Indirect Photochemical Formation of Carbonyl Sulfide (COS) and

Subscriber access provided by Kaohsiung Medical University

Environmental Processes

Indirect Photochemical Formation of Carbonyl Sulfide (COS) and Carbon Disulfide (CS) in Natural Waters: Role of Organic Sulfur Precursors, Water Quality Constituents, and Temperature 2

Mahsa Modiri Gharehveran, and Amisha Dilip Shah Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b01618 • Publication Date (Web): 25 Jul 2018 Downloaded from http://pubs.acs.org on July 26, 2018

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.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

Environmental Science & Technology

1

2

Indirect Photochemical Formation of Carbonyl Sulfide

3

(COS) and Carbon Disulfide (CS2) in Natural Waters:

4

Role of Organic Sulfur Precursors, Water Quality

5

Constituents, and Temperature

6 Mahsa Modiri Gharehveran§ and Amisha D. Shah§,†,*

7

8 9 10 11 12

§

†

Lyles School of Civil Engineering, Purdue University, West Lafayette, Indiana, USA

Division of Environmental and Ecological Engineering, Purdue University, West Lafayette, Indiana, USA

13 14

* Corresponding author phone: 765-496-2470; fax: 765-494-0395;

15

e-mail: [email protected]

16 17 18

7613 words

19

1 Scheme

20

2 Figures

21

1 ACS Paragon Plus Environment


22 23 24

Abstract

25

critical precursors to sulfate aerosols, which enable climate cooling. COS and CS2 stem from the

26

indirect photolysis of organic sulfur precursors in natural waters, but currently the chemistry

27

behind how this occurs remains unclear. This study evaluated how different organic sulfur

28

precursors, water quality constituents, which can form important reactive intermediates (RI), and

29

temperature affected COS and CS2 formation. Nine natural waters ranging in salinity were

30

spiked with cysteine, cystine, dimethylsulfide (DMS) or methionine and exposed to simulated

31

sunlight over varying times and water quality conditions. Results indicated that COS and CS2

32

formation increased up to 11× and 4×, respectively, after 12 h of sunlight while diurnal cycling

33

exhibited varied effects. COS and CS2 formation were also strongly affected by the DOC

34

concentration, organic sulfur precursor type, O2 concentration, and temperature while salinity

35

differences and CO addition did not play a significant role. Overall, important factors in forming

36

COS and CS2 were identified, which may ultimately impact their atmospheric concentrations.

Carbonyl sulfide (COS) and carbon disulfide (CS2) are volatile sulfur compounds that are

37 38


Page 2 of 29

Page 3 of 29

39 40


TOC Art

41 42 43 44 45 46 47 48 49 50 51 52



53

54

Introduction Carbonyl sulfide (COS) and carbon disulfide (CS2) are important trace gases in the

55

atmosphere because they can contribute to the stratospheric sulfate aerosol layer which enables

56

climate cooling1 and ozone destruction.2 COS can reach the stratosphere due to its > 1 year

57

tropospheric lifetime1 while CS2 can react with hydroxyl radicals (•OH) to form COS.3 The

58

ocean is one major source of COS and CS2, but previous studies have varied in accurately

59

predicting their flux from the ocean (39-639 Gg as S/year4,5 for COS and 90-700 Gg as CS2/year

60

for CS26,7). Consequently, large uncertainties remain in balancing their global budgets,8 which

61

suggests that a greater understanding is still needed towards how they form in natural waters.

62

One major pathway for forming COS and CS2 in water occurs through indirect photolysis

63

from sunlight9–11 since organic sulfur precursors typically present in natural waters (e.g. cysteine)

64

do not undergo direct photolysis.1 Sunlight is important since COS and CS2 concentrations have

65

been found to decrease by up to 0.5× (i.e. 50% decrease) and 0.8× (i.e. 20% decrease) with

66

decreasing sunlight intensity and increasing ocean depth of 1 to 3 m and 0 to 10 m,

67

respectively.9,10 COS formation also follows a pronounced diurnal cycle in surface waters where

68

a 2× increase is reached in early afternoon.12 Current COS flux models incorporate this light

69

effect but only do so using two parameters, the ocean surface sunlight intensity and the seawater

70

absorbance at 360 nm (UV360),13 a surrogate for the chromophoric dissolved organic matter

71

(CDOM) content.

