Embed Size (px)
Citation preview
Molecular beam photoionization study of acetone and acetoned 6Wayne M. Trott, Normand C. Blais, and Edward A. Walters Citation: The Journal of Chemical Physics 69, 3150 (1978); doi: 10.1063/1.437009 View online: http://dx.doi.org/10.1063/1.437009 View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/69/7?ver=pdfcov Published by the AIP Publishing Articles you may be interested in An ab initio structural and spectroscopic study of acetone—An analysis of the far infrared torsional spectra ofacetoneh 6 and d 6 J. Chem. Phys. 98, 2754 (1993); 10.1063/1.464157 Molecular beam photoionization study of S2 J. Chem. Phys. 84, 778 (1986); 10.1063/1.450576 Molecular beam photoionization study of HgCl2 J. Chem. Phys. 78, 37 (1983); 10.1063/1.444512 Molecular beam photoionization study of carbon disulfide, carbon disulfide dimer and clusters J. Chem. Phys. 73, 2523 (1980); 10.1063/1.440486 A photoionization study of carbon dioxide dimers in a supersonic molecular beam J. Chem. Phys. 68, 1768 (1978); 10.1063/1.435947
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.160.4.77 On: Sat, 20 Dec 2014 05:07:53
Molecular beam photoionization study of acetone and acetone-ds
Wayne M. Trott
Department of Chemistry. University of New Mexico, Albuquerque, New Mexico 87131
Normand C. Blais
University of California, Los Alamos Scientific Laboratory, Los Alamos, New Mexico 87545
Edward A. Walters
Department of Chemistry, University of New Mexico, Albuquerque, New Mexico 87131 (Received 26 May 1978)
High resolution photoionization efficiency curves have been obtained for CH3COCHt and CDJCOCDt using supersonic molecular beam sampling. As a result of adiabatic cooling during the nozzle expansion, sufficient concentrations of (CHJCOCHJ)l' (CDJCOCDJ)l' (CHJCOCHJ)J' and (CHJCOCHJ)4 were formed to permit the study of their photoion yield curves as well. Appearance potential curves have been determined for CHJCO+, CDJCO+, and (CHJCOCHJ).CHJCO+ fragments. The measured ionization potentials of acetone and acetone-d6 monomers are 9.694±0.006 and 9.695±0.006 eV, respectively. Transitions to higher vibrational levels in CHJCOCHt are seen at 320, 695, and 930-1370 cm~1 above threshold. The effect of perdeutero substitution is to reduce these frequencies to 260 and 660--1100 cm~l. Appearance potentials of CHlCO+ and CDJCO+ fragments are observed at 1O.52±0.02 and 1O.56±0.02 eV, respectively. The measured ionization energies for (CHJCOCH3)n' n = 1-4, are found to decrease linearly as a function of l!n. Observed ionization thresholds for (CHJCOCHJb (CH3COCHJ)J' and (CHJCOCHJ)4 are 9.26±0.03, 9.1O±0.03, and 9.02±0.03 eV, respectively. Within experimental resolution, the ionization potentials of (CHJCOCHJ)l and (CDJCOCD3)l are identical. The appearance potential of the process (CHlCOCHJ)l -{CHJCOCH3)·CHlCO+ +CHl + e ~ is found to be 1O.08±0.05 eV. By consideration of appropriate thermodynamic cycles, a lower bound for the acetone dimer ion binding energy is calculated to be 0.538 eV (12.4 kcal/mole) and the desolvation energy of (CHlCOCH3)·CH3CO+ is estimated to be 0.544 eV (12.5 kcal/mole).
I. INTRODUCTION
The application of high resolution molecular beamphotoionization mass spectrometry to complex organic molecules remains a largely unexplored possibility. Except for monomer and cluster studies of ethylene1
and smaller alcohols, 2.S this method has generally been confined to experiments involving parent molecules with five atoms or less. 4- 14 Departing from this trend, the present study is concerned with the photoionization of several species produced in supersonic nozzle expansions of acetone and acetone-d6 vapor. As the simplest aliphatic ketone, the ionization and dissociative properties of acetone are of Significant interest as being representative of a sizable homologous class of substances. Consequently, this compound has been extensively studied by various methods including electron impact, 15-S0 photo ionization, Sl-S9 photoe lectron spectroscopy, 4O-5S photoelectron-photoion coincidence spectroscopy, 54,55 vacuum ultraviolet absorption spectroscopy,56 Penning ionization,57.58 charge transfer spectra,59 and electron energy loss spectroscopy.60 The more recent of these experiments have produced fairly consistent first and higher ionization potential values for the parent species and appearance potential values for the acetyl ion. Little additional ionic state information is known, however.
The effective resolution of conventional photoionization mass spectrometric methods is limited by the thermal vibrational and rotational energy spread of the sample molecules. By introducing the target gas through a
supersonic nozzle expansion, it is possible to reduce the rotational envelope of a poly atomic sample from approximately 3/2 kT or 0.04 eV at room temperature to 0.01 eV or less. 11.1S Substantial vibrational relaxation also occurs. Using nozzle beam sampling techniques, we have measured more precisely the photoionization efficiency curves for CH3COCH; and CHsCO' in order to obtain such data as low energy ionic vibrational frequencies, Franck-Condon factors for direct ionization, and effects of thermal relaxation on threshold values. Photoion yield curves have been determined for CDsCOCD; and CDsCO' as well. In addition, we report photoionization mass spectrometric measurements of (CDsCOCDs)i, (CHsCOCHs);' n = 2-4, and (CHsCOCHs) . CHsCO'.
II. EXPERIMENTAL
The apparatus used in this study consists of a moderate resolution mass spectrometer which has been modified to include a supersonic molecular beam sampling system and a photon ionization source. Figures 1 and 2 illustrate the various elements in the experimental design.
