Infrared Chemiluminescence Study of the Reaction ... - ACS Publications

Subscriber access provided by - Access paid by the | UCSB Libraries

A: Kinetics and Dynamics

Infrared Chemiluminescence Study of the Reaction of Hydroxyl Radical with Formamide and the Secondary Unimolecular Reaction of Chemically Activated Carbamic Acid Nadezhda I. Butkovskaya, and Donald W. Setser J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.8b01512 • Publication Date (Web): 03 Apr 2018 Downloaded from http://pubs.acs.org on April 4, 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 34 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

Infrared Chemiluminescence Study of the Reaction of Hydroxyl Radical with Formamide and the Secondary Unimolecular Reaction of Chemically Activated Carbamic Acid N. I. Butkovskaya*1 and D. W. Setser2 1

Semenov Institute of Chemical Physics, Russian Academy of Sciences, 119991 Moscow, Russian Federation; 2Department of Chemistry, Kansas State University, Manhattan, Kansas 66506, United States *

Corresponding author: [email protected]; Tel.: +7 495 939 7316

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 34

2

Abstract Reactions of OH and OD radicals with NH2CHO and ND2CHO were studied by Fourier-transform infrared emission spectroscopy of the product molecules from a fast-flow reactor at 298 K. Vibrational distributions of the HOD and H2O molecules from the primary reactions with the C-H bond were obtained by computer simulation of the emission spectra. The vibrational distributions resemble those for other direct H-atom abstraction reactions, such as with acetaldehyde. The highest observed level gives an estimate of the C-H bond dissociation energy in formamide of 90.5±1.3 kcal mol-1. Observation of CO2, ammonia, and secondary water chemiluminescence gave evidence that recombination of OH and NH2CO forms carbamic acid (NH2COOH) with excitation energy of 103 kcal mol-1, which decomposes through two pathways forming either NH3 + CO2 or H2O + HNCO. The branching fraction for ammonia formation was estimated to be 2-3 times larger than formation of water. This observation was confirmed by RRKM calculation of the decomposition rate constants. A new simulation method was developed to analyze infrared emission from NH3, NH2D, ND2H and ND3. Dynamical aspects of the primary and secondary reactions are discussed based on the vibrational distributions of CO2 and those of H/D isotopes of water and ammonia.


Page 3 of 34 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


3

1. Introduction The reaction of formamide with hydroxyl radicals is of practical interest for two reasons. First, formamide is considered as a hazardous air pollutant present in the atmosphere because of its use in the agrochemical and pharmaceutical industries and as a solvent in the production of textiles.1,2 The atmospheric fate of amides strongly depends on the reactions with OH radicals, and knowledge of the rate coefficients and products is needed to predict the composition of the atmosphere. Besides, formamide is a participant in the complex prebiotic chemical reactions leading from simple precursors to key biological molecules like proteins and peptides3,4, and reactions of formamide and its derivatives both in gas and liquid phase present theoretical and biochemical interest. Systematic investigation of the atmospheric fate of amides started only a few years ago. Using a vacuum ultraviolet flash-photolysis/resonance-fluorescence technique, it was established that the room temperature OH + NH2CHO reaction proceeds exclusively by abstraction of the aldehydic hydrogen 5,6. Both studies used high-level computations of the stationary points on the potential energy surface (PES) to define the transition state and the hydrogen bonded complexes that exist in the entrance and exit channels. These calculations also provide a C-H bond dissociation energy of formamide, for which an experimental value has not been reported. In the present work we analyzed the infrared chemiluminescent spectra from a fast-flow reactor at 298 K over the 2000-4000 cm-1 range from the reaction of NH2CHO and ND2CHO with OH and OD radicals. The aim was to get more insight into the reaction mechanism by obtaining vibrational distributions of the product H2O and HOD molecules, using methods described earlier7 and advanced in the recent study of methylamines8. The spectra contained not only H2O and HOD emission, but also a strong CO2 emission, indicating the occurrence of a fast secondary reaction. Reactions with ND2CHO showed that, besides CO2, the secondary reaction produced deuterated water and ammonia. The observed emission implies that recombination of NH2CO and OH forms activated carbamic acid ACS Paragon Plus Environment


Page 4 of 34

4

(NH2COOH), which decomposes producing either NH3 + CO2 (a) or H2O + HNCO (b). The latter process is one step in the synthesis of urea, where carbamic acid is an important transient intermediate911

