Comprehensive Three-Dimensional Gas Chromatography with Time

Dec 30, 2016 - E-mail: [email protected]. Abstract. Abstract Image. Development of comprehensive, three-dimensional (3D) gas chromatography ...
0 downloads 22 Views 2MB Size
Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES

Article

Comprehensive Three-Dimensional Gas Chromatography with Time-of-Flight Mass Spectrometry Nathanial E. Watson, H. Daniel Bahaghighat, Ke Cui, and Robert E. Synovec Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b04112 • Publication Date (Web): 30 Dec 2016 Downloaded from http://pubs.acs.org on January 2, 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 34

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

Comprehensive Three-Dimensional Gas Chromatography with Time-of-Flight Mass Spectrometry

Nathanial E. Watson,a,b H. Daniel Bahaghighat,a,b Ke Cuia and Robert E. Synoveca,*

(a) Department of Chemistry, University of Washington, Box 351700, Seattle, WA 98195, U.S.A (b) Department of Chemistry and Life Science, United States Military Academy, West Point, NY 10996, U.S.A

Manuscript prepared for consideration to publish in Analytical Chemistry Revision of AC-2016-04112K

December 29, 2016

* CORRESPONDING AUTHOR: Tel: +1-206-685-2328; fax: +1-206-685-8665 EMAIL: [email protected]

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

1 2

Abstract Development of comprehensive, three-dimensional (3D) gas chromatography with time-

3

of-flight mass spectrometric detection (GC3 – TOFMS) is described. This instrument provides

4

four dimensions (4D) of chemical selectivity, and includes significant improvements to total

5

selectivity (mass spectrometric and chromatographic), peak identification and operational

6

temperature range relative to previous models of the GC3 reported. The new instrumental design

7

and data output is evaluated and illustrated via two samples, a 115-component test mixture and a

8

diesel fuel spiked with several compounds, for the purpose of illustrating the chemical selectivity

9

benefits of this instrumental platform. Useful approaches to visualize the 4D data are presented.

10

The GC3 – TOFMS instrument experimentally achieved total peak capacity nc,3D ranging from

11

5,000 to 9,600 (̅ = 7,000,  = 1,700) for 10 representative analytes for 50 min separations with

12

component dimensional peak capacities averaging 406, 3.6 and 4.9 for 1D, 2D, and 3D

13

respectively. Particularly, GC3 – TOFMS achieved a combined 2D × 3D peak capacity ranging

14

from 10 to 26 (̅ = 17.6,  = 5.0), which is similar to what is achieved by 2D alone in a GC × GC

15

operating at equivalent modulation period conditions. The analytical benefits of employing three

16

varied chemical selectivities in the 3D separation coupled with TOFMS are illustrated through

17

the separation and detection of 1,6-dichlorohexane and cyclohexyl isothiocyanate as part of the

18

diesel fuel analysis.

19 20 21

KEYWORDS: comprehensive, three-dimensional, gas chromatography, time-of-flight mass

22

spectrometry, peak capacity

23

2 ACS Paragon Plus Environment

Page 2 of 34

Page 3 of 34

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

24 25

Analytical Chemistry

Introduction The promise of comprehensive two-dimensional (2D) gas chromatography (GC × GC) as

26

described by Giddings1 and realized by Philips and Liu2 has been largely recognized within the

27

twenty-first century analytical laboratory. The discourse has moved beyond conceptualization, to

28

broad implementation and optimization. Multiple reviews are available on the topic covering

29

many aspects of the field from application and instrumentation,3,4 to data analysis.5 The most

30

current research focuses on optimizing the total peak capacity of the 2D separation while

31

maintaining the full information imparted by the primary separation dimension.4,6–11

32

Enhancement of the chemical selectivity is also provided through the development and use of

33

novel stationary phases and important applications.4,12–14 Indeed, GC × GC benefits greatly from

34

the large 2D peak capacity that can be regularly provided, in the range of ~ 4,000 to 7,000,9–11,15

35

while a relatively long duration, high efficiency one-dimensional (1D) GC separation provides a

36

peak capacity in the range of ~ 600 to 900.16–19 In addition, GC × GC provides the added benefit

37

of increased chemical selectivity through the use of the second dimension separation.

38

In order to further enhance chemical selectivity, it is intriguing to consider higher order

39

instruments, specifically, to provide comprehensive three-dimensional (3D) separations.

40

Although the development of comprehensive 3D separation techniques is still in its infancy,

41

there are some notable examples based upon various combinations of GC, liquid

42

chromatography (LC), and capillary electrophoresis (CE). In particular LC × LC × CE,20 LC ×

43

GC × GC,21 and GC × GC × GC,22,23 have all been reported. All of these “classical”

44

comprehensive 3D separations produce a data cube in which each of the separated sample

45

components literally “hover” in 3D space at a set of three retention time coordinates. Various

46

intriguing options abound, for data reduction from the 3D space to 2D and 1D domains,22 to

3 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

47

allow the analyst the opportunity to take advantage of the added chemical selectivity of the 3D

48

separation by focusing on critical regions of interest in the 3D separation. In the strictly GC field,

49

a few related designs have been reported that both used a flow switching device, such as GC x

50

2GC,24 in which two GC × GC separations are simultaneously produced, and a hybrid GC

