Investigation of the Photochemical Reactivity of Soot Particles Derived

Feb 16, 2015 - In the current study, the heterogeneous reaction of NO2 with soot and biosoot surfaces was investigated in the dark and under illuminat...
0 downloads 0 Views 2MB Size
Subscriber access provided by MICHIGAN STATE UNIVERSITY | MSU LIBRARIES

Article

Investigation of the Photochemical Reactivity of Soot Particles Derived from Biofuels Towards NO. A Kinetic and Product Study. 2

Manolis N. Romanias, Philippe Dagaut, Yuri Bedjanian, Aurea Andrade-Eiroa, Roya Shahla, Karafas S Emmanouil, Vassileios C. Papadimitriou, and Apostolos Spyros J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/jp511468t • Publication Date (Web): 16 Feb 2015 Downloaded from http://pubs.acs.org on February 18, 2015

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 free 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 accessible to all readers and 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.

The Journal of Physical Chemistry A 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 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

1

Investigation of the Photochemical Reactivity of Soot Particles Derived from

2

Biofuels Towards NO2. A Kinetic and Product Study.

3 4 5 6 7 8 9 10 11

Manolis N. Romanías*, 1 Philippe Dagaut*, 1 Yuri Bedjanian, 1 Auréa Andrade-Eiroa, 1 Roya Shahla, 1 Karafas S. Emmanouil, 2 Vassileios C. Papadimitriou, 3 and Spyros Apostolos.2 1. Institute de Combustion, Aérothermique, Réactivité et Environnement (ICARE)CNRS, 1C Avenue de la Recherche Scientifique, 45071, Orléans, France. 2. NMR Laboratory, Department of Chemistry, University of Crete, Vassilika Vouton, 71003, Heraklion, Crete, Greece. 3. Laboratory of Photochemistry and Chemical Kinetics, Department of Chemistry,

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

University of Crete, Vassilika Vouton, 71003, Heraklion, Crete, Greece.

Journal of Physical Chemistry A

*

Corresponding authors:

34

Manolis N. Romanías, Tel.: +33 238255492, Fax: +33 238696004, e-mail address:

35

[email protected]

36 37 38

Philippe Dagaut: Tel.: +33 238255466, Fax: +33 238696004, e-mail address: [email protected]

39 40 41 42 1 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

Abstract

43 44

In the current study, the heterogeneous reaction of NO2 with soot/bio-soot

45

surfaces was investigated in the dark and under illumination relevant to the atmosphere

46

conditions (JNO2 = 0.012 s-1). A flat-flame burner was used for preparation and collection

47

of soot samples from premixed flames of liquid fuels. The biofuels were prepared by

48

mixing

49

(CH3(CH2)6COOCH3) (iii) anhydrous diethyl carbonate (C2H5O)2CO and 2,5 dimethyl

50

furan (CH3)2C4H2O additive compounds in conventional kerosene fuel (JetA-1).

51

Experiments were performed at 293 K using a low-pressure flow-tube reactor (P = 9

52

Torr) coupled to a quadrupole mass spectrometer. The initial and steady-state uptake

53

coefficients, γ0 and γss respectively, as well as the surface coverage, Ns, were measured

54

under dry and humid conditions. Furthermore the branching ratios of the gas phase

55

products NO (~80-100%) and HONO ( JetA-1/MOC JetA-1> JetA-1/DEC ≈ JetA-1/1-BNOL.

349 350

Figure 5. γss of NO2 as a function of soot type under a) dry conditions (0.0032% RH)

351

and b) 5% of RH. Experiments were performed at T = 293 K and 66 ppbv of NO2. Solid

352

and open symbols denote results under dark and UV light irradiation conditions (JNO2 =

353

0.012 s-1), respectively. Note that for JetA-1/1-BNOL sample under dry conditions,

354

JetA-1/DEC and JetA-1/2,5-DMF under 5.5% of RH there is overlapping of the data

355

points in the graph. The BET surface areas measured were used for γss calculations.

356 13 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

357

A closer analysis of the results obtained for γ0 and those of γss reveals that the

358

observable differences between the soot samples have been attenuated upon NO2

359

processing of the soot surfaces. In particular, the initial uptake of NO2 on JetA-1/2,5-

360

DMF particles was by a factor of 10 higher than on JetA-1/1-BNOL bio-soot, while for

361

the steady state uptake a factor of 4.6 was obtained; similar attenuation was recorded for

362

the other samples. The former results indicate that processing of the BC/BrC particles

363

with NO2 lead to depletion of the most active surface sites existing on the freshly

364

emitted soot samples.

