analogues.132
732 Although current studies well demonstrate encouraging results on eaq- destruction of PFAS in
734 1) The dilemma of ARPs for realistic treatment is that eaq- generated is highly reactive
735 toward both target PFAS compounds and many water matrix constituents, which are
736 typically much more abundant than PFAS at trace concentrations. Consequently, the
737 fraction of eaq- allocated for the PFAS degradation is limited, leading to a low treatment
738 efficiency. Therefore, novel approaches to minimizing the scavenging effects of different
739 co-existing species (e.g. H+, DO, and NOM) for eaq- are highly demanded. Particularly,
740 alleviation of the H+ inhibition can reduce the reliance of ARPs on an alkaline pH and
741 thus avoid costly pH adjustment. Another strategy is to improve the yield of eaq- using
743 2) Production and toxicity of PFAS transformation products after the ARP treatment need to
744 be focused. And effects of solution chemistry parameters and operating conditions on
745 formation of these unwanted byproducts deserve in-depth evaluation. Attention should be
746 paid to production of the transformation products produced from the reactions of eaq- with
747 both PFAS and water matrix constituents (e.g. NOM) and to formation of the byproducts
38
749 3) Can ARPs be combined with other physical/chemical processes and even biological
750 methods to build more efficient multiple barriers for PFAS in water? For example, the
751 PFAS reduction products generated from the ARP treatment may be efficiently degraded
752 by certain chemical oxidation processes if abundant reducing functional groups remain
756 impacts is essential to address the PFAS pollutions. However, it is inappropriate now to
757 carry out techno-economic assessment (TEA) and life-cycle analysis (LCA) for ARPs
758 and compare the technologies with exiting PFAS treatment options, considering that the
759 emerging treatment process is still at the early stage. With the advances in the ARP
760 chemistry as well as the technology development and optimization, the TEA and LCA
761 will be needed at a future and more-developed phase. During these assessments, residual
763
764
765 AUTHOR INFORMATION
767 Yang Deng, CELS 220, Department of Earth and Environmental Studies, Montclair State
770 Notes
39
772
773 ACKNOWLEDGMENT
774 This project was supported through the Career Development Grant of Montclair State University
775 (MSU). Junkui Cui was sponsored under the Doctor Assistantship of PhD Program in
776 Environmental Science and Management at MSU. Panpan Gao worked at MSU with the support
777 from the MoE 111 Project (No. B18049). Special thanks to MSU Office of International
778 Engagement. We greatly appreciate valuable comments from three anonymous reviewers.
779
780 REFERENCES
40
797 groundwater at a fire-training area at Wurtsmith Air Force Base, Michigan, USA. Journal
798 of Environmental Monitoring 2003, 5, (2), 341-345.
799 8. Naile, J. E.; Khim, J. S.; Wang, T.; Chen, C.; Luo, W.; Kwon, B.-O.; Park, J.; Koh,
800 C.-H.; Jones, P. D.; Lu, Y., Perfluorinated compounds in water, sediment, soil and biota
801 from estuarine and coastal areas of Korea. Environmental Pollution 2010, 158, (5),
802 1237-1244.
803 9. Moody, C. A.; Martin, J. W.; Kwan, W. C.; Muir, D. C.; Mabury, S. A., Monitoring
804 perfluorinated surfactants in biota and surface water samples following an accidental
805 release of fire-fighting foam into Etobicoke Creek. Environmental science & technology
806 2002, 36, (4), 545-551.
807 10. USEPA Basic Information on PFAS. https://www.epa.gov/pfas/basic-information-
808 pfas
809 11. Gounaris, V.; Anderson, P. R.; Holsen, T. M., Characteristics and environmental
810 significance of colloids in landfill leachate. Environmental science & technology 1993,
811 27, (7), 1381-1387.
812 12. Braun, J. M.; Chen, A.; Romano, M. E.; Calafat, A. M.; Webster, G. M.; Yolton,
813 K.; Lanphear, B. P., Prenatal perfluoroalkyl substance exposure and child adiposity at 8
814 years of age: The HOME study. Obesity 2016, 24, (1), 231-237.
