Public Notices

Absolute absorption cross-section measurements of CO2 in the ultraviolet from 200 to 206 nm at 295 and 373 K

Description
Laboratory measurements of the absolute absorption cross-section of CO2 at the temperatures 295 and 373 K have been made between 200 and 206 nm using cavity ring-down spectroscopy. Below 205 nm, the cross-section at 373 K is significantly larger than
Categories
Published
of 5
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Absolute absorption cross-section measurements of CO 2  inthe ultraviolet from 200 to 206 nm at 295 and 373 K A. Karaiskou  a,c , C. Vallance  d , V. Papadakis  a , I.M. Vardavas  a,b , T.P. Rakitzis  a,b,* a Institute of Electronic Structure and Laser, Foundation for Research and Technology-Hellas (IESL-FORTH), 711 10 Heraklion-Crete, Greece b Department of Physics, University of Crete, P.O. Box 2208, 71003 Heraklion, Greece c Department of Chemistry, University of Crete, Leof. Knossou, 71409 Heraklion, Greece d Physical and Theoretical Chemistry Laboratory, Department of Chemistry, University of Oxford, South Parks Road, Oxford OX1 3QZ, UK  Received 22 September 2004; in final form 20 October 2004Available online 6 November 2004 Abstract Laboratory measurements of the absolute absorption cross-section of CO 2  at the temperatures 295 and 373 K have been madebetween 200 and 206 nm using cavity ring-down spectroscopy. Below 205 nm, the cross-section at 373 K is significantly larger thanat 295 K, whereas beyond 205 nm measurements at both temperatures yield cross-sections approximately equal to the Rayleigh scat-tering cross-section, within experimental error. The present measurements should resolve a long-standing discrepancy between pre-viously published data on this system.   2004 Elsevier B.V. All rights reserved. 1. Indroduction Carbon dioxide is an important trace constituent of the present day atmosphere of the earth due to its actionas a greenhouse gas. In the Early Precambrian atmos-phere, prior to the development of life on earth, CO 2 was, most likely, present in much higher concentrations,and its greenhouse action may have played a vital role inpreventing the earth from freezing in the presence of amuch cooler young main sequence sun [1,2]. CO 2  is alsohighly relevant to other planetary atmospheres, in par-ticular those of Venus and Mars, which have CO 2 atmospheres at pressures of 96 bar and 6 mbar,respectively.Photolysis of CO 2  is possible at wavelengths shorterthan the 227.5 nm dissociation limit. This is a minorprocess in the earth  s present atmosphere, due to effi-cient shielding by O 2  absorption below 200 nm, in theregion of the Schumann–Runge bands (175–200 nm)and continuum (137–175 nm), and above 200 nm inthe region of the Herzberg dissociation continuum(185–242 nm). However, in a CO 2 -rich planetary atmos-phere containing little O 2 , the situation is reversed, withpreferential photodissociation of CO 2  above 200 nmshielding O 2  from photodissociation via absorption inthe Herzberg continuum. In such atmospheres, even arelatively small absorption cross-section for CO 2  above200 nm (close to the Rayleigh scattering limit) may re-sult in significant shielding of O 2  at these wavelengths.Several measurements of the CO 2  absorption cross-section,  r (CO 2 ), have been reported below   200 nm. r (CO 2 ) is large in this region, and the agreement be-tween the various measurements has, in general, beenvery good [3–10]. Above 200 nm, where  r (CO 2 ) dimin-ishes to the limit of the Rayleigh-scattering cross-section,  r R , there have been fewer measurements, andagreement between them has been poor. Of particularnote, Ogawa [6] reported a significant tail in theabsorption between 200 and 216 nm, within whichthe measured values were around three times largerthan  r R , and made no measurements beyond 216 nm 0009-2614/$ - see front matter    2004 Elsevier B.V. All rights reserved.doi:10.1016/j.cplett.2004.10.073 * Corresponding author. Fax: +30 2810 391305. E-mail address:  ptr@iesl.forth.gr (T.P. Rakitzis).www.elsevier.com/locate/cplettChemical Physics Letters 400 (2004) 30–34  to investigate where  r R  was reached. The raw data be-yond 200 nm from a study by Shemansky [7] alsoshowed an apparent absorption tail past 250 nm, wellbeyond the dissociation threshold of    227.5 nm. How-ever, various experimental checks, including a calibra-tion using N 2  at 253.6 nm, showed the tail to bespurious, leading Shemansky to conclude that theextinction coefficient beyond 204 nm is due almost en-tirely to Rayleigh scattering. Shemansky also suggestedthat his, and possibly Ogawa  s, observation of anabsorption tail beyond 204 nm was due to scatteredshort-wavelength radiation, an artifact of the experi-mental technique used. DeMore and Potapoff  [8] con-firmed Shemansky  s conclusions by measuring  r (CO 2 )between 200 and 204 nm, but carried out experimentsonly at very high pressures, from 21.4 to 48.7 atm.They showed that absorption above 203 nm could beaccounted for entirely by Rayleigh scattering. The tem-perature dependence of   r (CO 2 ) near 300 K has beendetermined only for wavelengths shorter than 200nm, though longer wavelength (>200 nm) data areavailable for higher temperatures of around 1500 K[11].Cavity ring-down spectroscopy (CRDS) [12–16] is themost sensitive technique available for absorption meas-urements from the near UV to the infrared, a spectral re-gion for which high reflectivity ( R  > 0.999) mirrors areavailable. Though the production of mirrors of thisquality for wavelengths much below 200 nm exceedsthe capabilities of current technology, mirrors with R  0.99 are available at 200 nm, making CRDS consid-erably more sensitive than single-pass absorption tech-niques in this wavelength region (see, for example, theapplication of deep UV CRDS by Sneep et al. [17]).One of the great advantages of CRDS is that by meas-uring the exponential decay constant of the light withinthe cavity (which is independent of the energy of theexciting radiation), rather than the total light exitingthe cavity, shot-to-shot fluctuations in laser pulse inten-sity do not appear as noise in the final spectrum. This,together with the long path lengths achievable, providesthe enhanced sensitivity of CRDS over standard absorp-tion spectroscopy.We report absolute measurements of   r (CO 2 ) between200 and 206 nm at temperatures of 295 and 373 K. 2. Experimental procedure Fig. 1 depicts a schematic overview of the experimen-tal set up. The optical cavity was 1.35 m long, consistingof two highly reflecting concave dielectric mirrors (CVI),with  R  0.985 from 200 to 210 nm. The mirrors had a 4m radius of curvature and a diameter of 13.5 mm, andwere mounted inside a vacuum chamber pumped by arotary pump and equipped with a capacitance manome-ter (either an MKS baratron or a psi-tronix digitalmanometer) for absolute pressure measurements. Theresidual pressure in the evacuated cavity was 0.2 mbar.The 200–206 nm light was generated by frequency tri-pling the output of a 10 Hz Nd:YAG-pumped MOPOsystem (Spectra Physics 730DT10) using consecutiveKDP and BBO crystals. The resulting light pulses were  5 ns in duration, with a bandwidth of    0.2 cm  1 and an energy of    5  l J. During a measurement, thegas of interest was admitted to the cavity to the desiredpressure, a small amount of the laser beam was injectedinto the cavity through one of the end mirrors and thelight exiting the opposing mirror was detected using aphotomultiplier tube (Hamamatsu) fitted with a 10 cmbiconcave lens. Ring-down traces were acquired on adigital oscilloscope (LeCroy) and transferred to a PCvia a GPIB/LabVIEW interface. The waveforms wereaveraged over 100 laser shots and fitted to an exponen-tial function to determine the ring-down time constant.Data sets yielding anomalously high  v 2 values (e.g. frommechanical vibrations or laser jitter) were rejected, andthe measurements automatically repeated. The meas-ured ring-down times were converted into cross-sectionsas described in Section 3. Sample ring-down signals forthe evacuated and CO 2 -filled cavity are shown in Fig. 2.The small magnitude of   r (CO 2 ) (  10  24 cm 2 ) de-manded that great care be taken to avoid false contribu-tions to the signal from other trace species present in thevacuum system. Signal contamination from residualwater vapour in the chamber proved to be particularlyproblematic. Admission of CO 2  to the cavity appearedto redistribute adsorbed H 2 O from the chamber wallsto the mirrors, reducing their effective reflectivity andshortening the ring-down time accordingly. The effect,illustrated in Fig. 3, was dependent on CO 2  pressure,and was initially mistaken for enhanced absorption by Fig. 1. Schematic set up of the CRDS experiment. A. Karaiskou et al. / Chemical Physics Letters 400 (2004) 30–34  31  CO 2 . Fig. 3a shows three sets of measured cross-sectionsat 224 nm as a function of pressure under two differentsets of experimental conditions. For the first two meas-urements, no desiccant traps were used, the cell was notbaked prior to beginning measurements, and the mirrorswere maintained at ambient temperature. The solidsquares are points recorded as the pressure was in-creased from 0 up to 2000 mbar, and show an anoma-lous increase in  r (CO 2 ) with pressure, from near  r R (  5  ·  10  25 cm 2 ) at 200 mbar, up to a value of 2  ·  10  24 cm 2 at higher pressures. The open squaresare points recorded as the pressure was reduced backdown from 2000 mbar to zero and show the measuredcross-section remaining at the higher value, with no re-turn to the lower values previously measured at lowerpressures. If the two lowest pressure measurements areomitted, the cross-section appears to be constant at avalue of    2  ·  10  24 cm 2 , significantly higher than thetrue value. Trace contamination therefore has the poten-tial to produce a constant but anomalously high cross-section over a limited pressure range. In the third setof measurements shown in Fig. 3a (solid triangles), thecell was baked to remove any residual H 2 O, and themeasurements returned to  r R  over the entire range of pressures sampled.To minimize the effects of trace contaminants on r (CO 2 ) measurements, the cell was baked at   100   Cfor about 12 h before each measurement, the mirrorswere heated independently to   80   C during measure-ments to minimize water adsorption onto their surfaces,and desiccant traps were used at the inlets and outlets of the cell. We often observed an increase in the emptyring-down time after warming the mirrors, which weinterpreted as evidence for desorption of impurities(such as water or pump oil) from the mirror surfaces.Fig. 3b shows  r (CO 2 ) measured as a function of pres-sure (both increasing and decreasing) at 203 nm withthese precautionary measures in place. As required, r (CO 2 ) is constant for all pressures sampled. We empha-size the importance of removing all traces of water fromthe cavity to obtain reliable  r (CO 2 ) measurements.Several further steps were taken to maximize the reli-ability of the measurements. The ring-down time of theevacuated cavity was measured both before and aftereach photoabsorption cross-section measurement.Tracking the baseline ring-down time in this way pre-cluded contributions to the signal from unnoticed tracecontaminants. Perhaps the most convincing test of thereliability of our measurements is that for wavelengthsbeyond 205 nm (e.g. 210 and 220 nm) we were able tomeasure correctly  r R  for both CO 2  and N 2  (measuredat the beginning and end of each CO 2  run).Both CO 2  and N 2  cross-section measurements wereconducted at 295 K (room temperature) and 373 K.For the higher temperature measurements, the chamberwas wrapped in resistively heated tape and heated until 0250500750100012501500175020000.00.51.01.52.02.5      σ    (   t  o   t  a   l   )   (   1   0   -    2   4    c  m    2    ) Pressure (mbar) 02004006008001000120014001600012345 Pressure (mbar)      σ    (   t  o   t  a   l   )   (   1   0   -    2   4    c  m    2    )   σ (Increasing Pressure)  σ (Decreasing Pressure) (a)(b) Fig. 3. (a)  r (CO 2 ) under different conditions: with water contamina-tion (1) increasing in CO 2  pressure (solid squares) and (2) decreasing inCO 2  pressure (open squares); (3) without water contamination (solidtriangles), compared to  r R  (solid line). (b)  r (CO 2 ) at 203 nm for 373 Kfor both increasing and decreasing pressure, without watercontamination. 