Replacing the Argon ICP: Nitrogen Microwave Inductively Coupled

Subscriber access provided by UNIV OF LOUISIANA

Article

Replacing the Argon ICP: Nitrogen Microwave Inductively Coupled Atmospheric-Pressure Plasma (MICAP) for Mass Spectrometry Matthias Schild, Alexander Gundlach-Graham, Ashok Menon, Jovan Jevtic, Velibor Pikelja, Martin Tanner, Bodo Hattendorf, and Detlef Günther Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b03251 • Publication Date (Web): 17 Oct 2018 Downloaded from http://pubs.acs.org on October 22, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Replacing the Argon ICP: Nitrogen Microwave Inductively Coupled Atmospheric-Pressure Plasma (MICAP) for Mass Spectrometry Matthias Schild,1 Alexander Gundlach-Graham,1 Ashok Menon,2 Jovan Jevtic,2 Velibor Pikelja,2 Martin Tanner,3 Bodo Hattendorf1*, and Detlef Günther1* 1. Department of Chemistry and Applied Biosciences, ETH Zurich, Vladimir-Prelog-Weg 1, 8093 Zurich, Switzerland 2. Radom Corporation, 10521 W Forest Home Ave, Milwaukee, Wisconsin, USA 3. TOFWERK AG, Uttigenstrasse 22, 3600 Thun, Switzerland *[email protected], [email protected]

Abstract We combine a recently developed high-power, nitrogen-sustained microwave plasma source—the Microwave Inductively Coupled Atmospheric-Pressure Plasma (MICAP)—with time-of-flight mass spectrometry (TOFMS) and provide the first characterization of this elemental mass spectrometry configuration. Motivations for assessment of this ionization source are scientific and budgetary: unlike the argon-sustained Inductively Coupled Plasma (ICP), the MICAP is sustained with nitrogen, which eliminates high operating costs associated with argon-gas consumption. Additionally, use of a commercial grade magnetron for microwave generation simplifies plasma-powering electronics. In this study, we directly compare MICAP-TOFMS performance with that of an argon-ICP as the atomic ionization source on the same TOFMS instrument. Initial results with the MICAP source demonstrate limits of detection and sensitivities that are, for most elements, on par with those of the ICP-TOFMS. The N2-MICAP source provides a much “cleaner” background spectrum than the ICP; absence of argon-based interferences greatly simplifies analysis of isotopes such as

40Ca, 56Fe,

and

75As,

which typically suffer from spectral

interferences in ICP-MS. The major plasma species measured from the N2-MICAP source include NO+, N2+, N+, N3+, O2+, N4+, and H2O+; we observed no plasma-background species above mass-to-charge 60. Absence of troublesome argon-based spectral interferences is a compelling advantage of the MICAP source. For example, with MICAP-TOFMS, the limit of detection for arsenic is less than 100 ng L-1 even in a 1% NaCl solution; with ICP-MS, 35Cl40Ar+ interferes with 75As+ and arsenic analysis is difficult-toimpossible in chlorine-containing matrices.

