Absolute integrated intensities of vapor-phase hydrogen peroxide (H2O2) in the mid-infrared at atmospheric pressure

We report quantitative infrared spectra of vapor-phase hydrogen peroxide (H(2)O(2)) with all spectra pressure-broadened to atmospheric pressure. The data were generated by injecting a concentrated solution (83%) of H(2)O(2) into a gently heated
of 10
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
  ORIGINAL PAPER  Absolute integrated intensities of vapor-phase hydrogenperoxide (H 2 O 2 ) in the mid-infrared at atmospheric pressure Timothy J. Johnson  &  Robert L. Sams  & Sarah D. Burton  &  Thomas A. Blake Received: 3 March 2009 /Revised: 9 April 2009 /Accepted: 15 April 2009 /Published online: 12 May 2009 # Springer-Verlag 2009 Abstract  We report quantitative infrared spectra of vapor- phase hydrogen peroxide (H 2 O 2 ) with all spectra pressure- broadened to atmospheric pressure. The data were generated by injecting a concentrated solution (83%) of H 2 O 2  into agently heated disseminator and diluting it with pure N 2 carrier gas. The water vapor lines were quantitativelysubtracted from the resulting spectra to yield the spectrumof pure H 2 O 2 . The results for the  ν 6  band strength (includinghot bands) compare favorably with the results of Klee et al.(  J Mol. Spectrosc.  195:154, 1999) as well as with theHITRAN values. The present results are 433 and 467 cm -2 atm − 1 (±8 and ±3% as measured at 298 and 323 K,respectively, and reduced to 296 K) for the band strength,matching well the value reported by Klee et al. ( S  =467 cm − 2 atm − 1 at 296 K) for the integrated band. The  ν 1 +  ν 5  near-infrared band between 6,900 and 7,200 cm − 1 has anintegrated intensity  S  =26.3 cm − 2 atm − 1 , larger than previous-ly reported values. Other infrared and near-infrared bandsand their potential for atmospheric monitoring are discussed. Keywords  Infrared.Fouriertransform infrared.Quantitative.Bandstrengths.Hydrogenperoxide Introduction The Pacific Northwest National Laboratory (PNNL) contin-ues to build a database of quantitative vapor-phase infrared(IR) data for ambient monitoring: The data are from gases pressure-broadened to 760 Torr, recorded at moderately highresolution(0.1cm − 1 ), and with a broad spectral bandwidth soas to be optimized for tropospheric monitoring, either  passive or active [1  –  3]. The signal-to-noise ratio of thespectrometer system is optimized for the long-wave IR, 800  –  1,300 cm − 1 , but each spectrum has a spectral range of 600 cm − 1 or below to 6,500 cm − 1 or above. The data arerecorded as a series of single measurements, with each burden having the analyte pressurized with pure N 2  to760 Torr, and then averaged to form a composite [1]. Thespectra are recorded at 0.1 cm − 1 resolution so as to resolveall possible vibrational band spectral features, e.g., the Q branches of polyatomic molecules that may have Lorentzianhalfwidths of 1 cm − 1 or less.During the construction of the IR database, it wasrecognized that H 2 O 2  has recently garnered great interest amongst the spectroscopic community for two reasons. First,it is well known that H 2 O 2  and other organic peroxides are powerful oxidizers. They have long been used, for example,as dilute solutions in medicine and dentistry, e.g., as oralantiseptics. But as a strong oxidant, H 2 O 2  can sustainradical reactions, and concentrated peroxide solutionstherefore have the potential to form powerful explosives[4]. Second, it has more recently been recognized that H 2 O 2  plays a very significant role in global atmospheric chem-istry, as a stratospheric reservoir species for HO  x  [5, 6], but  also more recently elevated concentrations have beenassociated with biomass burning [7, 8]. In the stratosphere, H 2 O 2  is well established as a reservoir molecule for boththe hydroxyl and hydroperoxyl radicals (HO and HO 2 =HO  x ) as has been discussed by several authors [5, 6, 9]. This is seen in the following two reactions:H 2 O 2  þ  h  ν  !  OH  þ  OH  ð 1 Þ Anal Bioanal Chem (2009) 395:377  –  386DOI 10.1007/s00216-009-2805-xT. J. Johnson ( * ) : R. L. Sams : S. D. Burton :  T. A. BlakePacific Northwest National Laboratory,P.O. Box 999, Richland, WA 99352, USAe-mail:  OH  þ  H 2 O 2  !  H 2 O  þ  HO 2  ð 2 Þ where reaction 2 exceeds reaction 1 consuming OH. Interms of source production, H 2 O 2  comes about predomi-nantly in the stratosphere via the HO 2  self-reaction [10, 11]: HO 2  þ  HO 2  !  