51

instrument that coupled GC × GC sequentially with a third GC separation for the purpose of

52

excising the components eluting from a critical region of the GC × GC for more detailed

53

separation on the third GC separation dimension.25

54

The previous developments of GC × GC × GC, referred to herein as GC3, were fruitful

55

and encouraging.22,23 In these initial reports, an FID was implemented for detection, and two

56

diaphragm valves served as modulators, one valve between the primary column 1D and

57

secondary column 2D separations, and the other valve between the secondary column 2D and

58

tertiary column 3D separations, respectively. In the first GC3 – FID report, a 3D peak capacity of

59

3,500 was achieved, and the high quality trilinear data was demonstrated to leverage the benefits

60

of chemometric analysis using PARAFAC.22 In the subsequent report, the GC3 – FID instrument

61

was demonstrated to provide unique insight into visualizing various chemical compound classes

62

by leveraging the enhanced chemical selectivity of the 3D separation, in particular using an ionic

63

liquid stationary phase column for one of the three separation dimensions.23 In this second report

64

the peak capacity production was 4-fold better than the initial report, going from about 45

65

peaks/min to 180 peaks/min, with a 3D peak capacity of 3,600 achieved in only a 20 min

66

separation of diesel fuel.23 While use of the FID was satisfactory for the initial development of

67

GC3, limitations of the FID for providing confident analyte identification are obvious. Also, the

68

diaphragm valves had to be face mounted on the wall of the GC oven in order to partially

69

overcome the temperature limitations of these valves.26 The face mounted diaphragm valves

4 ACS Paragon Plus Environment

Page 4 of 34

Page 5 of 34

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

70

were able to reliably function up to about 250 °C, which was suitable for demonstrating GC3, but

71

not suitable in the long term for wider adoption of this separation technology. In order for GC3 to

72

be more rigorously evaluated as a separation technology platform, and to gain interest in the

73

separations field, it is imperative that GC3 be combined with a more informative method of

74

detection such as a time-of-flight mass spectrometry (TOFMS),27,28 and to use modulators that

75

overcome the temperature limitations of the face mounted diaphragm valves.

76

Herein, we describe significant improvements to the previously described GC3 – FID

77

instrument by replacing the FID with TOFMS. The GC3 – TOFMS instrument reported herein

78

includes the major additions of a high-temperature diaphragm valve that has been recently

79

demonstrated to reliably function to 325 °C,29,30 to serve as one of the two modulators, and the

80

stock thermal modulator within the commercial instrument platform with the TOFMS to serve as

81

the other modulator. These improved components bring the benefits of increased operating

82

temperature, decreased sample splitting, and added mass spectral selectivity with peak

83

identification. The GC3 – TOFMS instrument also increases the order of the data, naturally

84

producing fourth order data, which opens up new data analysis options unavailable at lower

85

levels of data dimensionality. This report provides a proof-of-principle for GC3 – TOFMS, and

86

demonstrates the benefits of the added chemical selectivity afforded by three chemical stationary

87

phases while maintaining if not exceeding the total separation peak capacity of GC × GC –

88

TOFMS. Instrumental design considerations are made, anchored in comprehensive multi-

89

dimensional separation theory, to design the GC3 – TOFMS instrument to maximize the

90

combined peak capacity of the 2D and 3D separation, which nominally produces a GC × GC –

91

TOFMS separation at every modulation along the 1D separation. Two samples were used for

5 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

Page 6 of 34

92

demonstration purposes, a test mixture of 115 compounds, and a diesel fuel spiked with non-

93

native compounds which we focus on to highlight the selectivity benefits of GC3 – TOFMS.

94

Basic Principles and Instrumental Design

95

For GC × GC, the ideal peak capacity is given by

96

, =  ∗  

97

where 1nc and 2nc are the 1D and 2D peak capacities, respectively. With the addition of a third

98

separation dimension, the ideal peak capacity for GC3 is given by

99

, =  ∗   ∗ 

(1)

(2)

100

where 3nc is the 3D peak capacity. At unit resolution, Rs = 1, eq 2 can be expressed as,

101

, = 

102

where 2t is equivalent to the modulation period, 1Pm, for coupling the 1D to 2D separations, and 3t

103

is equivalent to the modulation period, 2Pm, for coupling the 2D and 3D separations. The nominal

104

peak width-at-base (4σ) for each dimension is given by 1w, 2w, and 3w, respectively.

105 106

   

 

(3)

In terms of modulation periods instead of separation run times of 2D and 3D, eq 3 can be rearranged and expressed as,        

107

, =

108

The arrangement of eq 4 facilitates use of the modulation ratio MR, the ratio of the peak width-at-

109

base for a given separation relative to the modulation period coupling the given separation to a

110

subsequent separation.8 Substitution of 1MR and 2MR, into eq 4 results in the following,

111

, = 

(4)

     





(5)

6 ACS Paragon Plus Environment

Page 7 of 34

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

112

Peak width minimization on all dimensions per eq 4 is critical to optimize nc,3D, particularly 3w

113

for the 3D separation. However, peak width minimization must be balanced with proper selection

114

of 1Pm and 2Pm to provide suitable 1MR and 2MR, in order to provide a comprehensive 3D

