Subscriber access provided by UNIV LAVAL
Article
Determination of Organic Partitioning Coefficients in Water-Supercritical CO2 Systems by Simultaneous In Situ UV and Near-Infrared Spectroscopies David A. Bryce, Hongbo Shao, Kirk J. Cantrell, and Christopher J. Thompson Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b00641 • Publication Date (Web): 26 Apr 2016 Downloaded from http://pubs.acs.org on May 9, 2016
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.
Environmental Science & Technology 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 27
Environmental Science & Technology
Determination of Organic Partitioning Coefficients in Water-Supercritical CO2 Systems by Simultaneous In Situ UV and Near-Infrared Spectroscopies David A. Bryce,† Hongbo Shao,‡ Kirk J. Cantrell, and Christopher J. Thompson* Pacific Northwest National Laboratory, P.O. Box 999, Richland, WA 99352, USA
ACS Paragon Plus Environment
1
Environmental Science & Technology
Page 2 of 27
1
ABSTRACT
2
CO2 injected into depleted oil or gas reservoirs for long-term storage has the potential to
3
mobilize organic compounds and distribute them between sediments and reservoir brines.
4
Understanding this process is important when considering health and environmental risks, but
5
little quantitative data currently exists on the partitioning of organics between supercritical CO2
6
and water. In this work, a high-pressure, in situ measurement capability was developed to assess
7
the distribution of organics between CO2 and water at conditions relevant to deep underground
8
storage of CO2. The apparatus consists of a titanium reactor with quartz windows, near-infrared
9
and UV spectroscopic detectors, and switching valves that facilitate quantitative injection of
10
organic reagents into the pressurized reactor. To demonstrate the utility of the system,
11
partitioning coefficients were determined for benzene in water/supercritical CO2 over the range
12
35-65 °C and approximately 25-150 bar. Density changes in the CO2 phase with increasing
13
pressure were shown to have dramatic impacts on benzene's partitioning behavior. Our
14
partitioning coefficients were approximately 5-15 times lower than values previously determined
15
by ex situ techniques that are prone to sampling losses. The in situ methodology reported here
16
could be applied to quantify the distribution behavior of a wide range of organic compounds that
17
may be present in geologic CO2 storage scenarios.
18
ACS Paragon Plus Environment
2
Page 3 of 27
Environmental Science & Technology
19
INTRODUCTION
20
Growing energy demands due to expanding global population and prosperity have resulted in
21
increased CO2 emissions into the atmosphere. Several methods, including energy conservation,
22
use of alternative energy sources, and long-term storage of CO2 have been suggested to reduce
23
emissions of greenhouse gasses, and it is generally accepted that some combination of these
24
methods will be employed to control growing atmospheric CO2 levels.1 One promising approach
25
that is gaining public acceptance is geologic carbon sequestration (GCS).2-4 In GCS, CO2 is
26
captured from a large point source such as a power plant or factory, liquefied, and injected into
27
subsurface geologic features such as deep saline aquifers, depleted oil and natural gas reservoirs,
28
and un-mineable coal seams.4 Use of CO2 in enhanced oil recovery (EOR) applications is well
29
documented;1, 5, 6 however, much of the utilized CO2 is recovered. Many of the same reservoirs
30
where EOR is practiced could be used for storage of CO2 collected from power plants.4, 7
31
At typical GCS storage depths (>800 m), CO2 will be held at pressures and temperatures
32
above the critical point (73.8 bar, 31.1˚C).8 In the supercritical phase, CO2 is an effective solvent
33
for non-polar organic compounds and other contaminants, a trait that has been well documented
34
and employed in industry for many years.9-12 This poses unique contamination risks for a
35
supercritical CO2 (scCO2) storage program where residual organic matter is present such as in
36
depleted oil reservoirs. Many of the compounds found in crude oil are toxic or carcinogenic,
37
including benzene and various benzene derivatives such as toluene, ethyl benzene and xylene
38
(BTEX), as well as polycyclic-aromatic hydrocarbons (PAHs). In the event of a leak, stored CO2
39
could potentially escape into overlying groundwater aquifers,1, 13 carrying organics extracted
40
from the storage reservoir.14
ACS Paragon Plus Environment
3
Environmental Science & Technology
41
Page 4 of 27
To better assess the risks of contaminant mobilization, it is necessary to develop a
42
fundamental understanding of pertinent organic contaminant behavior at conditions (temperature,
43
pressure and ionic strength of brine solutions) similar to those found in possible storage sites.15
44
By understanding how organic contaminants behave in binary systems of scCO2 and water/brine,
45
predictions may be made regarding contamination risks associated with leak scenarios. The
46
distribution of an organic contaminant in each of these two phases is described by the
47
partitioning coefficient (K).
