Carbon Fiber Ionization Mass Spectrometry for the Analysis of

Subscriber access provided by READING UNIV

Article

Carbon Fiber Ionization Mass Spectrometry for the Analysis of Analytes in Vapor, Liquid, and Solid Phases Min-Li Wu, Te-Yu Chen, Yen-Chun Chen, and Yu-Chie Chen Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b03736 • Publication Date (Web): 20 Nov 2017 Downloaded from http://pubs.acs.org on November 20, 2017

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 36 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

Carbon Fiber Ionization Mass Spectrometry for the Analysis of Analytes in Vapor, Liquid, and Solid Phases Min-Li Wu,˧ Te-Yu Chen,˧ Yen-Chun Chen, Yu-Chie Chen* Department of Applied Chemistry, National Chiao Tung University, Hsinchu 300, Taiwan

˧

These two authors contributed equally to this work.

*

Corresponding author E-mail: [email protected] Fax: +886-3-5723764 Phone: +886-3-5131527

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

Abstract Various ionization methods in mass spectrometry (MS) are available for the analysis of analytes with different properties. Nonetheless, using a single ionization method to analyze mixtures containing analytes with different polarities and volatiles in different phases at atmospheric pressure remains a challenge. Exploring an ionization method that can ionize small organics and large biomolecules with different properties for MS analysis is advantageous. Carbon fiber ionization mass spectrometry (CFI-MS) that uses a carbon fiber bundle as the ion source is useful for the analysis of small organics with low polarities. Voltage needs to be applied on the carbon fiber bundle to initiate corona discharge for ionization of analytes. In this study, we explore the suitability of using CFI-MS in the analysis of analytes in vapor, liquid, and solid phases by eliminating direct electric contact on the carbon fiber (~10 m). We demonstrate that CFI-MS is useful for analyzing not only small and low-polarity organics but also polar biomolecules, such as peptides and proteins. The limit of detections of analytes with high polarities such as dodecyl trimethyl ammonium bromide and bradykinin are estimated to be ~16 pM and ~53 pM, respectively. Ionization mechanisms, including corona discharge and electrospray, are involved in the ionization of analytes with low to high polarities. Furthermore, sesame oil containing aroma volatiles and compounds with different polarities is used as a model sample to demonstrate the capability of the developed ionization method to provide comprehensive chemical information from a complex sample. In addition, the feasibility of using the developed method for quantitative analysis of non-polar, medium and high polarity analytes is also demonstrated. The sensitivity of the developed method toward analytes with high polarity is higher than those with low polarity. The method precision was estimated to be ~7.8%. 2


Page 2 of 36

Page 3 of 36 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 Mass spectrometers are useful and sensitive analytical tools1 that can be used to obtain accurate molecular weights and structural information on analytes of interest. Only gas-phase ions can be detected by mass spectrometers. Therefore, ionization techniques that can convert samples from the condensed phase to the gas phase for ionization are important in mass spectrometry (MS) analysis. Electrospray ionization (ESI)2,3 is commonly used to ionize analytes with medium to high polarities. By applying a high voltage on the electrospray emitter, the sample solution eluted from the ESI emitter can be readily ionized through ESI processes. Moreover, atmospheric pressure chemical ionization (APCI)4–7 is suitable for ionizing analytes with low to medium polarities. In APCI, a high voltage is applied to an APCI metal needle that can initiate ionization through corona discharge.7 The selection of ionization techniques in MS mainly depends on the properties of analytes. A universal ionization technique that is suitable for analytes with low to high polarities, including vapor samples, small organics, and large biomolecules, remains unavailable. Although many ambient MS ionization techniques8–22 have been explored in the past two decades, ionization techniques that can serve as a universal ion source to analyze analytes possessing diverse polarities and operating in open-air conditions have not been established to date. Carbon materials are electrically conductive substrates. Studies have revealed the possibility of using carbon fibers23,24 and carbon tube impregnated papers25 as substrates to develop ionization methods. For example, Guo and co-workers24 proposed an ionization technique called carbon fiber ionization (CFI)-MS that can assist in the ionization of small organics and trace compounds from human breath through corona discharge; a carbon fiber bundle was used as the ion source. A high 3



voltage was applied directly to the carbon fiber bundle, which was placed close to the inlet of the mass spectrometer. A high voltage is required to initiate ionization in CFI-MS. This ionization method focuses on the analysis of small analytes with low-polarity organics. The use of CFI-MS in the analysis of large and highly polar biomolecules is not reported. We have previously presented two ESI derived ionization methods,26-32 which are suitable for the analysis of small organics and large biomolecules without the need for any direct electric-contact on the sample introduction site. It was discovered that the sample droplet deposited on a dielectric substrate can be directly used as the ESI emitter. That is, ESI can occur as long as the sample droplet is placed close to the inlet of the mass spectrometer, which is applied with a sufficiently high voltage. The apex of the sample droplet can form an apparent Taylor cone for generation of electrospray. Gas phase ions of analytes derived from the electrospray can be readily acquired by the mass spectrometer during this ionization process. On the basis of these ionization techniques, an ionization method derived from CFI-MS is explored in this study. We propose that ionization can readily occur when a carbon fiber deposited with samples is placed close to the inlet of the mass spectrometer that is applied with a high voltage. Any direct electric contact on the carbon fiber is not required. Presumably, APCI-like ionization, i.e. corona discharge,33 can be involved in the ionization of nonpolar and low-polarity analytes. Using this similar setup, electrospray derived from the sample droplet containing polar analytes placed on the carbon fiber can be generated for MS analysis. Accordingly, a universal ionization technique suitable for analytes with low to high polarities and with a wide mass range can be developed simply by using a carbon fiber as the ion source. Small organics with different polarities and large polar biomolecules were used as model 4


Page 4 of 36

Page 5 of 36 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


samples to demonstrate the usefulness of this universal ionization technique.

5



Experimental The details of chemicals and instrumentation used in this study are described in Supporting Information. Preparation of carbon fibers Carbon fibers derived from pencil leads and carbon tubes were used to assist ionization in CFI-MS. Pencil leads were sharpened to have tips with different diameters. The smallest diameter of the pencil lead that can be shaped by a sharpener was ~50 μm. The sharpened pencil lead was rinsed with methanol and deionized water before use. To have carbon fibers with the diameter < 50 μm, we isolated thin carbon fibers from a carbon tube (Figure S1A). That is, we soaked a carbon tube in the solvent (10 mL) containing water and acetone (1:1, v/v) under sonication for 30 min. Thin carbon fibers from the carbon tube could be dissociated and isolated (Figure S1B). Tweezers were used to separate thin carbon fibers from the resultant soaking carbon tube. The isolated carbon fibers were then rinsed with methanol and deionized water. The tip diameter of the resultant carbon fibers was 7-10 μm. Setup of CFI-MS Scheme 1 shows the setup of the CFI-MS. The fiber was placed either horizontally (Scheme 1A) or vertically (Scheme 1B) to the inlet of the mass spectrometer by wooden tweezers. Although mass spectra can be readily obtained by placing the carbon fiber in either ways, the background ions derived from the carbon fiber placed vertically were fewer than that placed horizontally to the inlet of the mass spectrometer. The distance between the tip and the inlet of the mass spectrometer was ~6 mm. There was no direct electric contact made on the carbon fiber when conducting MS analysis. A sample droplet was deposited on the tip of the carbon fiber. When conducting MS analysis of solid samples, ~2 L of sample solution was loaded 6


Page 6 of 36

Page 7 of 36 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


on the thin carbon fiber (~10 m) first. After solvent evaporation, the fiber loaded with the dried sample was placed close to the MS inlet. After the mass spectrometer was switched-on, mass spectra were immediately acquired by the mass spectrometer. If liquid samples were analyzed directly, CFI-MS mass spectra were immediately acquired after the carbon fiber was deposited with the sample droplet (2-5 L). Scheme 1C shows the typical setup of CFI-MS for analysis of vapor samples. A carbon fiber (length: ~ 1 cm; tip diameter: ~ 10 m) was placed close (~ 1 mm) to the inlet of the mass spectrometer. The samples in either solid or liquid forms containing compounds with high volatilities were placed under the tip of the carbon fiber. Once the mass spectrometer was switched on, mass spectra were immediately recorded by the mass spectrometer. Additional experimental details are described in Supporting Information.

7



Results and Discussion Using benzo[a]pyrene as the model sample Benzo[a]pyrene (MW= 252 Da), a non-polar compound, was initially used as a model sample. After drying the sample (~2 L, 10-6 M) on the carbon fiber, the carbon fiber was placed close (~6 mm) to the inlet of the mass spectrometer (Scheme 1A). Mass spectra were readily acquired after switching on the mass spectrometer. Figure 1A shows the resultant mass spectrum of benzo[a]pyrene, and the inset displays the microscopic image of the carbon fiber. The diameter of the carbon fiber was ~10 m. The ion peak at m/z 252 contributed by benzo[a]pyrene ([M+.]) dominated the mass spectrum. The signal-to-noise (S/N) ratio was as high as ~1663. The result indicated that CFI-MS can be readily conducted without applying any direct electric contact on the carbon fiber. The electric voltage applied on the inlet of the mass spectrometer was sufficient to facilitate ionization of analytes from the carbon fiber. The ionization mechanism was presumed to be similar to that observed in corona discharge.4–7 The details of the putative ionization mechanism will be discussed later. In addition, we also observed that the ion intensity of benzo[a]pyrene was decreased as the diameter of the carbon fiber was increased (Figures 1B-1D). There were no analyte ions observed in the mass spectrum as the diameter of the carbon fiber was increased to ~500 m (Figure 1E). It is not surprising because corona discharge can be easily induced by a thinner carbon fiber than a thicker one under the same experimental condition. Optimization of experimental parameters In addition, the effects of the distance between the carbon fiber and the inlet of the mass spectrometer and the voltage applied on the inlet of the mass spectrometer were also investigated. Benzo[a]pyrene (m/z 252) was used as the model sample to 8


Page 8 of 36

Page 9 of 36 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


examine these effects. If the distance was too short (< 6mm) (Figure S2A), the background ions became quite apparent and suppressed the ion peak derived from analytes. However, if the distance was farer than 8 mm, no ion signals derived from benzo[a]pyrene at m/z 252 can be observed (Figure S2D). Thus, the optimized distance between the carbon fiber and the inlet of the mass spectrometer was 6-8 mm. In addition, the higher the voltage applied on the inlet of the mass spectrometer, the higher the ion intensity was obtained (Figure S3). The voltage should not be lowered than 3000 V. Otherwise, ion peaks derived from analytes were barely observed in the resultant mass spectra. More discussion is provided in Supporting Information. Analysis of the samples with different polarities We examined whether the proposed ionization method is suitable for analytes with different polarities. This approach is also suitable for the analysis of medium-polarity analytes containing functional groups, such as carboxylate and amino groups. Figures 2A and 2B show the mass spectra obtained from dodecanoic acid and hexadecylamine, respectively. Figure 2A is dominated by the peak at m/z 199 derived from deprotonated dodecanoic acid ([M-H]-= 199). Namely, our approach can also be operated in negative ion mode. Figure 2B is dominated by the peak at m/z 242 contributed by protonated hexadecylamine. Although Figures 2A and 2B were obtained from the samples dried on the carbon fiber, the mass spectra of dodecanoic acid and hexadecylamine can also be acquired by directly depositing the sample droplet on the carbon fiber for ESI analysis. Similar to ESI processes,29 ions obtained from liquid samples were derived from electrospray when the droplet was placed close to the inlet of the mass spectrometer. Table S1 in Supporting Information shows the summarized results of a number of compounds that we have examined, and Figures S4–S28 in Supporting Information display their corresponding mass spectra. 9



Some analytes can be analyzed in solid and liquid phases, whereas other analytes may only be analyzed in one phase. The polarity, vapor pressure, and molecular weights of analytes (Table S1) were involved in determining detectable forms using the current ionization technique. Figures 2C, 2D, and 2E show the mass spectra of 2-phenylethylamine ([M+H]+= 122), benzylmalonic acid ([M-H]-= 193), and histamine ([M+H]+= 112), respectively. In addition, the fragment ions at m/z 105, 149, and 95 derived from 2-phenylethylamine, benzylmalonic acid, and histamine by losing a NH3, a COO, and a NH3, respectively, were also observed in the corresponding mass spectra. Moreover, we also discovered that analytes with a high vapor pressure can be directly analyzed by placing the samples under the carbon fiber, which is placed close to the inlet of the mass spectrometer. Figures 2F, 2G, and 2H show the mass spectra of atrazine ([M+H]+= 216), acenaphthene ([M+.= 154], and methyl cinnamate ([M+H]+= 163), respectively, which were prepared in different solvents (1 mL) and placed underneath the carbon fiber for CFI-MS analysis. The ion peaks derived from atrazine prepared in solid (Figure S12B) and liquid (Figure S13B) forms can also be observed in the resultant CFI mass spectra (Table S1). However, analytes that possess high vapor pressures, such as acenaphthene only appeared in the CFI mass spectra when they were placed under the carbon fiber (Table S1). No ion peaks were observed in the mass spectra when these samples were directly deposited on the carbon fiber. Presumably, analytes with high vapor pressures evaporate quickly. Therefore, analyte vapor is immediately diluted by air after being deposited on the carbon fiber. The amounts of samples surrounding the carbon fiber were insufficient for obtaining mass spectra. Alternately, the vapor that continuously evaporated from the sample under the carbon fiber continuously provided a sufficient amount of analyte vapor for CFI-MS analysis. These results indicated that our CFI-MS approach 10


Page 10 of 36

Page 11 of 36 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


can be used to analyze analytes with different polarities. Generally, mass spectra can be obtained from non-polar analytes that are dried on a carbon fiber. Analytes with low to medium polarities may be analyzed with the current approach if they are prepared in solid and liquid phases. If analytes with medium polarity also possess certain vapor pressures, mass spectra can be obtained directly from their vapors. Analysis of large and polar biomolecules We also demonstrated that our approach can be used to analyze large and polar biomolecules. We deposited a sample droplet (~2 μL) on a carbon fiber placed close to the inlet of the mass spectrometer. Mass spectra were immediately recorded after switching on the mass spectrometer. Figures 3A and 3B show the resultant mass spectra of bradykinin and cytochrome c, respectively, using our CFI-MS approach. Multiply charged ions derived from bradykinin and cytochrome c dominated the mass spectra, which are similar to those obtained from conventional ESI-MS. Figure 3C shows the photograph of the setup. For easy deposition of sample droplets, the carbon fiber was placed horizontally to the inlet of the mass spectrometer. These results demonstrated that our CFI-MS approach is suitable for the analysis of small organics as well as large and polar biomolecules. Analysis of volatiles from real samples Moreover, this current ionization approach can quickly detect high volatiles from real samples by simply placing the samples under the carbon fiber, which was placed close to the inlet of the mass spectrometer. The main aroma/odor compositions from such samples dominated the CFI mass spectra. The details of the results obtained from the samples including mothballs (Figure S29A), menthae piperitae oil (Figure S29B), and spoiled canned sardines (Figure S29C) were discussed in Supporting Information. For instance, mothballs are mainly composed of naphthalene. When we placed a 11



mothball under the carbon fiber, which was put close to the front of the inlet of the mass spectrometer (inset photograph in Figure S29A), the ion peak at m/z 128 derived from the cation radical of naphthalene immediately appeared and dominated the mass spectrum (Figure S29A). The results indicated that our approach can be readily used to obtain the mass spectra derived from the compounds that result in the unique aroma/odor of real samples without conducting additional sample pretreatment. Analysis of a mixture containing analytes with low to high polarities As demonstrated above, the developed CFI-MS is suitable for the analysis of analytes with low to high polarities. To demonstrate the feasibility of the current approach in ionizing a heterogeneous mixture containing analytes with different polarities, a sample containing benzo[a]pyrene, ametryn, and arginine possessing polarities from low to high was used as the model sample. A sample droplet containing the mixture was deposited on the tip of carbon fiber that was placed close to the inlet of the mass spectrometer. After the sample was dried on the carbon fiber, the mass spectrometer was switched on, and mass spectra were immediately recorded. Figure 4A shows the resultant EICs at m/z 252 (blue curve), 228 (red curve), and 175 (black curve) representing the cation radical of benzo[a]pyrene, protonated ametryn, and protonated arginine, respectively. Zero intensity was observed at the EIC of m/z 175 at the first 0.6 min. Figure 4B shows the resultant mass spectra obtained from the time points between 0 and 0.6 min. The ion peaks at m/z 252 and 228 dominated the mass spectra. A methanol droplet (10 μL) was then deposited from the top of the carbon fiber, allowing the solvent to flow down to the tip end of the carbon fiber at the time point of 0.6 min. As the solvent flowed through the carbon fiber, the polar analytes were immediately dissolved into the solvent droplet, and ionization occurred 12


