Anal. Chem. 2000, 72, 2468-2474
Speciation of Sodium Nitrate and Sodium Nitrite Using Kiloelectronvolt Energy Atomic and Polyatomic and Megaelectronvolt Energy Atomic Projectiles with Secondary Ion Mass Spectrometry Michael J. Van Stipdonk,† Dina R. Justes,‡ Crista M. Force, and Emile A. Schweikert*
Department of Chemistry, Texas A&M University, P.O. Box 30012, College Station, Texas 77843-3012
The negative-ion mass spectra produced by kiloelectronvolt energy (CsI)nCs+ (n ) 0-2) and megaelectronvolt energy 252Cf fission fragment projectile impacts on NaNO3 and NaNO2 were collected and compared. The mass spectra generated by impacts of the kiloelectronvolt polyatomic primary ions on NaNO3 were markedly different from those derived from the fission fragment impacts, featuring higher relative intensities of nitrate (NO3-) specific secondary ions (those that reflect the sample stoichiometry). The most prominent secondary ion (SI) peaks produced from NaNO3 by the kiloelectronvolt energy projectiles were NO3- and Na(NO3)2-, both of which relate directly back to the chemical composition of the starting material. Likewise, the most prominent peaks produced by the kiloelectronvolt energy polyatomic projectile impacts on NaNO2 were NO2- and Na(NO2)2-. The fission fragment projectiles produced SI spectra from NaNO3 that were dominated by signals characteristic more of NaNO2, indicating that the megaelectronvolt energy ions induce considerable degradation of the nitrate solid. In addition, the fission fragment projectile produced relative negative SI intensity distributions that are remarkably similar to those reported in earlier studies of the use of laser desorption to produce SI signals from NaNO3. Of the projectiles examined in this study, the 20 keV (CsI)Cs+ projectile generated negative-ion mass spectra that best differentiated NaNO3 and NaNO2, primarily by producing a base peak in the NaNO3 spectrum that was unambiguously representative of the original sample stoichiometry. It is now well established that replacing atomic with polyatomic or cluster primary projectiles in secondary ion mass spectrometry (SIMS) significantly improves the yields of polyatomic and * Corresponding author. Phone: (409) 845-2343. Fax: (409) 845-1655. E-mail:
[email protected]. † Present address: Department of Chemistry, Wichita State University, Wichita, KS 67260-0051. ‡ Present address: Department of Chemistry, The Pennsylvania State University, University Park, Pennsylvania 16802.
2468 Analytical Chemistry, Vol. 72, No. 11, June 1, 2000
molecular secondary ions (SIs).1-13 In the regime of single-ion impacts, cluster ions such as Aun+ and (CsI)nCs+ ranging in size from two to seven atoms are known to produce nonlinear yield enhancements,2-4 which are defined as an increase in yield per projectile atom when atomic and polyatomic projectiles are compared at the same incident velocity. Using beams of cluster projectiles such as ReO4- and SF5+, SI yields have also been shown to increase per unit surface area damage8-10 when compared to those of atomic primary ions at a similar impact energy or momentum. Therefore, the use of polyatomic projectiles