365

Concerning the surface coverage of soot, Ns, it was found to be ~1013 molecule

366

cm-2, in agreement with the literature,11, 12 independent on RH and irradiation conditions,

367

(Figure 6 and Table 1). JetA-1/1-BNOL and JetA-1/DEC bio-soot surfaces displayed

368

similar coverages, (2-3) × 1013 molecule cm-2, while the maximum Ns value was

369

recorded on JetA-1/2,5-DMF surfaces, Ns ~ 8 × 1013 molecule cm-2. However, it should

370

be noted that Ns values should be considered as lower limits since saturation of the

371

surface (that would indicate the end of the reaction) was not achieved within the time

372

scale of the experiment.

373

374 375

Figure 6. Surface coverage of different types of soot upon their exposition to 66 ppbv of

376

NO2 at 293 K for 45 min. Solid and open symbols corresponds to experiments under dry

14 ACS Paragon Plus Environment

Page 14 of 30

Page 15 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

377

and humid conditions, 0.0032% and 5% RH respectively. The BET surface areas

378

measured were used for Ns calculations.

379 380

3.2.3 Product yield determination

381

In addition to the kinetic measurements presented above, the yield of reaction

382

products

formed

were

determined

according

to

the

following

expression:

383

∆[product]/∆[ΝΟ2]. NO and HONO were the only gas phase products detected and their

384

distribution is summarized in Figure 7. The formation of nitro containing surface

385

products, (Nitro-PAHs, R-NO, R-NO2, RONO2, etc., R=alkyl group), should not be

386

excluded, although their yield seems to be low, at least under the conditions of the

387

present study. One should note that in our latest study,14 the formation of several

388

nitroPAHs compounds was verified upon processing of kerosene BC/BrC particles with

389

NO2.

390

Under dry conditions, NO yields ranged from 80 to 100 % while HONO

391

production was limited to < 5% for all soot surfaces, except JetA-1/1-BNOL bio-soot

392

where it reached 10%, and JetA-1/2,5-DMF, where HONO was not produced. Similar

393

results were obtained under humid conditions where NO was the major reaction product.

394

However, HONO production was more pronounced in presence of H2O. Indeed, under

395

dark conditions HONO yield increases up to 20% for pure kerosene and JetA-1/2,5-

396

DMF soot surfaces, while it slightly increased for JetA-1/1-BNOL, >10%, ~5% for

397

JetA-1/MOC and ~10% for JetA-1/DEC soot. This tendency could be expected since

398

heterogeneous production of HONO is favoured on any humid surface that interacts

399

with NO2.35,

400

conditions. Note that similar behavior was observed in previous studies from our group

401

where the interaction of NO2 with mineral oxide aerosols was studied.37, 38 Finally, the

402

UV irradiation of soot surfaces caused a systematic increase of NO yields, probably due

403

to photodegradation of surface nitro-containing products leading to NO production.

36

Conversely, NO production rates were slightly lower under humid

404

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

405 406

Figure 7. Product yield for NO (squares) and HONO (circles), under dark (solid

407

symbols) and UV irradiation conditions (open symbols) upon exposition of soot surfaces

408

to 66ppbv of NO2 at 293 K under a) dry conditions and b) 5% RH.

409 410

3.3 Solid phase analysis of soot.

411

One of the major tasks of the current study was the solid phase analysis of the soot

412

surfaces. The objective was to correlate the chemical reactivity of soot towards NO2

413

observed for each fuel, with the chemical composition and the existing functional groups

414

on soot. To this end the analysis of solid soot surfaces was realized, with (i) Raman

415

spectroscopy, (ii) NMR and (iii) HPLC for the determination of PAHs compounds on

416

soot. 16 ACS Paragon Plus Environment

Page 16 of 30

Page 17 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

417

The Journal of Physical Chemistry

3.3.1 Raman spectra

418

The Raman spectra acquired for conventional and bio-kerosene soot samples are

419

presented in Figure 8. For sake of clarity, obtained spectra are presented only in the

420

wavelength range of interest, 900 and 1800 cm-1 where two broad bands appear, with

421

peaks centered at 1330 cm-1 and 1585 cm-1 and correspond to Defective (D) and Graphitic

422

(G) bands, respectively. In order to extract valuable information about the structure and

423

the chemical composition of the soot films, the Raman spectra were deconvoluted.

424

Different fitting methods have been proposed, using Gaussian, Lorentzian, or

425

combination of those functions for peaks fitting..39-42 In addition, the number of the fitting

426

components used is also critical. When using too many components for the same Raman

427

peak, the residual might represent the original spectrum well but most of the produced

428

peaks without any physical meaning. In the literature, up to 10 fitting peaks have been

429

used to deconvolute Raman spectrum of carbon materials. 40, 43, 44

430

The ratio of the intensities of the D and G bands, ID/IG has been used as an

431

indicative parameter of the degree of disordering of the carbon material.45 High values of

432

ID/IG indicate low graphitic-like structure of the material.45 The best deconvoluted fit was

433

achieved using a Lorentzian function with two components. The JetA-1/MOC soot

434

sample was the only exception where a “D-side” band, SD, was observed at ~1200 cm-1,

435

thus a 3 component Lorentzian function was employed to achieve the best fit. G band can

436

stem from in two different origins.41, 46 The carbon stretch vibrations of benzene and in

437

general condensed aromatic rings at ~1588 cm-1 and C=C sp2 stretching vibrations of

438

olefinic or conjugated carbon chains.46 Regarding the origin of the D band, the presence

439

of heavy aromatic compounds with large aromatic rings existing in soot (≥ 6) is

440

considered as the major source.40,

441

observed in all soot samples is due to a wide range of aromatic rings size.40 Therefore, an

442

increased ID/IG ratio can be attributed to the formation of aromatic compounds having 6

443

or more fused benzene rings and provide information about the aromatic/olefinic ratio of

444

the soot sample (the higher the ID/IG, the higher the aromatic/olefinic ratio). The presence

445

of the SD peak after Raman deconvolution of JetA-1/MOC sample is attributed to sp2 and

446

sp3 structures. Compounds such as aryl-alkyl ethers, and methyl carbon dangling to

447

aromatics rings contributes to these wavenumbers and are considered as the major source

448

of this band.40, 41, 48

47, 48

In addition, the wide range of the D bands

17 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

449

450 451

Figure 8. Raman spectra of soot/bio-soot samples studied (λ= 780 nm, focused area =10

452

µm2, resolution, R= 2 cm-1). The spectra were focus for presentation purposes. 18 ACS Paragon Plus Environment

Page 18 of 30

Page 19 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

453 454

The values of the ID/IG ratios obtained for reference JetA-1 and bio soot samples

455

are presented in Table 1. JetA-1/2,5-DMF appears with the maximum ID/IG value

456

followed by pure kerosene, JetA-1/MOC, JetA-1/1-BNOL and JetA-1/DEC soot surfaces,

457

1.45, 1.41, 1.28, 1.23 and 1.11 respectively. One can note that samples with high

458

chemical reactivity appear with the maximum values of ID/IG ratio. This is an indication

459

that the heavy aromatic fraction of soot can possibly influence BC/BrC particles

460

reactivity. However, it should be mentioned that the absolute concentrations of both the

461

aromatic and the aliphatic fractions should be determined.

462

3.3.2 NMR analysis

463

The 1H NMR spectra of the chloroform soluble fraction of the soot samples and

464

the type of chemical groups associated with each signal are presented in Figure S3. One

465

can note that the soot spectra have similar features, which can be explained by the

466

presence of kerosene as the major (80% vol) component in all fuels. Table 2 summarizes

467

the soot characterization data obtained by 1H NMR and reports the proton density of

468

several types of structural units of soot, the CH2/CH3 ratio and the aromatic/aliphatic ratio

469

for the apolar fraction of pure and bio-soot samples.

470

The region δ=0.5 – 2.0 corresponds to aliphatic methyl (-CH3) and methylene (-

471

CH2-) soot protons, and the presence of large signals in this region indicates the high

472

amount of alkanes in pure and bio soot samples. The relative ratio of CH2/CH3 groups

473

can be used to assess the amount of carbon branching in the soot samples. Except from

474

the JetA-1/MOC soot (2.6) the other soot samples were found to have very similar

475

branching ratios (3.3±0.1), indistinguishable from the value of the pure kerosene soot

476

(3.4). This indicates that branching is not substantially affected by the additive compound

477

used to prepare the bio soot. The high values of branching ratio suggest the existence of

478

alkanes with large and straight chains.

479

The signals in the region between δ 2.0 – 2.8 of the NMR spectra represent the

480

presence of methyl and ethyl-substituted aromatic compounds. As can been seen in Table

481

2 these compounds are present in higher amounts in the butanol and the dimethyl furan

482

soot, compared to pure kerosene soot, while smaller amounts are observed for the JetA-

483

1/MOC and JetA-1/DEC soot samples. Figure S3 shows that the combustion of all fuels

484

leads to the production of R-CH (R is nitrogen or an oxygen atom) protons in the region δ

485

2.8 – 4.6 in which the nitrogen atom is chemically connected to aromatic rings.

486

According to Table 2 the JetA-1/1-BNOL soot contains the largest amount of these 19 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

487

compounds, followed by the JetA-1/2,5-DMF and JetA-1/DEC soots that contain similar

488

amounts to JetA-1, and the JetA-1/MOC soot which appears to contain the smallest

489

amount of N-CH compounds. In the region δ 4.6 – 6.2 all the spectra exhibit peaks

490

indicating the presence of very small amounts of vinyl C=CH protons. The low signal in

491

this region indicates the small amount of olefins present in the apolar extracts of all

492

samples.

493

A major feature of the spectra of all soot samples is the large amount of signal in

494

the aromatic region δ 6.5 – 9.4 indicating that the majority of protons in soot have a large

495

aromatic character, with 55-69% of the total protons appearing in this region. It should

496

also be noted that the presence of peaks with chemical shift 9.5 < δ < 8.0 in the NMR

497

spectra of the soot samples indicates that a significant amount of PAHs is present in the

498

soot samples. Finally the peaks with chemical shifts δ 9.5 – 10.8 indicate that the

499

combustion of all fuels leads to the production of aldehydes.

500 501

3.3.3 HPLC analysis

502

The fresh collected soot samples were analysed with a HPLC chromatograph

503

coupled with a UV/Visible photodiode array detector prior to their exposure to NO2. The

504

objective of this series of experiments was to analyze the PAHs content of the soot

505

samples and find relations between measured concentration of classes of compounds and

506

the observed reactivity of the samples towards NO2. Seven different classes of PAHs

507

were separated and more than 200 hundred compounds were identified. The presentation

508

of a detailed list of PAHs compounds identified is a subject of ongoing publications from

509

our lab and was beyond the scope of the present study. We only present the results that

510

are correlated with the observed chemical reactivity of soot with NO2. The HPLC

511

analysis showed that the relative abundance of nitro-PAHs pre-existing on soot (prior to

512

its exposure to NO2) is the key factor influencing the chemical reactivity. In particular, in

513

Figure 9 the concentrations of the 10 Nitro-PAHs compounds identified are displayed.

514

JetA-1/1-BNOL (~5800 ng/mg soot) and JetA-1/DEC (~4000 ng/mg soot) were the soot

515

samples with the highest nitro-PAHs concentrations, followed by Jet A1 (~1800 ng/mg

516

soot). Regarding the JetA-1/MOC and JetA-1/2,5-DMF samples, their nitro-PAHs

517

contents are significantly lower, 400 ng and 169 ng per mg of soot, respectively. The

518

former observation clearly demonstrates that soot samples containing high concentrations

519

of nitro-PAHs prior to their reaction with NO2 have the lower reactivity towards NO2.

20 ACS Paragon Plus Environment

Page 20 of 30

Page 21 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

520

Although this trend looks like “quite logical” it has never been reported in the past and

521

could be a quite useful prediction factor for soot reactivity with NOx.

522 523

Figure 9. Concentrations of nitro-PAHs detected in soot samples prior to their exposure to

524

NO2.

525 526 527 528

4. Summary and Conclusions. In the framework of this study, the chemical reactivity of soot/bio-soot particles

529

towards NO2 has been studied and the chemical composition of the soot samples

530

produced were analysed using Raman spectroscopy, NMR and HPLC. Furthermore, the

531

BET surface areas were measured and the pore size distributions of the freshly produced

532

BC/BrC particles have been determined employing NLDFT. To the best of our

533

knowledge, this is the first experimental study where the reactivity, under relevant

534

atmospheric conditions, and the physicochemical properties of soot particles has been

535

investigated together.

536

The results showed that the additive compounds used to dope the kerosene fuel

537

influenced the physicochemical properties and reactivity of the corresponding bio-soot

538

samples produced. In particular, BET surface measurements and particle size distribution 21 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

539

showed that the combustion of JetA-1/2,5-DMF and JetA-1/DEC biofuels produced soot

540

particles with the highest and lowest surface area and pore volume, respectively (Figures

541

2, S1 and Table 1). Conversely, no significant differences have been observed between

542

the other Br/BrC particles.

543

The degradation of NO2 over the soot particles was found to depend on the

544

substrate BC/BrC sample. Using JetA-1 as reference fuel, JetA-1/2,5-DMF and JetA-

545

1/MOC appeared to be more reactive while for JetA-1/1-BNOL and JetA-1/DEC NO2

546

reactivity was significantly decreased. These differences were observed at the maximum

547

extent within the first minutes of the NO2-soot interaction, where γ0 was measured, and

548

were attenuated upon processing of soot particles for longer time, γss. This attenuation

549

was attributed to the depletion of the most reactive surface sites of soot within the first

550

minutes of the interaction. UV illumination of the surface didn’t accelerate the uptake

551

process at least to an extent that could be clearly measured and not be attributed to

552

experimental uncertainties. Regarding the products, NO was measured to be the major

553

one, ranging from 80% to 100%, while HONO production was limited to significantly

554

lower limits (