72

Given this, much is still unknown about the photochemical processes involved in COS and

73

CS2 formation in such waters and what specific precursors, reactive intermediates, and reaction

74

mechanisms are involved. This is especially evident given that nearshore waters exhibit up to

75

40× higher COS concentrations than open ocean waters.12,14,15 This may be due to the higher 4 ACS Paragon Plus Environment

Page 4 of 29

Page 5 of 29


76

CDOM concentrations found in the nearshore waters,16 but the influence of other factors remains

77

unclear. One factor may be the organic sulfur content which has mostly been quantified for

78

individual compound (e.g. cysteine and dimethylsulfide (DMS)) concentrations, which range

79

from pM to nM for freshwater and seawater.1,17 Alternatively, total dissolved organic sulfur

80

(DOS) content has been challenging to measure due to analytical constraints, but one study

81

indicated that it ranged in the low ( DMS (1.5 nM) > methionine (0.92 nM) (Fig. 2a). This trend was not solely

359

due to the dark controls since the dark values were then subtracted from the light-generated COS,

360

where the net formation again following the order of where cysteine (2.4 nM) > cystine (1.6 nM)

361

> DMS (1.5 nM) > methionine (0.92 nM) (Fig. 2a). Two studies also found that cysteine >

362

methionine by 2.8-3:1 (cysteine: methionine) when forming COS11,15. Alternatively, CS2

363

formation, while similarly matching COS in terms of its dark formation (only cysteine led to >

364

d.l. concentrations (Fig. 2b)), behaved differently with different precursors with light (Fig. 2b).

365

In this case, CS2 decreased in formation where cysteine (0.62 nM) > cystine (0.4 nM) >

366

methionine (0.23 nM) > DMS (< d.l) (Fig. 2b) but this order slightly changed for net formation

367

where cystine (0.38 nM) > cysteine (0.32 nM) > DMS (0.17 nM) > methionine (< d.l.) (Fig. 2b).

368

Other studies observed that CS2: (i) qualitatively formed from cystine10, (ii) did not form 10 or

369

formed at 1-4 nM with methionine 11, or (iii) decreased according to cysteine > methionine 11. In

370

parallel, no methionine decay was observed over time, unlike cysteine and DMS (Fig. 2).

371

Overall, these trends indicated that organic sulfur precursor type impacted COS and CS2

372

formation, but their sequences for net formation were not identical (thiols > disulfides >

373

thioethers for COS and disulfides > thiols > thioethers for CS2). In both cases, the thiol (cysteine)

374

consistently generated greater levels of COS and CS2 than the thioethers (DMS and methionine),



375

but the disulfide (cystine) formed higher levels of CS2 than the other precursors. Different

376

functional groups adjacent to the sulfur atom thus appear to alter COS and CS2 formation. These

377

functional groups are not directly linked to S oxidation number since the sulfur oxidation states

378

for COS and CS2 are -2 which is the same for cysteine, methionine, and DMS and in fact lower

379

than cystine (sulfur oxidation state of -1).47 Instead, these patterns are controlled by a more

380

complex set of reaction mechanisms (see later sections).

381 382

Role of O2

383

The presence of O2 ([O2]0 ≈ 8.9 mg/L at 20˚C) inhibited COS and CS2 formation under both

384

dark and light conditions in the no spike and cysteine-, DMS-, methionine-, and cystine- spiked

385

LA-B1 water (Fig. 2). This effect was especially dramatic with light where the presence of O2

386

decreased COS by 0.2× with no precursor added but decreased by 0.3, 0.6, 0.6, and 0.9× with

387

cysteine, cystine, DMS, or methionine, respectively (Fig. 2c). For methionine, the decrease

388

observed with O2 was considerably lower than that observed without the precursor (Fig. 2c).

389

This result indicated that O2 quenched methionine or the intermediates generated from it to some

390

extent, although this likely occurred to a small degree since no observable % methionine loss

391

occurred with O2 (Fig. 2c). Previously, studies found that O2 either decreased11,15 or increased

392

COS formation,9 but they used a 254 nm light source9,11. O2 also decreased CS2 levels to < d.l,
60% and 80%, respectively, with O2 (Fig. 2c) than without O2. Thus, O2 does

400

not appear to be directly involved in forming the COS and CS2, but more likely appears to be

401

indirectly involved by quenching the various reactants involved in their formation. For COS, this

402

is opposite to a previously proposed mechanism, although no direct experimental evidence was

403

provided in this case to support it.11

404 405

Role of CO

406

COS and CS2 concentrations remained relatively unchanged and exhibited no trend when

407

amending CO to the cysteine-spiked LA-B1 or cysteine-spiked synthetic water (Fig. 2). These

408

results were expected for CS2 since no oxygen is needed but less expected for COS, which was

409

proposed to form when thiols react with CO.30 Interestingly, a former study did find that the CO

410

concentration simultaneously decreased with increasing COS formation in a cysteine-spiked

411

natural water29. However, this study further proposed that this correlation was not due to having

412

CO serve as a reactant for COS, but instead having it serve as a competing product from a

413

common precursor29. One precursor tested was acetylacetonate which reacted with bisulfide to

414

form both CO and COS.29 Our results further support the claim that CO is not involved in COS

415

formation.

416 417 418

Proposed Pathways Based on these results, several photochemical and dark pathways for COS and CS2 formation

419

have been proposed that depend on the structure of the organic sulfur precursor (Scheme 1). The

420

photochemical reactions are: (i) not expected to involve O2 or CO, (ii) likely involve DOM-



421

based moieties, which are predicted to form key RIs (e.g. 3DOM*)) and for COS, contribute

422

oxygen, and (iii) less likely involve halides or carbonate species.

423

Cysteine: Cysteine is hypothesized to form considerable levels COS and CS2, as compared

424

to the other precursors, through several steps which involve either the thiol (-SH) or thiolate (R-

425

S-) moieties, given that its pKa of 8.425 falls within the pH range of the waters tested (Table S1).

426

In fact, the thiolate moiety is expected to be more reactive than the thiol since R-S- is a better

427

nucleophile than SH.48 Both the thiol and thiolate are then expected to undergo one electron

428

transfer with RIs potentially including 3DOM*, •OH, or ROO•, as observed previously in

429

biological matrices,48 to form R-S• (Scheme 1). Notably, the thiol may also undergo hydrogen

430

abstraction by the RI, as observed previously with 3DOM* and •OH (Scheme 1), 30,49 and

431

especially in low O2 environments50 (Scheme 1). Hydrogen abstraction has been shown to either

432

form R-S• or a carbon centered radical (-αC• ) by attacking the S-H or αC-H moiety,

433

respectively..49,51,52 αC• is predicted to form more readily than R-S• since the C-H bond exhibits a

434

lower bond dissociation energy than the S-H bond.49 Although, they can convert back and forth

435

by rearranging via hydrogen transfer (Scheme 1).53 After these radicals are formed, COS is

436

proposed to form through some sequence of: (i) hydrogen abstraction, (ii) disproportionation,

437

which is known to occur in peptides containing cysteine,51 and/or (iii) β-cleavage51 (Scheme 1).

438

Within this sequence, oxygen also becomes bonded to carbon through some unknown steps but

439

which likely include attack by various DOM-derived oxygen based radicals (e.g. •OH, RO•,

440

ROO•, and R-C•(O)29) (Scheme 1).

441

CS2 is also likely generated from the -αC• or R-S• radicals but where different steps are

442

required to form a S-C-S linkage and then generate CS2. One possibility is that the -αC• and/or R-

443

S• radicals undergo some sequence of reactions involving: (i) reaction with each other to form a


Page 20 of 29

Page 21 of 29


444

S-C bond, as previously proposed,11 (ii) hydrogen abstraction, (iii) disproportionation and/or (iv)

445

β-cleavage (Scheme 1). A second possibility is that the thioaldehyde (-C(S)HR1) moiety,

446

generated via the COS pathway, undergoes nucleophilic attack by another cysteine thiolate

447

moiety (R-S-), as also observed in multiple cysteine-containing peptide chains54, to form the S-C-

448

S bond. Following this, it becomes less evident how the S-C-S linkage then forms CS2. Clearly,

449

several bonds need to be cleaved, but this is not hypothesized to occur by radical pathways

450

induced by RIs since this is not supported by the CS2 diurnal cycling results. Instead, some long-

451

lasting intermediate is involved (e.g. , (R-S-)) (Scheme 1), but further research is required to

452

confirm its identity. In addition, it should be noted that CS2 has been observed to react with •OH

453

in the aqueous phase to form COS.55 However, the linear formation of CS2 during diurnal

454

cycling and the fact that CS2 formation after 12 h light exposure and 12 h of diurnal cycling

455

closely matched each other (Fig. 1b) indicated that this was not likely the dominate route for

456

forming COS.