. Occurrence of this reaction and its deutero-analogs was confirmed by observation of weak emission

of ammonia isotopes (NH3, ND3, NH2D, ND2H) and D2O. Actually, this is the first direct evidence of gas-phase generation of carbamic acid, a compound which has never been observed and characterized experimentally. Some experimental data were reported on formation of solid NH2COOH at low temperature, as a result of proton bombardment of a layer of solid NH3 covered by a layer of H2O + CO2 ice12. For a long time it was thought that carbamic acid is unstable in the gas-phase, however, investigation of the reversible decomposition of ammonium carbamate (NH4CO2NH2) accompanied by ab initio calculations suggested the presence of NH2COOH as an intermediate10. In our study, the recombination reaction gives NH2COOH with an excitation energy of 103 kcal mol-1, and the vibrationally excited molecules decompose at the experimental pressures of 0.5-1.0 Torr. The decomposition branching ratio was estimated by analysis of the relative intensities of isotopic water and ammonia products from OH and OD reactions with ND2CO. The vibrational distributions of H2O and HOD from the primary reaction of formamide with OH and OD are discussed with reference to calculated transition-state structures and properties of the exit channels6. The energy disposal to water molecule is compared to previously obtained data for related compounds, such as acetaldehyde13, formaldehyde14 and methyl amine8. The energy disposal pattern from the decomposition of carbamic acid is compared to that from vibrationally excited acetic acid15. The branching between (a) and (b) determined from the experiments was supported by RRKM calculated ratio of the unimolecular rate constants. 2. Experimental Methods The experimental methods and the procedure for analysis of the H2O, HOD and D2O emission spectra have been given in a recent paper8. Briefly, the infrared chemiluminescence spectra from the ACS Paragon Plus Environment

Page 5 of 34 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


5

vibrationally excited products generated in a fast-flow chemical reactor were recorded by a Fouriertransform infrared spectrometer (BIORAD FTS-60) with spectral resolution of 2 cm-1. The wavelength response of the liquid N2 cooled InSb detector was calibrated with a standard black-body source. The spectrometer chamber and the tube connecting the observation window with the spectrometer were continuously flushed with dry air to remove water vapor and carbon dioxide. A quartz filter (Spectrogon SP-4300) was installed in front of the detector, which increased the sensitivity in the 24004000 cm-1 range. Removal of the filter for detection of the luminescence below 2400 cm-1 led to an increase of the N/S ratio and carbon dioxide background signal. To diminish the latter during full-range measurements, the spectrometer and connecting tube were flushed with dry N2. The 4 cm diameter Pyrex flow-reactor with NaCl observation window used dry Ar as a carrier gas; typical flow velocities of 120-250 m/s at corresponding pressures 1.0-0.5 Torr gave reaction times of 0.28 and 0.15 ms, respectively, for observation immediately after addition of formamide. The OH or OD radicals were produced 30 cm upstream of the observation window via the H(D) + NO2 reaction; the H(D) atoms were generated by a microwave discharge of H2(D2)/Ar mixture. The degree of H2(D2) dissociation was measured as 50±5%. Standard conditions for recording spectra from the primary reaction were [OH] = (6-8) × 1012 molecule cm-3 and [NO2] = (3-5) × 1013 molecule cm-3 at pressure 0.5 -0.7 Torr. Formamide (NH2CHO, Aldrich) and formamide-N,N-d2 (ND2CHO, CDN Isotopes; 99.6% D) were introduced into the reactor by Ar flow over a liquid sample. Concentration of formamide, ≈ 8×1012 molecule cm-3, was estimated by the rate of sample evaporation. Comparison of spectra recorded at different pressures showed that spectra acquired between 0.5 and 0.7 Torr were suitable for analysis without vibrational relaxation. For investigation of secondary reactions, the OH(OD) and NO2 concentrations were varied over the (1.3-12) × 1012 and (0.9-19) × 1013 molecule cm-3 range, respectively, giving [NO2]/[OH] from ~4 to 112. Changes were observed for the CO2 spectra only when the concentration of NO2 was at least 20 ACS Paragon Plus Environment