815 13. Taniyasu, S.; Kannan, K.; Horii, Y.; Hanari, N.; Yamashita, N., A survey of
816 perfluorooctane sulfonate and related perfluorinated organic compounds in water, fish,
817 birds, and humans from Japan. Environmental Science & Technology 2003, 37, (12),
818 2634-2639.
819 14. So, M.; Miyake, Y.; Yeung, W.; Ho, Y.; Taniyasu, S.; Rostkowski, P.; Yamashita,
820 N.; Zhou, B.; Shi, X.; Wang, J., Perfluorinated compounds in the Pearl river and
821 Yangtze river of China. Chemosphere 2007, 68, (11), 2085-2095.
822 15. Nakayama, S. F.; Strynar, M. J.; Reiner, J. L.; Delinsky, A. D.; Lindstrom, A. B.,
823 Determination of perfluorinated compounds in the Upper Mississippi River Basin.
824 Environmental science & technology 2010, 44, (11), 4103-4109.
41
825 16. Murakami, M.; Kuroda, K.; Sato, N.; Fukushi, T.; Takizawa, S.; Takada, H.,
826 Groundwater pollution by perfluorinated surfactants in Tokyo. Environmental science &
827 technology 2009, 43, (10), 3480-3486.
828 17. Houtz, E. F.; Higgins, C. P.; Field, J. A.; Sedlak, D. L., Persistence of
829 perfluoroalkyl acid precursors in AFFF-impacted groundwater and soil. Environmental
830 science & technology 2013, 47, (15), 8187-8195.
831 18. Xiao, F.; Simcik, M. F.; Halbach, T. R.; Gulliver, J. S., Perfluorooctane sulfonate
832 (PFOS) and perfluorooctanoate (PFOA) in soils and groundwater of a US metropolitan
833 area: Migration and implications for human exposure. Water research 2015, 72, 64-74.
834 19. Barzen-Hanson, K. A.; Roberts, S. C.; Choyke, S.; Oetjen, K.; McAlees, A.;
835 Riddell, N.; McCrindle, R.; Ferguson, P. L.; Higgins, C. P.; Field, J. A., Discovery of 40
836 classes of per-and polyfluoroalkyl substances in historical aqueous film-forming foams
837 (AFFFs) and AFFF-impacted groundwater. Environmental science & technology 2017,
838 51, (4), 2047-2057.
839 20. Mak, Y. L.; Taniyasu, S.; Yeung, L. W.; Lu, G.; Jin, L.; Yang, Y.; Lam, P. K.;
840 Kannan, K.; Yamashita, N., Perfluorinated compounds in tap water from China and
841 several other countries. Environmental science & technology 2009, 43, (13), 4824-4829.
842 21. Hu, X. C.; Andrews, D. Q.; Lindstrom, A. B.; Bruton, T. A.; Schaider, L. A.;
843 Grandjean, P.; Lohmann, R.; Carignan, C. C.; Blum, A.; Balan, S. A., Detection of poly-
844 and perfluoroalkyl substances (PFASs) in US drinking water linked to industrial sites,
845 military fire training areas, and wastewater treatment plants. Environmental science &
846 technology letters 2016, 3, (10), 344-350.
847 22. Guelfo, J. L.; Adamson, D. T., Evaluation of a national data set for insights into
848 sources, composition, and concentrations of per-and polyfluoroalkyl substances
849 (PFASs) in US drinking water. Environmental pollution 2018, 236, 505-513.
850 23. Busch, J.; Ahrens, L.; Sturm, R.; Ebinghaus, R., Polyfluoroalkyl compounds in
851 landfill leachates. Environmental Pollution 2010, 158, (5), 1467-1471.
42
852 24. Huset, C. A.; Barlaz, M. A.; Barofsky, D. F.; Field, J. A., Quantitative
853 determination of fluorochemicals in municipal landfill leachates. Chemosphere 2011, 82,
854 (10), 1380-1386.
855 25. Benskin, J. P.; Li, B.; Ikonomou, M. G.; Grace, J. R.; Li, L. Y., Per-and
856 polyfluoroalkyl substances in landfill leachate: patterns, time trends, and sources.