A molecular beam is obtained by expanding a gas through a 0.004" diameter nozzle at stagnation pressures of up to 2000 torr. The nozzle exhaust chamber is pumped by two 4" ring jet booster pumps which maintain a vacuum of -lO-s torr. The high intensity central portion of the expansion is extracted with a 0.018" diameter conical skimmer into a second differential pump-
3150 J. Chern. Phys. 69(7), 1 Oct. 1978 0021-9606/78/6907-3150$01.00 © 1978 American Institute of Physics
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.160.4.77 On: Sat, 20 Dec 2014 05:07:53
Trott, Blais, and Walters: Photoionization of acetone 3151
TO LN2 BEAM TRAP
ELE'CTROSTATIC QUADRUPOLE
10
TO MASS ANALYZER
FIG. 1. Nozzle beam sampling and ion source: (1) ion box at + 10 kV (see text), (2) 12 x 10 MO to provide uniform acceleration of ions over 10 kV, (3) tuning fork chopper, (4) secondary skimmer, (5) electromagnetic beam flag, (6) primary skimmer and skimmer support, (7) nozzle, (8) nozzle support and bevel gears for nozzle advance, (9) nozzle exhaust chamber, (10) second differential pumping chamber; (PI) 575 l/sec (see text), (P 2) 700 l/sec (see text), (P 3) 1400 l/sec 6" diffusion pump.
ing region. This chamber is pumped by a 4" diffusion pump to a pressure of less than 10-5 torr. The expansion is further collimated by a O. 040" conical skimmer before entering the ionization chamber. Provision is made to flag the beam. A liquid nitrogen-cooled surface is used to trap those molecules which escape ionization. The main chamber pressure is maintained at less than 3 x 10-7 torr when the beam is running. On the basis of electron impact measurements on several compounds, beam densities are estimated to be 1-3 x 1012
molecule/cc. in the ionization region.
In the present work, an intense expansion of CH3COCH3
was produced by seeding acetone vapor at 20-25 °c in H2. Total stagnation pressures were varied between 350-750 torr in order to maximize the concentration of given components in separate experiments. Inasmuch as some enhancement of cluster ion intensities relative to the parent could be achieved by increasing the beam stagnation pressure, the ratio of Hz to acetone used was generally higher for the polymer studies. However, the functional dependence of cluster ion signal and H2 source pressure was not found to be particularly strong. Considerable care was taken to maintain a constant acetone source temperature and total stagnation pressure throughout each run. Because mass analysis was employed during the whole course of the present study, Mallinckrodt Spectrar grade acetone (99.5% minimum purity) was used without further purification; 99.5 atom % D acetone-d6, obtained from Aldrich Chemical Co., Inc., was used directly for the isotope work.
The many-line spectrum of H2 produced by a McPherson Model 630 dc capillary discharge lamp is used as a photon source. In view of the performance results of cann,61 the standard 6 mm discharge capillary is replaced by one with a 3 mm bore. In order to extend the photon wavelength range below 1040 A (the transmission
limit for lithium fluoride), the light source-monochromator system is required to operate windowless. Consequently, the entrance arm of a 0.5 m Seya-Namioka vacuum ultraviolet monochromator has been modified to include two stages of differential pumping. With the lamp operated at 1 torr H2 and with the entrance slit at 200 /J., the normal operating pressure in the monochromator is 2.5 X 10-5 torr.
The principal optical component is a MgF 2 coated, 2400 line/mm holographic grating blazed at 1200 A. Light dispersed from the grating is focused on the monochromator exit slit and refocused into a line image in the center of the ionizer by a platinum-coated 80° angle of incidence alignment mirror. With 200 /J. entrance and exit slits, the resolution is approximately 1.5 A or 12 meV at 1200 A. The photon beam is essentially unattenuated by the molecular beam and the transmitted photon intensity is determined by using a photomultiplier to monitor the fluorescence from a sodium salicylate coated window. The quantum efficiency of a fresh coating of sodium salicylate is known to exhibit the following consistent behavior: A constant yield from 400-1250 A, followed by a generally monotonic increase of - 2<Jl, up to 1500 A.62 A picoammeter is used to measure the current output from the photomultiplier. Signals are typically one to two orders of magnitude higher than the photomultiplier dark current.
The ionizer consists of a gold-plated box which contains entrance and exit ports for the dispersed ultra-
5
6
FIG. 2. Schematic diagram of photoionization mass spectrometer: (A) light source-monochromator system, (B) ionization chamber, (e) mass analyzer and detector; (1) McPherson Model 630 dc capillary discharge lamp, (2) monochromator entrance slit, (3) first differential pumping chamber, (4) second differential pumping chamber, (5) 0.5 m Seya-Namioka vacuum ul,traviolet monochromator, (6) grating, (7) monochromator ex-it slit, (8) grazing incidence alignment mirror, (9) micrometer for mirror adjustment, (10) molecular beam crosses photon beam at right angles, (11) ion box, (12) photomultiplier tube, (13) sodium salicylate coated window, (14) uniform acceleration region, (15) electrostatic quadrupole, (16) mass spectrometer entrance slit, (17) drift tube, (18) 60°-sector electromagnet, (19) mass spectrometer exit slit, (20) electron multiplier; (Pt ) 47l/sec Roots blower, (P z) 700 l/sec 4" diffusion pump, (P3) 700 l/sec 4" diffusion pump, (P4) 1400 l/sec 6" diffusion pump, (P,) 700 l/sec 4" diffusion pump.
J. Chern. Phys., Vol. 69, No.7, 1 October 1978
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.160.4.77 On: Sat, 20 Dec 2014 05:07:53
3152 Trott, Blais, and Walters: Photoionization of acetone
violet light, a filament-collector arrangement to provide electron bombardment capability, and an inner surface to which a small repelling voltage (+ 0-20 V) may be applied. Extracted positive ions are uniformly accelerated by a + 10 kV potential into a pair of quadrupole lenses6s•64 which focus the ions into a line image at the mass spectrometer entrance slit.
Mass analysis is accomplished by a 60° -sector electromagnet with a 17.750" radius of curvature. The drift tube entrance and exit slits are opened wide enough to maximize ion transmission while maintaining sufficient resolution to cleanly separate 1 amu in 300. Conventional pulse counting techniques are used for ion detection. The pressure in the drift tube and detection chamber is maintained at less than 2 x 10"7 torr.