Page 6 of 34

6

times higher than that of OH, while most of the experiments were carried out with (4-7)-fold excess; therefore, other secondary reactions were not considered (a possible reaction of NH2CO with NO2 is discussed in the Supporting Information). 3. Methods of Analysis 3.1. H2O, HOD, D2O and CO2 emission spectra. The experimental spectra were modeled as a superposition of the emission from the vibrational v1,3v2 states of H2O (D2O) (∆v3 = -1 and ∆v1 = -1 transitions), and v3v1,2 states of HOD (∆v3 = -1 transitions). Representation by v1,3 and v1,2 is a consequence of the collisional coupling of the resonant ν1 and ν3 levels of H2O (and D2O) and the ν1 and 2ν2 levels of HOD by collisions with the Ar carrier gas (equilibration between the resonant levels occurs at approximately gas kinetic rate16). Construction of the model emission bands was described earlier7,8. The bands are for 298 K rotational distribution within the vibrational state. Least-squares fitting of the reference spectral bands to the experimental spectrum gives vibrational distributions of these states. Summation over the bending (bending + O-D stretch in case of HOD) states gives the so-called stretching populations P1,3(v1,3) of H2O and D2O, and P3(v3) of HOD. Summation over the v1,3 levels gives the overall bending distribution of H2O and D2O. In the present work all bands of H2O and HOD, including hot and combination ones, were renewed using the recent HITRAN data17. The HOD lines absent in HITRAN were partly taken from the calculation of Voronin et al.18. The improved description of the 003 hot band of HOD and the equilibrated 003/102/201/300 band of H2O made it possible to determine low populations of the v3 = 3 state of HOD and v1,3 = 3 state of H2O. The hot and combination bands of D2O were constructed using the theoretical line list of Shirin et al.19, whereas earlier modeling was based on shifting the fundamental bands. To estimate the population of the non-emitting states, the surprisal analysis20 was used. Adaptation of surprsal analysis to H2O, HOD and D2O for estimation of relative populations of P1,3(0) and P3(0) was described in Refs. 7 and 8, and the plots used in this work are shown in the ACS Paragon Plus Environment

Page 7 of 34 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


7

Supporting Information. Analysis of CO2 spectra has been fully described elsewhere13,15. Model spectra for least-squares simulation of the experimental spectra were calculated as a superposition of ∆v3 = -1 transitions from (v1,v2l,v3) combination bands with rigid-rotor approximation for ro-vibrational line intensities. In the present work, positions of the line centers were corrected for Fermi-resonance interaction within v1-2v2 polyads, using expressions from Suzuki21. 3.2. NH3, NH2D, ND2H and ND3 emission spectra. This work is the first attempt to observe and analyze emission spectra from ammonia molecules produced by a gas-phase reaction. Spectroscopic constants of ammonia deutero-isotopes available in literature and used in the analysis are given in the Supporting Information. From the six normal modes, ∆v = -1 transitions of N-H and N-D stretches (3 modes in each isotope) fall into the spectral range of our detector. An important feature of ammonia vibrational levels is inversion splitting that complicates spectra analysis. NH3 emission was modeled using ν1 and ν3 fundamental bands constructed using line positions and intensities from the HITRAN database. Conversion from absorption to emission intensities was done as for H2O7. By analogy to H2O, the states with excited symmetric stretching vibration ν1 (band center ν0 = 3336.1 cm-1) and doubly degenerate antisymmetric stretching vibration ν3 (ν0 = 3443.6 cm-1) were assumed to be collisionally equilibrated. The same assumption was made for the ν1 (ν0 = 2420.1 cm-1) and ν3 (ν0 = 2563.9 cm-1) modes of ND3. The spectra were modeled as a superposition of the ∆v3 = -1 and ∆v1 = -1 transitions from the excited vibrational v1,3v2v4 states of NH3 (ND3), where v2 and v4 refer to inversion and doubly degenerate bending vibration, respectively. The fundamental emission bands from the deutero-isotopic ammonia molecules were calculated using tables of absorption lines of stretching transitions reported by Snels et al.22-24. Besides the v1,3 bands of ND322, there are isolated ν1 bands of ND2H (N-H stretch) and NH2D (N-D stretch) and collisionally ACS Paragon Plus Environment