115

separation to ensure quantitative data.8,31–33 Here we strive to fully optimize the 3D peak capacity

116

production of the instrument while not degrading the quantitative precision due to valve-based

117

modulator undersampling. A MR of 2 or greater is needed to ensure the %RSD due to valve-

118

based modulation is between 1% - 2%.8,31–33 Moreover, the negative implications of statistical

119

overlap and modulator induced band broadening are greatest at low MR values.7,11,34–36 Equation 5 provides a useful vehicle to theoretically estimate and provide guidance for

120 121

experimentally maximizing the peak capacity that could be achieved with GC3. Assuming

122

implementation of conditions to achieve a MR of 2 for both modulation interfaces and an average

123

3

124

achieve this impressive result could be a 1w of 6 s, with a 1Pm of 3 s (separation time on 2D) and

125

2

126

individual peak capacities would then be 1nc of 500, 2nc of 6, and 3nc of 5. It is noteworthy that

127

the product of 2nc by 3nc is a theoretical peak capacity of 30, which is much higher than typically

128

obtained for the 2D dimension of GC × GC,9–11,16,34 with the added benefit of an additional

129

dimension of separation providing more selectivity. Doing the same calculation, with conditions

130

producing a MR of 2.5 for both modulation interfaces while holding the 1t and 3w constant, the

131

resulting nc,3D drops from 15,000 to 9,600. Thus, the challenge in instrument design and

132

experimental implementation is to produce low MR, at or approaching a value of 2, for both

133

modulation steps. Experimentally meeting this challenge is investigated in this report.

w of 50 ms, an nc,3D of 15,000 could be achieved for a 1t of 50 min. An example of conditions to

w of 500 ms, coupled with 2Pm of 250 ms (separation time on 3D) and 3w of 50 ms. The

134

7 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

135 136

Experimental A Pegasus 4D GC × GC – TOFMS (LECO Corporation, St. Joseph, MI) with an

137

integrated Agilent 6890N Gas Chromatograph (Agilent Technologies, Santa Clara, CA, USA)

138

was modified to produce a GC3 – TOFMS instrument as shown in Figure 1. One high-speed, six

139

port diaphragm valve (Valco Instruments Company Inc, Houston, TX, USA), upgraded by the

140

manufacturer to operate at a maximum temperature of 325 °C,26,29,30 and fitted with a 5 µL

141

sampling loop was installed in the GC oven. The high-temperature valve was utilized as the

142

modulator between 1D and 2D separations, and the stock thermal modulator was implemented

143

between the 2D and 3D separations. The 3D column was contained within the secondary oven in

144

the commercial instrumental platform to provide a small temperature offset.

145

Three columns with different stationary phases were installed in column 1, column 2 and

146

column 3 positions as depicted in Figure 1 for the 1D, 2D, and 3D separations, respectively. A 30

147

m, 250 µm inner diameter (id), 0.50 µm film thickness (5% phenyl)-methyl polysiloxane

148

stationary phase column (Rtx-5; Restek, Bellefonte, PA, USA) was installed as column 1. A 3.5

149

m, 180 µm id, 0.18 µm film thickness polyethylene glycol stationary phase column (Rtx-Wax;

150

Restek, Bellefonte, PA, USA) was installed as column 2. Finally, a 1 m, 100 µm id, 0.1 µm film

151

thickness trifluoropropyl-methyl polysiloxane stationary phase column (Rtx-200; Restek,

152

Bellefonte, PA, USA) was installed as column 3. The orthogonality of these phases was

153

previously evaluated in detail and the phases were shown to be complementary.22 Wrap around

154

was also intentionally applied to ensure maximal use of the separation space.22

155

Liquid injections of 1 µL of both a test mixture of 115 compounds and diesel spiked with

156

non-native compounds were made with an Agilent 7683 auto-injector. The composition of the

157

115 compound test mixture, as well as the list of non-native compounds spiked into the diesel are

8 ACS Paragon Plus Environment

Page 8 of 34

Page 9 of 34

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

158

provided in Supporting Information. The inlet was operated at 250 °C in split mode with a split

159

flow of 25 mL/min. Effluent from 1D was injected onto 2D with the high-temperature diaphragm

160

valve set to actuate for 400 ms at a modulation period Pm = 3 s. Effluent from 2D was transferred

161

to 3D via the thermal modulator with a hot pulse time of 120 ms at a Pm = 250 ms, as previously

162

reported.9 The 1D column was operated at a constant volumetric flow of 0.5 mL/min. The 2D and

163

3

164

program held at 25 psi for the first 3 min and ramped to 35 psi at a rate of 0.211 psi/min and held

165

at the final pressure for 1 min. This program resulted in an approximate volumetric flow on 2D

166

and 3D columns of 1 mL/min. The 1D column was operated at less than the optimum flow rate in

167

order to slightly widen the peaks in the 1D dimension to ensure sufficient peak width and

168

sampling rate at the valve-based modulator. Detection was accomplished with the TOFMS.

169

Effluent was passed through a 0.33 m, 280 °C transfer line into the TOFMS where it was

170

analyzed at 200 Hz between 33 Da to 334 Da. The GC oven was operated with a temperature

171

program starting at 60 °C and held at that temperature for 3 min. The oven was then ramped to

172

250 °C at a rate of 4 °C/min. The final temperature was held for 1 min. The secondary oven was

173

held at a constant 10 °C offset above the primary oven temperature. The 115-component test

174

mixture and spiked diesel were both analyzed in triplicate.

175

D columns were controlled via the auxiliary pressure controller under a ramped pressure

All data were collected using ChromaTOF 3.32 and transferred to MATLAB 2016a (The

176

Mathworks, Natick, MA, USA) using in-house software (peg2mat3p8).37 All chromatograms

177

were baseline corrected in a 1D fashion and folded into a four-way array (4D data). Compounds

178

were identified through a library search utilizing MS Search 2.0 (NIST, Gaithersburg, MD,

179

USA). All multi-dimensional visualization was achieved using functions and utilities included

180

with MATLAB 2016a. The peak width and retention time, tR, of 10 representative compounds

9 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

181

were measured by Gaussian Curve Fitting utilizing the Curve Fitting Toolbox available as an

182

add-on application for MATLAB 2016a.

183

Results and Discussion

184

Data collected for adamantane using the GC3 – TOFMS instrument are presented in

185

Figures 2A-D, with adamantane serving as a representative analyte in the 115-component test

186

mixture. All subfigures were created using m/z 136, a highly selective m/z for this compound.

187

Figure 2A depicts the raw, baseline corrected data vector at m/z 136 showing the expected results

188

for a 3D separation, analogous to a 2D separation. The 1D peak is sampled with a 1Pm of 3 s and

189

the resultant 2D peaklets are then sampled by 3D with a 2Pm of 250 ms. The term “peaklets”

190

refers to a set of peaks in a given separation dimension produced by modulating the analyte peak

191

eluting from the preceding dimension, eg., 3D peaklets resulting from modulating a given 2D

192

peak.10,34 Figure 2B shows detail of the region in Figure 2A with the highest signal intensity, so

193

one can draw their own conclusions regarding the peak widths and peak capacity on the 3D

194

dimension. Finally, Figures 2C & D provide the peak profiles for adamantane on the 1D and 2D

195

dimensions, after summing the signal from the remaining two axes. Figures 2B-D illustrate how

196

peak widths were measured for subsequent determination of modulation ratios and separation

197

peak capacity on each separation dimension for adamantane and other representative analytes.

198

As indicated in Figure 2B, a 3wb of 50 ms was measured. Using a 2Pm of 250 ms, a 3nc of 5.0 was

199

determined for adamantane. Figures 2C & D were used to measure a 1wb of 9.6 s and a 2wb of 0.7

200

s, but were first fitted to a Gaussian profile to do so. With a 2Pm of 3 s, the 2nc was 4.3. Finally,

201

with these measured peak widths, a 1MR of 3.2, and a 2MR of 2.8 were determined.

202

Following the same approach illustrated with adamantane, additional measurements and

203

figures-of-merit calculations for five other representative compounds from the separation of the

10 ACS Paragon Plus Environment

Page 10 of 34

Page 11 of 34

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

204

115-component test mixture and four representative compounds from the spiked diesel fuel

205

separation, vide infra, are summarized in Table 1. The ten compounds in Table 1 were chosen to

206

represent a wide variety of chemical functional classes including varying degrees of saturated,

207

unsaturated and aromatic hydrocarbons along with alcohol and ketone functionality. Though this

208

is but a small portion of the functionality encountered in a complex chemical analysis, this set of

209

compounds provides a good breadth of chemical functionality and depicts the wide range of

210

chromatographic performance that can be expected across varied compound classes.

211

The data portions depicted in Figure 2 were collected as part of the analysis of the test

212

mixture. Figure 3 delves beyond adamantane into the additional benefits gained through the use

213

of the improved GC3 instrument with TOFMS detection. These improvements include mass

214

spectral selectivity and peak identification, and increased maximum temperature while still

215

maintaining the high chromatographic selectivity afforded through the use of three

216

complementary stationary phases. The complete set of images in Figure 3 serve as an illustrative

217

depiction of what becomes possible during the analysis of a truly complex sample. Figure 3A

218

depicts the full separation of the test mixture, but simplifies the data by summing the 2D and 3D

219

chromatographic dimensions and mass spectral dimension onto 1D (Rtx-5). This provides a

220

figure analogous to a traditional GC-MS total ion current (TIC) chromatogram. Immediately the

221

benefits of mass spectral peak identification become apparent with even the relatively simple test

222

mixture. Figure 3B zooms to a region of interest in Figure 3A where the potential for improved

223

selectivity of GC3 – TOFMS can be explored; the compounds butyl-benzene, 1-octanol, 1-

224

decene, 2-nonanone, decane and adamantane all elute here in the order listed. Even though most

225

of the compounds are baseline resolved on the 1D separation, they were selected to provide

226

clarity in the following illustration.