857 Environmental science & technology 2012, 46, (21), 11532-11540.
858 26. Yan, H.; Cousins, I. T.; Zhang, C.; Zhou, Q., Perfluoroalkyl acids in municipal
859 landfill leachates from China: Occurrence, fate during leachate treatment and potential
860 impact on groundwater. Science of the Total Environment 2015, 524, 23-31.
861 27. Lang, J. R.; Allred, B. M.; Field, J. A.; Levis, J. W.; Barlaz, M. A., National
862 estimate of per-and polyfluoroalkyl substance (PFAS) release to US municipal landfill
863 leachate. Environmental science & technology 2017, 51, (4), 2197-2205.
864 28. USEPA EPA’s Per- and Polyfluoroalkyl Substances (PFAS) Action Plan (EPA
865 823R18004); 2019.
866 29. Gehringer, P.; Eschweiler, H., Ozone/electron beam process for water treatment:
867 design, limitations and economic considerations. Ozone: science & engineering 1999,
868 21, (5), 523-538.
869 30. Yoon, S. H.; Abdel-Wahab, A.; Batchelor, B. In Advanced Reduction Processes
870 for Hazardous Waste Treatment, Qatar Foundation Annual Research Forum
871 Proceedings, 2011; Bloomsbury Qatar Foundation Journals: 2011; p EVP16.
872 31. Vellanki, B. P.; Batchelor, B.; Abdel-Wahab, A., Advanced reduction processes: a
873 new class of treatment processes. Environmental engineering science 2013, 30, (5),
874 264-271.
875 32. Gordon, S.; Hart, E.; Matheson, M.; Rabani, J.; Thomas, J., Reactions of the
876 hydrated electron. Discussions of the Faraday Society 1963, 36, 193-205.
877 33. Gordon, S.; Hart, E. J.; Matheson, M. S.; Rabani, J.; Thomas, J., Reaction
878 constants of the hydrated electron. Journal of the American Chemical Society 1963, 85,
879 (10), 1375-1377.
43
880 34. Hart, E. J.; Gordon, S.; Thomas, J., Rate Constants of Hydrated Electron
881 Reactions with Organic Compounds1. The Journal of Physical Chemistry 1964, 68, (6),
882 1271-1274.
883 35. Keene, J., The absorption spectrum and some reaction constants of the hydrated
884 electron. Radiation Research 1964, 22, (1), 1-13.
885 36. OZAWA, T.; SETAKA, M.; YAMAMOTO, H.; KWAN, T., On the reaction of the
886 sulfite radical anions with thioureas. Chemical and Pharmaceutical Bulletin 1974, 22,
887 (4), 962-964.
888 37. Liu, X.; Yoon, S.; Batchelor, B.; Abdel-Wahab, A., Degradation of vinyl chloride
889 (VC) by the sulfite/UV advanced reduction process (ARP): Effects of process variables
890 and a kinetic model. Science of the Total Environment 2013, 454, 578-583.
891 38. Liu, X.; Yoon, S.; Batchelor, B.; Abdel-Wahab, A., Photochemical degradation of
892 vinyl chloride with an Advanced Reduction Process (ARP) - Effects of reagents and pH.
893 Chemical Engineering Journal 2013, 215, 868-875.
894 39. Vellanki, B. P.; Batchelor, B., Perchlorate reduction by the sulfite/ultraviolet light
895 advanced reduction process. Journal of Hazardous Materials 2013, 262, 348-356.
896 40. Jung, B.; Nicola, R.; Batchelor, B.; Abdel-Wahab, A., Effect of low- and medium-
897 pressure Hg UV irradiation on bromate removal in advanced reduction process.
898 Chemosphere 2014, 117, 663-672.
899 41. Botlaguduru, V. S. V.; Batchelor, B.; Abdel-Wahab, A., Application of UV-sulfite
900 advanced reduction process to bromate removal. Journal of Water Process Engineering
901 2015, 5, 76-82.
902 42. Bensalah, N.; Nicola, R.; Abdel-Wahab, A., Nitrate removal from water using UV-
903 M/S 2 O 4 M advanced reduction process. International Journal of Environmental
904 Science and Technology 2014, 11, (6), 1733-1742.
905 43. Moussavi, G.; Jiani, F.; Shekoohiyan, S., Advanced reduction of Cr(VI) in real
906 chrome-plating wastewater using a VUV photoreactor: Batch and continuous-flow
907 experiments. Separation and Purification Technology 2015, 151, 218-224.