Introduction

1 ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The argon-sustained Inductively Coupled Plasma (hereafter referred to as ICP) is a widely used atomization and ionization source for both optical emission spectrometry (OES) and mass spectrometry (MS).[1] ICP-MS was first reported in 1980[2] and is widely used for trace and ultra-trace elemental analysis; it offers high-sensitivity, isotopic detection, large linear dynamic range and flexible coupling with different sample introduction systems. State-of-the art instruments provide sensitivity of up to 1010 cps µg-1 L, low absolute detection limits (attograms), and a large linear dynamic range (109–1012);[3] however, recent advances in ICP-MS have been largely based on advances in sample-introduction systems and mass analyzer designs. Here, we present the evaluation of a novel nitrogen-sustained microwave inductively coupled atmospheric-pressure plasma (N2-MICAP) source combined with time-of-flight mass spectrometry (TOFMS). Operation of the ICP for elemental analysis has not changed significantly since its introduction in the 1960s.[4,5] Despite success and widespread use of ICP-MS, use of the ICP has several persistent drawbacks, such as need for sophisticated RF-power generators, incompatibility with ambient (air-borne) aerosols, high argon gas consumption (> 15 L min-1), and argon-based polyatomic interferences that complicate mass-spectral analysis. Approaches to overcome ICP drawbacks include, among others, gas-reaction cells to selectively attenuate argon-based interferences,[6] low-flow plasma torches to reduce argon consumption,[7] and gasexchange devices to introduce ambient particles.[8] Various research groups have attempted to replace the ICP with microwave plasma (MWP) sources or other atomic ionization sources. In Table 1, we record first reports of MWPs with a focus on combinations with MS. MWP sources have been used with argon, helium, oxygen, nitrogen and air; however, most MWP research is split between two general designs: low-power helium-sustained MWPs (He-MWPs) and low- and high-power nitrogen MWPs (N2-MWPs). HeMWPs are well-suited for the detection of high-IE elements, such as the halogens and nonmetals. However, due to relatively low gas temperatures and power densities, He-MWPs are not capable of efficient vaporization and atomization of liquid or solid aerosols, and are thus not a viable replacements of the ICP.[9] Compared to He and Ar, nitrogen is a readily available and inexpensive gas: in fact, switching from Ar to N2 for ICP operation represents a cost savings of roughly 70% (see SI for details). Several N2-MWPs designs have been investigated to replace the ICP as sources for both OES and MS. Based on the Okamoto MIP, Hitachi developed a commercial high-power N2-MIP-MS instrument—the Hitachi P-7000—in the 1990s and produced two additional instrument models (P-6000 and P-5000) into the early 2000s.[10] Performance of Okamoto microwave induced plasma (MIP)-MS instruments has been reported to be on par with ICP-MS.[11-15]


Page 2 of 25

Page 3 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Here, we report the first combination of the N2-MICAP with mass spectrometry; this source has previously been reported in combination with OES.[16,17] We use “MICAP” synonymously with “N2MICAP” because all experiments reported here use nitrogen as the plasma gas. Despite different naming conventions, it is important to remember that both the conventional argon-ICP and MICAP are, in fact, atmospheric-pressure inductively coupled plasma sources, though the means of generating the inductive field that transfers power to the plasma are different. To generate a plasma, the MICAP uses a tuned dielectric resonator ring made of advanced technical ceramic (Al2O3) that, when immersed in a microwave field, exhibits a bulk polarization current around the ring that oscillates at microwave frequency. This polarization current induces an oscillating electric field in the region of the plasma torch. The dielectric resonator ring can be thought of as analogous to the load coil of a conventional argon-ICP. However, unlike in ICP-MS, the dielectric resonator ring never has a net electric potential and energy coupling into the plasma is purely inductive. In the ICP, inductive coupling dominates, but capacitive coupling due to high voltage across the induction coil causes a plasma offset potential, which may affect the ion energy distribution and induce a secondary discharge to the ICP-MS interface.[18] Specific resonator/coil designs[19,20] or use of a shielded torch[21] had to be employed to minimize the plasma potential in ICP-MS.

Experimental The MICAP source was combined with TOFMS in the laboratory of Trace Element and Micro Analysis at ETH Zurich. To compare performance of ICP and MICAP as elemental MS sources, we carried out successive experiments with the same TOFMS instrument. In this analysis, we strived to keep operating conditions as similar as possible.

Microwave-sustained Inductively Coupled Atmospheric-pressure Plasma (MICAP) Source The MICAP ion source for mass spectrometry (Radom Corp, USA) consists of a cavity magnetron (1500 W Panasonic 2M262A-07AC), an aluminum waveguide to direct microwaves to the resonator ring, an inductive iris to provide impedance matching, the dielectric resonator ring, the torch assembly, an aluminum ring to tune the dielectric ring’s resonance frequency, and a ventilation shaft for removing hot gas from the plasma-source box. A flat quartz disk with a central hole 20-mm in diameter (just larger than the torch diameter) was fixed between the end of the torch and the sampler plate. The hot gas from between the quartz disk and sampler plate was extracted via the house ventilation. Power to the magnetron was supplied by a custom 4kV DC switching power supply (Magdrive 2000, Dipolar AB, Sweden) and controlled by a personal computer. For all experiments, a conventional quartz ICP torch (Spectro Arcos SOP-Torch, 1.8-mm injector, Precision Glassblowing, USA) was used. Optimum position of the plasma torch inside the dielectric resonator was obtained with a 3D translation stage; however, the MICAP torch



