Proof of concept of high-temperature comprehensive two-dimensional

Oct 10, 2017 - In this work a reversed-phase high-temperature GC×GC–TOFMS approach for the qualitative and quantitative analysis of crude oils can ...
0 downloads 8 Views 1MB Size
Subscriber access provided by - Access paid by the | UCSB Libraries

Article

Proof of concept of high-temperature comprehensive twodimensional Gas Chromatography time-of-flight Mass Spectrometry for two-dimensional simulated distillation of Crude Oils Maximilian Karl Jennerwein, Markus S. Eschner, Thomas M. Gröger, Thomas Wilharm, and Ralf Zimmermann Energy Fuels, Just Accepted Manuscript • Publication Date (Web): 10 Oct 2017 Downloaded from http://pubs.acs.org on October 10, 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.

Energy & Fuels 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 23

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

Energy & Fuels

1

Proof of concept of high-temperature comprehensive two-dimensional Gas Chromatography

2

time-of-flight Mass Spectrometry for two-dimensional simulated distillation of Crude Oils

3

Maximilian K. Jennerweina,b,c*, Markus S. Eschnera, Thomas M. Grögerb, Thomas Wilharma,

4

Ralf Zimmermannb,c

5

a

6

b

7

Oberschleißheim, Germany

8

c

9

Chemistry, Rostock, Germany

ASG Analytik Service Gesellschaft mbH, Neusäß, Germany Helmholtz Zentrum München, German Research Center for Environmental Health,

University of Rostock, Institute of Chemistry, Division of Analytical and Technical

10

11

Abstract

12

In this work a reversed-phase high-temperature GC×GC–TOFMS approach for the qualitative

13

and quantitative analysis of crude oils can be presented. The proposed setup provides the best

14

utilization of the two-dimensional separation space for carbon numbers between C10 and

15

C60. Visual Basic Script (VBS) was successfully applied for data processing in order to

16

achieve comprehensive classification of the main compound classes. On this basis, crude oils

17

from different origin could be compared by their composition. Real distillation cuts following

18

ASTM D 2892 and ASTM 5236 were applied for the development of area based templates

19

representing virtual boiling point cuts. By this approach a quantification of an artificial crude

20

oil sample with defined initial boiling point was evaluated versus the quantitative result

21

according to ASTM D 7169 (simulated distillation for high boiling samples, hereinafter 1D-

22

SimDist) and by this a two-dimensional simulated distillation (2D-SimDist) was successfully

23

developed.

ACS Paragon Plus Environment

Energy & Fuels

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

Introduction

26

Crude oil takes a central role in human society and economy. But while crude oil reservoirs as

27

well as the quality of the produced oils are decreasing worldwide, the consumption of the

28

main petrochemical products is still continuously increasing. Dealing with this issue is a

29

major challenge for the petroleum industry. Heavier cuts and blends of different kinds of

30

feedstock have to be upgraded using catalytic cracking and hydrocracking in order to satisfy

31

this rising demand for fuels with a steady quality [1 - 7].

32

The qualitative and quantitative analysis of different kinds of heteroatomic compounds within

33

crude oil is of high relevance. The problems caused by these undesirable compounds are

34

commonly known for their toxicity, carcinogenicity and mutagenicity for humans and animals

35

on the one and for catalyst poisoning, fouling and corrosion during refinery processes on the

36

other hand. Not only knowledge of the total heteroatomic content, but also of the chemical

37

structures of the corresponding compounds is crucial for the production of petrochemical

38

products. In recent studies with the focus on heteroatomic compounds in crude oils, scientists

39

applied different sophisticated approaches in order to gain deeper insight in the composition

40

of these compound classes. Hereby, the application of high resolution mass spectrometry

41

using different ionization techniques is very common. Fourier transform ion cyclotron

42

resonance mass spectrometry (FT-ICRMS) is used in most cases [8 – 12] often also in

43

combination with previous HPLC separation [8, 10]. Also comprehensive two-dimensional

44

gas chromatography is used in combination with different detection techniques [13 – 16].