The photoionization efficiency is determined by the ratio of ion signal to photon signal as a function of wavelength. The lamp intensity remains stable to ± 2% at each point. Known hydrogen atom and molecule emission lines are used for wavelength calibration. 65.66
We find background corrections to be quite small. With the molecular beam flagged, the ion counting background rate is about 0.02-0.1 count/sec. Thus, reproducible ion signals of 0.03-0.05 count/sec are detectable. Since a holographically recorded diffraction grating is used, stray light levels are expected to be substantially reduced. Under high resolution conditions, it is possible to reproduce in substantial detail the Hz spectrum given by Samson. 66
Two different monochromator entrance and exit slit widths were used, depending upon the intensity of the specie investigated. Operating conditions for the discharge lamp were typically 0.8 torr Hz pressure and 450 mA discharge current; 200 IJ. slits were used for the acetone and acetone-d6 monomer studies and counts were collected for 500-600 sec at approximately 1 A intervals. Counting rates in the region 1240-1280 A varied from - 15 to - O. 5 count/sec at threshold (at Lyman O!,
9.611
1290
9.686 9.762 E(eV) 9.840
••• I P = 9\.694 ± 0.006 ev ....... ., ..
Cl •• -Qj ., • .....
• • 1280 1270 12~
).(A)
9918 9998 10.080
1230
FIG. 3. Photoion yield curve for (CHsCOCHs)+ from 1240-1285 A.
9.611 9.686
I I o 260
1290 1280
9.762
E(eV)
9.840
Q 660-1100
em-I
1270 x(A) 1260
9.918 9.996
1250 1240
FIG. 4. Photoion yield curve for (CDsCOCDs)+ from 1248-1282 A.
counting rates could be varied from about 200 to 650 count/sec, depending upon lamp pressure and nozzle expansion conditions). For cluster and fragment ion studies, 500 IJ. monochromator entrance and exit slits were used, resulting in a resolution of approximately 3.75 A or 30 meV at 1200 A. Readings were taken at 1. 5-2. 5 A intervals for 600-900 sec. Typical counting rates ranged from about 0.03-5 count/sec. In each run, a reference point was repeated periodically to account for any drift in the mass peak, collection efficiency, etc.
III. RESULTS
A partial mass spectrum for an acetone-Hz molecular beam (600 torr total stagnation pressure) was obtained using 100 eV electron bombardment. Signals corresponding to acetone monomer and cluster ions were observed at m/ e 58, 116, 174, 232, 290, and 348. Fragment ions at m/e 43 and 101 were seen in concentrations x 3 - x 4 those of the monomer and dimer ion, respectively. In addition, small peaks corresponding to ions of the form (CHaCOCHa)nH" n = 1-6, were observed at m/e 59, 117, 175, 233, 291, and 349. These ions presumably originate from high energy fragmentation of cluster species or ion-molecule reactions involving some of the numerous parent, cluster, and fragment ions formed; i. e., (CHaCOCHa):, (CHsCOCHa)n . CHaCO" CH;, H2, H+, etc. Protonated acetone and acetone cluster ions have been previously reported in field ionization67 and ion-molecule reaction66- 74 studies.
In contrast, the photoionization mass spectrum for the acetone-Hz beam was markedly less complex in the region 1100-1300 A. At 500 torr total stagnation pressure, ions of the form (CHsCOCHs):, n = 1-4, were observed at Lyman °! in the apprOximate ratio 100:40:5:3. Signals at m/ e 59, 117, 175, etc., were seen only in those concentrations which would be expected on the basis of natural isotopiC abundances of lSC and D. At 1160 A, the CHsCO+ and (CHaCOCHs) . CHaCO+ ions (m/ e 43 and 101) could be seen in very low concentrations.
J. Chern. Phys., Vol. 69, No.7, 1 October 1978
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.160.4.77 On: Sat, 20 Dec 2014 05:07:53
Trott, Blais, and Walters: Photo ionization of acetone 3153
TABLE 1. Summary of energetic information for acetone and acetone-d6 parent, fragment, and cluster ions.
Ionization or appearance potential (eV)
Ion m/e This Work PI literaturea PES literatureb
Ionic vibrational structure
CH3COCH3 58 9. 694± 0.006 9. 71±0. 0132
9.69±0.0l35 9.709 ± 0.00540
9. 70±0. 0143
9.72 ± 0.0148
320±50 cm- I
695 ±50 cm-! 930-1370 cm-! 260±50 cm-I
660-1100 cm-! CD3COCDj 64 9.695±0.006
CH3CO· 43 10.52±0.02 10.33,33 10.3734 (298 OK) 10.4533 ,34 (OOK)C
10. 4232 (298 OK)
CD3CO' 46 10.56±0.02
(CH3COCH3)i 116 9.26 ± O. 03
(CD3COCD3>2 128 9.25±0.03
(CH3COCH3)j 174 9.10±0.03
(CH3COCH3); 232 9.02±0.03
(CH3COCH3)
• CH3CO· 101 10.08±0.05
"Selected photoionization literature values. tJselected photoelectron spectroscopy literature values. "rhis value determined by adding the calculated internal energy of acetone at room temperature to the measured 298 OK acetyl ion appearance potential.
The relative Simplicity of the photoionization spectra was taken as strong evidence that only simple ionization and fragmentation processes occur. 73
Photoion yield curves obtained in the present work are shown in Figs. 3-10. With the exception of the acetone tetramer and acetone-ds dimer, each ion was investigated at least twice. The spectra were found to be reproducible in substantial detail. The measured ionization potentials and appearance potentials of the various species are given in Table I, along with ionic vibrational structure information which has been derived from the experimental curves.