44
908 44. Xie, B.; Shan, C.; Xu, Z.; Li, X.; Zhang, X.; Chen, J.; Pan, B., One-step removal
909 of Cr (VI) at alkaline pH by UV/sulfite process: reduction to Cr (III) and in situ Cr (III)
910 precipitation. Chemical Engineering Journal 2017, 308, 791-797.
911 45. Yazdanbakhsh, A.; Eslami, A.; Moussavi, G.; Rafiee, M.; Sheikhmohammadi, A.,
912 Photo-assisted degradation of 2, 4, 6-trichlorophenol by an advanced reduction process
913 based on sulfite anion radical: Degradation, dechlorination and mineralization.
914 Chemosphere 2018, 191, 156-165.
915 46. Park, H.; Vecitis, C. D.; Cheng, J.; Choi, W.; Mader, B. T.; Hoffmann, M. R.,
916 Reductive defluorination of aqueous perfluorinated alkyl surfactants: effects of ionic
917 headgroup and chain length. The Journal of Physical Chemistry A 2009, 113, (4), 690-
918 696.
919 47. Qu, Y.; Zhang, C.; Li, F.; Chen, J.; Zhou, Q., Photo-reductive defluorination of
920 perfluorooctanoic acid in water. Water research 2010, 44, (9), 2939-2947.
921 48. Song, Z.; Tang, H.; Wang, N.; Zhu, L., Reductive defluorination of
922 perfluorooctanoic acid by hydrated electrons in a sulfite-mediated UV photochemical
923 system. Journal of hazardous materials 2013, 262, 332-338.
924 49. Qu, Y.; Zhang, C.-J.; Chen, P.; Zhou, Q.; Zhang, W.-X., Effect of initial solution
925 pH on photo-induced reductive decomposition of perfluorooctanoic acid. Chemosphere
926 2014, 107, 218-223.
927 50. Lyu, X.-J.; Li, W.-W.; Lam, P. K.; Yu, H.-Q., Insights into perfluorooctane
928 sulfonate photodegradation in a catalyst-free aqueous solution. Scientific reports 2015,
929 5, 9353.
930 51. Zhang, C.; Qu, Y.; Zhao, X.; Zhou, Q., Photoinduced reductive decomposition of
931 Perflurooctanoic acid in water: effect of temperature and ionic strength. CLEAN–Soil,
932 Air, Water 2015, 43, (2), 223-228.
933 52. Gu, Y.; Dong, W.; Luo, C.; Liu, T., Efficient reductive decomposition of
934 perfluorooctanesulfonate in a high photon flux UV/sulfite system. Environmental science
935 & technology 2016, 50, (19), 10554-10561.
45
936 53. Sun, Z.; Zhang, C.; Chen, P.; Zhou, Q.; Hoffmann, M. R., Impact of humic acid
937 on the photoreductive degradation of perfluorooctane sulfonate (PFOS) by UV/Iodide
938 process. Water research 2017, 127, 50-58.
939 54. Gu, J.; Ma, J.; Jiang, J.; Yang, L.; Yang, J. X.; Zhang, J. Q.; Chi, H. Z.; Song, Y.;
940 Sun, S. F.; Tian, W. Q., Hydrated electron (e(aq)(-)) generation from phenol/UV:
941 Efficiency, influencing factors, and mechanism. Applied Catalysis B-Environmental
942 2017, 200, 585-593.
943 55. Gu, Y.; Liu, T.; Wang, H.; Han, H.; Dong, W., Hydrated electron based
944 decomposition of perfluorooctane sulfonate (PFOS) in the VUV/sulfite system. Science
945 of the Total Environment 2017, 607, 541-548.
946 56. Sun, Z.; Zhang, C.; Xing, L.; Zhou, Q.; Dong, W.; Hoffmann, M. R.,
947 UV/nitrilotriacetic acid process as a novel strategy for efficient photoreductive
948 degradation of perfluorooctanesulfonate. Environmental science & technology 2018, 52,
949 (5), 2953-2962.