48
Timko et. al. provides a summary of available data for organic contaminant partitioning
49
coefficients in water-CO2 systems.16 There are several limitations in the available data set. First,
50
partitioning coefficients have been measured for a relatively small number of compounds.
51
Notably absent are ethyl benzene and the xylene isomers, as well as any representative long-
52
chain saturated hydrocarbons, such as dodecane. Additionally, the scope of conditions
53
investigated show considerable variation, resulting in gaps in the data. Moreover, the impact of
54
ionic strength effects on K values does not appear to have been investigated. Finally, the
55
available data for benzene exhibits considerable uncertainty.17, 18 Some of this uncertainty may
56
be due to the use of ex-situ measurement techniques in one or both phases. For relatively non-
57
volatile compounds, errors caused by the escape of organic vapors during sampling may be
58
small; but for highly volatile compounds such as benzene, such errors could be significant.
59
In this work, our objectives were to 1) develop a fully in situ measurement platform for
60
measuring organic partitioning coefficients in scCO2-water systems, 2) demonstrate the use of
61
the apparatus by measuring K for benzene in scCO2-water over a range of temperatures and
62
pressures, 3) compare our results with previous studies and address uncertainties in the benzene
63
data set, and 4) explore the limitations of the system and its application to other organic
ACS Paragon Plus Environment
4
Page 5 of 27
Environmental Science & Technology
64
compounds. We constructed a unique apparatus that utilizes two spectroscopic detectors to
65
quantify organic concentrations. Measurements in the CO2-rich phase are made in a custom
66
reactor by near-infrared spectroscopy (NIRS), while simultaneous aqueous-phase concentrations
67
are determined by circulating the aqueous fluid past an in-line UV detector. The system employs
68
a high-pressure titration capability that allows quantitative additions of a contaminant of interest
69
to the pressurized reactor. This enables real-time observation of the contaminant’s partitioning
70
behavior. Additionally, due to the high precision of the injection volume, reliable mass-balance
71
checks may be performed to validate measured concentrations and ensure that a third, non-
72
aqueous phase has not formed. We initially applied the system to quantify the partitioning
73
behavior of benzene due to inconsistencies in previous measurements and benzene’s low EPA
74
drinking water standards and associated health risks.20-23 We also conducted some scoping
75
measurements with toluene that demonstrate some of the limitations of our approach.
76
EXPERIMENTAL SECTION
77
Theory. The partitioning coefficient at a given temperature and pressure, Kt,p, is defined as the
78
ratio of mole fraction in the CO2 phase (yi) to that in the water phase (xi).18, 24 These mole
79
fractions can be determined using the concentrations of organic measured in each phase and the
80
fluid densities. When expressed as a function of measured concentrations, Kt,p may be found
81
using:
82
Kt,p =
CCO2 ቆCH2 O +
ρH O 2 ቇ MWH O 2 ρCO 2 ቇ CH2 O ቆCCO2 + MWCO 2
(1)
83
where CCO2 and CH2O represent the organic concentration in the carbon dioxide and water phases
84
respectively, ρCO and ρH 2
2O
represent densities of the solvent (CO2 or water) at the experimental
ACS Paragon Plus Environment
5
Environmental Science & Technology
Page 6 of 27
85
temperature and pressure, and MWCO2 and MWH2 O are the molecular weights of the solvents,
86
CO2 and water respectively. This equation assumes that the organic concentrations are low, and
87
the pure-component densities are a good approximation for the actual solution densities.