Page 8 of 34

8

equilibrated ν3a3b combined bands calculated from the symmetric ν3a and antisymmetric ν3b stretches23,24. The combination and hot bands were approximated by shifting the ν1 and ν3 fundamentals. Positions of the hot and combination bands of NH3 and, where possible, of ND3 were determined using the known band centers25-31; those of NH2D and ND2H were approximated with the help of anharmonic constants32-34. In ammonia, with strong centrifugal distortion and multiple rotationinversion interactions, a strong distortion of rotational structure is expected. However, the present-day spectroscopic database for ammonia deutero-isotopes does not enable more accurate construction of the bands. Another peculiar complication is that the hot and combination bands can be shifted to both sides with respect to the fundamentals, unlike H2O and HOD, where emission from the combination states is always red-shifted. The inversion-combination bands of NH2D and ND2H are all predicted to be blueshifted33,34. The absorption band-sum intensities for the stretching vibrations of all four ammonia isotopic molecules, Sv, were given by Koops et al.35 The band-sum intensities of the equilibrated v1,3 emission bands of NH3 and ND3 relate to that of H2O and D2O as 0.13 and 0.14, respectively. More details are given in the Supporting Informaiton. It is important to note that the ro-vibrational emission transitions of the CO2 and isotopic NH3 and H2O bands used for modeling in the present study correspond to actual intensities obtained by a uniform conversion of the absolute absorption intensities. Hence, the ratio of the sum probabilities gives the concentration ratio. 4. Results 4.1. Chemical systems. Typical full-range spectra from the OH (and OD) + NH2CHO and OH (and OD) + ND2CHO reactions recorded without a filter and corrected for the detector response are presented in Figure 1. These spectra reflect the main characteristics of the four isotopic systems studied:


Page 9 of 34 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


9

System 1. OH + NH2CHO NH2CHO + OH → H2O + NH2CO

(1)

NH2CO + OH → NH2COOH → NH3 + CO2

(2a)

→ H2O + HNCO

(2b)

System 2. OD + NH2CHO NH2CHO + OD → H-OD + NH2CO

(1D)

NH2CO + OD → NH2COOD → NH2-D + CO2

(2Da)

→ H-OD + NHCO

(2Db)

System 3. OH + ND2CHO ND2CHO + OH → H2O + ND2CO

(1d)

ND2CO + OH → ND2COOH → ND2-H + CO2

(2da)

→ HO-D + DNCO

(2db)

System 4. OD + ND2CHO ND2CHO + OD → H-OD + ND2CO

(1Dd)

ND2CO + OD → ND2COOD → ND3 + CO2

(2Dda)

→ D2O + DNCO

(2Ddb)

The spectrum from System 1 (Figure 1a) consists of a mixture of the primary and secondary H2O emission (3200-3900 cm-1) with addition of NH3 emission in the 3000-3500 cm-1 region. All spectra include CO2 emission in the 2000-2400 cm-1 region. The spectrum from System 2 (Figure 1b) includes the primary HOD mixed with the secondary HOD for both 3200-3900 cm-1 and 2400-2900 cm-1 regions of the spectrum; the same regions include N-H and N-D stretch bands of NH2D. System 3 (Figure 1c) gives emission from the primary H2O (3200-3900 cm-1) contaminated with N-H stretch band of ND2H; the 2400-2900 cm-1 region consists of emission from secondary HOD (O-D stretch) and ND2H (N-D stretches) free from contribution from the primary reaction. Only System 4 (Figure 1d) produces a spectrum of the pure primary product, H-OD (3200-3900 cm-1), which does not have overlap from any secondary product. The secondary products, ND3 and D2O, emit in the 2300-2900 ACS Paragon Plus Environment


10

cm-1 region, where contribution from the primary HO-D must also be present. In viewing Figure 1, one should remember that the emission strengths of CO2 (001), HOD (001 and 100), NH3 (1000), and ND3 (1000) relate as 100:(20.1 and 3.5):2.4:0.32. 0,02

a) NH2CHO + OH

CO2

H2O

0,01

NH3

0,00 0,02

H-OD

b) NH2CHO + OD

Intensity (a.u.)