11 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

Modern enhancements to 3D visualization allow for significant improvement relative to

227 228

previous reports on GC3 by way of graphic access to the full 4D selectivity. Figure 3C (also

229

Figure S1 in Supplemental) provides a realization of these improvements, a 3D isosurface plot of

230

the same region shown Figure 3B. The software connects the dots in 3D space where the signal

231

intensity achieves a user selected value creating a cloud representing the data. The addition of

232

color allows for various m/z to be depicted on the same figure. Additional “contours” can be

233

added at different intensities by varying opacity to achieve something akin to a 2D contour plot

234

in 3D space. This feature is not depicted here due to the challenge of presenting such detail on a

235

static 2D surface. Figure 3C includes m/z 56 (1-octanol, 1-decene and decane), m/z 58 (2-

236

nonanone), m/z 91 (butyl-benzene), and m/z 136 (adamantane) at intensities of 500, 1000, 2000,

237

and 800 ion counts respectively. The m/z were chosen due to their selectivity for the analytes

238

eluting within the region depicted. Figure 3D shows the chemical selectivity provided by GC3 for the 2D separation on the

239 240

2

D and 3D dimensions, isolated from the 1D separation for the same analytes per m/z shown in

241

Figure 3C. In Figure 3D the region of the 1D separation between 19.0 and 21.5 min is summed

242

leaving a 2D contour plot showing the chromatographic profiles for the same compounds as in

243

Figs. 3B & C on 2D and 3D. Note that 2-nonanone and decane are overlapped in all three

244

chromatographic dimensions, as a result of applying wraparound to maximize peak capacity

245

usage, and only the addition of the TOFMS enabled their resolution. The added selectivity of

246

TOFMS detection will become particularly important as we move on to a truly complex mixture.

247

The limit-of-detection (LOD) for GC3 – TOFMS was evaluated using adamantane, which

248

was found to be a representative analyte for the other compounds in the test mixture. The

249

injected concentration LOD was ~ 10 ppm at m/z 136. A 30:1 split was applied for injection onto

12 ACS Paragon Plus Environment

Page 12 of 34

Page 13 of 34

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

250

the 1D column, and ~ 15% of the material eluting from the 1D column was transferred by the

251

high-temperature valve to the 2D column, for an overall split of ~ 210:1. Thus, the detected mass

252

LOD was ~ 40 pg (S/N = 3). This LOD is suitable for this proof-of-principle report, however

253

future designs would benefit by having a thermal modulator for both modulation stages.

254

The more rigorous evaluation of GC3 – TOFMS is provided by the analysis of diesel fuel,

255

in this study spiked with a mixture of non-native compounds (see Supplemental Material for

256

spiked compound table). Figure 4 showcases a portion of the results of this analysis. Figure 4A

257

depicts a reduction of the dimensionality down to a 1D chromatogram akin to Figure 3A, and

258

again provides a familiar benchmark separation. Due to the magnitude of the detector response

259

generated by the diesel sample, Figure 4A is the analytical ion chromatogram (AIC) constructed

260

from summing only m/z 41, 43, 53, 55, 74, and 91. Indeed, use of the TIC overwhelms the

261

eluting compounds, reducing the apparent resolution to near zero in a plot of the 1D separation

262

resulting from the presence of significantly more baseline noise in the entire GC3 separation

263

space, which obfuscates the summation of signal of all m/z across the 2D and 3D separations.

264

Figure 4B further showcases the true power of GC3 – TOFMS. Here, another isosurface

265

plot is shown using the same function described for Figure 3C. The complexity of these spiked

266

diesel data allows for a more detailed visualization of the potential of GC3 – TOFMS to tackle

267

challenging chemical analysis problems. In this case, the signal at m/z 41, 43, 53 and 91 are

268

displayed at intensities of 5000, 5000, 1000, and 2500 ion counts, respectively. The intensities at

269

each contour are varied to scale the size of the clouds so the different m/z do not obfuscate each

270

other, to ensure analyte peaks at lower S/N are visible. Considering that the primary composition

271

of diesel fuel is a mixture of hydrocarbons with various degrees of unsaturation and aromaticity

272

we have included four m/z that are generally selective for those compound classes. Usefully, the

13 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

273

phenomenon whereby homologous series of compounds of the same functional groups elute

274

along a line during 2D GC analysis is shown to exist in three dimensions as well. The green (m/z

275

43) alkane band is most obvious, but the red (m/z 41) mono-unsaturated hydrocarbon and yellow

276

(m/z 53) di-unsaturated hydrocarbon series are also apparent though both wrap around on 2D.

277

The varied molecular mass and structures of the aromatics depicted in blue (m/z 91) do not create

278

a linear spatial array of peaks in the same fashion as the various alkanes, alkenes, and alkynes;

279

however, the aromatics do separate nicely as a group along the 3D separation dimension,

280

showcasing the additional chromatographic selectivity by completely isolating the aromatic

281

compounds from the bulk of the hydrocarbons present. This is useful for fingerprinting complex

282

samples which may have a variety of compound classes present.

283

The benefit of having three dimensions of chromatographic selectivity is illustrated in

284

Figures 4C & D. In Figure 4C is presented a contour plot of the 1D separation versus the 2D

285

separation using the AIC summation of m/z 41, 43, 53, 55, 74, and 91, by summing the signal

286

along the 3D time axis. Significantly, the various alkane, alkene, and alkyne bands are generally

287

recognizable down the diagonal of the plot as demonstrated in Figure 2B, but several of the

288

spiked non-native compounds eluting in this region are not well resolved in this 2D view. The

289

mass channels m/z 55 and 74 were added to the AIC in Figure 4C to better target two of the non-

290

native spiked compounds detailed in Table 2: 1,6-dichlorohexane and cyclohexyl isothiocyanate,

291

which have selective m/z of 74 and 55, respectively. Neither are well resolved in Figure 4C, but

292

are very well separated by the 3D separation dimension as shown in Figure 4D. Figure 4D

293

depicts the 1D separation versus the 3D separation for the same portion of the 1D separation

294

between 22 and 28 min at m/z 55 and 74, but in this case summing the signal along the 2D time

295

axis. Significantly, utilization of the two selective m/z coupled with the chromatographic

14 ACS Paragon Plus Environment

Page 14 of 34

Page 15 of 34

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

296

selectivity of the 3D separation enables the clear separation and detection of these two

297

compounds. For complex mixture analysis such as the spike diesel, GC3 – TOFMS provides the

298

opportunity for novel classification and molecular fingerprinting strategies. The added chemical

299

selectivity presents the opportunity to view different planes within the 4D array that represent

300

different compound classes. As an example, the views in Figure 4C & D respectively depict

301

fingerprints of major compound classes and electron rich chemicals in diesel. Using the GC3 experimental parameters implemented coupled with peak width

302 303

measurements summarized in Table 1, key figures-of-merit were calculated. For the ten

304

representative analytes from the two samples studied, the experimentally achieved total peak

305

capacity nc,3D ranged from 5,000 to 9,600 for GC3 – TOFMS (̅ = 7,000,  = 1,700), which is on

306

par with, if not exceeding, previous GC3 – FID reports.22,23 This is very competitive with state-

307

of-the-art GC × GC, which provides nc,2D ranging from approximately 4,000 to 7,000.9–11,15 The

308

benefits of GC3 – TOFMS stem from the additional chemical selectivity along the 3D

309

chromatographic dimension (relative to GC × GC – TOFMS) and addition of mass spectral

310

selectivity (relative to GC3 – FID), while maintaining a similar total peak capacity. Particularly,

311

GC3 – TOFMS achieves a combined 2D × 3D peak capacity ranging from 10 to 26 (mean = 17.6,

312

2 = 5.0) for the representative analytes described in Table 1 which is similar to what is achieved

313

by 2D alone in a GC × GC operating at an equivalent PM. When one considers the benefits of the

314

added chemical selectivity afforded by 3D with a third stationary phase, the usefulness of GC3 is

315

apparent. Looking into the data in Table 1, and upon reflection of the theoretical peak capacity

316

calculations presented earlier, one can see evidence for where future improvements should be

317

made to increase nc,3D. The experimental average of 1MR was 2.6 (s=0.5), while the average for

318

2

MR was 3.6 (s=0.9). These two MR are higher than a more desirable MR of 2. Since the MR were

15 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

319

measured through reconstruction of the modulated peaks, modulator induced band broadening

320

has already been accounted for,34 therefore application of eq 5 appropriately estimates peak

321

capacity, albeit without the statistical correction.7,35,36 In terms of peak capacities, the averages

322

were 406, 3.6 and 4.9 for 1D, 2D, and 3D respectively. In particular, additional effort should be

323

placed in reducing the peak widths on the 2D separation to increase the peak capacity on 2D.

324

Experimentally obtaining a nc3D in the range of 10,000 to 15,000 certainly is very achievable.

325

Conclusions

326

The GC3 – TOFMS instrument may arguably achieve the maximum selectivity available

327

for a gas chromatographic instrument. Herein, we presented data which support this assertion.

328

The coupling of TOFMS with GC3 brings all the benefits inherent with mass spectrometry,

329

namely mass spectral peak identification and added chemical selectivity. The high-temperature

330

diaphragm valve completes the ensemble and brings the GC3 – TOFMS instrument into the same

331

temperature regime as other GC-based instruments. The new design achieves a nc,3D approaching

332

10,000 for select compounds and on average maintains a nc,3D of 7,000. Future studies will be

333

aimed at additional instrumental improvements and leveraging the 4D data structure to tackle

334

challenging chemical analysis problems using chemometric analysis tools.

335

Supporting Information

336

Table S1: 115-mix components

337

Table S2: Non-native spike components

338

Figure S1: A detailed deconstruction of Figure 3C.

16 ACS Paragon Plus Environment

Page 16 of 34

Page 17 of 34

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

339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384

Analytical Chemistry

Literature Cited (1) Giddings, J. C. Unified Separation Science; John Wiley & Sons, Inc.: New York, NY, 1991. (2) Liu, Z.; Phillips, J. B. J. Chromatogr. Sci. 1991, 29, 227–231. (3) Mondello, L. Comprehensive Chromatography in Combination with Mass Spectrometry; Wiley Series on Mass Spectrometry; John Wiley & Sons, Inc.: Hoboken, NJ, 2011. (4) Seeley, J. V.; Seeley, S. K. Anal. Chem. 2013, 85, 557–578. (5) Pierce, K. M.; Kehimkar, B.; Marney, L. C.; Hoggard, J. C.; Synovec, R. E. J. Chromatogr. A 2012, 1255, 3–11. (6) Blumberg, L. M.; David, F.; Klee, M. S.; Sandra, P. J. Chromatogr. A 2008, 1188, 2–16. (7) Davis, J. M.; Stoll, D. R.; Carr, P. W. Anal. Chem. 2008, 80, 461–473. (8) Khummueng, W.; Harynuk, J.; Marriott, P. J. Anal. Chem. 2006, 78, 4578–4587. (9) Fitz, B. D.; Wilson, R. B.; Parsons, B. A.; Hoggard, J. C.; Synovec, R. E. J. Chromatogr. A 2012, 1266, 116–123. (10) Pinkerton, D. K.; Parsons, B. A.; Anderson, T. J.; Synovec, R. E. Anal. Chim. Acta 2015, 871, 66–76. (11) Klee, M. S.; Cochran, J.; Merrick, M.; Blumberg, L. M. J. Chromatogr. A 2015, 1383, 151– 159. (12) Anderson, J. L.; Armstrong, D. W. Anal. Chem. 2005, 77, 6453–6462. (13) Payagala, T.; Zhang, Y.; Wanigasekara, E.; Huang, K.; Breitbach, Z. S.; Sharma, P. S.; Sidisky, L. M.; Armstrong, D. W. Anal. Chem. 2009, 81, 160–173. (14) Ho, T. D.; Zhang, C.; Hantao, L. W.; Anderson, J. L. Anal. Chem. 2014, 86, 262–285. (15) Mohler, R. E.; Dombek, K. M.; Hoggard, J. C.; Young, E. T.; Synovec, R. E. Anal. Chem. 2006, 78, 2700–2709. (16) Wilson, R. B.; Siegler, W. C.; Hoggard, J. C.; Fitz, B. D.; Nadeau, J. S.; Synovec, R. E. J. Chromatogr. A 2011, 1218, 3130–3139. (17) Wilson, R. B.; Hoggard, J. C.; Synovec, R. E. Anal. Chem. 2012, 84, 4167–4173. (18) Snijders, H.; Janssen, H.-G.; Cramers, C. J. Chromatogr. A 1995, 718, 339–355. (19) Snijders, H.; Janssen, H.-G.; Cramers, C. J. Chromatogr. A 1996, 756, 175–183. (20) Moore Jr, A. W.; Jorgenson, J. W. Anal. Chem. 1995, 67, 3456–3463. (21) Edam, R.; Blomberg, J.; Janssen, H.-G.; Schoenmakers, P. J. J. Chromatogr. A 2005, 1086, 12–20. (22) Watson, N. E.; Siegler, W. C.; Hoggard, J. C.; Synovec, R. E. Anal. Chem. 2007, 79, 8270– 8280. (23) Siegler, W. C.; Crank, J. A.; Armstrong, D. W.; Synovec, R. E. J. Chromatogr. A 2010, 1217, 3144–3149. (24) Bueno Jr., P. A.; Seeley, J. V. J. Chromatogr. A 2004, 1027, 3–10. (25) Mitrevski, B.; Marriott, P. J. Anal. Chem. 2012, 84, 4837–4843. (26) Sinha, A. E.; Johnson, K. J.; Prazen, B. J.; Lucas, S. V.; Fraga, C. G.; Synovec, R. E. J. Chromatogr. A 2003, 983, 195–204. (27) Stefanuto, P.-H.; Perrault, K. A.; Stadler, S.; Pesesse, R.; LeBlanc, H. N.; Forbes, S. L.; Focant, J.-F. Anal. Bioanal. Chem. 2015, 407, 4767–4778. (28) Welthagen, W.; Shellie, R. A.; Spranger, J.; Ristow, M.; Zimmermann, R.; Fiehn, O. Metabolomics 2005, 1, 65–73. (29) Freye, C. E.; Mu, L.; Synovec, R. E. J. Chromatogr. A 2015, 1424, 127–133. (30) Freye, C. E.; Synovec, R. E. Talanta 2016, 161, 675–680. 17 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

385 386 387 388 389 390 391 392 393 394

(31) Siegler, W. C.; Fitz, B. D.; Hoggard, J. C.; Synovec, R. E. Anal. Chem. 2011, 83, 5190– 5196. (32) Seeley, J. V. J. Chromatogr. A 2002, 962, 21–27. (33) Seeley, J. V.; Micyus, N. J.; Bandurski, S. V.; Seeley, S. K.; McCurry, J. D. Anal. Chem. 2007, 79, 1840–1847. (34) Pinkerton, D. K.; Parsons, B. A.; Synovec, R. E. J. Chromatogr. A 2016, 1476, 114–123. (35) Davis, J. M. Anal. Chem. 1993, 65, 2014–2023. (36) Davis, J. M.; Stoll, D. R.; Carr, P. W. Anal. Chem. 2008, 80, 8122–8134. (37) Pierce, K. M.; Hoggard, J. C. Anal. Methods 2014, 6, 645–653.