45

Nitrogen and sulfur containing compounds are in the focus of most studies, only few studies

46

address oxygen containing compounds. Two recent works can be reported in this context, Li

47

et al. accomplished a quantification of dibenzofurans and benzo[b]naphthofurans in crude oils

48

[17] and Rohwer et al. presenting a high temperature GC×GC analysis of oxidized paraffinic ACS Paragon Plus Environment

Page 2 of 23

Page 3 of 23

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

Energy & Fuels

49

products [18]. A sophisticated approach using several different detectors for the analysis of

50

shale oil was reported in 2015 which provided detailed information about the chemical

51

composition of this matrix [15]. The authors combined the results of FID. SCD, NCD and

52

TOFMS results and identified several heteroatomic compound classes in concentrations

53

different from common crude oil. However, beside of commonly known compound classes,

54

further heteroatomic compound classes can be found with increasing boiling point and carbon

55

number.

56

Next to the qualitative composition also the knowledge of the true boiling point (TBP)

57

distribution and composition of distillation feedstock is of high importance for refineries

58

when products with constant quality should be ensured. The determination of the boiling point

59

distribution can be achieved using the common simulated distillation following ASTM 7169,

60

but only rough information about the composition can be gained by this approach when a

61

CNS-SimDis analyzer is applied [19]. Some years ago the first steps were made towards a

62

correlation between GC×GC-TOFMS and common one-dimensional simulated distillation

63

(1D-SimDist) by Dutriez et al. for the characterization of vacuum gas oils (VGO) [3 – 5]. The

64

authors emphasize several limitations of the whole approach of a two-dimensional simulated

65

distillation (2-DSimDist), as there are: temperature limits of chromatographic columns,

66

injector discriminations, cracking of petroleum compounds at high temperatures and partial

67

overlapping of compound families. The thoroughly chosen normal-phase column combination

68

included a short 1st column with high phase ratio (10m DB1-HT; 0.32mm × 0.1µm)

69

consequently resulting in a long modulation period of 20 seconds and flame ionization

70

detection (FID). By this approach a rough classification of saturates and aromatics up to tetra-

71

aromatics could be achieved, but only partial distinction of n-, iso- and cyclic alkanes was

72

possible. Contrary to their considerations concerning PTV or on-column injection, hot split

73

injection was chosen for their method development, leading to the described discriminations

74

of high boiling compounds. Nevertheless, since the carbon numbers of the applied samples ACS Paragon Plus Environment

Energy & Fuels

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

75

range between C20 and a maximum of C60, approximately 100% of all vaporizable

76

compounds eluted from the column. Thus, quantification and correlation to 1D-SimDist could

77

be achieved by normalizing the total area to 100% and converting to weight%.

78

Discrimination of the system was taken under account by the application of response factors

79

for n-paraffin standard solutions. Apart from the mentioned short comings the authors could

80

show a good accordance of their developed method with 1D-SimDist following ASTM 2887

81

and conclude that as long the stability of stationary phases is not improved for high

82

temperatures, they have approached the elution limits for the use of GC×GC-TOFMS.

83

However, the major part of the above mentioned works have in common the utilization of the

84

so-called normal phase column combination. Already in 2006 Tran et al. showed in detail the

85

advantages of a reversed-phase column combination with a polar first dimension column and

86

a non-polar second dimension column for the analysis of crude oils, petrochemical products

87

and environmental samples like oil spills [20]. They concluded that due to the increased

88

retention of hydrocarbons in the second dimension, the two-dimensional separation space can

89

be used more effective and the separation itself can be improved. Only few scientists have

90

followed this concept so far especially for the analysis of crude oils. A recent work was

91

reported by Li et al. for the analysis of crude oils in 2015 including a detailed explanation of

92

the comprehensive two-dimensional elution profile [21].