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 10 of 34

0,01 NH2D

NH2-D, HOD 0,00 0,02

c) ND2CHO + OH

H2O

0,01

ND2-H

ND2H, HO-D 0,00 0,04

d) ND2CHO + OD 0,02 0,00 2000

H-OD

ND3, D2O, HO-D 2200

2400

2600

2800

3000

3200

3400

3600

3800

Wavenumbers (cm-1)

Figure 1. Spectra from the four reaction systems (corrected for the apparatus response) recorded without the filter at P = 0.7 Torr (τ = 0.2 ms), [OH] ≈ 7 × 1012 and [NO2] ≈ 5 × 1013 molecule cm-3. 4.2. Primary reaction of OH and OD radicals with formamide. Figure 2 demonstrates modeling the HOD spectrum from reaction (1Dd) of System 4 and H2O spectrum from reaction (1) of System 1. To acquire the spectra suitable for modeling the primary H2O formation, the concentration of OH was lowered in order to decrease the concentration of the secondary products. For HOD, a significant population exists in the first and second v3 stretching levels, P3(1:2) = 66.2: 33.8, with maximum bending numbers v2=4 in v3=1 state (25.8 kcal/mol) and v2=2 in v3=2 state (28.4 kcal/mol). The population of the level v2=5 in v3=1 state (29.5 kcal/mol) was less definite. At the same time, inclusion of the 032 (32.1 kcal mol-1) band made the fitting worse. The characteristic peaks of the v3 = 3 level without bending excitation (30.4 kcal mol-1) were not identified. Note, that the ACS Paragon Plus Environment

Page 11 of 34

11 300

a) H-OD

250 200

2

3

1

= v3

ND2CHO + OD

150

Intensity (a.u.)

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


100 50 0 3200 350

3400

b) H2O

300 250

3300

3500

3600

3700 2

3

3800 1

3900

= v1,3

NH2CHO + OH

200 150 100 50 0 3200

3300

3400

3500

3600

Wavenumbers (cm-1)

3700

3800

3900

Figure 2. a) HOD emission from ND2CHO + OD reaction ([OH] = 6.6 × 1012); b) H2O emission from NH2CHO + OH reaction ([OH] = 2.9 × 1012). Both spectra were recorded at P = 0.5 Torr (τ = 0.15 ms), and [NO2] = 3.8 × 1013 molecule cm-3. Model spectra are shown in red color. Vertical ticks indicate the centers of the fundamental bands. maximum vibrational energy delivered to HOD corresponds to the available energy and, hence, is a measure of the C-H bond dissociation energy in formamide. The average available energy is defined as = -∆Ho0 + Ea + ET, where -∆Ho0 is the reaction enthalpy, Ea is the activation energy and ET = 4RT, assuming that the thermal energy consists of rotational and translational components of reactants. Reaction (1) is either barrierless,6 or has a small barrier of 0.16 kcal mol-1 5 and, hence, -∆Ho0 = 4RT. If the maximum vibrational energy is determined as the average of the highest observed and the lowest not observed levels, then = 30.3±1.0 kcal mol-1. As for OD reactions is 0.3 kcal mol-1 larger than that for OH reactions (zero-point energy correction), we obtain for the NH2CHO + OH reaction = 30.0 kcal mol-1 and -∆Ho0 = 27.6 kcal mol-1. Since the reaction enthalpy is equal to D0(R-H) - D0(HO-H), taking D0(HO-H) = 118.08+0.03 kcal mol-1 36, we obtain D0(NH2C(O)-H) = 90.5+1.3 kcal mol-1. The latter is ≈3 kcal mol-1 higher than D0(CH3C(O)-H) = 87.9+0.3 kcal mol-1 36. Our experimental value can be compared to the values of 93.75 and 93.16 kcal mol-1 derived from the recent ab initio studies, which is 3-5 kcal mol-1 lower than those from the earlier theoretical study of ACS Paragon Plus Environment


Page 12 of 34

12

Kaur et al.37. We did not find experimental D0(NH2C(O)-H) values in the literature, but our value agrees with the result of an arrested-relaxation chemiluminescence study of the F + ND2CHO reaction38. There is an experimental D0(N-H) value39, 110+2 kcal mol-1, with can be compared with those from the above calculations: 115.05, 113.16 and 114.5 kcal mol-1 40. The difference between the experimental and theoretical values is exactly the same as in the case of D0(C-H). At first, the population of the P3(0) state of H-OD from OD + ND2CHO was determined from the surprisal plots calculated for = 30.3 kcal mol-1, using two models for the prior: model 1 treats the ND2CO radical as an atom and model 2 includes three rotations associated with the radical (see Supporting Information). Linear surprisal plots gave P3(0) = (29.0±2.3)% (model 1) and (37.2±2.4)% (model 2), where the error limits correspond to 1 kcal mol-1 uncertainty in . The renormalized full stretching distributions are P3(0-3) = 37.2: 41.5: 21.2:

Infrared Chemiluminescence Study of the Reaction ... - ACS Publications

Recommend Documents