H 2 O 2  þ  O 2  ð 3 Þ Certainly in the aqueous phase, but also in the vapor  phase, peroxide can decompose to water and oxygen [4],catalyzed by any of several metals or higher temperatures:H 2 O 2  !  H 2 O  þ  1 = 2 O 2  ð 4 Þ In terms of biomass burning, Rinsland et al. [8] havevery recently detected elevated H 2 O 2  mixing ratios as highas 1.7 ppb v  (1 ppb v =1 ×10 9  per unit volume) wheremeteorological back trajectories associated the air masswith a young biomass burning plume. Previously Lee et al.[12] had shown that H 2 O 2  mixing ratios can be as high asapproximately 10 ppb v  in biomass burning plumes andcorrelate well with CO and other known biomass burningspecies. It is also known that H 2 O 2  is an important atmospheric oxidant (e.g., sulfur compounds), but isvulnerable to both dry and wet deposition owing its highwater solubility. von Kuhlmann et al. [9] have shown that vapor-phase H 2 O 2  can also be regenerated via various photochemical pathways, and has a tropospheric lifetime,  τ  ,on the order of days, but there is large variability owing todeposition. Typical background concentrations in nonurbanatmospheres have been measured by multiple techniquesand have been found to typically range from approximately0.2 to 2.0 ppb v  using techniques including diode laser absorption spectroscopy, enzymatic fluorescence, and lumi-nol chemiluminescence [7, 13, 14]. With the exception of  these few works, relatively few researchers have studiedH 2 O 2  in the  vapor   phase; quantitative reference data, in particular, are quite sparse. H 2 O 2  is in fact found in theHITRAN [15] vapor-phase IR spectral database, but thereare only data for the strong  ν 6  band from 1,170 to1,380 cm -1 . Realizing that broadband IR spectroscopy isone of the few techniques available for remote sensing in anopen-path sensor configuration [3], and also that many of the lead salt and quantum cascade IR laser methods [16, 17] offer good detectivity for in situ sensing such as for explosives detection or smog chamber experiments, thegoal of the present work is to provide quantitative bandstrengths for as many H 2 O 2  bands as possible, particularlywith data resolved to 0.1 cm − 1 for atmospheric sensing.We thus describe our methods used to generate the H 2 O 2 vapor-phase mixtures, and how the quantitative spectra arethen derived by subtracting water lines. The resulting bandstrength data are compared with band strengths availablein the literature, which are essentially only for the strong  ν 6  band. However, we also discuss other H 2 O 2  bandstrengths and line strengths in the IR and near-IR (NIR)and their potential use for atmospheric or industrialmonitoring. Experimental One of the prerequisites of data collection for the PNNLgas-phase IR database is that the chemicals be in arelatively pure state. This sometimes means purificationvia distillation, or more often  “ drying ”  chemicals such asethers or ketones by removing the H 2 O impurities viaadsorption onto CaSO 4 . For some species there are stilltrace amounts of other gases, most notably H 2 O vapor or CO 2  gas, that must be subtracted from each of theindividual pressure-pathlength burdens; typically 10 to 12spectra are measured and the exact partial pressures of theanalyte and the impurity are calculated for each burden.The composite spectrum has all impurities removed in thisfashion. While tedious, impurity spectral subtraction presents no major difficulties so long as (1) one is confident there are no IR-transparent impurities, e.g., H 2 , N 2 , Ar . . .and (2) that any impurities are at sufficiently lowconcentration that their absorption features are not opticallysaturated; that is, they stay within the linear domain of Beer  ’ s law behavior [1]. Otherwise, subtracting a givenimpurity ro-vibrational feature may result in removal of that optical feature, but may result in undersubtraction or oversubtraction of a different feature for the same species. H 2 O 2  preparation For H 2 O 2  clearly the  “ impurity ”  of concern is water. Our experience had shown that for IR spectral subtraction to beeffective, it was necessary to achieve H 2 O 2  concentrationsof more than 50%. As 50% solutions are the strongest concentrations easily available owing to shipping regula-tions, and because 90% solutions were formerly commer-cially available with still-manageable hazard levels, it wasdeemed desirable to start with a 50% solution and obtainconcentrations in the 70  –  90% range. That is, for this studyachieving