0.20.40.60.81.01.21.405101520    I  n   t  e  n  s   i   t  y   (  a  r   b   i   t  r  a  r  y  u  n   i   t  s   ) Time ( µ s )  Vacuum decay rate CO 2  decay rate Fig. 2. Typical ring-down traces obtained for an evacuated cavity( s 0  = 254 ns), and for a cavity filled with 1.5 bar CO 2  at 373 K and206 nm ( s  = 212 ns).32  A. Karaiskou et al. / Chemical Physics Letters 400 (2004) 30–34  thermal equilibrium was reached. Measurements werecarried out at six different gas pressures ranging from350 to 1500 mbar. 3. Results and discussion The decay in intensity of laser light trapped within anoptical cavity follows a simple exponential decay:  I  ð t  Þ¼  I  0 exp ½ t  = s  ð 1 Þ for which the cavity ring-down time  s  is given by: s ¼  d c ½ð 1   R Þþ kl  ð 2 Þ where  d   is the cavity length,  c  is the speed of light,  R  isthe mirror reflectivity and  k   is the absorption coefficientof the absorber present over a pathlength  l   inside thecavity (in our case  l   =  d  ). The ring-down time for anempty cavity,  s 0 , is given by: s 0  ¼  d c ð 1   R Þ :  ð 3 Þ At 205 nm, the 1/ e  decay time of the empty cavity inour experimental configuration is 285 ns, correspondingto an effective mirror reflectivity of 98.5%. A time con-stant of this magnitude corresponds to a path lengthof 85 m, or 35 round trips of the cavity. With a laserbandwidth of 0.2 cm  1 ,   55 longitudinal modes arespanned in the 1.35 m cavity at each excitationwavelength.The absorption coefficient  j  is obtained from the val-ues of   s  and  s 0 : j ¼ 1 c 1 s   1 s 0    ð 4 Þ and is related to the absorption cross-section  r  (in unitsof cm 2 ) by  j  =  N  r , where  N   is the number of moleculesper cm 3 . In general,  r  contains contributions from bothabsorption and  r R . The pressure and temperaturedependence of   N   may be approximated using the idealgas law  N   ¼  N  0 ð T  0 = T  Þð  p  =  p  0 Þ ;  ð 5 Þ where  p  and  T   are the pressure and temperature of theabsorbing gas and the subscript refers to standardconditions.The minimum detectable absorption in a CRD exper-iment may be expressed in terms of   D s 0 , the standarddeviation of   s 0 , by: D j ¼ 1 c 1 s 0   1 s 0 þ D s 0   :  ð 6 Þ A ring-down time of 285 ± 3 ns at 205 nm yields aminimum detectable absorption of 1.3  ·  10  8 cm  1 , cor-responding to a minimum detectable cross-section of 0.5  ·  10  25 cm 2 at a pressure of 1 atm. This value is closeto an order of magnitude smaller than  r R  of CO 2 , indi-cating that the CRDS measurements should be at leastas sensitive as those described in previous studies. Thetrue experimental error is likely to be somewhat largerthan the limit set by the minimum detectable cross-section calculated above, primarily due to minor varia-tions in the alignment of the cavity.Fig. 4 shows the results of   r (CO 2 ) measurements atwavelengths in the range from 200 to 206 nm at 295 K(solid squares) and 373 K (open squares). The data arealso presented in Table 1. The error bars indicate 2 r from the mean of the measured ring-down times. Alsoshown is  r R  of CO 2 , calculated using the relation [18]: r R  ¼  6 þ 3 D 6  7 D    A  1 þ  B k 2    2 4 : 577  10  21 cm 2 k 4   ; ð 7 Þ where  k  is the wavelength in  l m,  A  and  B   are constants( A  = 4.39  ·  10  4 ,  B   = 6.4  ·  10  3 ) and  D  = 0.0805 is thedepolarization factor.The cross-sections from this work and from [6,7,9,10]are shown in Fig. 5. Ogawa  s measurements [6] show an 2002012022032042052060.00.51.01.52.02.53.03.54.0      σ    (   t  o   t  a   l   )   (   1   0  -    2   4   c  m    2    ) Wavelength (nm) σ  at 295K σ  at 373K σ R Fig. 4.  