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 (