concentrations in this range was sufficient enrichment to avoid water line subtraction problems, yet deemed reasonably safe. Obtaining still higher concentra-tions is technically difficult and increases the chances of accidental detonation. Moreover, at room temperature thesolutions are not stable, decomposing to H 2 O and O 2  unlessthey are refrigerated. Fractional crystallization, whereby thewater freezes at 0.00°C and the supernatant is rich in H 2 O 2 ,was one possibility for increasing the H 2 O 2  concentration, but is only successful at achieving concentrations up to 378 T.J. Johnson et al.  approximately 55%. Classical distillation under a vacuumwas deemed more appropriate: A commercial roto-vacuumdevice was used with an aspirator to achieve vacuum.The 50% solution of H 2 O 2  was obtained from Sigma-Aldrich, CAS no. 7722-84-1. To start, a Fourier transform(FT)-Raman spectral purity check of the stock liquid wascollected over the range from 50 to 3,600 cm − 1 Stokes shift and inspected for impurities using a previously describedinstrument [18]. The FT − Raman spectra have frequencyaccuracies shown to be better than 0.5 cm − 1 or less, andinspection of the data showed only H 2 O 2  and H 2 O bands.Prior to use, all components of the roto-vacuum devicewere soaked for hours (typically 24 h) in the 50% solutionto scavenge the glassware (and metal) parts for traceorganic residues. In a similar fashion, the syringes used toflow the solution into the disseminator T-piece were rinsedwith 50% solution and soaked for at least 2 h to scour for organics. The distillation was conducted by placingapproximately 20  –  25 ml of solution in the distillation flask and gradually warming the bath water over many minutesand eventually obtaining a temperature of 341 K. At thistemperature, under an aspirator vacuum, the neck of theflask was filled with condensate and small amounts werecollected in the condensation flask. The solution wasdistilled from about 50% by weight to between 82.8% for the 298 K spectrum and 82.5% for the 323 K spectrum. Thedensity of the solution was obtained by drawing 500µl intoa Hamilton 500-µl syringe and weighing the amount on a balance with a precision of 0.001 g. The concentration wascalculated from the measured density at room temperature(297 K) by using the formulae generated by Easton et al.[19]. A typical measured density was 1.352 g/ml, whichcorresponds to a solution that is 82.8% by weight. Theweight percent concentrations were converted to mole percent using the molecular masses of H 2 O 2 , 34.0147g/mol, and water, 18.0153 g/mol. The corresponding  mole  percent of H 2 O 2  for the 298 K solution was 71.8% and for the 323 K batch was 70.7%. Dissemination system and spectrometer For the present experiments, rather than use a static cellconfiguration, the IR spectra were recorded using a previously described disseminator flow system [20]. Thisis an IR long-path cell coupled to a liquid vaporizer whereby the liquid analyte is quantitatively delivered via asyringe pump into a stream of ultra-high-purity N 2  carrier gas, using a specially constructed heated vaporization piece[20]. The measurements were made in a customized Whitecell with the optical path set to 8.05 m (±0.5%). The cellhas a circulating liquid jacket, which can provide more precise temperature control for the gas. The gas temperatureis measured by placing a NIST-traceable temperature probewith an absolute accuracy of better than ±0.01 K directlyinto the gas, adjacent to the cell mirrors. One advantage of the flow system compared with a static cell for the H 2 O 2 measurements is that the time to vaporize the pure liquid isminimized. As soon as the analyte leaves the syringe tip, it is flash-vaporized by the disseminator block (held at 44 or 56°C for the two sets of measurements) and diluted to just afew parts per million in the ultra-high-purity N 2  carrier gas.Although there are additional wall contacts in the homog-enization process, these are minimized, and all flowcomponents were passivated with 50% solution prior tomeasurement, further reducing decomposition. For theconcentrated solutions, however, it was noted that withtime small bubbles developed inside the syringe owing tothe spontaneous reaction/decomposition leading to theformation of H 2 O liquid and O 2  gas (reaction 4). Althoughthe O 2  gas does enter the gas stream, its contribution to the ballast gas is negligible, and we still have a quantitativevalue for the number