IV. DISCUSSION
A. Acetone and acetone-d 6 monomer ions
The photoionization efficiency curves for acetone and acetone-ds parent ions are characterized by sharp thresholds and distinct steplike structure at higher energies. The well-resolved fine structure and the apparent absence of hot band effects are direct results of the efficient thermal relaxation achieved by sampling from a supersonic nozzle expansion. Within the estimated resolution of 1. 5 A, the observed first ionization onsets are very nearly step functions. This result would indicate that the rotational envelope for acetone has been narrowed to approximately a width of 12 meV or less through supersonic gas expansion; i. e., the rotational temperature of the sample is - 100 OK. Rotational temperatures of ~ 118 OK have been achieved in previous nozzle beam photoionization studies. n ,13 Similarly, vibrational temperatures are apparently lowered to the
extent that significant population of states occurs only within a narrow energy band above the ground state. The measured iOnization potential of CH3COCHs is in good agreement with previous photoionization and photoelectron spectroscopy results (see Table I). Ionization probably occurs at a 2p lone-pair electron site on the oxygen atom, 75 resulting in a large v = 0 to v' = 0 onset. 38
In addition, the observed photoion yield curve exhibits several fine structure features within 0.2 eV of threshold, including two abrupt steps at 320 and 695 cm-l and a broad rise from about 930-1370 cm- l • The estimated unce rtainty of these values is :!: 50 cm -1. These features presumably arise because sufficient excess energy is available to excite higher vibrational levels of the ion. Removal of a nonbonding electron should have a minimal effect on many low energy vibrational frequencies of the molecule since these frequently correspond to structural deformations. Consequently, it should be possible to correlate the observed ionic transitions with known low energy vibrations in acetone. 7S,77
Steps corresponding to measured methyl group torsional frequencies in neutral acetone "12 = 105.3 cm-l
and "24 = 108. 4 cm-l are not seen in this work. Since the ionic torSional modes are not totally symmetric (assuming CHsCOCHs retains Cav symmetry upon ionization), vibrational selection rules indicate that transitions to these levels from the neutral acetone ground state are symmetry forbidden. 78,79 Also, it is quite likely that the target molecules do not experience sufficient vibrational relaxation during the expansion to permit observation of these very low energy transitions. Indeed, these frequenCies lie nearly at the limit of the experi-
J. Chern. Phys., Vol. 69, No.7, 1 October 1978
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.160.4.77 On: Sat, 20 Dec 2014 05:07:53
3154 Trott, Blais, and Walters: Photoionization of acetone
E(eV) 10419 10507 10597 10688
1190
I I
CH3CO CH 3+h. - CH3CO·. CH3 +e-
CD3COCD3+hv- CD3CO++CD3 •• -
APCH3CO+1 -10.52 ±o.02.V
I 1180 1170 1160
A(A)
10781 I
o CH 3CO'
o CD3CO'
I 1150
loa76
1140
FIG. 5. Appearance potential curves for CH3CO· and CD3CO· from 1148-1183 A.
mental optical resolution. The onset of 320 cm-! appears to correspond with the A! symmetric C-C-C deformation mode (va), to which is assigned the 385 cm-I
band in the vibrational spectra of the neutral molecule. This ionic transition was tentatively identified in a previous photoelectron spectroscopy study.4a Several normal modes could conceivably give rise to the observed feature at 695 cm-!, including the symmetric C-C stretch (v7) and two out-of-plane skeletal fundamentals (V I9 ' vza)' In neutral acetone, these modes occur at 777, 530, and 484 cm- I
, respectively. Since the out-ofplane vibrations involve the carbonyl group directly, these frequencies could be substantially altered upon ionization. To a lesser extent, a shift is expected for the C-C stretch as well. Within the estimated resolution limits, a possible contribution from 2va must also be considered. We do not attempt a specific assignment for this transition; however, the vibrational selection rules favor the Al symmetric modes (v7 , 2va). In previous studies, an ionic vibrational feature has been observed at - 1200 cm-1 which has been assigned to either the C-O stretch32 or the methyl group symmetric deformation mode. 4a In the present work, this feature appears significantly broader than the lower energy transitions. It seems likely that the manifold at 930-1370 cm-1 arises from several unresolved vibrational excitations, possibly including the CH3 asymmetric and symmetric deformation modes (v4, v5, V15, ViS, v21), the CHa rocking modes (vs, vl1, v1S' v22 ), and the asymmetric C-C stretch (V17)' A broad peak centered at about 1300 cm-! in the acetone electron energy loss spectrum has been interpreted in a similar fashion. so Again, the Al symmetric modes (v4, vs, vs) are favored by the selection rules. At energies above 1400 cm-1
from threshold, the photolon yield curve becomes very complex. This is due principally to the increasing superposition of iOnization continua and the onset of dissociative processes. The occurrence of autoionization at 0.1-0.2 eV above threshold has also been suggested. 41
Within stated error limits, the measured ionization
potential of CD3COCD3 is identical to that for the acetone parent (see Table 1). Fine structure in the acetone -ds photoionization curve consists of a transition at 260 cm-I
and a broad rise from about 660-1100 cm-I • In the vibrational spectra of the neutral molecule, perdeutero substitution has the effect of reducing va by about 65 cm-I
and lowering the CH3 rocking and deformation modes by 160-400 cm-I
• 77 A reduction in v7 and [,/17 by about 100 and 200 cm -I, respectively, is also observed. These results favor the assignment of the 260 cm-I mode to va and the higher energy manifold to a composite of C -C stretches (v7, v17) and methyl gror,p excitations. Vibrational selection rules apply as in the case of CH3COCHs'
In view of the aforementioned wavelength dependence of sodium salicylate fluorescence quantum efficiency between 1240-1500 A, care must be exercised in interpreting the parent ion efficiency curves over a wide energy range; however, this effect is undoubtedly small « O. 5%) for individual features occurring in less than 30 meV. Consequently, it should be possible to derive fairly accurate relative Franck-Condon factors from the step heights of the observed structure near threshold. For CHaCOCH;, the measured relative transition probabilities for the first three onsets are 1. 00 : O. 35 : O. 12. The relative values for the first two steps in the CD3COCD; curve are about 1.00: O. 25. It must be emphasized that the initial thresholds may include some contribution by unresolved transitions to (and from) the low energy methyl group torsional modes.
B. CH 3CO+ and CD3CO+ fragment ions
Appearance potential curves for acetyl fragment ions from acetone and acetone-ds are shown in Fig. 5. The measured threshold value of 10.52 ± O. 02 eV for CH3CO' is 0.1-0.2 eV higher than previous room temperature determinations (see Table I) and is, in fact, 0.07 eV higher than the calculated 0 OK value of Murad and Inghram. 33 The relatively high appearance potential and the near absence of curvature at onset provide evidence of substantial thermal relaxation of the acetone parent molecules in the present work. In view of the apparent uncertainty in the earlier measurements, 32-34 the present determination strongly suggests that the 0 OK appearance potential is likely 10.54 eV or higher.