r (CO 2 ) between 200 and 206 nm at 295 K (solid squares) and at373 K (open squares), together with  r R  (solid line). Measurements(solid diamonds), and calculated (dashed line)  r R  (N 2 ) are also shown.Table 1 r (CO 2 ) at 295 and 373 K (2 r  error bars), compared to the Rayleighscattering cross-section,  r R k  (nm)  r 295  ( · 10  24 cm 2 )  r 373  ( · 10  24 cm 2 )  r R  ( · 10  24 cm 2 )200 1.51 ± 0.16 3.38 ± 0.20 0.85201 1.47 ± 0.05 3.05 ± 0.12 0.83202 1.34 ± 0.07 2.10 ± 0.05 0.81203 1.05 ± 0.06 2.07 ± 0.12 0.80204 0.73 ± 0.41 1.62 ± 0.22 0.78205 0.83 ± 0.22 0.95 ± 0.63 0.76206 0.58 ± 0.16 0.92 ± 0.40 0.74210 0.66 ± 0.10 – 0.68220 0.52 ± 0.12 – 0.55 A. Karaiskou et al. / Chemical Physics Letters 400 (2004) 30–34  33  anomalously high cross-section, which, based on ourexperiences, we believe may have arisen from residualH 2 O present in the experimental apparatus. The meas-ured cross-sections are seen to be strongly dependenton temperature in the wavelength range from 200 to203 nm.  r (CO 2 ) increases markedly with vibrationalexcitation, and the presence of a small fraction of mole-cules possessing one or two quanta in vibration is en-ough to account for the observed temperaturedependence [11]. At  k  = 204 nm, there is still some con-tribution to the absorption from (weakly) vibrationallyexcited CO 2 . At 205 and 206 nm, the  r R  limit has beenreached for both ground and vibrationally excited CO 2 ,and the temperature dependence has virtually disap-peared. The 373 K cross-sections agree qualitativelywith an extrapolation to 200–206 nm of Lewis andCarver  s measurements at wavelengths shorter than197 nm [9]. 4. Conclusions We have presented photoabsorption cross-sectionmeasurements for CO 2  over the range from 200 to 206nm. The results appear to support, qualitatively, the val-ues previously reported by Shemansky [7] and Demoreand Potapoff  [8], rather than those published by Ogawa [6]. Resolution of the discrepancy between these conflict-ing data has important consequences for photochemicalmodels of planetary atmospheres, since reliable CO 2 absorption cross-sections are required both for accuratecalculation of CO 2  photodissociation rates, and for aquantitative treatment of the effects of CO 2  absorptionon the photodissociation rates of other atmosphericspecies such as O 2  and O 3 . Acknowledgements We gratefully acknowledge support from theHGSRT program PENED 2001 (No. ED 01479) andHPRI-CT-1999-00074. References [1] J.H. Carver, I.M. Vardavas, Annales Geophysicae 12 (1994) 674.[2] J.H. Carver, I.M. Vardavas, Annales Geophysicae 13 (1995) 782.[3] P.G. Wilkinson, H.L. Johnston, J. Chem. Phys. 18 (1950) 1440.[4] E.C.Y. Inn, K. Watanabe, M. Zalikoff, J. Chem. Phys. 21 (1953)1648.[5] P.S. Nakata, K. Watanabe, F.M. Matsunaga, Sci. Light 14 (1965)54.[6] M. Ogawa, J. Chem. Phys. 54 (1971) 2550.[7] D.E. Shemansky, J. Chem. Phys. 56 (1972) 1582.[8] W.B. DeMore, M. Potapoff, J. Geophys. Res. 77 (1972) 6291.[9] B.R. Lewis, J.H. Carver, J. Quant. Spectrosc. Radiat. Transfer 30(1983) 297.[10] W.H. Parkinson, J. Rufus, K. Yoshino, Chem. Phys. 290 (2003)251.[11] R.J. Jensen, R.D. Guettler, J.L. Lyman, Chem. Phys. Lett. 277(1997) 356.[12] A. O  Keefe, D.A.G. Deacon, Rev. Sci. Instrum. 59 (1988) 2544.[13] P. Zalicki, R.N. Zare, J. Chem. Phys. 102 (1995) 2708.[14] J.J. Scherer, J.B. Paul, A. O  Keefe, R.J. Saykally, Chem. Rev. 97(1997) 1.[15] M.D. Wheeler, S.T. Newman, J. Orr-Ewing, M.N.R. Ashfold, J.Chem. Soc., Faraday Trans. 94 (1998) 337.[16] G. Berden, R. Peeters, G. Meijer, Int. Rev. Phys. Chem. 19 (2000)565.[17] M. Sneep, S. Hannemann, E.J. van Duijn, W. Ubachs, Opt. Lett.29 (2004) 1378.[18] I.M. Vardavas, J.H. Carver, Planet. Space Sci. 32 (1984) 1307. 160170180190200210220 10 -24 10 -23 10 -22 10 -21 10 -20 10 -19 σ R Ogawa Shemansky Lewis and Carver Parkinson et al. This work      σ     (  c  m    2    ) Wavelength (nm) Fig. 5. Comparison of the room-temperature  r (CO 2 ) from this workwith values taken from [6,7,9,10].34  A. Karaiskou et al. / Chemical Physics Letters 400 (2004) 30–34
Search
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks