Electrothermal Vaporization Sample Introduction ... - ACS Publications

Subscriber access provided by NEW YORK UNIV

Article

Electrothermal Vaporization Sample Introduction for Spaceflight Water Quality Monitoring via Gas Chromatography-Differential Mobility Spectrometry William T Wallace, Daniel B Gazda, Thomas F Limero, John Michael Minton, Ariel V Macatangay, Prabha Dwivedi, and Facundo M Fernandez Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b00055 • Publication Date (Web): 14 May 2015 Downloaded from http://pubs.acs.org on May 19, 2015

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 free 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 accessible to all readers and 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.

Analytical Chemistry 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 26

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

Electrothermal Vaporization Sample Introduction for Spaceflight Water Quality Monitoring via Gas Chromatography-Differential Mobility Spectrometry.

William T. Wallace1,*, Daniel B. Gazda1, Thomas F. Limero1, John M. Minton2, Ariel V. Macatangay3, Prabha Dwivedi4, and Facundo M. Fernández5,*

1

Wyle Science, Technology, and Engineering Group, Houston, TX 77058

2

University of Arkansas-Little Rock, Little Rock, AR 72204

3

NASA Johnson Space Center, Houston, TX 77058

4

Centers for Disease Control and Prevention, Atlanta, GA 30341

5

School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, GA 30332

Co-corresponding authors: William T. Wallace. [email protected], Ph: 281 483 2846. Fax: 281 483 3058 Facundo M. Fernández. [email protected]. Ph: 404 385 4432. Fax: 404 385 3399

ACS Paragon Plus Environment


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

Page 2 of 26

Abstract In the history of manned spaceflight, environmental monitoring has relied heavily on archival sampling. However, with the construction of the International Space Station (ISS) and the subsequent extension in mission duration up to one year, an enhanced, real-time method for environmental monitoring is necessary. The station air is currently monitored for trace volatile organic compounds (VOCs) using gas chromatography-differential mobility spectrometry (GC-DMS) via the Air Quality Monitor (AQM), while water is analyzed to measure total organic carbon and biocide concentrations using the Total Organic Carbon Analyzer (TOCA) and the Colorimetric Water Quality Monitoring Kit (CWQMK), respectively. As mission scenarios extend beyond low earth orbit, a convergence in analytical instrumentation to analyze both air and water samples is highly desirable. Since the AQM currently provides quantitative, compound-specific information for air samples and many of the targets in air are also common to water, this platform is a logical starting point for developing a multi-matrix monitor. Here, we report on the interfacing of an electrothermal vaporization (ETV) sample introduction unit with a ground-based AQM for monitoring target analytes in water.

The results show that each of the

compounds tested from water have similar GC-DMS parameters as the compounds tested in air. Moreover, the ETV enabled AQM detection of dimethlsilanediol (DMSD), a compound whose analysis had proven challenging using other sample introduction methods. Analysis of authentic ISS water samples using the ETV-AQM showed that DMSD could be successfully quantified, while the concentrations obtained for the other compounds also agreed well with laboratory results.

1 ACS Paragon Plus Environment

Page 3 of 26

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


Introduction Monitoring the quality of the spacecraft environment is required during human spaceflight to maintain the health of crew members and vehicle systems.

Environmental quality accounts for

contaminants found in air and water, microbial growth, acoustics, and exposure to radiation. Some major atmospheric constituents (O2 and CO2) have been monitored in real time for decades, but the majority of environmental monitoring has relied on archival sampling techniques, in which samples are collected from the spacecraft and returned to Earth for analysis.1 While environmental assessment from archival samples has been adequate for short missions, such as those on the Space Shuttle, longer missions aboard the International Space Station (ISS), in which sample return may surpass 6 months after collection, have emphasized the need for more extensive real-time environmental monitoring. This fact is especially true as NASA and other space agencies move towards more distant exploration missions, when there will be limited ground support and no opportunity for archival sample return. The current air and water monitors on board the ISS provide very specific data.

Major

constituents in the ISS atmosphere (O2, CO2, N2, H2, and CH4) are monitored by the Major Constituent Analyzer (MCA), a magnetic sector mass spectrometry-based instrument.2 The Air Quality Monitor (AQM) measures compound-specific concentrations of a targeted list of volatile organic compounds (VOCs) selected based on their potential toxicological effects on crew health or detrimental effects on specific ISS systems.3-7 The AQM combines gas chromatography with differential mobility spectrometry (GC-DMS) and is the successor to the Volatile Organic Analyzer.8-10 Water quality is monitored in real time by the Total Organic Carbon Analyzer (TOCA) and the Colorimetric Water Quality Monitoring Kit (CWQMK). The TOCA measures the total organic carbon (TOC) load in the ISS water systems but does not provide compound-specific information. The CWQMK monitors the levels of molecular iodine in water from the U.S. segment and ionic silver from the Russian segment. These biocides are used to inhibit microbial growth in the water and must be maintained within a strict concentration range to prevent adverse health effects such as thyroid dysfunction or argyria.11,12



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

Page 4 of 26

While the TOCA provides excellent trending data regarding total organics and the overall water quality on the ISS, the inability to determine which compounds are responsible for any TOC increases makes it difficult to implement effective mitigation plans. If, for example, a compound with no known adverse health effects causes a TOC increase, precious resources could be saved by not attempting to remove it. Events beginning in June 2010 highlight the need for specific chemical information about the presence of organic compounds in ISS water.13 When the TOC began to increase in the water being produced by the ISS Water Processor Assembly (WPA), archival samples had to be used to determine which compound(s) was responsible for the increase because there was no compound-specific information available. Surprisingly, no individual organic compounds were detected at significant levels in the standard ground analyses. However, during testing for glycols, an interfering peak appeared.14 After significant analytical effort, it was determined that the interfering compound was dimethylsilanediol (DMSD), a degradation product of silicon-based organics, including those found in ISS air.4,6,15 DMSD was found to account for greater than 90% of the TOC in the WPA product water. Despite its low to moderate toxicological concern for oral exposure,16 DMSD causes other issues for ISS operations, such as the need for early replacement of the multifiltration (MF) beds used to remove inorganic and non-volatile organic contaminants.17 Replacement of the MF beds in the WPA eventually led to a decrease in the measured TOC; however, the increase has routinely reoccurred.18 While archival analyses have shown that DMSD was the cause of these increases, it is risky to assume that DMSD is the only compound that typically causes an increase in TOC. The use of the AQM for both air and water monitoring would provide improved information over existing systems, while also fulfilling the needs of future exploration missions by reducing the amount of hardware needed for environmental monitoring. This approach, however, necessitates an inlet system to allow analytes in water to be volatilized before entering the AQM. A number of different techniques, among which are membrane extraction19-21 and electrospray,22-24 are available for liberating analytes from a liquid matrix and promoting them to the gas phase for analysis by ion mobility spectrometry, each with its

own

strengths

and

weaknesses.

Membrane

extraction,


for

example,

routinely

uses

Page 5 of 26

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


polydimethylsiloxane (PDMS) membranes for sampling, which can lead to recovery problems for targets of interest in ISS water. Additionally, the use of membranes could lead to losses of highly water soluble small polar organics, such as methanol, ethanol, and isopropanol. In the case of electrospray, the use of organic species in the solvent to promote ionization would preclude the detection of those target species. In the present work, an electrothermal vaporization (ETV) unit recently reported by Dwivedi et al.25 for direct analysis in real time-mass spectrometry analysis (DART-MS) is coupled to a laboratorybased AQM for evaluation. The ETV releases VOCs from the water matrix, allowing them to be separated by GC, ionized by a

63

Ni radioactive source, and differential-mobility-analyzed as gas-phase

ions. The processes required for AQM calibration is described and analytical results from the ETV-AQM are compared with standard laboratory methods for ISS archival water samples. The performance of the ETV-AQM indicates that this platform could indeed be used for trace VOC analysis in both air and water aboard the ISS. The mass and volume of the system could also make it suitable as a field deployable method for water analysis in remote locations on Earth.

Experimental Chemicals The target analytes used for this proof-of-concept study are a subset of those used by Dwivedi et al.25 These compounds are routinely found in the ISS atmosphere and are already monitored using the AQM on board the ISS. The current target compounds include methanol (MeOH), ethanol (EtOH), acetone (ATN), isopropanol (IPA), DMSD, trimethylsilanol (TMS), and 2-butanone (2BN). DMSD was obtained from Absolute Standards, Inc. (Hamden, CT) at a concentration of 70 ppm. MeOH, EtOH, IPA, and 2BN standards were > 98% purity (ChemService, West Chester, PA), while TMS was ≥ 98% purity (Sigma-Aldrich, St. Louis, MO). Dilutions for testing were prepared using ultrapure water (18.2 MΩ cm) from a Milli-Q Integral 5 water purification system (EMD Millipore, Billerica, MA).



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

Instrumentation The AQM used in the current experiments, the technology on which it is based, and typical methods of analysis have been described in detail previously.26,27 The AQM, built by Sionex, integrates a preconcentrator (Carbopack B and Carboxen 1000), a GC column (15 m x 0.25 mm, bonded DB5), a microDMx sensor, and a Windows XP computer in a 3 kg, 25.4 cm x 15.2 cm x 13.2 cm size package with a nominal peak power demand of 72 W. The instrument is controlled by Draper Expert software (v. 4.1.3) using customized methods for sampling times and heating rates. For the results presented here, 5 ssample times were used. Scrubbing cartridges in the unit clean the recirculating carrier gas of moisture and VOCs using 20/54-mesh molecular sieve (HCRMS) and Carboxen 569, respectively. An additional scrubbing cartridge containing the same materials was used to flush (purge) the pre-concentrator after samples had been collected in order to remove excess moisture. The ETV used for these studies is based on the design previously described by Dwivedi et al.25 Samples were introduced into the AQM through a short tube attached to the instrument and furnishing a Swagelok-compatible fitting on the other end, which was inserted into the ETV glass tube. The entire ETV-AQM setup is shown in Figure 1A with a close-up view of the ETV in Figure 1B. Current was supplied to the sample ribbon using a GW Instek PSM 6003 programmable power supply while clean zero air flowed through the ETV at a rate of ~ 250 cc min-1. This gas flow rate was empirically determined to be satisfactory for the purposes of this study; however, lower flow rates may also be acceptable.

Analytical Procedure For the tests described here, 1 µL of sample containing the target analytes was pipetted onto the ETV nichrome ribbon prior to starting the timing sequences of both the AQM and the ETV (Supporting Information, Figure S-1). After starting the flow of air through the ETV, the start buttons of the AQM and ETV power supply were depressed simultaneously. Following 10 seconds of the AQM pump running with the ETV at 0 A, the 3-way valve of the AQM opens to collect the sample VOCs for 5 seconds while 5 ACS Paragon Plus Environment

Page 6 of 26

Page 7 of 26

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 current through the nichrome sample ribbon is increased to 2 A for 1 second, 5A for 2 seconds, and 6 A for 2 seconds. At this point, the 3-way valve of the AQM closes the sample inlet and opens the purge inlet to remove excess water from the pre-concentrator. The ETV timing sequence is repeated to clean the ribbon between samples. Prior to testing of all samples, both a “clean” and “no sample” run were performed (Tables S-1 – S-3). During the clean run, DI water was pipetted onto the nichrome ribbon and an ETV heating cycle was performed. Simultaneously, the AQM pre-concentrator and GC column were heated to temperatures in excess of their normal usage conditions. Each of these steps was performed in order to remove any residual sample from the system. Subsequently, the no sample procedure (blank) occurred. In this run, the ETV heating cycle was performed with no sample, and the AQM was set to heat the pre-concentrator and GC column in the same manner as during a sampling session, only without the valve opening to introduce a sample. The results obtained from this procedure can be used to determine the cleanliness of the system prior to sample collection.

Results and Discussion Compound Characterization While GC-DMS is capable of providing compound-specific information on the composition of a mixture, the analytical parameters for the individual compounds (compensation field, separation field, retention time) must first be ascertained.

Generally, this process requires testing compounds

independently. While our previous experience of calibrating the AQMs for spaceflight air monitoring has provided guidance, our initial uncertainty regarding the species that would be observed for each of the compounds selected for water analysis (monomer/dimer, hydrated clusters, etc.) required retesting of all compounds. An ETV-GC-DMS run for a 25 ppm IPA standard is shown in Figure 2 as an example. GC and ETV-DMS parameters obtained for all selected compounds are shown in Table 1, where all voltages have been converted to field strength (Townsend, Td) in order to account for the temperature (~ 353 K) and pressure in the DMS cell. AQM identification of the target compounds (except for DMSD) also has 6 ACS Paragon Plus Environment


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

Page 8 of 26

been confirmed with mass spectral analysis.27 Note the retention times for ATN and IPA. On the particular GC column used for these studies, these two compounds essentially coelute, making identification in mixtures challenging. However, by increasing the separation field (SF), as shown in Table 1, the two compounds can be separated in DMS space, albeit slightly, in order to allow quantification. This result is a consequence of a limitation of the current AQM setup; the GC column only cools to approximately 308 K prior to sample runs, so many small polar organics elute in a small time window. A strength of DMS is shown in the table, however. Positive and negative ions are analyzed and simultaneously detected in DMS, so a great deal of information was collected in a single run. Here, most of the selected compounds were detected as positive ions. However, DMSD was detected much more easily in the negative mode, as previously observed by Dwivedi et al.25 Potential causes for the presence of DMSD in the negative mode are discussed in the Supporting Information, section S-1. During this compound characterization process, the TMS spectra showed 2 peaks, one at a compensation field (CF) of 0.11 Td (Table 1) and one at -0.21 Td, corresponding to dimer and monomer species, respectively. A concentration-dependency study for this compound indicated that the dimer species provided a wider dynamic range for monitoring, while the monomer peak provided greater sensitivity, as expected. Based on our previous experience with air monitoring, in which only the dimer species is generally present, and the fact that the required detection limit of 0.5 mg L-1

25

was achievable

using the dimer peak, further testing focused on this particular ion type. Table 2 shows the DMS parameters for those compounds selected as targets for both air and water analysis. In this table, the CF has been normalized to the applied SF in order to account for any changes in separation field over time. It can be seen that the required normalized CFs are astonishingly similar regardless of the sample matrix, even though the results for compounds in air were obtained several years prior to those in water samples. This indicates that the ionic species being detected through ETV introduction are likely the same species detected during air analysis. As the parameters for DMSD in air were not previously determined using the same headspace methods, they are not included in the table. 7 ACS Paragon Plus Environment

Page 9 of 26

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


It should be noted that the concentrations used for the comparison in Table 2 are quite different. For instance, AQM measurements using a standard air sample are usually in the range of 0.1 – 0.9 ppm, while the water samples were all measured at 25 ppm. Previous experience has shown that air-based analysis of compounds at the levels used for the current water-based samples would saturate the detector and, in some cases, could even saturate the sieve material used to clean the recirculation gas. However, using the water samples at 25 ppm leads to peak intensities that are similar to those obtained in air at 20fold lower concentrations. There are 2 possibilities to be considered as an explanation for these intensity differences. First, it is possible that a significant amount of analyte is being lost during the vaporization process. These losses might be expected, as neither the ETV itself nor the ETV-to-AQM connection are airtight (Figure 1). Secondly, there is a finite mass of analyte present in the 1 µL water samples, whereas an air sample will be pulled in continuously, thereby providing a greater mass of analyte being adsorbed on the pre-concentrator, even at a much lower concentration (Supporting Information, S-2). It is likely that the latter explanation is responsible for the similar peak intensities observed for the vastly different concentrations used in this testing. Data obtained using the “scanning” mode of the AQM, such as that shown in Figure 2, in which the CF is swept at a constant SF, is extremely useful for characterizing individual compounds, showing the presence of unknowns, and also allowing for the detection of co-eluting compounds if their specific compensation values are sufficiently separated. However, in most cases, the use of a single SF will not adequately allow for the detection of all compounds. This is reflected in the spectra shown in Figure 3. Here, using a SF of 86 Td, a mixture containing MeOH, IPA, TMS, DMSD, and 2BN was introduced into the AQM. As can be seen in Figure 3A, for 3 of the 5 species, the peaks for the individual compounds are well separated from any other peaks in the positive mode spectrum, while DMSD is shown in the negative mode spectrum (Figure 3B). However, at this SF, the MeOH peak overlaps with the reactant ion peak (RIP). The RIP signal is generally recognized as an overlapped signal consisting of a hydrated proton with a variable number of water molecules attached arising from the 63Ni ionization source.28 In this case, as seen in Figure 3C, the SF must be increased in order to unambiguously separate and identify 8 ACS Paragon Plus Environment


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

Page 10 of 26

MeOH. This increase in SF can lead to the loss of other compounds. Note the complete loss of IPA and MEK and the large decrease in intensity for TMS. While it would certainly be possible to perform multiple experiments at different SFs in order to acquire information on each of the compounds, this process would not be efficient, especially for a nonautomated setup. The AQM also possesses a “GC” mode that allows GC retention time windows to be correlated with compound-specific separation and compensation voltages determined from the scanning mode (here, we use the term “voltage” in place of “field” due to the fact that those are the parameters that are entered into the Draper Expert software). A short GC method is displayed in Figure 4 to demonstrate how 6 different compounds possessing widely different separation and compensation voltages are all able to be detected in a single experiment.

Calibration Composite chromatograms from each of the target compounds with increasing concentration are shown in the Supporting Information, Figure S-2.

For each of the compounds, an increase in

concentration initially leads to larger peak heights, but this is also followed by a widening of the peak itself. While initial attempts at calibration were conducted with a mixture containing six compounds (MeOH, EtOH, IPA, DMSD, TMS, and 2BN), it was found that the MeOH peak areas began to decrease at the highest concentrations tested. Based on earlier compound characterization testing at varying concentrations, this behavior was unexpected. However, previous calibration testing with air samples in our laboratory had shown that high concentrations of large molecules, such as siloxanes, could sterically hinder adsorption of smaller molecules on the AQM pre-concentrator.

In that particular case, the

siloxanes were removed from the calibration mixtures and tested separately, and calibration of the smaller molecules could then be successfully completed.7 With that in mind, new test mixtures were prepared for water testing that did not include DMSD, and the MeOH peak area with increasing concentration returned to its expected behavior. DMSD was then prepared as a separate calibration mixture (Supporting Information, S-3). 9 ACS Paragon Plus Environment

Page 11 of 26

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


For calibration, at least 3 replicates of each concentration were tested, and the average area for each compound and concentration calculated by the internal Draper Expert software. The data used for calibration curve construction are presented in the Supporting Information (Table S-4). Seven replicates of the lowest concentration standard were obtained for all compounds in order to better estimate limits of detection (LoD; Table S-5). The large LoD for EtOH is likely due to background EtOH in the laboratory atmosphere entering the AQM during sampling. After collection of the areas for the individual compounds, Curve Expert software (shareware v.1.4) was used to prepare calibration curves. For EtOH, DMSD, and TMS, the curves, while not perfectly linear, did not flatten out at the higher concentrations used, while those for MeOH, IPA, and MEK all showed saturation behavior. This flattening was expected, as saturation in the source region is a known complication for atmospheric pressure ion sources.29 In the current case, there is a finite reservoir of charge in the RIP from which the analytes can become ionized.

Comparison with Standard Laboratory Methods Following calibration of the ETV-AQM, archival samples returned from the ISS were analyzed in order to assess the accuracy of the instrument. The first sample to be tested was obtained from the U.S. segment potable water dispenser. This water had been processed through the WPA and would be expected to be free of most impurities. Table 3 shows the concentrations of the present subset of analytes measured in the Toxicology and Environmental Chemistry laboratories at NASA Johnson Space Center using GC-MS or LC-refractive index detection (LC-RID) along with the values obtained from the ETVAQM.18 Three aliquots from each sample were measured using the ETV-AQM; the average concentrations are listed in Table 3 for the compounds that were above their respective LoD. For EtOH, IPA, and 2BN, the results provided by the AQM were consistent with the lab techniques, while those for MeOH and TMS did not match. Note, however, that the average value for TMS is only slightly above its LoD, and that the value for methanol is less than 10% higher than the reporting limit of the GC-MS method. The DMSD concentration reported by the ETV-AQM for the same archival sample correlated 10 ACS Paragon Plus Environment


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

Page 12 of 26

very well with that from the lab techniques, with a percent error of ~ 15%. While this error may seem large, when compared to the initial required accuracy for the TOCA (25%)30 and for volatile organic analysis in air samples by the AQM (40%),3,6 the DMSD accuracy reported here is well within the acceptable range for in-flight monitoring hardware. Analysis of a second ISS archival sample, this one being a WPA product water sample that would also be expected to contain low levels of contaminants, showed similar ETV-AQM results (Table 3).18 Here, all of the compounds matched with results from lab techniques insofar as the LoD would allow. The ~ 29% error in the DMSD result is higher than preferable, but it is still within a range that would provide important information regarding increases in TOC levels seen by TOCA. In general, as the TOCA is focused on monitoring the product water of the U.S. segment and, as a compound-specific analyzer would be used to complement the TOCA when TOC increases are seen, the ability to analyze non-potable water samples (condensate or wastewater prior to processing in the WPA) is not necessary. However, these types of samples can provide important information regarding the ability of an analyzer such as the ETV-AQM to be used as a trending tool at high concentrations. Table S-4 in the Supporting Information presents the results of ETV-AQM analysis of 2 ISS humidity condensate samples. In the ISS Water Recovery System (WRS), this condensate would be combined with urine distillate from the Urine Processing Assembly (UPA) prior to feeding into the WPA. As such, these samples would be expected to have much higher concentrations of organics than the product or potable water. Results from lab techniques showed that all of the present subset of compounds are now detected and are above the minimum reporting limits for both samples. The accuracy for DMSD is very good, while the results for some of the other compounds are much less accurate. In particular, the concentrations of ethanol are so high in the archival samples that the calibration curve (which maxed out at 10 mg L-1) cannot even provide an estimated concentration. The low values of MeOH and IPA could be the result of preferential adsorption on the pre-concentrator by the high concentration of DMSD, whereby these compounds were not successfully trapped, as described earlier. The accuracy of TMS could likely be improved by using the monomer peak for detection, while the reason for the high 2BN results is unclear. 11 ACS Paragon Plus Environment

Page 13 of 26

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


As mentioned above, the majority of water quality monitoring on the ISS will continue to be performed by the TOCA. Only when TOC levels begin to rise in the product water would a secondary method for monitoring be desired. At that point, the major question to be answered would be, “Is this increase due to DMSD?” The results shown here indicate that an ETV-AQM combination could be used to answer that question. If testing with the ETV-AQM indicates that DMSD is not the cause of a TOC increase, then a question as to the exact nature of the contaminant would arise. In this case, the AQM could identify any typical target species and provide trending data on their concentrations. The AQM could also provide information (large/small, polar/non-polar, positive/negative mode) that could lead to at least a tentative identification of the class of compounds detected, if different from set targets. In essence, the exact concentration of a particular species is not as important as knowing which species are leading to a TOC increase, as the maximum TOC level for drinking water will be the same regardless of the species.

Conclusions In order to protect the health of crew members and systems during human spaceflight, it is important to monitor the air and water for potential trace contaminants. Using a ground-based model of a currently deployed air monitor, we successfully used ETV to promote trace volatile organic species in water into the gas phase for detection and quantification. This technique allowed for the identification of all of the target compounds in the tested water samples and indicated that the detected species were likely in the same form as those normally monitored in air. Multi-point calibration of the instrument and analysis of archival water samples from the ISS showed that the instrument was able to quantify dimethylsilanediol, a compound of particular interest, and was able to trend other compounds in the samples. While use of this electrothermal vaporization source for testing in microgravity will require additional design work that is currently underway in order to minimize the amount of crew member interaction required, it is obvious that electrothermal vaporization holds promise as a simple, laboratorybased method for analysis of volatile species in aqueous solutions with little sample preparation.



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

Page 14 of 26

Acknowledgements WTW, DBG, and TFL acknowledge funding under NASA contract # NAS 9-02078. The work of JMM was supported through the Arkansas Space Grant Consortium. WTW would like to thank Sarah Castro (NASA Johnson Space Center) for critical editing and Zachary Pickett for his work on our early DMSD studies. FMF thanks NASA for award number NNX13AF51G S02.