in SIMS should translate to improved molecular characterization and lower limits of detection when applied to surface analysis. To date, most fundamental studies in the area of sputtering by polyatomic projectiles have focused on the emission of SIs from organic targets of varying molecular weights. The application of polyatomic projectiles and SIMS to the molecular speciation of inorganic solids has yet to be fully explored. For this reason, we have begun a series of investigations of the SI mass spectra and secondary ion yields produced from inorganic solids by kiloelectronvolt energy atomic and polyatomic primary ions.14-17 A primary (1) Van Stipdonk, M. J.; Harris, R. D.; Schweikert, E. A. Rapid Commun. Mass Spectrom. 1996, 10, 1987. (2) Harris, R. D.; Van Stipdonk, M. J.; Schweikert, E. A. Int. J. Mass Spectrom. Ion Processes 1998, 174, 167. (3) M. Benguerba, M.; Brunelle, A.; Della-Negra, S.; Depauw, J.; Joret, H.; Le Beyec, Y.; Blain, M. G.; Schweikert, E. A.; Ben Assayag, G.; Sudraud, P. Nucl. Instrum. Methods Phys. Res., Sect. B 1991, 62, 8. (4) Blain, M. G.; Della-Negra, S.; Joret, H.; Le Beyec, Y.; Schweikert, E. A. Phys. Rev. Lett. 1989, 63, 1625. (5) Belykh, S. F.; Matveev, V. I.; Rasulev, Kh. U.; Samartsev, A. V.; Veryovkin, I. V. In Secondary Ion Mass Spectrometry, SIMS XI; Gillen, G., Lareau, R., Bennett, J., Stevie, F., Eds.; Wiley: Chichester, U.K., 1998; p 957. (6) Stapel, D.; Brox, O.; Benninhoven, A. Appl. Surf. Sci. 1999, 140, 156. (7) Kotter, F.; Benninghoven, A. Appl. Surf. Sci. 1998, 133, 47. (8) Kotter, F.; Niehuis, E.; Benninghoven, A. In Secondary Ion Mass Spectrometry, SIMS XI; Gillen, G., Lareau, R., Bennett, J., Stevie, F., Eds.; Wiley: Chichester, U.K., 1998; p 459. (9) Appelhans, A. D.; Delmore, J. D. Anal. Chem. 1989, 61, 1087. (10) Groenewold, G. S.; Delmore, J. E.; Olson, J. E.; Appelhans, A. D.; Ingram, J. C.; Dahl, D. A. Int. J. Mass Spectrom. Ion Processes 1997, 163, 185. (11) Szymczak, W.; Wittmaack, K. Nucl. Instrum. and Methods 1994, B88, 149. (12) Roberson, S.; Gillen, G. Rapid Commun. Mass Spectrom. 1998, 12, 1303. (13) Ada, E. T.; Hanley, L. Int. J. Mass Spectrom. Ion Processes 1998, 174, 231. (14) Van Stipdonk, M. J.; Harris, R. D.; Schweikert, E. A. Rapid Commun. Mass Spectrom. 1997, 11, 1794. (15) Van Stipdonk, M. J.; Santiago, V.; Schweikert, E. A. J. Mass Spectrom. 1999, 34, 554. 10.1021/ac991427v CCC: $19.00
© 2000 American Chemical Society Published on Web 04/27/2000
objective of the work thus far has been to measure and compare the amounts of artifact peak (chemical damage) production generated by atomic and polyatomic projectile impacts in order to choose the optimum projectile for surface analysis. In this report, we present the SI mass spectra of sodium nitrate (NaNO3) and sodium nitrite (NaNO2) produced by the impacts of kiloelectronvolt energy Cs+, (CsI)Cs+, and (CsI)2Cs+ as well as megaelectronvolt energy 252Cf fission fragment projectiles. The general ion emission patterns generated by these projectiles are compared to those of previously published reports using laser microprobe mass spectrometry (LMMS), which is receiving considerable attention as a tool for the speciation of inorganic solids.18-20 The goal in the present work was to determine which primary projectile or projectiles produce the highest relative intensities of analytically useful SIs (i.e., representative of the solid stoichiometry), the lowest chemical damage (at the level of discrete ion impacts), and the best differentiation between the two solids when used with time-of-flight (TOF) mass spectrometry. NaNO3 and NaNO2 were chosen for the present study because the characterization of nitrate solids using incident ionizing radiation presents a challenge. In particular, the interaction of R particles,21 γ-rays,22,23 and X-rays24 with nitrate solids has been shown to induce defect production and dissociation of the nitrate anion, leading to significant chemical degradation. Because the radiation damage induced by low-energy electrons25,26 and UVwavelength photons27,28 has also been studied, nitrate solids are good model systems for the study of ion emission and chemical damage created by the impact of energetic primary ions. These factors are particularly important to the use of polyatomic or cluster primary ions during the application of SIMS to surface analysis. The study of secondary emission following the interaction of ionizing radiation with sodium nitrate is also of practical interest because nitrates constitute a large fraction of radioactive waste material. This is an important point in light of the documented need for characterization and remediation within U.S. Department of Energy installations.29,30 In addition, the ability to speciate nitrogen/oxygen species on the surfaces of, for instance, aerosol particulates is important to studies in the field of atmospheric chemistry.31,32 (16) Van Stipdonk, M. J.; Justes, D. R.; Santiago, V.; Schweikert, E. A. Rapid Commun. Mass Spectrom. 1998, 12, 1639. (17) Van Stipdonk, M. J.; Santiago, V.; Schweikert, E. A.; Chusuei, C. C.; Goodman, D. W. Int. J. Mass Spectrom., in press. (18) Struyf, H.; Van Vaeck, L.; Poels, K.; Van Grieken, R. J. Am. Soc. Mass Specrom. 1998, 9, 482. (19) Van Vaeck, L.; Adriaens, A.; Adams, F. Spectrochim. Acta, Part B 1998, 53, 367. (20) Poels, K.; Van Vaeck, L.; Gijbels, R. Anal. Chem. 1998, 70, 504. (21) Logan, S. R.; Moore, W. J. J. Phys. Chem. 1963, 67, 1042. (22) Hochanadel, C. J.; Davis, T. W. J. Chem. Phys. 1957, 27, 333. (23) Cunningham, J. J. Phys. Chem. 1961, 65, 628. (24) Cunningham, J.; Heal, H. G. Trans. Faraday Soc. 1958, 54, 1355. (25) Knutsen, K.; Orlando, T. M. Surf. Sci. 1996, 348, 143. (26) Shin, J.-J.; Langford, S. C.; Dickinson, J. T.; Wu, Y. Nucl. Instrum. Methods Phys. Res., Sect. B 1995, 103, 284. (27) Knutsen, K.; Orlando, T. M. Phys. Rev. B 1997, 55, 13246. (28) McCarthy, M. I.; Peterson, K. A.; Hess, W. P. J. Phys. Chem. 1996, 100, 6708. (29) Bajic, S. J.; Luo, S.; Jones, R. W.; McClelland, J. F. Appl. Spectrosc. 1995, 49, 1000. (30) Campbell, J. A.; Stromatt, R. W.; Smith, M. R.; Koppenaal, D. W.; Bean, R. M.; Jones, T. W.; Strachan, D. M.; Babad, H. Anal. Chem. 1994, 66, 1208A. (31) Andreae, M. O.; Crutzen, P. J. Science 1997, 276, 1052. (32) Lu, Q.-B.; Madey, T. E. J. Chem. Phys. 1999, 111, 2861.
EXPERIMENTAL SECTION Mass Spectrometry. Negative-ion mass spectra from kiloelectronvolt energy atomic and polyatomic and megaelectronvolt energy atomic ion impacts were acquired using a custom-built, dual TOF mass spectrometer, of which the configuration and the operation have been discussed in detail elsewhere.33 The experiments described in this report were carried out in the event-byevent bombardment and detection mode at the limit of singleprojectile impacts. (CsI)nCs+ (n ) 0-2) primary ions were produced by 252Cf fission fragments passing through an aluminized Mylar foil coated with a vapor-deposited layer of cesium iodide. Fission fragments (∼100 MeV energy) that passed through the source foil and traversed the primary ion TOF region to strike the sample allowed the collection of plasma desorption and kiloelectronvolt ion induced mass spectra in a single experimental run. The (CsI)nCs+ primary ions were accelerated to 20 keV impact energy. To ensure that less than one projectile of a given m/z ratio impacted the sample target per start pulse, the primary ion transmission was decreased to ∼10% using low-transmission grids. After traversing the drift region, the primary ions successively struck the sample target. Secondary electrons emitted from the target surface following primary ion impacts were steered by a weak magnetic field into a dual microchannel plate (MCP) detector. It is assumed that each primary ion impact stimulates the emission of at least one secondary electron, allowing a relative measure of the primary ion dose on a particular target during the course of an experiment. The SIs emitted from the target surface were accelerated to -7 keV, separated in the secondary TOF region, and detected using a second MCP detector. A coincidence-counting protocol was used to simultaneously record and arrange the secondary ions detected from each relevant primary ion into individual secondary ion mass spectra.34,35 Therefore, the mass spectra produced by different projectiles could be directly compared because the transmission and detection efficiencies and the target surface conditions remained constant throughout the course of the experiment. Because low primary ion doses were used to collect the mass spectra from NaNO3 and NaNO2, charge compensation at the surface using low-energy electrons was not necessary. The use of electron flood guns is common in the characterization of insulating surfaces by SIMS. The electrons are used to neutralize the positive charge induced at the surface due to ion beam implantation and sputtering effects. As discussed later, the use of electron flood guns poses a potential problem to the analysis of nitrate solids, which are susceptible to radiation damage. Sample Preparation. Sample targets were prepared using 0.3 M aqueous solutions of NaNO3 and NaNO2. The solid salts were purchased from Aldrich (St. Louis, MO) and used as received. A 10 µL aliquot of salt solution was applied to a stainless steel sample support and allowed to dry in a dark fume hood at ambient temperature. At this concentration, homogeneous coverage of the (33) Van Stipdonk, M. J.; Harris, R. D.; Schweikert, E. A. Rapid Commun. Mass Spectrom. 1996, 10, 1987. (34) Blain, M. G.; Della-Negra, S.; Joret, H.; Le Beyec, Y.; Schweikert, E. A. J. Vac. Sci. Technol., A 1990, 8, 2265. (35) Van Stipdonk, M. J.; Schweikert E. A.; Park, M. A. J. Mass Spectrom. 1997, 32, 1151.
Analytical Chemistry, Vol. 72, No. 11, June 1, 2000
2469
stainless steel support was achieved with minimal caking or crust formation. The reproducibility of the mass spectra and ion intensity trends produced by the suite of projectiles were checked by comparing experimental runs using three individual targets prepared from the same salt solutions. RESULTS AND DISCUSSION Mass Spectra from Sodium Nitrate. In a SIMS study of ion emission from NaNO3, Groenewold and co-workers demonstrated that the negative-ion mass spectra produced by kiloelectronvolt energy Cs+ and ReO4- projectile impacts on sodium nitrate were more representative of the sample stoichiometry than the positiveion mass spectra.10 The same has been shown for the characterization of NaNO3 by LMMS.18 Due to the observations noted above and the nature of our experimental procedure (secondary electrons are required to count the number of incident projectiles), only the negative SI mass spectra from NaNO3 generated by the kiloelectronvolt and megaelectronvolt projectiles are discussed here. Figure 1 compares the negative SI mass spectra produced from the same sodium nitrate target by (a) 20 keV Cs+, (b) 20 keV (CsI)Cs+, and (c) 252Cf fission fragment projectiles. The negative-ion mass spectrum produced by the 20 keV (CsI)2Cs+ projectile was qualitatively similar to the one produced by (CsI)Cs+ (spectrum not shown). The SI peak intensities in Figure 1 were normalized using the number of projectile impacts and illustrate the qualitative and quantitative differences in the mass spectra produced by the kiloelectronvolt atomic, polyatomic, and 252Cf fission fragment projectiles. Each primary projectile produced several SIs representative of the molecular composition of sodium nitrate, including peaks at mass-to-charge ratios (m/z) 62 {NO3-} and 147 {Na(NO3)2-}. Secondary ions that contain nitrite units, such as those at m/z 46 {NO2-}, m/z 131 {Na(NO2)(NO3)-}, and m/z 115 {Na(NO2)2-}, were also observed. It is apparent from the spectra in Figure 1 that the relative intensities of the secondary ions bearing nitrate and nitrite subunits differ depending on the projectile used. The relative abundance of secondary ions containing the nitrite anion was significantly higher in the spectrum generated by 252Cf fission fragment impacts. In an earlier report, we investigated the relative SI yields resulting from the bombardment of NaNO3 by kiloelectronvolt atomic and polyatomic projectiles in a range of impact energies.16 The yields of SIs bearing nitrite subunits increased relative to those containing nitrate as the projectile complexity (number of atoms) increased. The yield changes were less sensitive to the overall impact energy, and the conversion of nitrate to nitrite during ion formation was linked to changes in the density of energy deposited by the projectile impacts. In this context, it is not surprising that the fission fragment, which deposits a zone of high collisional and electronic excitation density along the nuclear track, produced the highest relative abundance of nitrite-bearing ions from NaNO3. The 252Cf fission fragments also produced prominent peaks at m/z 85, 101, and 117 from NaNO3; the same peaks were present at much lower relative intensity in the mass spectra produced by the kiloelectronvolt energy atomic and polyatomic projectiles. In a report of the characterization of inorganic solids, including 2470 Analytical Chemistry, Vol. 72, No. 11, June 1, 2000
Figure 1. Negative secondary ion mass spectra produced from NaNO3 by the impacts of (a) 20 keV Cs+, (b) 20 keV (CsI)Cs+, and (c) ∼100 MeV 252Cf fission fragment projectiles. Incident ion doses were of the following magnitudes: Cs+, 106 ions/cm2; (CsI)Cs+, 105 ions/cm2; 252Cf fission fragments, 105 ions/cm2. Secondary ion intensities were normalized using the incident ion dose and are plotted relative to the most abundant peak in the spectrum.
NaNO3, using LMMS18 the SI peaks at m/z 85 and 101 were assigned compositions of {NaNO3-} and {NaNO3‚O-}, respectively. We assume that the peak at m/z 117 produced by the fission fragment projectile represents the addition of an O atom to the cluster ion at m/z 101. In the laser desorption study, a peak at m/z 117 was either not observed or not reported. Mass Spectra from Sodium Nitrite. Figure 2 shows the mass spectra produced from the same sodium nitrite sample using (a) 20 keV Cs+, (b) 252Cf fission fragments, and (c) 20 keV (CsI)Cs+. As with the nitrate sample, the mass spectrum produced by the 20 keV (CsI)2Cs+ projectile was qualitatively similar to the spectrum produced by (CsI)Cs+ (spectrum not shown). Regard-
Table 1. Metastable Dissociation Pathways Discerned and Suggested Composition Assignments for Selected Secondary Ions Sputtered from NaNO3 and NaNO2 by 252Cf Fission Fragment Projectilesa precursor ion {NaNO4-} {NaN2O4-} {NaN2O5-} {NaN2O6-}
w w w w
dissocn products
proposed compn
NaO20 + NO2NaNO20 + NO2NaNO20 + NO3NaNO30 + NO3-
(NaO2)NO2(NaNO2)NO2(NaNO2)NO3(NaNO3)NO3
a The pathways were determined using the ion-neutral correlation method and a reflectron-TOF mass spectrometer, which is described in detail in ref 34.
Figure 2. Negative secondary ion mass spectra produced from NaNO2 by the impacts of (a) 20 keV Cs+, (b) 20 keV (CsI)Cs+, and (c) ∼100 MeV 252Cf fission fragment projectiles. Incident ion doses were similar to those used to generate the spectra in Figure 1. Secondary ion intensities were normalized using the incident ion dose and are plotted relative to the most abundant peak in the spectrum.
less of the primary projectile, the overall SI intensities from the NaNO2 target were lower than those observed from NaNO3. Although the relative intensities of chemical background peaks (primarily cluster species containing carbon and carbon + hydrogen) were higher in the nitrite sample spectrum, secondary ion peaks attributable to the nitrite solid were clearly discerned. The fission fragment, (CsI)Cs+, and (CsI)2Cs+ projectiles generated prominent secondary ion peaks at m/z 46 {NO2-} and 115 {Na(NO2)2-}. The Cs+ projectile, however, produced a lowintensity peak at m/z 46 at the dose used in these experiments (ca. 5 × 106 ions/cm2), and an SI peak representing {Na(NO2)2-} was not observed. The fission fragment projectile impacts on
NaNO2, as with NaNO3, also produced prominent SI peaks at m/z 85, 101, and 117. Confirmation of Secondary Ion Composition. To assess the accuracy with which sample-specific SIs observed in the negativeion mass spectra reflect the composition of NaNO3 and NaNO2, it was necessary to establish, as clearly as possible, the correct ion composition. In all cases, the nominal molecular weights of the SIs measured using kiloelectronvolt and megaelectronvolt ion impacts and TOF-SIMS in the present study matched those obtained using LMMS and high-resolution Fourier transform ioncyclotron resonance (FT-ICR) mass analysis.18-20 To assign molecular compositions to the negative SIs observed in the spectra of the two solids, duplicate samples of NaNO3 and NaNO2 were tested in a reflectron-TOF instrument using coincidence counting mass spectrometry and the ion-neutral correlation method.36,37 The ion-neutral correlation method is a “tandem-TOF” technique that allows the composition and structure of sputtered ions to be determined via metastable dissociation reactions.38-40 A detailed description of the instrument (which features a 252Cf primary ion source) and the experimental protocol used is beyond the scope of this report and is provided elsewhere.38 The metastable dissociation pathways for the sample-specific secondary polyatomic ions sputtered from NaNO3 and NaNO2 samples by 252Cf fission fragments are shown in Table 1. It is assumed that the cluster ions sputtered by kiloelectronvolt energy primary ions follow the same dissociation pathways as when they are produced by 252Cf fission fragment impacts. Neither the NO2nor NO3- SIs were metastable on the time scale probed by the reflectron instrument (∼10-8-10-5 s following fission fragment impact). The dissociation of the {Na(NO3)2-} and {Na(NO2)2-} ions produced NO3- and NO2- product ions, respectively. The dissociation pathway of the latter is of particular interest. In one report describing the characterization of NaNO3 using LMMS,18 the composition of the ion at m/z 115 was assigned as {(NaNO3)NO-}. The metastable dissociation results reported here indicate that the composition is better described as {(NaNO2)NO2-}. (36) Della-Negra, S.; LeBeyec, Y. Anal. Chem. 1985, 57, 2036. (37) Van Stipdonk, M. J.; Schweikert, E. A. Nucl. Instrum. Methods Phys. Res., Sect. B 1996, 112, 68. (38) Van Stipdonk, M. J.; Justes, D. R.; English, R. D.; Schweikert, E. A. J. Mass Spectrom. 1999, 34, 677. (39) Bouchonnet, S.; Denhez, J.-P.; Hoppilliard, Y.; Mauriac, C. Anal. Chem. 1992, 64, 742. (40) Tang, X.; Ens, W.; Standing, K. G.; Westmore, J. B. Anal. Chem. 1988, 60, 1791.
Analytical Chemistry, Vol. 72, No. 11, June 1, 2000
2471
Table 2. Relative Intensities of Sample-Specific Secondary Ions Sputtered from NaNO3 and NaNO2 by 20 KeV (CsI)nCs+ (n ) 0-2) and 252Cf Fission Fragment Projectilesa primary projectile secondary ion
Cs+
(CsI)Cs+
OOHNO2NO3(NaO2)NO2(NaO2)NO3Na(NO2)2Na(NO2)(NO3)Na(NO3)2-
30 33 7 16 1 1 3 10 20
NaNO3 17 17 58 95 21 12 30 66 100
OOHNO2(NaO2)NO2(NaO2)NO3Na(NO2)2-
15 20 1.5