density of H 2 O 2 . It is assumed that thetotal number of H 2 O 2  and H 2 O molecules leaving thesyringe tip per unit time is the same as the total number flowing through the White cell per unit time, i.e., that thereis a stoichiometric conversion of H 2 O 2  to H 2 O according toreaction 4. [While H 2 O 2  can also be photolyzed in the(upper) atmosphere via reaction 1, the effect is negligible inthe dark gas cell.] The number density of H 2 O molecules iscalculated from the IR spectrum and is subtracted from thetotal to yield the number of H 2 O 2  absorbers.A Bruker IFS 66v/S vacuum spectrometer [1] was usedover the 520-7,500-cm − 1 range with an external mercurycadmium telluride detector for the White cell. Thespectrometer hardware characteristics have been previouslydocumented [1, 21  –  23]. These papers also contain details of measurement parameters, as well as modifications toredress sundry artifacts, including ghosting,  “ warm aper-ture ”  [23], and detector nonlinearity [24, 25] phenomena. Data reduction The PNNL method typically measures 10 to 12 separate burdens ranging over approximately 2 orders of magnitude,with  systematic  errors in absorbance of approximately 7%for well-behaved molecules. For problematic species suchas H 2 O 2 , the values can be higher. To account for any of several known nonlinearity phenomena in the fit at eachwavelength channel, the individual burdens are alsoweighted according to  T  2 (where the transmittance  T  =  I  /   I  o ).The multiple measurement with weighted data approach hasseveral advantages in that the signal-to-noise ratio isenhanced, especially where the high-burden measurements bring out a better signal-to-noise ratio for weak bands, and The absolute infrared cross sections of vapor-phase hydrogen peroxide at atmospheric pressure 379  also for the strong bands, where the weighting scheme brings out a better fidelity to account for Beer  ’ s lawsaturation or detector nonlinearity effects. The statisticalanalysis is also useful at discerning chemical impurities,since their IR signatures typically do not scale with the fit.The present data were reduced in the same manner [1]. For the individual burdens, the water mixing ratios werequantified using the PNNL IR database. These vapor- phase-measured values were used to correct for the loss/ conversion of H 2 O 2  during the elution process as describedabove. The molar concentration for 298 K was measured to be 62.05% and that for 323 K was measured to be 60.75%.The molar concentrations represent a  relative  loss of approximately 14% (absolute loss approximately 10%) of H 2 O 2  from the time of the srcinal density measurement tothe time the vapor-phase mixing ratio value was deter-mined. Using the adjusted molar concentrations, we scaledthe H 2 O 2  spectra by 1.612 for 298 K and 1.646 for 323 K,i.e., the water features were removed by spectral subtractionto derive the H 2 O 2  spectra presented in this paper. Results Figure 1 presents our broadband IR spectrum of H 2 O 2  from600 to 4,300 cm − 1 . The H 2 O 2  spectrum was first reportedmore than a half century ago by Giguère [26], and we usemany of his assignments. We note, however, that the present data are quantitative with decadic absorbancevalues and the ordinate corresponds to an optical depth of 1 ppm H 2 O 2  through a path of 1 m, with the number density adjusted to 296 K and 1 atm of N 2  total pressure.The reported spectra, both for 298 and for 323 K, represent the weighted average of ten individual H 2 O 2  measurementsfor each temperature. We have labeled the positions of thesix vibrational fundamentals bands [26], where the torsionalmode  ν 4  is beyond the range of the current spectrometer [27], but its first overtone 2  ν 4  is seen at the red end of thespectrum. (Note that for 2  ν 4  and 3  ν 4  the levels indicate the τ  =1 level only; see the paper by Camy-Peyret et al. [27].)Also labeled are the  ν 5 +  ν 4 ,  ν 1 +  ν 4,  ν 2 +  ν 6 , as well as the  ν 4 +  ν 6  combination bands, all of which are active in the  C  2 (actually  C  2h † double group) point group [28]. Redingtonet al. [29] showed that the ground-state structure has thehydrogens neither   cis  nor   trans , but rather with a dihedralangle of 119.8 o , with a  cis  barrier that is several timeshigher than the  trans  barrier. However, it was establishedthat to a good approximation one can consider H 2 O 2  as asymmetric rotor to separate the rotational levels into overalland internal