Supporting Information Discussion of the detection of dimethylsilanediol in the International Space Station (ISS) air (S-1), differences between air and water concentrations (S-2), and the effect of DMSD on the coadsorption of smaller molecules on the AQM preconcentrator (S-3).

Tables with Air Quality Monitor (AQM)

instrument settings (Table S-1, Table S-2, Table S-3), table showing data used to prepare calibration curves (Table S-4), table showing limits of detection of target compounds (Table S-5), table with results for electrothermal vaporization (ETV)-AQM testing of ISS condensate samples (Table S-6), figure showing timing sequence and current ramp of ETV-AQM (Figure S-1), and figure showing composite chromatograms of target compounds with increasing concentrations (Figure S-2).

This material is available free of charge via the Internet at http://pubs.acs.org.


Page 15 of 26

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


Figure Captions Figure 1: (A) Photograph of the ETV-AQM setup. (B) Close-up view of ETV coupling with important elements noted.

Figure 2: Differential mobility topographic charts obtained at a separation field of 87 Td. (A) Deionized water used to show the spectra obtained with no trace contaminants. (B) Spectra obtained with 25 ppm isopropanol added to the ETV ribbon. The z-scale for both figures is constant.

Figure 3: Effect of separation field on the ability of the AQM to identify species in mixtures. Mixture containing methanol, isopropanol, dimethylsilanediol, trimethylsilanol, and 2-butanone added to the ETV. (A) Positive mode spectrum at a separation field of 87 Td. Note that methanol is not well-separated from the RIP at this separation field. (B) Negative mode spectrum obtained simultaneously to (Fig. 3 A). (C) Spectrum arising from the same mixture used in (Fig. 3 A,B) but at a separation field of 115 Td. This separation field allows for the detection of methanol, but other compounds are lost due to the high field.

Figure 4: GC trace for a water sample showing detection of 6 compounds in a single analytical run. For detection in GC mode, unique separation (SV) and compensation voltages (CV) are set within each compound’s GC retention time to produce high selectivity and sensitivity. The retention times, SV, and CV for the individual compounds are noted below their peaks in the chromatogram.



Table 1: Retention times of target compounds and compensation fields required for their detection at various separation fields (1 Td = 10-17 V cm2).

Separation Field (SF) Compound (mode)

Retention Time (s)

87 Td

96 Td

106 Td

115 Td

Methanol (+)

60.7

-0.85

-1.12

-1.50

-1.92

Ethanol (+)

71.8

-0.32

-0.43

-0.54

ND

Isopropanol (+)

81.9

-0.11

-0.11

-0.11

ND

Acetone (+)

82.9

-0.05

-0.05

ND

ND

Dimethylsilanediol (-)

103.1

-0.16

-0.21

-0.21

-0.59

Trimethylsilanol (+)

117.3

0.11

0.16

0.16

0.27

2-Butanone (+)

135.5

0.05

0.05

0.11

ND


Compensation Field (CF/Td)

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

Page 16 of 26

Page 17 of 26

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


Table 2: Comparison of the compensation fields required for analysis of target compounds in air and water matrices. The compensation fields have been normalized to the regularly used separation fields to aid in comparison.

Compound

CF/SF in Air

CF/SF in Water

Methanol

-0.0171

-0.0167

Ethanol

-0.0037

-0.0037

Acetone

0.0000

-0.0006

Isopropanol

-0.0006

-0.0012

Trimethylsilanol

0.0023

0.0023

2-Butanone

0.0006

0.0006



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

Page 18 of 26

Table 3: Analysis of “clean” ISS archival samples by the ETV-AQM and a comparison with the results obtained using standard laboratory methods. The relative standard deviation (% RSD) for 3 replicates are listed in parentheses for the compounds with concentrations exceeding their limits of detection.

Potable Water Dispenser 7/30/2013 Compound Methanol

Product Water 8/19/2013

Units ETV-AQM GC-MS / LC-RID ETV-AQM µg L-1

GC-MS / LC-RID

219 (8.7)

< 200

189 (14.1)

< 200

< 478

< 200

< 478

< 200

ND

< 200

< 56

< 200

4390 (10.1)

3800

6832 (6.5)

5300

137 (86.6)

ND

< 130

ND

< 233

Electrothermal Vaporization Sample Introduction ... - ACS Publications

Recommend Documents