18 ACS Paragon Plus Environment

Page 18 of 34

Page 19 of 34

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

Figure Captions Figure 1. Schematic of the major components of the GC3 – TOFMS instrument. A hightemperature diaphragm valve29,30was utilized as the modulator to interface the 1D column separation to the 2D column separation. The thermal modulator was used interface the 2D column separation to the 3D column separation. Figure 2. (A) Raw, baseline corrected signal for adamantane at m/z 136, with the 1Pm of 3 s indicated (1D to 2D). (B) Zoom of (A) to the highest intensity peaklet and next adjacent peaklet on 3D, indicating the 2Pmof 250 ms (2D to 3D) and a 3wb of 50 ms. (C) Reconstructed plot of the 1 D peak, created by summing the signal along the 2D and 3D dimensions. These data are fitted with a Gaussian profile and the fit parameters µ and σ were utilized to calculate the chromatographic figures-of-merit reported in Table 1 as explained in the Experimental. For adamantane, a 1wb of 9.6 s was determined. (D) Reconstructed plot of the 2D peak, created in the same fashion as (C). Likewise, a 2wb of 0.7 s was determined. Figure 3. (A) Output from the GC3 – TOFMS for the 115 component test mixture. Raw, baseline corrected total ion current (TIC) signal summed onto the 1D (Rtx-5) axis showing similar detail as one would gain from a traditional GC-MS separation. (B) Zoom to a region of interest composed of, in order of elution, butyl-benzene, 1-octanol, 1-decene, 2-nonanone, decane, and adamantane. 2-nonanone and decane are overlapped on 1D. (C) Isosurface plot of the region depicted in (B) depicting the three separation dimensions on orthogonal axes. The colors represent four different m/z: 56 (black), 58 (red), 91 (blue), and 136 (green). Retention data are included in Table 1. See also Figure S1 in Supplemental for each m/z on separate axes. (D) Contour plot of 3D versus 2D for the region depicted in (B) summed along 1D utilizing the same m/z and colors as (C). Figure 4. (A) Output from GC3 – TOFMS for diesel spiked with a mixture of non-native compounds. Reconstructed 1D chromatogram of the spiked diesel created by summing 2D and 3D onto 1D using only signals of m/z 41, 43, 53, 55, 74 and 91. (B) Isosurface plot depicting the full selectivity of the instrument. Output from m/z 41 (red, indicative of mono-unsaturated hydrocarbons), m/z 43 (green, indicative of saturated hydrocarbons), m/z 53 (yellow, indicative of di-unsaturated hydrocarbons), and m/z 91 (blue, indicative of aromatic hydrocarbons) are overlaid and depict grouping by chemical functionality. (C) Traditional 2D contour plot of 2D versus 1D of the spiked diesel using the same m/z. (D) Contour plot of 3D versus 1D for the region between 22 and 28 minutes showing selectivity benefits and total resolution of 1,6dichlorohexane (DCH) and cyclohexyl isothiocyanate (CHI).

19 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

Page 20 of 34

Table 1. Chromatographic peak measurements and figures-of-merit for compounds representing a variety of chemical functionality in both the 115-component test mixture separation and the spiked diesel fuel separation. 115-component test mixture 1-decene 2-nonanone decane

butylbenzene 19.5

1-octanol 20.0

20.4

20.7

7.2 2.4

5.9 2.0

7.8 2.6

1

416

506

2

2.2

2 2

Spiked Diesel limonene 1,6cyclohexyl dichlorohexane isothiocyanate 18.4 22.6 26.7

adamantane

pyridine

20.7

20.9

8.2

8.1 2.7

6.3 2.1

9.6 3.2

5.4 1.8

9.3 3.1

7.3 2.4

9.8 3.3

385

368

477

311

556

323

411

306

1.5

1.7

1.5

0.9

2.2

0.7

1.8

0.6

1.0

0.9

1.3

0.7

0.8

0.7

0.7

1.2

0.6

0.9

1.0

3.5

5.3

2.7

3.3

3.0

2.8

4.8

2.5

3.7

3.9

2

3.5

2.3

4.4

3.6

4.0

4.3

2.5

4.8

3.2

3.1

3

165

195

100

85

85

140

155

50

225

215

3

53

58

47

59

50

50

58

46

47

49

3

nC

4.7

4.3

5.3

4.3

5.0

5.0

4.3

5.4

5.3

5.1

nC, 3D

6803

4973

9106

5635

9639

6821

5987

8352

7010

4850

1

tR (min)

1

w (s) MR

1

nC tR (s) w (s) MR nC tR (ms) w (ms)

20 ACS Paragon Plus Environment

Page 21 of 34

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

Figure 1 Figure 1 190x254mm (96 x 96 DPI)

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

Figure 2A Figure 2A 190x254mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 22 of 34

Page 23 of 34

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

Figure 2B Figure 2B 190x254mm (96 x 96 DPI)

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

Figure 2C Figure 2C 190x254mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 24 of 34

Page 25 of 34

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

Figure 2D Figure 2D 190x254mm (96 x 96 DPI)

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

Figure 3A Figure 3A 190x254mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 26 of 34

Page 27 of 34

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

Figure 3B Figure 3B 190x254mm (96 x 96 DPI)

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

Figure 3C Figure 3C 190x254mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 28 of 34

Page 29 of 34

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

Figure 3D Figure 3D 190x254mm (96 x 96 DPI)

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

Figure 4A Figure 4A 190x254mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 30 of 34

Page 31 of 34

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

Figure 4B Figure 4B 190x254mm (96 x 96 DPI)

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

Figure 4C Figure 4C 190x254mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 32 of 34

Page 33 of 34

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

Figure 4D Figure 4D 190x254mm (96 x 96 DPI)

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

TOC Figure 190x254mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 34 of 34