950 57. Sun, M.; Zhou, H.; Xu, B.; Bao, J., Distribution of perfluorinated compounds in
951 drinking water treatment plant and reductive degradation by UV/SO 3 M process.
952 Environmental Science and Pollution Research 2018, 25, (8), 7443-7453.
953 58. Bao, Y.; Deng, S.; Jiang, X.; Qu, Y.; He, Y.; Liu, L.; Chai, Q.; Mumtaz, M.; Huang,
954 J.; Cagnetta, G., Degradation of PFOA Substitute: GenX (HFPO–DA Ammonium Salt):
955 Oxidation with UV/Persulfate or Reduction with UV/Sulfite? Environmental science &
956 technology 2018, 52, (20), 11728-11734.
957 59. Bentel, M. J.; Yu, Y.; Xu, L.; Li, Z.; Wong, B. M.; Men, Y.; Liu, J., Defluorination of
958 Per-and Polyfluoroalkyl Substances (PFASs) with Hydrated Electrons: Structural
959 Dependence and Implications to PFAS Remediation and Management. Environmental
960 science & technology 2019.
961 60. Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B., Critical review of
962 rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl
963 radicals (! OH/! M in aqueous solution. Journal of physical and chemical reference data
964 1988, 17, (2), 513-886.
46
965 61. Trojanowicz, M.; Bojanowska-Czajka, A.; Bartosiewicz, I.; Kulisa, K., Advanced
966 oxidation/reduction processes treatment for aqueous perfluorooctanoate (PFOA) and
967 perfluorooctanesulfonate (PFOS)–a review of recent advances. Chemical Engineering
968 Journal 2018, 336, 170-199.
969 62. Huang, L.; Dong, W.; Hou, H., Investigation of the reactivity of hydrated electron
970 toward perfluorinated carboxylates by laser flash photolysis. Chemical Physics Letters
971 2007, 436, (1-3), 124-128.
972 63. Tian, H.; Gao, J.; Li, H.; Boyd, S. A.; Gu, C., Complete defluorination of
973 perfluorinated compounds by hydrated electrons generated from 3-indole-acetic-acid in
974 organomodified montmorillonite. Scientific reports 2016, 6, 32949.
975 64. Tian, H.; Gu, C., Effects of different factors on photodefluorination of
976 perfluorinated compounds by hydrated electrons in organo-montmorillonite system.
977 Chemosphere 2018, 191, 280-287.
978 65. Larsen, R. E.; Glover, W. J.; Schwartz, B. J., Does the hydrated electron occupy
979 a cavity? Science 2010, 329, (5987), 65-69.
980 66. Devonshire, R.; Weiss, J. J., Nature of the transient species in the
981 photochemistry of negative ions in aqueous solution. The Journal of physical chemistry
982 1968, 72, (11), 3815-3820.
983 67. Hart, E. J., The hydrated electron. Science 1964, 146, (3652), 1664-1664.
984 68. Herbert, J. M.; Coons, M. P., The hydrated electron. Annual review of physical
985 chemistry 2017, 68, 447-472.
986 69. Kevan, L., Solvated electron structure in glassy matrixes. Accounts of Chemical
987 Research 1981, 14, (5), 138-145.
988 70. Casey, J. R.; Kahros, A.; Schwartz, B. J., To be or not to be in a cavity: the
989 hydrated electron dilemma. The Journal of Physical Chemistry B 2013, 117, (46),
990 14173-14182.
991 71. Swallow, A. J., Radiation chemistry. An introduction. 1973.
992 72. Schwarz, H. A., Free radicals generated by radiolysis of aqueous solutions.
993 Journal of Chemical Education 1981, 58, (2), 101-105.
47
48
1023 86. Long, F. H.; Lu, H.; Shi, X.; Eisenthal, K. B., Femtosecond studies of electron
1024 photodetachment from an iodide ion in solution: The trapped electron. Chemical Physics
1025 Letters 1990, 169, (3), 165-171.
1026 87. Long, F. H.; Shi, X.; Lu, H.; Eisenthal, K. B., Electron photodetachment from
1027 halide ions in solution: Excited-state dynamics in the polarization well. The Journal of
1028 Physical Chemistry 1994, 98, (30), 7252-7255.
1029 88. Sheu, W.-S.; Rossky, P. J., The electronic dynamics of photoexcited aqueous
1030 iodide. Chemical physics letters 1993, 202, (3-4), 186-190.