Murad and Inghram have used their extrapolated 0 OK acetyl ion appearance potential of 10.45 eV to derive several thermodynamic quantities. 33,34 In these derivations, the kinetic shift in the fragmentation of acetone to form CH3CO· was estimated to be 0.25 eV. Subsequent calculations based on quasi equilibrium theory indicate that this value is likely to be 0.01 eV or less. aD Since the present study favors an appearance potential as much as 0.1 eV higher than 10.45 eV, a cumulative error of up to 0.35 eV for the deduced thermochemical information may have resulted from the low threshold value and overestimated kinetic shift. We do not recalculate these thermodynamic quantities since we have no certain means of determining the 0 OK appearance potential.
The observed threshold for the dissociation of ace-
J. Chern. Phys., Vol. 69, No.7, 1 October 1978
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.160.4.77 On: Sat, 20 Dec 2014 05:07:53
Trott, Blais, and Walters: Photoionization of acetone 3155
..... Z Q
9.184
1350
E(eV) 9.537
1300 ).(A)
9.918
1250
FIG. 6. Photoion yield curve for (CH3COCH3>2 from 1239-1351 A.
tone-d6 to form CDsCO' is 10. 56± 0.02 eV. This represents a secondary isotope effect in the appearance potential of about 0.04 eV upon perdeuteration. This effect was clearly observed in three separate measurements of the CHsCO' and CDsCO' onsets. By comparison, a primary isotope effect of 0.092 eV has been seen in the formation of HCO' and DCO' from H2CO and D2CO, respectively. 81
C. Acetone and acetone-d6 cluster ions
The photoionization efficiency curve for (CHsCOCHs)i is shown in Fig. 6. In contrast to the sharp thresholds obtained in the monomer work, the photoion yield of the dimer exhibits a weak onset and generally increases slowly as a function of excess energy. This behavior has been observed in previous cluster studies405 •
11 and is best explained by the possibility of a substantial change in the potential energy surface upon ionization. For example, Mulliken estimates a 1. 55 A reduction in the equilibrium location of the potential well minimum for Xe2 - xei. 82 Under these circumstances, the FranckCondon factor for direct ionization may be quite small. Indeed, the possibility exists that the adiabatic transi-
9.184
1350
E(eV) 9537
IP=9.25±0.03eV
1300 ).(A)
9918
1250
FIG. 7. Photoion yield curve for (CD3COCD3)2 from 1270-1349 A.
E(eV) 8.984 9.050 9.116 9.184 9.252 9.322 9.393
JP=910±003eV
1380 1370 1360 1350 1340 1330 1320
).(A)
FIG. 8. Photoion yield curve for (CH3COCH3l3 from 1334-1369 A.
tion will not be seen above background. 4 Within the sensitivity of the present experiment, the ionization potentialof (CHsCOCHs)2 is observed at 9. 26± 0.03 eV. Much of the fine structure in the dimer ion curve is reproducible; however, we do not attempt specific assignments in view of the low resolution employed in this measurement, the obvious complexity in the vibrational structure of this system, and the unknown contribution to the photoion yield above threshold due to fragmentation of higher clusters. 4 Because of the ratios of dimer ions to higher cluster ions observed in the present work (see Sec. III), the latter effect is expected to be small.
The ionization potential determination for (CHsCOCHs)2 permits the calculation of the dimer ion binding energy by consideration of the following reactions:
(1)
CHsCOCHs + CHsCOCHs - (CHsCOCHs)2 (2)
(CHsCOCHs)2 - (CHsCOCHs)! + e- (3)
(CHsCOCHs)! + e- - CHsCOCH; + CHsCOCHs + e-. (4)
The sum of the enthalpies for (2), (3), and (4) equals the observed monomer ionization potential of 9.694 ± O. 006 eV. No experimental value for the neutral dimer binding energy has been reported. However, an estimate of this quantity may be made using known sec-0nd virial coefficient data. Braun and Leidecker have calculated the water dimer binding energy83 (BE) by assuming an ideal two component (monomer and bound dimer) mixture of H20 vapor for kT« BE. Under similar assumptions, the energy of formation ~u(T) for Reaction (2) at fixed values of T and P is given by
(5)
where bo is the so-called "excluded volume" of the gas, 84 B 2 (T) are the experimental second virial coeffiCients, and (llu(T) > is averaged over all internal states of the dimer. As T nears 0 OK, (llu(T) > approaches the bind-
J. Chern. Phys., Vol. 69, No.7, 1 October 1978
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.160.4.77 On: Sat, 20 Dec 2014 05:07:53
3156 Trott, Blais, and Walters: Photo ionization of acetone
8984 9. 9.116 E(eV) 9.184 9252 9.322
.l!! (CH3CO CH3)4 + Iw - (CH3CO CH3): + e-. ., :J ,., .. ~ :e (p= 9.02 ±O.03 eV ..s Z § J: ~ s
t t t t H z g
1380 1370 1360
t t t r t
1350 ).(A)
t I I! H j
1340 1330
.93
1320
FIG. 9. Photoion yield curve for (CHaCOCHa), from 1324-1379 A.
ing energy of the dime ric specieso Hence. by plotting the available In[bo -Bz(T)] data for acetone as a function of l/T, the binding energy may be estimated from the low temperature slope of Eq. (5). We have applied this procedure to several reported sets of Bz(T) data85 - 88
and have obtained calculated BE values in the range 1.6 -2.4 kcal/mole. The data of Anderson et ale 89 indicate a substantially higher value. Taking the average of these determinations, we estimate the acetone dimer binding energy to be about 2.4 kcal/mole or 0.104 eV. USing this value, Reactions (1)-(4) give a dimer ion binding energy of 0.538 eV or 12.4 kcal/mole. The uncertainty in this quantity is probably about ± 1 kcal/mole. In view of the possibility that the dimer adiabatic ionization potential is not observed, 12.4 kcal/mole must be considered a lower bound.
The photoion yield curve for (CDsCOCDs)i is shown in Fig. 7. Within the resolution of the present work, the measured ionization potential of the acetone-da dimer is identical to that of (CHsCOCHs)2' The (CHaCOCHa)i and (CDsCOCDa); curves seem to exhibit considerable
E(eV) 9918 9.998 10.080 10.162 10.246
APICl1,COCHYCH3CO+ 010.08 ± 0.011 011
1250 1240 1230 1210
10. 10419
1200 1190
FIG_ 10. Appearance potential curve for (CHaCOCHa)' CH3CO+ from 1190-1241 A.
~ 9.6
<1 i= 9.4 Z 1LI
~ 92
z o i= 9.0 « N Z 8.8 Q
ru ~ ~ M M M ro M M lin
FIG. 11. Plot of the observed ionization energies for (CHaCOCH3)n' n = 1-4, as a function of lin.
Similarity in their fine structure, suggesting that the observed transitions are fairly insensitive to isotopic substitution. For reasons previously stated, we do not attempt specific assignments.
Figures 8 and 9 contain the photoionization efficiency curves near threshold for (CHaCOCHa); and (CHaCOCH3);.
Ionization potentials for these species are observed at 9.10:1: O. 03 and 9.02 ± O. 03 eV, respectively. Again, it is not certain that these values represent the adiabatic transitions. The extremely low count rates (0.03-0.1 count/sec) in these measurements prevent any useful analysis of the fine structure.
With positive ion energetic information available for polymers up to the tetramer, it becomes possible to investigate the relationship between cluster size and bulk properties. For example, it has been observed that the ionization potential for an alkali metal aggregate containing as few as three or four atoms approaches the calculated effective work function of a continuous metal drop of the same size. 90- 92 In contrast, the ionization potentials for (Hg)n do not fall rapidly toward the work function with increasing n. 9a For compounds such as acetone, results of this type are likely to be of significant interest in that they have a direct bearing on nucleation and condensation phenomena and possibly even on liquid structure.
In the present study, we observe a smooth and systematic decrease in ionization potential with increasing acetone cluster size. As indicated in Fig. 11, the measured ionization energies for (CHaCOCHa)n fall squarely on a straight line when plotted as a function of l/n. It is interesting to compare this functional dependence to that which would be expected on the basis of a simple independent systems model94 for the cluster ion system. Using a straightforward perturbation approach, the energy levels for the individual monomeric units of a linear cluster ion are considered to undergo splitting due to weak nearest neighbor interactions. Long-range interactions are set equal to zero. The much smaller splittings due to the relatively weak interactions in the neutral cluster are ignored. Under these assumptions, the secular equation describing the degenerate manifold for cluster ion size n should be as follows:
J. Chern. Phys., Vol. 69, No.7, 1 October 1978
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.160.4.77 On: Sat, 20 Dec 2014 05:07:53
Trott, Blais, and Walters: Photoionization of acetone 3157
D = n
a-A {3 o o {3 a-A {3 0
o o
{3 a-A {3
o {3 a-A =0 (6)
where a represents the monomer ionization potential and {3 is the nearest neighbor interaction parameter. It can be shown that the minimum transition energy A = a - 2{3 COS(7T /(n + 1». To the extent that the above model accurately represents a cluster ion system, this implies a linear decrease of the ionization potential as a function of cos (7T/(n +1)). Since COS(7T/(n+l)) is very nearly a linear function of lin for n = 2-10, this mOdel produces a good correlation to our data for the first few members of (CHsCOCHs)n' Of course, this analysis does not establish a specific structure for the acetone cluster ions; however, it is intriguing that a satisfactory correlation of ionization potential with cluster size can be achieved using this simple picture.
If the ionization energies for (CHsCOCH3)n, n> 4, continue to decrease linearly as a function of 1/n, Fig. 11 predicts a bulk ionization potential of about 8.8 eV. A value of 8.86 eV is calculated from a plot of ionization potential vs COS(7T/(n + 1». This suggests that the ionization energy of acetone is reduced by about 0.85 eV upon condensation. In view of the uncertainty as to whether our measured cluster ion thresholds correspond to adiabatic transitions, 0.85 eV is probably a lower bound. This value compares favorably to relaxation shifts of 1. 0-1. 65 eV recently observed in the ultraviolet photoemission spectra of seven gases condensed on a MoS2 substrate. 95,96
The photoionization effiCiency curve for the final ion investigated in the present work, (CHsCOCH3)' CHsCO., is shown in Fig. 10. Despite the very low count rate for this specie (0.03-0.3 count/sec), a definite onset is seen at 10.08 ± O. 05 eV. This is taken to be the appearance potential for the process
We are not certain as to the origin of the very few counts observed between 1230-1240 A. The de solvation energy aHa, of (CHsCOCH3) • CHsCO' can be estimated from the enthalpies of Reactions (2) and (7) and the following:
CHsCOCHs + CHsCOCHs - CH3COCHs + CHaCO' (8)
+CHs +e-
(9)
+CHsCO' +CHs+e-.
Using the experimental appearance potentials for (7) and (8) and - 0.104 eV for the enthalpy of (2) as before, aHa is calculated to be 0.544 eV or 12.5 kcal/mole. The similarity between this value and the dimer ion binding energy calculated from Reactions (1)-(4) would seem to suggest that the methyl groups in acetone playa minimal role in the bonding of cluster ions.
ACKNOWLEDGM ENTS
The authors wish to thank Dr. Richard Keller for suggesting the independent systems model for the cluster ion system and Dr. James Stine for his instructive calculations on the same. We also wish to express our appreciation to other members of Group CNC-2 at LASL for their helpful comments on various aspects of this work. Special thanks extends to Dr. Berlyn Brixner of Group M -1 at LASL for the design and fabrication of the grazing incidence alignment mirror. One of us (W.M.T.) acknowledges funding from the Associated Western Universities, Inc. This research was done with support from the U. S. Department of Energy.
Ip. W. Tiedemann, S. T. Ceyer, C. Y. Ng, D. J. Trevor, P. L. Kronebusch, B. H. Mahan, and Y. T. Lee, "Energetics of Molecule Ions by the Molecular Beam Photoionization Method," 175th ACS National Meeting, Anaheim, CA, March 13-17, 1978.
2G. G. Jones, Ph. D. thesis, The University of WisconSin, 1975.
