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