1031 89. Long, F. H.; Lu, H.; Eisenthal, K. B., Femtosecond studies of the presolvated
1032 electron: An excited state of the solvated electron? Physical review letters 1990, 64,
1033 (12), 1469.
1034 90. Shi, X.; Long, F. H.; Lu, H.; Eisenthal, K. B., Femtosecond electron solvation
1035 kinetics in water. The Journal of Physical Chemistry 1996, 100, (29), 11903-11906.
1036 91. Sheu, W.-S.; Rossky, P. J., Electronic and solvent relaxation dynamics of a
1037 photoexcited aqueous halide. The Journal of Physical Chemistry 1996, 100, (4), 1295-
1038 1302.
1039 92. Tang, Y.; Shen, H.; Sekiguchi, K.; Kurahashi, N.; Mizuno, T.; Suzuki, Y.-I.;
1040 Suzuki, T., Direct measurement of vertical binding energy of a hydrated electron.
1041 Physical Chemistry Chemical Physics 2010, 12, (15), 3653-3655.
1042 93. Elliot, A. J., A pulse radiolysis study of the reaction of OH with I2 and the decay
1043 of 8 M2 Canadian Journal of Chemistry 1992, 70, (6), 1658-1661.
1044 94. Lian, R.; Oulianov, D. A.; Shkrob, I. A.; Crowell, R. A., Geminate recombination
1045 of electrons generated by above-the-gap (12.4 eV) photoionization of liquid water.
1046 Chemical physics letters 2004, 398, (1-3), 102-106.
1047 95. Sahul, K.; Sharma, B., Gamma radiolysis of nitrilotriacetic acid (NTA) in aqueous
1048 solutions. Journal of radioanalytical and nuclear chemistry 1987, 109, (2), 321-327.
1049 96. Mialocq, J.; Amouyal, E.; Bernas, A.; Grand, D., Picosecond laser photolysis of
1050 aqueous indole and tryptophan. The Journal of Physical Chemistry 1982, 86, (16),
1051 3173-3177.
49
1052 97. Tian, H.; Guo, Y.; Pan, B.; Gu, C.; Li, H.; Boyd, S. A., Enhanced photoreduction
1053 of nitro-aromatic compounds by hydrated electrons derived from indole on natural
1054 montmorillonite. Environmental science & technology 2015, 49, (13), 7784-7792.
1055 98. Guo, C.; Zhang, C.; Sun, Z.; Zhao, X.; Zhou, Q.; Hoffmann, M. R., Synergistic
1056 impact of humic acid on the photo-reductive decomposition of perfluorooctanoic acid.
1057 Chemical Engineering Journal 2019, 360, 1101-1110.
1058 99. Hori, H.; Hayakawa, E.; Einaga, H.; Kutsuna, S.; Koike, K.; Ibusuki, T.;
1059 Kiatagawa, H.; Arakawa, R., Decomposition of environmentally persistent
1060 perfluorooctanoic acid in water by photochemical approaches. Environmental science &
1061 technology 2004, 38, (22), 6118-6124.
1062 100. Nohara, K.; Toma, M.; Kutsuna, S.; Takeuchi, K.; Ibusuki, T., Cl atom-initiated
1063 oxidation of three homologous methyl perfluoroalkyl ethers. Environmental science &
1064 technology 2001, 35, (1), 114-120.
1065 101. Bertin, M.; Martin, I.; Duvernay, F.; Theulé, P.; Bossa, J.-B.; Borget, F.;
1066 Illenberger, E.; Lafosse, A.; Chiavassa, T.; Azria, R., Chemistry induced by low-energy
1067 electrons in condensed multilayers of ammonia and carbon dioxide. Physical Chemistry
1068 Chemical Physics 2009, 11, (11), 1838-1845.
1069 102. Lafosse, A.; Bertin, M.; Azria, R., Electron driven processes in ices: Surface
1070 functionalization and synthesis reactions. Progress in Surface Science 2009, 84, (5-6),
1071 177-198.