457

Moreover, in the dark, cysteine is hypothesized to form COS and CS2 via two possible routes

458

where: (i) R-S• is generated through metal-catalyzed auto-oxidation.56 Trace concentrations of

459

metals are expected in these waters, even though < d.l. concentrations of Cu, Pb, Hg were

460

measured (Table S1) or (ii) through other pathways, since cysteine can react with DOM in the

461

dark.57–59

462

Cystine: Cystine is believed to form COS and CS2 through two major proposed pathways.

463

The first pathway suggests that its C-S bond can undergo homolytic bond cleavage when

464

reacting with various sensitizers (e.g. n, π* and π, π* type- 3DOM* sensitizers) to form the RSS•

465

intermediate, as similarly observed for other disulfides60. In the second pathway, cystine could

466

react with 3DOM* through a one electron transfer to form the disulfide radical anion, RSSR•-, 21 ACS Paragon Plus Environment


467

which is known to readily dissociate and form R-S-, and R-S•.48,52,61 Once R-S- and R-S• are

468

formed, pathways similar to cysteine would be adopted. Both proposed pathways support the fact

469

that cystine formed greater levels of CS2 than other precursors, since the two reactive S moieties

470

formed from one cystine molecule would be in close proximity to each other.

471

DMS and Methionine: The mechanisms driving DMS and methionine to form COS and CS2

472

are less clear, although their attached methyl groups and the absence of R-S- implies that forming

473

R-S• is more difficult. This is especially true if R-S• is formed through homolytic bond cleavage,

474

where DMS and methionine would generate less stable radicals as compared to cystine62.

475

Alternatively, hydrogen abstraction could occur at the C-H bond adjacent to the sulfur group,50

476

leading to a carbon centered radical. In addition, methionine has previously been reported to

477

react with 3DOM* to form a sulfur centered radical cation, R-S•+.63 More detailed investigations

478

are needed to assess how the carbon centered radical. and/or R-S•+ would then form COS or CS2.

479 480 481 482

Environmental Implications

483

constituents, and temperature affected COS and CS2 formation. These efforts were made to

484

better elucidate which factors were important and how this could affect their volatilization into

485

the atmosphere and inform global sulfur models. This is a critical issue since these models only

486

use the ocean surface sunlight intensity and UV36013,36 to predict COS photoproduction rates.

487

While our results further confirmed that DOM and UV360 affected COS, other influential factors

488

for COS and CS2 formation included: (i) length of sunlight exposure, especially for CS2 which

489

required only a brief period of light to continue formation, (ii) O2 concentration, which can vary

490

from 4 to 9 mg/L in surface waters depending on salinity and temperature.64 (iii) temperature,

This study aimed to link how sunlight exposure, various organic precursors/water quality


Page 22 of 29

Page 23 of 29


491

which can fluctuate from -1.9˚C to 30˚C depending on latitude and seasonal variations65 and (iv)

492

organic sulfur precursor type. In addition, these findings provided greater mechanistic insights

493

towards the key RIs and other radicals that are involved such as those derived from DOM and

494

quenched by O2 such as 3DOM*,24,33 R•,34 and sulfur-centered radicals (e.g. R-S•)11. One next

495

research step is to further elucidate the role of individual RIs in more controlled, synthetic-based

496

systems. Moreover, one major impedance in linking COS and CS2 formation to a more robust set

497

of water quality parameters falls in the inability to quantify total DOS content in natural waters.

498

Recent work has made significant progress in this area but has only estimated upper boundary

499

limits (

Indirect Photochemical Formation of Carbonyl Sulfide (COS) and

Recommend Documents