93

The here presented work shows a new column combination and GC×GC –TOFMS set up,

94

optimized for high-temperature measurements comprising a temperature range that has not

95

been reported so far. The thoroughly chosen column set up and GC parameters allow a

96

comprehensive analysis of the vaporable fraction of high boiling samples such as crude oils,

97

heavy fuel oils and vacuum gas oils. In contrary to previous studies mass spectrometry was

98

applied for detection in order to provide a detailed classification of the major compound

99

classes using VBS for data processing. Automated classification tools like VBS are necessary,

100

despite the advantageous separation performance of GC×GC, since the number of overlapping ACS Paragon Plus Environment

Page 4 of 23

Page 5 of 23

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

Energy & Fuels

101

substance classes is increasing along with the complexity of samples to be examined. By the

102

application of predefined functions and variable search parameters within VBS it is possible

103

to search mass spectra fast and reliable for decisive criteria and assign to corresponding

104

substance classes and the generation of virtual boiling point cuts. In addition, measurements

105

using high-resolution TOF (HRTOF) with preliminary GC×GC separation were applied. By

106

this approach, the higher mass resolution of the instrument provided the possibility to

107

investigate the presence of further compound classes, which could not clearly be identified

108

based on nominal mass and elution region. Furthermore, a correlation between standard

109

simulated distillation methods following ASTM D 7169 could be established. For this

110

approach different distillation fractions were produced in house following standard distillation

111

processes ASTM D 2892 and D 5236, respectively. Finally, the concept of a two-dimensional

112

simulated distillation using GC×GC –TOFMS, which combines the information of the boiling

113

point distribution and the classification of different compound groups could be realized and

114

evaluated vs. real boiling point cuts.

115

Experimental

116

Samples

117

Two barrels of a light crude oil (CPC Blend) were provided by Gunvor Raffenerie Ingolstadt

118

GmbH. The whole amount of crude oil was homogenized and portioned in several 5L

119

canister. These portioned light crude oil samples were used for the production of different

120

distillation fractions following ASTM D 2892 and D 5236 standard distillation processes.

121

Additional to the CPC Blend several other crude oil samples from different origin were used

122

for the development of the qualitative analysis using Visual Basic Script, including Arabian

123

light crude oil, Norwegian Troll crude oil, Nigerian Furcado crude oil, Mittelplate crude oil

124

and a blend of North African crude oils.

ACS Paragon Plus Environment

Energy & Fuels

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 23

125

Methods

126

Several portions of the CPC Blend were distillated following ASTM D2892 and ASTM D

127

5236, whereby different distillation cuts were generated for further method development. The

128

distillation processes were performed on two different distillation units, Petrodist® 100CC for

129

distillations following ASTM D 2892 and Petrodist® 200CC for distillations following

130

ASTM D 5236 from the company Pilodist® in Meckenheim, Germany. Distillation process

Temperature parameter

V1

V2

V3

V5

70°C

80°C

90°C

15°C

Start temperature fraction collection End temperature fraction ASTM D2892

collection

370°C 380°C 390°C

160°C

Boiling point range of each fraction

30°C

145°C

Start temperature fraction collection

370°C 380°C 390°C

-

580°C 560°C 570°C

-

End temperature fraction ASTM D5236

collection Boiling point range of each fraction

30°C

-

131

Table 1: Overview of the three different distillation processes for the production of narrow

132

crude oil fractions and one distillation cut with initial boiling point of 160°C

133

Distillation parameters are given in the result section and detailed information is given in the

134

supplemental data. According to the specification of the applied systems in theory there is no

135

minimum limitation for the temperature difference of distillation cuts, but from experience it

136

is known that the amount of overlap is increasing with narrowing distillation cuts. This effect ACS Paragon Plus Environment

Page 7 of 23

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

Energy & Fuels

137

could also be observed and determined for the produced cuts by simulated distillation

138

following ASTM D 7169 and is discussed in detail in the results and discussion part. In order

139

to achieve a good compromise between the narrowness of single distillation cuts and the

140

amount of overlap, boiling point ranges of 30°C were chosen, starting from 70°C, 80°C and

141

90°C (V1 – V3), respectively (see also table 1). Additional, a distillation was performed for

142

the production of only two cuts with boiling temperatures below and above 160°C,

143

respectively. By this approach, an optimized artificial sample was generated for the evaluation

144

of a two-dimensional simulated distillation.

145

The sample and the different cuts were diluted using carbon disulfide resulting in 1%(m/m)

146

solutions and afterwards analyzed using GC-FID following ASTM D7169 for a quantitative

147

characterization of the crude oil samples.