1072 103. Wang, Y.; Zhang, P., Photocatalytic decomposition of perfluorooctanoic acid
1073 (PFOA) by TiO2 in the presence of oxalic acid. Journal of hazardous materials 2011,
1074 192, (3), 1869-1875.
1075 104. Erkoç, V2: Erkoç, F., Structural and electronic properties of PFOS and LiPFOS.
1076 Journal of Molecular Structure: THEOCHEM 2001, 549, (3), 289-293.
1077 105. Paul, A.; Wannere, C. S.; Schaefer, H. F., Do linear-chain perfluoroalkanes bind
1078 an electron? The Journal of Physical Chemistry A 2004, 108, (43), 9428-9434.
50
1079 106. Bao, Y.; Cagnetta, G.; Huang, J.; Yu, G., Degradation of hexafluoropropylene
1080 oxide oligomer acids as PFOA alternatives in simulated nanofiltration concentrate:
1081 Effect of molecular structure. Chemical Engineering Journal 2020, 382, 122866.
1082 107. Bao, Y.; Huang, J.; Cagnetta, G.; Yu, G., Removal of F–53B as PFOS alternative
1083 in chrome plating wastewater by UV/Sulfite reduction. Water research 2019, 163,
1084 114907.
1085 108. Bentel, M. J.; Yu, Y.; Xu, L.; Kwon, H.; Li, Z.; Wong, B. M.; Men, Y.; Liu, J.,
1086 Degradation of Perfluoroalkyl Ether Carboxylic Acids with Hydrated Electrons:
1087 Structure–Reactivity Relationships and Environmental Implications. Environmental
1088 Science & Technology 2020.
1089 109. Hasson, V.; Nicholls, R., Absolute spectral absorption measurements on
1090 molecular oxygen from 2640-1920 AA. II. Continuum measurements 2430-1920 AA.
1091 Journal of Physics B: Atomic and Molecular Physics 1971, 4, (12), 1789.
1092 110. Sörensen, M.; Frimmel, F. H., Photodegradation of EDTA and NTA in the
1093 UV/H2O2 process. Zeitschrift für Naturforschung B 1995, 50, (12), 1845-1853.
1094 111. Larson, R. A.; Stabler, P. P., Sensitized photooxidation of nitrilotriacetic and
1095 iminodiacetic acids. Journal of Environmental Science & Health Part A 1978, 13, (8),
1096 545-552.
1097 112. Ma, H.; Allen, H. E.; Yin, Y., Characterization of isolated fractions of dissolved
1098 organic matter from natural waters and a wastewater effluent. Water research 2001, 35,
1099 (4), 985-996.
1100 113. Croue, J.-P.; Korshin, G. V.; Benjamin, M. M., Characterization of natural organic
1101 matter in drinking water. American Water Works Association: 2000.
1102 114. Frimmel, F., Characterization of natural organic matter as major constituents in
1103 aquatic systems. Journal of Contaminant Hydrology 1998, 35, (1-3), 201-216.
1104 115. Shon, H.; Vigneswaran, S.; Snyder, S. A., Effluent organic matter (EfOM) in
1105 wastewater: constituents, effects, and treatment. Critical reviews in environmental
1106 science and technology 2006, 36, (4), 327-374.
51
1107 116. Jin, P.; Jin, X.; Bjerkelund, V. A.; Østerhus, S. W.; Wang, X. C.; Yang, L., A study
1108 on the reactivity characteristics of dissolved effluent organic matter (EfOM) from
1109 municipal wastewater treatment plant during ozonation. Water research 2016, 88, 643-
1110 652.
1111 117. Andreottola, G.; Cannas, P., 2.4 Chemical and Biological Characteristics of
1112 Landfill Leachate. Landfilling of waste: leachate 1992, 1, 65.
1113 118. Shouliang, H.; Beidou, X.; Haichan, Y.; Liansheng, H.; Shilei, F.; Hongliang, L.,
1114 Characteristics of dissolved organic matter (DOM) in leachate with different landfill
1115 ages. Journal of Environmental Sciences 2008, 20, (4), 492-498.