rotational coordinates, and then consider theasymmetry perturbation for the overall levels. Essentiallyall of the bands show a systematic doubling [26], with thetorsional barrier so small that the O-H groups are nearlyfreely rotating about the O-O bond at room temperature.The  ν 4  torsional mode has large rotational spacings owingto the relatively small moment of inertia for the axis Fig. 1  Broadband infraredspectrum of H 2 O 2  vapor. The  y -axis is quantitative and the ab-sorbance corresponds to anoptical depth of 1 ppm H 2 O 2 through an optical path of 1 m.Band assignments are discussedin the text 380 T.J. Johnson et al.  through the plane of the molecule. The  ν 5  and  ν 1  bands(OH antisymmetric, symmetric stretches, respectively)overlap near 3,614 cm − 1 . The  ν 2 +  ν 6  band is relativelystrong for a combination band, and was assigned byRedington et al. [29], displaying a prominent   R Q 0  subbandnear 2,658 cm − 1 as discussed below.It is also seen in Fig. 1 that the  ν 6  band, which has beenused for remote sensing, is indeed the strongest band in thespectrum. Our results indicate that the integrated bandintensities are 433 cm − 2 atm − 1 for the 298 K data and467 cm − 2 atm − 1 for the 323 K data, where Napierianlogarithm units are used. These are both in goodagreement with the HITRAN values as discussed below[15]. The random error estimates on these values areapproximately ±8% for the 298 K data and ±3% for the323 K data. Although the integrated band strength shouldnormally be invariant with temperature, we put morecredence in our 323 K data, not because of the better agreement with the HITRAN values [15], but because of the smaller error bars and the greatly reduced adhesion/ sticking phenomena at 298 K that can be understood asfollows. Empirically, it is well known that as well asdissociating easily, H 2 O 2  is a  “ sticky ”  molecule, easilysorbing or adhering to experimental surfaces [26], similar to other species, such as HNO 3  or SOCl 2  [30, 31]. For the 298 K data, the experiment forces one to decreasethe temperature of all components downstream from thedisseminator device, including the gas cell. For the 298 K data there is thus an inherently much greater adhesion tosystem surfaces. This in turn means that even after manyminutes of flowing, the system will not have fullyequilibrated. Moreover, after a high-burden measurement,it can take many minutes, or even hours, for residual H 2 O 2 to fully desorb from the cooler surfaces, thus affecting theensuing measurement through contamination. For the323 K measurements, on the other hand, the exponentialvapor-pressure dependence with temperature greatlyreduces all of these effects, allowing for more rapid dataacquisition and more reliable values at significantly higher concentrations (factors of approximately 3  –  5 times),which in turn greatly increases the signal-to-noise ratioand decreases all sorption or desorption  “ memory ”  effects.The experimental results from the present measurementsfor the absolute intensity of the  ν 6  band are plotted versusthe HITRAN values in Fig. 2. As seen, the agreement isexcellent, with the difference plotted as the lower trace. For this plot the HITIRAN values [15] were plotted with theline width set to 0.1 cm − 1 and the optical depth set to1 ppm-m so as to correspond to the PNNL values. Theresulting calculated HITRAN spectrum has been verticallyoffset for clarity. We note that the HITRAN data are in fact from the HITRAN2004 update: Those band intensities arescaled to the 1991 work of May [32], but the originalcompilation of lines was actually taken from an earlier work by Hillman et al. [33], which did not include all thetorsional levels, and this partially accounts for the imperfect fit of some of the line profiles.In addition to the HITRAN2004 data, there have also been more recent high-resolution measurements of the  ν 6  band, as made by Klee et al. [34], which give very similar  Fig. 2  Comparison of the  ν 6  band of H 2 O 2  vapor from theHITRAN database ( top trace )and from the Pacific Northwest  National Laboratory (PNNL)experimental values ( middletrace ). Both plots correspond toan optical depth of 1 ppm m at 0.1-cm − 1 resolution. TheHITRAN spectrum [15] has been vertically offset for clarity.The  bottom trace  is the differ-ence between the two spectraThe absolute infrared cross sections of vapor-phase hydrogen peroxide at atmospheric pressure 381
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