box—including magnetron and dielectric resonator—are stationary at the vacuum interface. Controlled positioning of the resonator and torch box with respect to the MS interface is critical because all conductive surfaces around the resonator contribute to the effective resonance frequency. An overview sketch of the MICAP-TOFMS instrument is provided in Figure 1; a more detailed view is provided in Figure S1. Plasma ignition was accomplished in a computer controlled sequence of around 15 sec. A small volume (~100 mL, 1.5 L min-1 Ar for 4 sec) of argon gas was introduced coincident with a spark to the intermediate gas to facilitate propagation of the ignition spark to the dielectric resonator region. Nitrogen and argon gas flows (gas purity 99.996%, PanGas AG, Switzerland) were controlled by built-in mass-flow controllers of the MICAP control unit (Radom Corp., Wisconsin, USA). Because the dielectric resonator ring experiences no conductive current, ohmic heating is minimized and it can be operated without active cooling. In Table 2, we provide routine operating parameters of the MICAP-TOFMS instrument. As seen in Figure 1b, the N2 plasma had pink background emission and was noticeably dimmer than an argon-ICP. The MICAP is an annular-shaped plasma with physical structure similar to that of a conventional ICP.[16] Stable operation of the plasma was obtained for powers of 1100 W to 1500 W, though gas flows had to be reduced for low-power operation.

Vacuum Interface and Time-of-Flight Mass Spectrometer The MICAP source was mounted to the existing mass analyzer interface, which is based on that of the ELAN-6000 ICP-MS instrument (PerkinElmer/Sciex, Ontario, Canada). To accommodate the MICAP source, we modified the ICP-TOFMS instrument through use of a smaller orifice-diameter (0.8 mm) sampler cone and an additional rotary vane pump connected to the first stage of the vacuum interface to maintain a pressure of 2.1 mbar (combined pumping speed of ˜75 m3 h−1, E2M40 and E2M28, Edwards, UK). Also, a grounded photon stop (3-mm diameter) was installed just beyond the skimmer in the third vacuum stage. The TOFMS system (TOFWERK AG, Switzerland) coupled to ICP has been previously described.[22] Readers interested in a detailed explanation of ICP-TOFMS are referred elsewhere.[23-25] The instrument utilizes a quadrupole RF notch filter to selectively attenuate abundant species upstream of the TOF mass analyzer. With an ICP source and a wet plasma, 14N+, 16O+, 16O +, 2

and 40Ar+ are routinely notched; however, the only species that required attenuation with the MICAP

source was 14N16O+. The ion optics consist of a cylindrical lens at positive potential between the skimmer cone and a negatively biased entrance aperture of the notch filter assembly. This configuration typically acts as a chromatic lens, in a sense that higher potentials need to be applied for optimized transmission of ions of increasing kinetic energy.[26] For TOFMS operation the potential at this lens is static, which led to preferential transmission of ions of a specific energy range.

Sample Introduction


Page 4 of 25

Page 5 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Liquid sample introduction was investigated with two setups: “wet” sample introduction via a concentric pneumatic nebulizer (MicroMist, Glass Expansion Inc., MA, USA) combined with a noncooled cyclonic spray chamber and “desolvated” sample introduction via an Apex Q (Elemental Scientific, NE, USA) desolvation system, using the same nebulizer, a heated cyclonic spray chamber and a condenser. For both introduction systems, sample was transported to the nebulizer by peristaltic pump and the liquid uptake rate was about 500 µL min-1.

Samples Multi-element solutions were made from single-element ICP stock solutions (Merck AG, Germany and Inorganic Ventures, USA). Dilutions were made with sub-boiled 1% HNO3 in ultra-high purity (UHP) water (Millipore, Billerica, Massachusetts, USA), and were carried out gravimetrically (Mettler AE240, Mettler-Toledo, Switzerland). Final concentrations ranged from 1 ng L-1 to 1 µg L-1; standard solutions had equal nominal concentrations of all analytes except for calcium which was ten times higher. For investigation of the sodium matrix effect, multi-element solutions of 10 µg L-1 nominal concentration were made in NaCl (Trace-grade, Sigma Aldrich) solutions from 1 µg g-1 to 100 mg g-1 and 1% HNO3.