3J. W. Taylor, G. R. Parr, and G. G. Jones "Photoionization Mass Spectrometry USing Molecular Beam Sampling and Synchrotron Radiation," Vac. Ultraviolet Radiat. Phys., Proc. Int. Conf., 4th 1974, pp. 197-198.
4G. G. Jones and J. W. Taylor, J. Chern. Phys. 68, 1768 (1978).
5C . Y. Ng, D. J. Trevor, P. W. Tiedemann, S. T. Ceyer, P. L. Kronebusch, B. H. Mahan, and Y. T. Lee, J. Chern. Phys. 67, 4235 (1977).
6C. Y. Ng, P. W. Tiedemann, B. H. Mahan, and Y. T. Lee, J. Chern. Phys. 66, 5737 (1977).
lC. Y. Ng, P. W. Tiedemann, B. H. Mahan, and Y. T. Lee, J. Chern. Phys. 66, 3985 (1977).
BC. Y. Ng, D. J. Trevor, B. H. Mahan, and Y. T. Lee, J. Chern. Phys. 66, 446 (1977).
9C. Y. Ng, D. J. Trevor, B. H. Mahan, and Y. T. Lee, J. Chern. Phys. 65, 4327 (1976).
10C. Y. Ng, B. H. Mahan, and Y. T. Lee, J. Chern. Phys. 65, 1956 (1976).
tiC. Y. Ng, Ph. D. thesis, The University of California, Berkeley, CA, 1976.
12G. R. Parr and J. W. Taylor, Int. J. Mass Spectrom. Ion Phys. 14, 467 (1974).
13G. R. Parr and J. W. Taylor, Rev. Sci. Instrum. 44, 1578 (1973).
14G. R. Parr, Ph. D. thesis, The University of WisconSin, 1973.
15G. Mouvier and R. Hernandez, Org. Mass Spectrom. 10, 958 (1975).
16R. A. W. Johnstone, F. A. Mellon, and S. D. Ward, Int. J. Mass Spectrom. Ion Phys. 5, 241 (1970).
11M. A. Haney and J. L. Franklin, Trans. Faraday Soc. 65, 1794 (1969).
lBK • Hirota and M. Hatada, Kinet. 8, 748 (1967). ISV. K. Potapov and D. N. Shlgorin, Zh. Fiz. Khlm. 40, 200
(1966). 2oD. N. Shlgorin, A. D. Filyugina, and V. K. Potapov, Teor.
Eksperim. Khim. 2, 554 (1966). 21F. H. Dorman, J. Chern. Phys. 42, 65 (1965). 22A. Foffani, S. Pignataro, B. Cantone, and F. Grasso, Z.
Phys. Chern. (Frankfurt) 42, 221 (1964). 23A. Foffani, S. Pignataro, B. Cantone, and F. Grasso,
Nuovo Cimento 29, 918 (1963). 241. Kanomata, Bull. Chern. Soc. Jpn. 34, 1864 (1961). 25J. R. Majer, C. R. Patrick, and J. C. Robb, Trans. Fara-
J. Chem. Phys., Vol. 69, No.7, 1 October 1978
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.160.4.77 On: Sat, 20 Dec 2014 05:07:53
3158 Trott, Blais, and Walters: Photoionization of acetone
day Soc. 57, 14 (1961). 26A. B. King and F. A. Long, J. Chern. Phys. 29, 374 (1958). 21R. I. Reed and J. C. D. Brand, Trans. Faraday Soc. 54,
478 (1958). 28K. Higasi, I. Omura, and H. Baba, Nature 178, 652 (1956). 211. Omura, K. Higasi, and H. Baba, Bull. Chern. Soc. Jpn.
29, 504 (1956). 30J. D. Morrison and A. J. C. Nicholson, J. Chern. Phys.
20, 1021 (1952). 311. Omura, T. Kaneko, Y. Yamada, and T. Kondo, Mass
Spectrosc. 16, 349 (1968). 32V. K. Potapov, A. D. Filyugina, D. N. Shigorin, and G. A.
Ozerova, Dokl. Akad. Nauk SSSR 180, 398 (1968). 33E. Murad and M. G. lnghram, J. Chern. Phys. 40, 3263
(1964). 34E. Murad and M. G. lnghram, J. Chern. Phys. 41, 404
(1964). 35K. Watanabe, T. Nakayama, and J. Mottl, J. Quant. Spec
trosc. Radiat. Transfer 2, 369 (1962). 36H. Hurzeler, M. G. lnghram, and J. D. Morrison, J. Chern.
Phys. 28, 76 (1958). 31F. I. Vilesov and A. N. Terenin, Dokl. Akad. Nauk SSSR
115, 744 (1957). 38K. Watanabe, J. Chern. Phys. 26, 542 (1957). 39K. Watanabe, J. Chern. Phys. 22, 1564 (1954). 4oR. Hernandez, P. Masclet, and G. Mouvier, J. Electron
Spect. Related Phenom. 10, 333 (1977). 41V. K. Potapov, G. V. Karachevtsev, and M. M. Lipei,
Khim. Vys. Energ. 11, 107 (1977). 42V. Y. Young and K. L. Cheng, J. Chern. Phys. 65, 3187
(1976). 43K. Kimura, S. Katsumata, T. Yamazaki, and H. Wakaba
yashi, J. Electron Spect. Related Phenom. 6, 41 (1975). 44G. Hentrich, E. Gunkel, and M. Klessinger, J. Mol. Struct.
21, 231 (1974). 45J. Kelder, H. Cerfontain, B. R. Higginson, and D. R. Lloyd,
Tetrahedron Lett. 1974, 739. 46 H. Ogata, J. Kitayama, M. Koto, S. KOjima, Y. Nihei, and
H. Kamada, Bull. Chern. Soc. Jpn. 47, 958 (1974). 41W. -C. Tam, D. Yee and C. E. Brion, J. Electron Spect.
Related Phenom. 4, 77 (1974). 4BC. R. Brundle, M. B. Robin, N. A. Kuebler, and H. Basch,
J. Am. Chern. Soc. 94, 1451 (1972). 49G. V. Karachevtsev, V. K. Potapov, and V. V. Sorokin,
Dokl. Akad. Nauk SSSR 206, 144 (1972). 50A. W. Potts, T. A. Williams, and W. C. Price, Faraday
Discuss. Chern. Soc. 104-15 (1972) 54. 51 B . J. Cocksey, J. H. D. Eland, and C. J. Danby, J. Chern.
Soc. B 1971, 790. 52M. J. S. Dewar and S. D. Worley, J. Chern. Phys. 50, 654
(1969). 53M. 1. Al-Joboury and D. W. Turner, J. Chern. Soc. 1964,
4434. 54R. Stockbauer, Int. J. Mass Spectrom. Ion Phys. 25, 89
(1977). SSC. s. T. Cant, C. J. Danby and J. H. D. Eland, J. Chern.
Soc. Faraday Trans. II 71, 1015 (1975). 56A . B. F. Duncan, J. Chern. Phys. 3, 131 (1935). s1D. s. C. Yee and C. E. Brion, J. Electron Spect. Related
Phenom. 8, 377 (1976). S8V. Cermak, Collection Czech. Chern. Commun. 33, 2739
(1968). 59M . A. Slifkin and A. C. Allison, Nature 215, 949 (1967). 60R . H. Huebner, R. J. Celotta, S. R. Mielczarek, and C. E.
Kuyatt, J. Chern. Phys. 59, 5434 (1973). 61M . w. P. Cann, Appl. Opt. 8, 1645 (1969). 62 J. A. R. Samson, J. Opt. Soc. Am. 54, 6 (1964). 63C ._S . Lu and H. E. Carr, Rev. Sci. lnstrum. 33, 823 (1962). 64C. F. Giese, Rev. Sci. lnstrum. 30, 260 (1959). 6SG . Herzberg and L. L. Howe, Can. J. Phys. 37, 636 (1959). 66J. A. R. Samson, Techniques of Vacuum Ultraviolet Spec-
troscopy (Wiley, New York, 1967), p. 139. 61J. H. Beynon, A. E. Fontaine, and B. E. Job, Z. Natur
forsch. Teil A 21, 776 (1966). 68Z. Luczyllski and H. Wincel, Int. J. Mass Spectrom. Ion
Phys. 23, 37 (1977). 69A. S. BlairandA. G. Harrison, Can. J. Chern. 51, 703
(1973). 10L . W. Sieck and P. Ausloos, Radiat. Res. 52, 47 (1972). 11K. A. G. MacNeil and J. H. Futrell, J. Phys. Chern. 76,
409 (1972). 12 J. O. Terry and T. O. Tiernan, Proc. 16th Annual Confer
ence on Mass Spectrometry and Allied Topics, ASTM Committee E14, Pittsburgh, PA, 1968, p. 33.
73V . K. Potapov, Dokl. Akad. Nauk SSSR 183,386 (1968). 14M. s. B. Munson, J. Am. Chern. Soc. 87, 5313 (1965). 1SR . S. Mulliken, J. Chern. Phys. 3, 564 (1935). 16J . Chao and B. J. Zwolinski, J. Phys. Chern. Ref. Data 5,
319 (1976). 17G . Dellepiane and J. Overend, Spectrochim. Acta 22, 593
(1966). 78J. W. Rabalais, Principles of Ultraviolet Photoelectron
Spectroscopy (Wiley, New York, 1977), Chap. 3. 19G . Herzberg, Molecular SPectra and Molecular Structure
III. Electronic Spectra and Electronic Structure of Polyatomic Molecules (Van Nostrand, Princeton, NJ, 1966).
80M . L. Vestal, in Fundamental Processes in Radiation Chemistry, edited by P. Ausloos, (lnterscience, New York, 1968, pp. 59-118.
Blp. M. Guyon, W. A. Chupka, and J. Berkowitz, J. Chern. Phys. 64, 1419 (1976).
82R . S. Mulliken, J. Chern. Phys. 52, 5170 (1970). 83C . Braun and H. Leidecker, J. Chern. Phys. 61, 3104 (1974). 84J . O. Hirschfelder, C. F. Curtiss, and R. B. Bird, Molec-
ular Theory of Gases and Liquids (Wiley, New York, 1954). 85R . F. Hajjar, W. B. Kay, andG. F. Leverett, J. Chern.
Eng. Data 14, 377 (1969). 86J . H. Dymond and E. B. Smith, The Virial Coefficients of
Gases (Clarendon, Oxford, 1969), p. 76. 81J. D. Lambert, G. A. H. Roberts, J. S. Rowlinson, and V.
J. Wilkinson, Proc. R. Soc. London Ser. A 196, 112 (1949). 88J . S. Rowlinson, Trans. Faraday Soc. 45, 974 (1949). 89L . N. Anderson, A. P. Kudchadker, and P. T. Eubank, J.
Chern. Eng. Data 13, 321 (1968). 90 p . J. Foster, R. E. Leckenby, and E. J. Robbins, J. Phys.
B 2, 478 (1969). 91 E . J. Robbins, R. E. Leckenby, and P. Willis, Adv. Phys.
16, 739 (1967). 92A . Herrmann, E. Schumacher, and L. Waste, J. Chern.
Phys. 68, 2327 (1978). 93 p . J. Harbour, J. Phys. B 4, 528 (1971). 94W. T. Simpson, Theories of Electrons in Molecules (Pren
tice-Hall, Englewood Cliffs, NJ, 1962), Chap. 4. 9SW . E. Spicer, in Optical Properties of Solids: NewDevel
opments, edited by B. O. Seraphin (American Elsevier, New York, 1976), pp. 667-671.
96K . Y. Yu, J. C. McMenamin, and W. E. Spicer, Surf. Sci. 50, 149 (1975).
J. Chern. Phys., Vol. 69, No.7, 1 October 1978
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
130.160.4.77 On: Sat, 20 Dec 2014 05:07:53
![%D7%AA%D6%B7%D6%BC%D7%A8%D6%B0%D7%A9 · D7%AA%D6%B7%D6%BC%D7%A8%D6%B0%D7%A9 ... • Jonah 1:3 (Jonah 1:3 [5]), ... Aššur Babylon E](https://img.pdfslide.us/doc/110x75/5b0448167f8b9a2d518d665e/d7aad6b7d6bcd7a8d6b0d7a9-d7aad6b7d6bcd7a8d6b0d7a9-.jpg)