1116 119. Zhao, R.; Jung, C.; Trzopek, A.; Torrens, K.; Deng, Y., Characterization of
1117 ultraviolet-quenching dissolved organic matter (DOM) in mature and young leachates
1118 before and after biological pre-treatment. Environmental Science: Water Research &
1119 Technology 2018, 4, (5), 731-738.
1120 120. Kang, S.-H.; Choi, W., Oxidative degradation of organic compounds using zero-
1121 valent iron in the presence of natural organic matter serving as an electron shuttle.
1122 Environmental Science & Technology 2008, 43, (3), 878-883.
1123 121. Scott, D. T.; McKnight, D. M.; Blunt-Harris, E. L.; Kolesar, S. E.; Lovley, D. R.,
1124 Quinone moieties act as electron acceptors in the reduction of humic substances by
1125 humics-reducing microorganisms. Environmental science & technology 1998, 32, (19),
1126 2984-2989.
1127 122. Wang, W.; Zafiriou, O. C.; Chan, I.-Y.; Zepp, R. G.; Blough, N. V., Production of
1128 hydrated electrons from photoionization of dissolved organic matter in natural waters.
1129 Environmental science & technology 2007, 41, (5), 1601-1607.
1130 123. Lester, Y.; Sharpless, C. M.; Mamane, H.; Linden, K. G., Production of photo-
1131 oxidants by dissolved organic matter during UV water treatment. Environmental science
1132 & technology 2013, 47, (20), 11726-11733.
1133 124. Zhang, D.; Yan, S.; Song, W., Photochemically induced formation of reactive
1134 oxygen species (ROS) from effluent organic matter. Environmental science &
1135 technology 2014, 48, (21), 12645-12653.
52
1136 125. Li, C.; Zheng, S. S.; Li, T. T.; Chen, J. W.; Zhou, J. H.; Su, L. M.; Zhang, Y. N.;
1137 Crittenden, J. C.; Zhu, S. Y.; Zhao, Y. H., Quantitative structure-activity relationship
1138 models for predicting reaction rate constants of organic contaminants with hydrated
1139 electrons and their mechanistic pathways. Water Research 2019, 151, 468-477.
1140 126. Spalding, R. F.; Exner, M. E., Occurrence of nitrate in groundwater—a review.
1141 Journal of environmental quality 1993, 22, (3), 392-402.
1142 127. USEPA Monitoring and Assessing Water Quality - Volunteer Monitoring.
1143 http://water.epa.gov/type/rsl/monitoring/vms510.cfm
1144 128. Gonzalez, M. G.; Oliveros, E.; Wörner, M.; Braun, A. M., Vacuum-ultraviolet
1145 photolysis of aqueous reaction systems. Journal of Photochemistry and Photobiology C:
1146 Photochemistry Reviews 2004, 5, (3), 225-246.
1147 129. Bichsel, Y.; Von Gunten, U., Formation of iodo-trihalomethanes during
1148 disinfection and oxidation of iodide-containing waters. Environmental Science &
1149 Technology 2000, 34, (13), 2784-2791.
1150 130. Hua, G.; Reckhow, D. A.; Kim, J., Effect of bromide and iodide ions on the
1151 formation and speciation of disinfection byproducts during chlorination. Environmental
1152 Science & Technology 2006, 40, (9), 3050-3056.
1153 131. Plewa, M. J.; Wagner, E. D.; Richardson, S. D.; Thruston, A. D.; Woo, Y.-T.;
1154 McKague, A. B., Chemical and biological characterization of newly discovered iodoacid
1155 drinking water disinfection byproducts. Environmental science & technology 2004, 38,
1156 (18), 4713-4722.
1157 132. Richardson, S. D.; Fasano, F.; Ellington, J. J.; Crumley, F. G.; Buettner, K. M.;
1158 Evans, J. J.; Blount, B. C.; Silva, L. K.; Waite, T. J.; Luther, G. W., Occurrence and
1159 mammalian cell toxicity of iodinated disinfection byproducts in drinking water.
1160 Environmental Science & Technology 2008, 42, (22), 8330-8338.
1161 133. Pan, Y.; Li, W.; An, H.; Cui, H.; Wang, Y., Formation and occurrence of new polar
1162 iodinated disinfection byproducts in drinking water. Chemosphere 2016, 144, 2312-
1163 2320.
1164
53