Data Processing TOF data were acquired using TOFdaq Recorder (TOFWERK AG) software with spectral acquisition rate of 3.3 Hz (10,000 TOF spectra summed on the digitizer). Post-acquisition data analysis, including mass calibration, baseline fitting and subtraction, and mass-spectral peak integration was done in Tofware (ver. 2.5.10, http://www.tofwerk.com/tofware/). Details of influence of baseline fitting and subtraction for ICP-TOFMS can be found elsewhere.[25,27] Mass-spectral peak data were exported as CSV files from Tofware and further processing (e.g. blank subtraction, calibration, limits of detection calculations) and plotting was performed in R (https://www.rproject.org). Limits of detection (LODs) were estimated from 3 times the standard deviation of the blank signal divided by the sensitivity.

Safety High power (1500 W) microwave radiation from the magnetron can pose a major health concern. Due to the short wavelength of microwaves, even smallest gaps in the metal housing can lead to potentially hazardous leakage. The dielectric resonator cavity used for the MICAP source was designed to effectively block microwave leakage. For basic microwave safety readers are referred to chapter 4 on microwave safety in Jankowski and Reszke’s book on MIPs.[28] Like an ICP, the MICAP is a source of ultraviolet radiation and care should be taken not to look directly into the plasma.



Page 6 of 25

Results and Discussion When we first combined the MICAP source with the TOFMS instrument, we observed a dramatically elevated, m/z-independent, baseline in the mass spectrum. This elevated baseline was likely due to emission from N2 metastable species[29] downstream of the skimmer. To reduce this background, we installed the photon stop, which is an original component of the ELAN-6000 ICP-MS interface.[30] As seen in Figure S2, the inclusion of the photon stop lowers the MICAP-TOFMS background by almost two orders of magnitude with limited effect on isotopic sensitivities.

ICP-TOFMS vs. MICAP-TOFMS: Dynamic Range, Sensitivities, and LODs Figure 2 shows mass spectra of ICP-TOFMS and MICAP-TOFMS for the same 10 µg L-1 multi-element solution and sample introduction system, while Figure S4 shows a spectrum for the same solution at 1 µg L-1 with desolvated sample introduction. In Figure S3 we also provide a full spectrum of the MICAP-TOFMS for a 1% HNO3 blank. As expected, all major Ar-species, such as Ar+, ArH+, ArC+, ArN+, ArO+, and Ar2+, were absent from the MICAP mass spectrum; additionally, many low-mass oxygen species (O+, OH+, H2O+) were more intense with the ICP. The most abundant MICAP background species are NO+, N2+, and N+. Not obvious from Figure 2 is that the MICAP source had a substantially lower background ion flux than the ICP. Despite the fact that with the MICAP only NO+ had to be attenuated in the notch filter, the total background ion current was at least four-times lower with the MICAP compared to the ICP, where Ar+ and O+ required attenuation to protect the detector. In Figure 3, we compare abundance-normalized sensitives of ICP- and MICAP-TOFMS with wet sample introduction, and Table 3 lists LODs obtained with the two ion sources. As seen in Figure 3, analyte sensitivities were slightly lower for MICAP-TOFMS for almost all elements (except Cs). This is considered to be related mostly to the more pronounced mass discrimination of the MICAPTOFMS. We consider this to be due to the energy-dependent ion-transfer through optics and a stronger dependence of ion kinetic energies on mass from the supersonic expansion of N2 instead of Ar.[18] This effect is a matter of current investigations. Contrary to sensitivity differences found between the MICAP- and ICP-TOFMS setups, LODs for the two systems were comparable for almost all elements due to lower background levels for MICAP-TOFMS compared to ICP-TOFMS. Substantial differences in LODs between the ICP- and MICAP-TOFMS instrument setups are only found for Mg, Ca, and Fe. In the case of Mg, lower mass discrimination of the ICP-TOFMS instrument yielded higher sensitivity and thus a better LOD. For Ca and Fe, using wet plasma conditions, we obtained over an order of magnitude lower LODs with the MICAP source because major isotopes of these elements (40Ca and

56Fe)

are accessible with

MICAP-TOFMS. For all elements measured, a linear dynamic range (LDR) of 105 to 106 was determined for MICAP-TOFMS with roll-off on the high concentration end caused by TOFMS6 ACS Paragon Plus Environment

Page 7 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


detector saturation; this matches the LDR of ICP-TOFMS. Example calibration curves are available in Figure S7. In comparison to previous work with Okamoto-type MIP-MS instruments,[11,31] the MICAP-TOFMS provides ~1 order of magnitude higher LODs for light elements (< m/z 45) and comparable LODs for heavier elements (Table S1). Major differences between performance of the MICAP-TOFMS and the MIP-MS are likely due to the mass analyzer and ion optics configuration rather than plasma source. For MICAP-TOFMS, desolvated sample introduction leads to a five- to ten-fold increase in sensitivity for most elements, a reduction of oxide levels, and has little effect on abundances of plasma background species, which results in improvement of LODs also by one order of magnitude. In Table 3, we also provide a comparison of percent oxides, nitrides, and doubly charged ions in the ICP and MICAP under standard operating conditions. Generally, we find that the MICAP source is more prone to oxide formation than the ICP. Nitride species formation is roughly ten times greater with the MICAP compared to the ICP, however, as will be discussed in the next sections, metal-nitride abundances appeared stable for a wide range of plasma conditions.

Spectral Characterization of MICAP-TOFMS In Figure 4, we provide the background mass spectrum for MICAP-TOFMS with wet sample introduction of a 1% HNO3 solution. Significant background plasma species occur between m/z 14–60; background species above m/z 60 are most likely impurities in the HNO3 blank (Fe, Ni, Co, and Zn). In Table S2, we provide a survey of all background species observed with MICAP-TOFMS along with analyte isotopes they potentially interfere with and the mass resolving power required to overcome each interference. Apart from the dominant NO+, the major background species include N2+, N+, N3+, O2+, N4+, O+, and H2O+; all other species typically have peak intensities of less than 2 cps. The MICAP background spectra show some similarity to those observed in “cold plasma” ICP-MS experiments[32], in which NO+ appears as one of the major species. However, we have not observed high abundance of (Hydr-)Oxide ions characteristic of cold plasma ICP-MS,[33] which indicates that the gas temperature of the MICAP is similar to that of conventional ICP operation. Furthermore, elements with ionization energies (IEs) higher than NO (IE=9.26 eV) are present in the MICAP-TOFMS spectra, indicating that N2+ plays a similar role as Ar+. Nitrogen-based interferences are also more easily resolved than argon-based interferences because nitrogen has a positive mass defect:[34] On the TOFMS instrument, a RM (∆m at full-width half-maximum) of 1500–2500 is commonly achieved, which allows for identification or separation of many observed nitrogen-based molecular ions.



Optimization of MICAP-TOFMS As part of our investigation of MICAP-TOFMS, we did a multi-parameter optimization of nebulizer gas flow and plasma power for identical TOFMS and ion optics settings. In Figure 5, we plot selected element- and plasma-species responses to changes in nebulizer gas flow and plasma power. Responses of all species studied are provided in Fig. S5 and S6. As seen in Fig. 5a, two general trends were observed when changing nebulizer gas flow: 1) sensitivity for elements with higher first ionization energies (e.g. As, IE=9.79 eV) was higher with lower nebulizer gas flow and 2) high-m/z ions optimized at higher gas flows. At lower nebulizer gas, the plasma is hotter and thus more efficiently ionizes high-IE elements. At the same time, the change in gas temperature leads to mass-dependent changes in ion transmission through the ion optics where ions with higher m/z optimize at higher flow rate under the conditions used. In a similar manner as with ICP-MS, increased nebulizer gas flow leads to higher oxide abundances (e.g. CeO+/Ce+ and ThO+/Th+); interestingly however, as shown in Fig. S5, this trend was not observed for nitride species. Changing the magnetron power for a fixed nebulizer flow rate showed a similar behavior. Higher power led to an increase in plasma temperature and steadily increasing analyte signals. Only isotopes that optimized at higher gas flow rate were found to show an intermediate maximum, caused by the mass dependency of the ion kinetic energies.

Matrix Tolerance of MICAP-TOFMS A common preconception about microwave plasmas is that they have lower tolerance to liquidsample introduction and suffer much more than ICPs from matrix effects. This idea dates back to the early Beenakker cavity,[35] Surfatron,[36,37] and MPT[38] systems, which all are inherently low-tomedium power designs (

Replacing the Argon ICP: Nitrogen Microwave Inductively Coupled

Recommend Documents