Page 12 of 36

Page 13 of 36 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


immediately through ESI processes. Figure 4C shows the mass spectra acquired between the time points at 0.6 and 0.8 min. The peaks at m/z 175 and 228 derived from arginine and ametryn, respectively, dominated the mass spectrum. The ion peak at m/z 252, contributed by benzo[a]pyrene, disappeared. This disappearance is understandable because arginine was ionized through the ESI process, such that methanol was able to dissolve arginine from the carbon fiber for generation of electrospray. However, benzo[a]pyrene was ionized through corona discharge. The ionization of benzo[a]pyrene was inefficient when the solvent was deposited on the carbon fiber. Ametryn possesses medium polarity, so it can be ionized either in solid or liquid phase. These results demonstrated that CFI-MS can ionize analytes with different polarities from the same sample and is a universal ion source for analytes with different polarities. Examination of limit of detection The limit of detection (LOD) of analytes with different polarities was examined. Figure S30A shows the mass spectrum of the sample containing benzo[a]pyrene (m/z 252, 10-8 M), in which the sample (2 L) was dried on the carbon first before placing the fiber in front of the mass spectrometer for MS analysis. Figures S30B, S30C, S30D, S30E, and S30F show the mass spectra of the samples including dodecyl trimethyl ammonium bromide (M+ (m/z 228), 10-8 M), prometon ([M+H]+ (m/z 226), 10-8 M), bradykinin ([M+2H]2+ (m/z 531), 10-9 M), and insulin ([M+5H]5+ (m/z 1163), 10-9 M), in which the samples were detected right after deposition of the sample droplet (2 L) on the carbon fiber. Figure S30F shows the mass spectrum of the vapor derived from methyl cinnamate ([M+H]+ (m/z 163), 10-4 M, 2 mL). The setup was the same as shown in Scheme 1C. The signal to noise (S/N) ratios for the ion peaks derived from the molecular/pseudomolecular ions of the analytes were shown 13



in the mass spectra. Accordingly, the LODs of benzo[a]pyrene, dodecyl trimethyl ammonium bromide, prometon, bradykinin, insulin, and the vapor derived from methyl cinnamate were estimated to be ~7 nM, ~16 pM, ~210 pM, ~53 pM, ~110 pM, and ~1.7 M, respectively, on the basis of the S/N ratio equal to 3,. The LOD of methyl cinnamate was estimated to be ~1.7 M (Figure S30F), which is much higher than other analytes. It is because that only vapor evaporated from the liquid methyl cinnamate sample was detected. Quantitative analysis In addition, the possibility of using the current ionization method for quantitative analysis was also examined. The samples containing benzo[a]pyrene (M+.= 252) at different concentrations (2-800 M) spiked with prometon (5 M) ([M+H]+= 226) as the internal standard. The mass spectra were obtained after the samples deposited on the carbon fiber were dried. Figure S31A shows the resultant plot by plotting the relative intensity of benzo[a]pyrene to prometon (I252/I226) versus the concentration of benzo[a]pyrene. Three replicates were conducted for each data point. Apparently, the dynamic linear range was found between 2 and 100 M (inset in Figure S31A, Y= 0.176X0.017, R2= 0.997), indicating the feasibility of using the current ionization method in quantitative analysis. Figure S31B shows the corresponding CFI mass spectra obtained from the linear dynamic range (2-100 M). These results demonstrated the potential of using the current approach for quantitative analysis. Benzo[a]pyrene is a non-polar compound. To investigate the figures of merit of our method towards analytes with different polarities, we further conducted quantitative analysis by using prometon and octyl trimethylammonium bromide, which possess medium and high polarity, respectively, as the model samples. Liquid samples were used in quantitative analysis. The results showed that the linear dynamic ranges of 14


Page 14 of 36

Page 15 of 36 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


prometon and octyl trimethylammonium bromide were 0.02-1 M (inset in Figure S32A) and 0.03-0.5 M (inset in Figure S33A), respectively. Figures S32B and S33B

show

the

corresponding

mass

spectra

of

prometon

and

octyl

trimethylammonium bromide, respectively, obtained in the linear dynamic range. The slopes of the calibration curves obtained from the analytes of prometon (Figure S32A, Y= 2.302X 0.004, R2= 0.998) and octyl trimethylammonium bromide (Figure S33A, Y= 2.591X0.003, R2= 0.991) were 2.302 and 2.591, resepctively, which were much higher than that obtained from non-polar analytes, e.g. benzo[a]pyrene (slope= 0.176). That is, the sensitivity of the current method towards analytes with higher polarity is higher than towards non-polar anayltes. Evaluatoin of method precision In addition, we also investigated the precision of the current method. Two persons used CFI-MS to analyze the same sample, i.e. benzo[a]pyrene (7.5 M) spiked with prometon (5 M), for multiple times within five days. Table S2 shows the summarized results. After 60 runs, the precision was estimated to be ~7.8% in terms of relative standard deviation%, indicating the developed method has an acceptable precison. Analysis of complex real samples To further demonstrate that CFI-MS can be used to obtain mass spectra from complex samples containing compounds with different polarities, sesame oil was selected as the sample. Sesame oil, which can enhance the flavor of food, is frequently used in Asian cuisine.34 The main compositions in sesame oil are aroma volatiles, fatty acids, and triglycerides.35,36 Highly volatile aroma compounds in sesame oil are usually detected by electron ionization MS,35 whereas compounds with low to medium polarities, such as triglycerides and fatty acids, are commonly detected 15



by APCI-MS36 and ESI-MS,37 respectively. Figures S34A and S34B show the mass spectra of the sample obtained by placing the sesame oil sample under the carbon fiber observed in the low- (m/z 100–500) and high-mass regions (m/z 500–1000), respectively. The setup is similar to that shown in Scheme 1C. Figures S34C and S34D show the mass spectra of the same sesame oil (1%) prepared in water and acetone (2:1, v/v) obtained by depositing the sample droplet (~2 L) on the carbon fiber in the low- (m/z 100–500) and high-mass regions (m/z 500–1000), respectively. When the sample was still in liquid phase, the carbon fiber deposited with the sample was immediately placed in front of the mass spectrometer for MS analysis. The mass spectra obtained from the sample in vapor (Figure S34A) and liquid phases (Figure S34C) look similar. No unique ions appeared in the high-mass region (Figures S34B and S34D). The peaks at m/z 109 and 123 were presumably contributed by the aroma compounds pyrazines and pyridines,35 respectively. The possible structures are shown in Scheme S1 according to a previous report.35 These aroma compounds were observed in the CFI mass spectra in their vapor and liquid phases because of their medium polarities and volatilities. In addition, when analyzing the liquid sesame oil in the negative ion mode, the ions derived from fatty acids were observed in the mass spectra (insets in Figures S34C). The ion peaks at m/z 255, 279, 281, and 283 were derived from palmitic acid (P), linoleic acid (L), oleic acid (O), and stearic acid (S),36 respectively. Figures S34E and S34F show the mass spectra of the same sesame oil sample (1%) obtained by drying 2 L of the same oil sample on the carbon fiber. After solvent evaporation, the carbon fiber was placed in front of the mass spectrometer, and the mass spectrometer was switched on to acquire mass spectra. No ions were observed in the low-mass region (Figure S34E). Nevertheless, fatty acids derived from P, L, O, and S appeared at m/z 255, 279, 281, and 283 appeared in the 16


Page 16 of 36

Page 17 of 36 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


negative ion mode mass spectrum (inset in Figure 34E). These results are similar to those discussed in Table S1. That is, compounds with medium polarities and certain volatilities can be detected by CFI-MS from their liquid and solid phases. Moreover, four apparent ion groups appearing at m/z ~616, m/z ~642, m/z ~880, and ~904 dominated the mass spectrum in the high-mass region (Figure S34F). Figure S35 shows the zoom-in mass spectrum. These peaks were derived from the intact triglycerides (Figure S35A) and their fragments by losing one fatty acid from the intact triglycerides (Figure S35B).36 For example, the peaks at m/z 876, 878, 880, and 882 corresponded to protonated LLnLn, LLLn, LLL, and LLO (Figure S35A), respectively. Figure S35C shows the corresponding chemical compositions of the triglycerides. We only observed the ions derived from fatty acids appearing in the CFI mass spectrum of sesame oil operated at the negative ion mode, while triglycerides derived from sesame oil were only observed in the CFI mass spectrum operated at the positive mode. For comparison, we also used conventional ESI-MS and electron ionization (EI)-MS to analyze the same sesame oil. When using conventional ESI-MS to analyze the sesame oil sample, only the ions derived from fatty acids at m/z 255, 279, 281, and 283 corresponding to deprotonated P, L, O, and S, respectively, were observed in the mass spectrum (Figure S36A). No ions derived from triglycerides were observed. On the other hand, more ions were observed in the EI mass spectrum of the same sesame oil sample, but the mass spectrum was dominated by the fragments derived from fatty acids and triglycerides (Figure S37). These results showed that our current method can provide more comprehensive chemical information than other conventional MS methods when using the complex sesame oil as the model sample. Namely, our CFI-MS can be potentially used as a universal MS method for characterization of 17



Page 18 of 36

complex samples containing compounds with different polarities. Ionization mechanism According to Table S1, the cation radicals of PAHs dominate the CFI mass spectra because PAH cationic radicals are stable. Presumably, corona discharge occurs around the tip of the carbon fiber when the fiber is undergone a sufficiently high electric field provided by the inlet of the mass spectrometer. That is, analytes (A) on the surface of the carbon fiber was evaporated to the air and simultaneously ionized through corona discharge. In addition, our results (Table S1) show that protonated analyte ions derived from the analytes with certain polarities were common in the CFI mass spectra. The ionization mechanism of the samples dried on the carbon fiber is also assumed to be similar to corona discharge.4–7,33 The moisture (trace water vapor) in atmospheric pressure around the carbon fiber was charged through corona discharge followed by formation of protons (H3O+) (reaction 1). The generated protons in the gas phase interact with analytes (A) derived from the carbon fiber in the air to generate protonated analyte ions ([A+H]+) (reaction 2). The moisture in our laboratory was ~50%. Thus, although no additional solvent was supplied to provide proton sources in our approach, trace water in the air provides a sufficient proton source for analytes during ionization processes.

H2O+(g)+ H2O(g) H3O+(g)+ HO(g) H3O+(g) + A(g) [A+H]+(g) + H2O(g)

(1) (2)

To further validate this assumption, we placed a bottle containing boiling heavy water (D2O) underneath the carbon fiber when conducting CFI-MS analysis. A droplet (2 μL) of 2,2-bipyridine (10−6 M) prepared in ethanol was deposited on the carbon 18


Page 19 of 36 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


fiber. After drying the sample, the carbon fiber was placed close to the inlet of the mass spectrometer for CFI-MS analysis. Figure S38A shows the resultant mass spectrum, which was dominated by the protonated 2,2-bipyridine molecular ion peak at m/z 157. However, when a bottle containing boiling heavy water was placed underneath the carbon fiber, which had deposited with the same sample as used to obtain Figure S38B, the ion peak at m/z 158 dominated the resultant mass spectrum. These results indicated that 2,2-bipyridine was ionized by trapping a D+; therefore, its molecular ion was shifted to 158. These results confirmed that the proton source utilized to ionize compounds with low to medium polarities from the dried sample on the carbon fiber was mainly contributed by trace water vapor in the air. Moreover, the setup used to obtain Figured S38A and S38B as shown in Figure S38C can be used to identify whether the molecular ions are derived from the cation radicals or protonated molecular ions of analytes. According to Table S1, the cation radicals derived from non-polar PAH samples such as pyrene and benzo[a]pyrene with low vapor pressure (< 4.50  10-6 mmHg), can only be observed in the CFI mass spectra from their solid phase deposited on the carbon fiber. However, if PAH samples possess high vapor pressure (≥ 2.1510-3 mmHg), e.g. naphthalene (Figure S17A), azulene (Figure S17B), and acenaphthene (Figure S17C), their cation radicals derived from their vapor dominated the resultant CFI mass spectra. No ions were observed when the samples are deposited on the carbon fiber either in solid and liquid samples. It is understandable because only ~2 L of sample was deposited on the carbon fiber when conducting the CFI-MS analysis from their solid and liquid samples, the analytes quickly evaporated and diluted in the air owing to their high vapor pressure. Thus, no sufficient ions derived from only a small volume of samples (~2 L) can be acquired by the mass 19



spectrometer. On the other hand, when conducting the CFI-MS analysis of the vapor evaporated from a relatively large volume of liquid samples containing such analytes, the vapor can be continuously supplied for CFI-MS analysis. In addition, protonated long-chain alkyl amines (CnH2n+1NH2, n= 12, 14, and 16), which possess certain vapor pressure (10-3-9.7510-1 mmHg) (Table S1) and polar amino groups, can be observed in the CFI mass spectra from their solid (Figures S6A-6C), liquid (Figures S7A-7C), and vapor samples (Figures S8A-8C). However, as the carbon chain length is increased to 18, e.g. stearylamine, the protonated molecular ions derived from stearylamine could only be observed in the CFI mass spectra of the solid (Figure S6D) and liquid (Figure S7D) samples. No molecular ions derived from the stearylamine vapor were observed in the CFI mass spectrum. It is because that the vapor pressure -5

of stearylamine is reduced to 4.4010 mmHg; thus, no sufficient amounts of the analytes can be evaporated to the air for CFI-MS analysis. Analytes such as gallic acid (Figure S27E) have high polarity and relatively low vapor pressure (7.3210-11 mmHg), they can only be ionized from their liquid phase. Given that such analytes possess polar functional groups, their protonated molecular ions usually dominate the CFI mass spectra. On the basis of Table S1, we can conclude if the analytes with polarity and vapor pressure to a certain extent, they can be ionized from their solid, liquid, and vapor phases. Protonated molecular ions derived from analytes containing polar functional groups usually dominate the CFI mass spectra. If analytes are relatively non-polar with the chemical structures like non-polar PAHs, their cation radicals dominate the CFI mass spectra. Furthermore, if analytes possess low vapor pressure, the analytes may be ionized from their solid and liquid phases. Whether the resultant molecular ions of analytes derived from their cation radicals or protonated molecular ions are dependent on their chemical structures. The polarity and vapor 20


Page 20 of 36

Page 21 of 36 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


pressure of analytes determinate what phases of the samples should be prepared for CFI-MS analysis.

21



Conclusions We successfully demonstrated a facile and straightforward ionization approach that can be used to analyze analytes with low to high polarities. By simply placing a carbon fiber in front of the inlet of the mass spectrometer, analytes deposited on the carbon fiber can be readily ionized. Direct electric contact on the carbon fiber is not required. Thus, the ionization method can be easily established in any MS laboratories. Nevertheless, commercial mass spectrometers may have different voltage settings on the inlet and the emitter of the mass spectrometer from those we used in this study. That is, it is worth mentioning that the same mass spectra must be obtained by the application of high voltage to the emitter with the inlet at the ground potential when employing the current approach. The key component in this approach is the thin carbon fiber, which is relatively inexpensive. A carbon tube (inner diameter: 1 mm; outer diameter: 2 mm; length: 1 m) used to generate carbon fiber only costs USD 1 per meter. One carbon tube can make million carbon fibers (~10 m) in terms of calculation of the surface area of the cross section. That is, the cost of one piece of carbon fiber is extremely low. In addition, we discussed the putative ionization mechanisms for analytes with different polarities. Various mechanisms, including corona discharge and electrospray, are involved in this universal ionization method for non-polar, low-polarity, and polar analytes. Although no additional electric contact is applied on the carbon fiber, the carbon fiber functions as a suitable ion source for analytes with different polarities. Polar analytes can be ionized by depositing a sample droplet on the carbon fiber through ESI processes. The developed ionization technique is also suitable for the analysis of large molecules, such as peptides and proteins. In addition, the CFI mass spectra usually have low background. The LOD of benzo[a]pyrene, which is a non-polar compound, is as low as ~7 nM (2 L) (i.e. ~14 22


Page 22 of 36

Page 23 of 36 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


fmol), while the LOD of prometon with medium polarity is estimated to be ~0.21 nM. The LODs of analytes with high polarities such as dodecyl trimethyl ammonium bromide and bradykinin are lowered to ~16 pM and ~53 pM, respectively. Our results also show that the sensitivity of this developed MS method toward analytes with higher polarity is higher than those with lower polarity. In addition, we have demonstrated the feasibility of using the developed method for quantitative analysis of analytes with different polarity. Moreover, the method precision is evaluated to be as low as ~7.8%. To the best of our knowledge, the developed setup is one of the simplest ion sources ever reported and is suitable for the analysis of diverse analytes. Moreover, depending on the analytes of interest in a complex sample, the sample can be selectively analyzed in vapor, liquid, or solid phase using the CFI setup. Comprehensive chemical information can be readily obtained from the sample without conducting tedious sample pretreatment. Namely, the developed CFI-MS can be practically used as a universal ion source for analytes having different polarities with a wide mass range. Thus, we believe that the developed MS method should have potential to be further popularized.

23



Acknowledgement We thank the Ministry of Science and Technology of Taiwan (MOST 105-2113-M-009-022-MY3) for financial support of this research.

24


Page 24 of 36

Page 25 of 36 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


Supporting Information Additional experimental details (materials and reagents, instrumentation, optimization of experimental conditions, analysis of analytes with low to medium polarities in solid phase, analysis of large biomolecules in liquid phase, analysis of a mixture containing analytes with low to high polarities, and study of ionization mechanism) and additional results (optimization of experimental parameters, analysis of real samples from their vapor phase, Tables S1 & S2, and Figures S1-S38).

25



References (1) Kandiah, M.; Urban. P. L. Chem. Soc. Rev. 2013, 42, 5299-5322. (2) Yamashita, M.; Fenn. J. B. J. Phys. Chem. 1984, 88, 4451-4459. (3) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science, 1989, 246, 64-71. (4) Horning, E.; Horning, M.; Carroll, D.; Dzidic, I.; Stillwell, R. Anal. Chem. 1973, 45, 936-943. (5) Horning, E. C., Carroll, D.I., Dzidic, I., Haegele, K.D., Horning, M.G., Stillwell, R.N. J. Chromatogr. Sci. 1974, 12, 725-729. (6) Carroll, D.I.; Dzidic, I.; Stillwell, R.N.; Horning, M.G.; Horning, E.C. Anal. Chem. 1974, 46, 706-710. (7) Carroll, D. I., Dzidic, I., Stillwell, R. N., Haegele, K. D., Horning, E. C. Anal. Chem. 1975, 47, 2369-2373. (8) Cooks, R. G.; Ouyang, Z.; Takats, Z.; Wiseman, J. M. Science. 2006, 311, 1566-1570. (9) Lebedev, A. T. Russ. Chem. Rev. 2015, 84, 665-692. (10) Urban, P. L.; Chen, Y.-C.; Wang, Y.-S. Time-Resolved Mass Spectrometry: From Concept to Applications. Chapter 2, 2016, Wiley (ISBN: 978-1-118-88732-5) (11) . Cody, R. B; Laramée, J. A.; Durs,t H. D. Anal. Chem. 2005, 77, 2297–2302. (12) Haddad, R.; Milagre, H. M.; Catharino, R. R.; Eberlin, M. N. Anal Chem. 2008, 80, 2744-50. (13) Shiea, J;. Huang, M.-Z.; HSu, H.-J.; Lee, C.-Y.; Yuan, C.-H.; Beech, I.; Sunner, J. Rapid Commun. Mass Spectrom. 2005, 19, 3701–3704. (14) Robb, D.; Covey, T.; Bruins, A. Anal. Chem. 2000, 72, 3653–3659. (15) Kauppila, T. J.; Kuuranne, T.; Meurer, E. C.; Eberlin, M. N.; Kotiaho, T.; 26


Page 26 of 36

Page 27 of 36 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


Kostiainen, R. Anal. Chem. 2002, 74, 5470–5479. (16) Cooks, R. G.; Ouyang, Z.; Takats, Z.; Wiseman, J. M. Science. 2006, 311, 1566-70. (17) Meher, A. K.; Chen, Y.-C. 2017, 6, S0057-S0057. (18) Chen, H. W.; Venter, A.; Cooks, R. G. Chem. Commun. 2006, 19, 2042-2044. (19) Zhu, L.; Gamez, G.; Chen, H. W.; Chingin, K.; Zenobi, R. Chem. Commun. 2009, 5, 559-561. (20) Chang, C.-H.; Urban, P. L. Analytical Chemistry. 2016, 88, 8735-8740. (21) Hiraoka, K. J. Mass Spectrom. 2004, 39, 341-350. (22) Trimpin, S.; Inutan, E. D.; Herath, T. N.; McEwen, C. N. Mol. Cell. Proteomics. 2009, 9, 362–367. (23) Liu, J.; Ro, K.W.; Busman, M.; Knapp, D. R. Anal. Chem. 2004, 76, 3599-3606 (24) Wu, M.-X.; Wang, H.-Y.; Zhang, J.-T.; Guo, Y.-L. Anal. Chem. 2016, 88, 9547-9553. (25) Narayanan, R.; Sarkar, D.; Cooks, R.G.; Pradeep T. Angew. Chem. Int. Ed. 2014, 53, 5936-5940. (26) Hsieh, C.-H.; Chang, C;-H.; Urban, P. L.; Chen, Y.-C. Anal. Chem. 2011, 83, 2866-2869. (27) Hsieh, C.-H.; Chao, Chin-Sheng; Mong, K.-K. T.; CHen, Y.-C. J. Mass Spectrom. 2012, 47, 586-590. (28) Hsieh, C.-H.; Meher, A. K.; Chen. Y.-C. PLoS One, 2013, 8, e66292. (29) Meher, A. K.; Chen, Y. C. J. Mass Spectrom. 2015, 50, 444-450. (30) Meher, A. K.; Chen, Y.-C. RSC Adv. 2015, 5, 94315-94320. (31) Meher, A. K.; Chen, Y.-C. Anal. Chim. Acta. 2016, 937, 106-112. (32) Meher, A. K.; Chen, Y.-C. Anal. Chem. 2016, 88, 9151-9157. 27



(33) Horning, E.; Carroll, D.; Dzidic, I.; Haegele, K.; Horning, M.; Stillwell, R. A. J. Chromatogr. Sci. 1974, 12, 725-729. (34) Bedigian, D. Econ. Bot. 2004, 58, 3, 329-353. (35) Shimoda, M.; Shiratsuchi, H.; Nakada, Y.; Wu, Y.; Osajima, Y. J. Agric. Food Chem. 1996, 44, 3909-3912 (36) Jakab, A.; Nagy, K.; Héberger, K.; Vékey, K.; Forgács,E. Rapid Commun. Mass Spectrom. 2002, 16, 2291-2297. (37) Kurata, S.; Yamaguchi, K.; Nagai, M. Anal. Sci. 2005, 21, 1457-1465.

28


Page 28 of 36

Page 29 of 36 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


Scheme 1. CFI-MS setup by placing the carbon fiber (A) horizontally and (B) vertically to the inlet of the mass spectrometer. (C) The setup for vapor samples.

29



Figure Legends Figure 1. Effects of the tip diameter of the carbon fiber in CFI-MS analysis. Mass spectra obtained by depositing a sample droplet (2 μL) containing benzo[a]pyrene on the carbon fiber with the tip diameters of (A) ~ 8 μm , (B) ~ 54 μm, (C) 270 μm, (D) ~ 373 μm, and (E) ~538 μm followed by CFI-MS analysis. All the mass spectra were obtained by placing the carbon fibers in front of the mass spectrometer after the deposited samples were dried. The concentration of benzo[a]pyrene on the carbon fiber to obtain Panel (A) was 10-6 M, while the concentration of benzo[y]pyrene on the carbon fibers to obtain Panels (B)-(E) was 10-4 M. The inset in each Panel shows the microscopic image of the carbon fiber that was used to obtain the corresponding mass spectrum. Figure 2. Analysis of solid, liquid, and vapor samples by CFI-MS. CFI mass spectra of the solid samples including (A) dodecanoic acid ([M-H]-= 199) (10-3 M), and (B) hexadecylamine ([M+H]+= 242) (10-4 M) dried on the carbon fiber. CFI mass spectra obtained from the liquid samples (~5 μL) including (C) 2-phenylethylamine ([M+H]+= 122) (10-5 M), (D) benzylmalonic acid ([M-H]-= 193), and (E) histamine ([M+H]+= 112). After the sample droplet (2 μL) deposited on the carbon fiber, CFI-MS analysis was conducted immediately. CFI mass spectra of vapor samples including (F) atrazine ([M+H]+= 216]) (10-4 M), (G) acenaphthene ([M+.= 154] (10-3 M), and (H) methyl cinnamate ([M+H]+= 163) (10-3 M) prepared in ethanol. The sample solutions were placed underneath the carbon fiber. Figure 3. Analysis of biological samples. (A) CFI mass spectrum obtained by deposing a sample droplet (~2 μL) containing cytochrome c (10-6 M) prepared in the solvent of acetonitrile/water (1:1, v/v) containing 1% acetic acid on the carbon fiber (~10 m) followed by CFI-MS analysis. (B) CFI mass spectrum obtained by deposing 30


Page 30 of 36

Page 31 of 36 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


a sample droplet (~2 μL) containing bradykinin (10-6 M) prepared in the solvent of acetonitrile/water (1:1, v/v) on the carbon fiber (~10 m) followed by CFI-MS analysis. The blue arrow indicates where the carbon fiber is. Figure 4. Analysis of the mixture containing analytes with different polarities. The mixture contained benzo[a]pyrene (10-5 M), ametryn (10-5 M), and arginine (10-4 M) dissolved in their suitable solvents followed by diluted with methanol. A sample droplet (~2 L) containing the mixture was deposited on the tip of carbon fiber, placed vertically to the inlet of the mass spectrometer. After solvent evaporation, the mass spectrometer was switched on. (A) EICs at m/z 175 (black), 228 (red), and 252 (blue) corresponding to protonated arginine, protonated ametryn, and benzo[a]pyrene cation radicals, respectively. Mass spectra obtained at the time ranges of (B) 0 to 0.6 min and (C) 0.6 min to 0.8 min. At the time point of 0.6 min, a methanol droplet (10 μL) was loaded from the top of carbon fiber that allowed the solvent to flow through the tip of carbon fiber to form a droplet for MS analysis.

31



Figure 1

32


Page 32 of 36

Page 33 of 36 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 2

33



Figure 3

34


Page 34 of 36

Page 35 of 36 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 4

35



Graphical abstract


Page 36 of 36

Carbon Fiber Ionization Mass Spectrometry for the Analysis of

Recommend Documents