148

All high temperature two-dimensional gas chromatography time-of-flight mass spectrometry

149

(HT- GC×GC –TOFMS) analyses were performed using a Pegasus 4D (Leco, St.Joseph, MI)

150

with an Agilent Technologies 7890A gas chromatograph (Palo Alto, CA) equipped with a 2nd

151

oven and a non-moving quad-jet dual-stage modulator. Furthermore, GC×GC in combination

152

with high resolution TOFMS was applied using a Leco Pegasus HRT. All GC×GC and

153

TOFMS and HR-TOFMS parameters are given in the supplemental data. The different crude

154

oils and the distillation cuts were diluted in dichloromethane resulting in solutions of

155

10%(m/m) and afterwards analyzed using the developed HT- GC×GC –TOFMS method.

156

Data evaluation was performed using Leco ChromaTOF 4.50.8.0 with build in GC×GC

157

Scripts option for VBS. Different scripts have been developed during previous works [22] for

158

the qualitative and quantitative analysis of middle distillate. The selectivity of these scripts is

159

sufficient regarding the manageable complexity of middle distillates but had to be adapted for

160

highly complex matrices like crude oils. In order to validate the applied scripts, measurements

ACS Paragon Plus Environment

Energy & Fuels

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

161

of the investigated samples were performed using a GC×GC –high-resolution TOF system.

162

By this approach it was furthermore possible to identify also oxygen containing compounds.

163

For the development of a two-dimensional simulated distillation, the produced distillation cuts

164

were measured using the developed HT- GC×GC –TOFMS analysis method. Based on these

165

measurements it was possible to generate templates within the 2D plot with a boiling point

166

range of 30°C of every segment. Upper and lower boundaries of each segment were

167

determined by evaluation of the boiling points of single substances, first of all the boiling

168

points of n-paraffins. Afterwards, based on the developed templates, one template with

169

boiling point ranges of 10°C for every segment could be developed.

170

Results and Discussion

171

The precondition for the conjunction of qualitative information based on GC×GC and

172

simulated distillation is a classification of the main compound classes within crude oils. This

173

could be achieved by the application of mass spectrometry for detection and comparison of

174

obtained mass spectra with MS databases, e.g. NIST database. Automated classification can

175

be developed according to the results of a database alignment. ChromaTOF® provides an

176

interface for Visual Basic Script programming of compound group classification which can be

177

also linked to area based classification in the 2D separation space. The completely developed

178

classification method was then used for a qualitative analysis and comparison of different

179

crude oils. The qualitative results of the HT- GC×GC –TOFMS and data evaluation using

180

VBS are given in table 2 and corresponding exemplary chromatograms are given in the

181

supplemental data. By this approach and with the used parameters approximately 99% of all

182

detected compounds, based on area(%), could be classified. The remaining unknown

183

compounds are still a challenge since the obtained spectra are difficult for interpretation and

184

in most cases only low similarity and probability could be achieved by NIST library

185

comparison and homologues series are not recognizable. A validation of the applied scripts ACS Paragon Plus Environment

Page 8 of 23

Page 9 of 23

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

Energy & Fuels

186

and the identification of yet unknown compounds was achieved through measurements using

187

GC×GC separation in combination with high resolution TOFMS. Arabian Mittelplate CPC Blend Light (Saudi Troll (Iceland) (Kazakhstan) Arabia) (Norway) n-/iso-Paraffins 49.66% 49.01% 52.13% 28.62% Naphthenes 18.84% 19.63% 14.43% 21.98% Dinaphthene 4.41% 8.42% 4.05% 10.39% Polynaphthene 0.07% 2.20% 0.10% 2.37% Hopanes/Steranes 0.41% 0.14% 0.13% 0.31% Alkylbenzenes 13.48% 10.18% 16.21% 23.68% Naphthenobenzenes 4.62% 4.15% 3.46% 5.05% Naphthalenes 3.19% 3.40% 3.39% 4.04% Various Diaromatics 0.43% 0.51% 0.41% 1.22% Fluorenes 0.35% 0.55% 0.55% 1.01% Tri-aromatics 0.23% 0.44% 0.31% 0.56% Pyrenes/Fluoranthenes 0.01% 0.02% 0.03% 0.05% Tetra-aromatics