88
Apparatus. Partitioning coefficient measurements were made using a custom-built apparatus
89
that consists of a high-pressure reactor, near infrared and UV spectroscopic detectors, and a high-
90
pressure reagent injection system (Figure 1). The reactor (Parr Instrument Company) was
91
machined from grade 3 titanium with a pressure rating of 200 bar at 75 °C and an internal
92
volume of approximately 104 mL. Two 0.5" thick, 1" diameter quartz windows were installed on
93
opposite sides of the reactor for in situ spectral measurements of the reactor's contents. A third
94
quartz window with the same dimensions was installed flush with the internal floor of the reactor
95
to allow visual inspection of a glass-coated magnetic stir bar during experiments. Protection from
96
over pressurization was provided by a rupture disc rated for 193 bar that was mounted on a port
97
at the head of the reactor. Internal temperature monitoring along the vertical dimension was
98
facilitated by a linear array of 5 T-type thermocouples mounted in a 1/8" OD titanium sheath that
99
was fed through a port at the top of the reactor. Pressure monitoring was performed by a pressure
100
transducer (Druck model PTX 500) that was also mounted at the head of the reactor. The
101
thermocouples, pressure sensor, and the UV detector (discussed below) were interfaced to a 24-
102
bit USB data acquisition board (Measurement Computing Corporation model USB-2416-4AO).
103
Pressure, temperature, and UV absorbance data were acquired, displayed graphically, and logged
104
by custom software written using Embarcadero C++ Builder Professional. Heat was supplied to
105
the reactor by two cartridge heaters that were mounted in recessed sockets in the reactor's
106
bottom. Regulation of the heaters was performed by a proportional-integral-derivative controller
107
(Omega CNi32 series), based on the temperature input of a T-type thermocouple mounted in the
ACS Paragon Plus Environment
6
Page 7 of 27
Environmental Science & Technology
108
wall of the reactor 2.75" from the bottom. The reactor was seated in a custom-built, aluminum
109
stand that was attached to the base of the near-infrared spectrometer's sample compartment. A
110
motor-driven, magnetic stirring assembly was installed underneath the reactor in the base of the
111
stand for mixing the reactor's contents.
112
The reagent-injection system is comprised of a syringe pump, SS tubing, switching valves, an
113
HPLC pump, and reagent and waste reservoirs (components are shown to the left of the reactor
114
in Figure 1). The syringe pump (Teledyne Isco model 260D) was operated in constant-pressure
115
mode to deliver pressurized CO2 to the reactor and provide a driving-force for organic injections
116
as described below. Two 6-port, 2-position switching valves (Rheodyne model 7000) direct fluid
117
flow through the system. The first valve was plumbed with a 1/16" OD PEEK tubing loop for
118
reproducible, fixed-volume additions of organic reagents into the reactor. For experiments with
119
benzene, the tubing had an ID of 0.010" and an internal volume of approximately 11 µL. A
120
LabAlliance Series I HPLC pump was plumbed in-line with the valve and a glass reservoir
121
containing an organic reagent of interest. To fill the tubing loop, the valves were switched to
122
direct flow from the HPLC pump through the loop and to the waste reservoir, and the pump was
123
operated long enough to thoroughly purge and fill the loop. Injections into the reactor were
124
performed by first switching the first valve to place the tubing loop in-line with the syringe
125
pump, and switching the second valve to direct flow into the reactor. Next, the pressure of the
126
syringe pump was set to approximately 100 psi above the current reactor pressure, causing CO2
127
to push the tubing loop's contents into the reactor. Due to changes in the viscosities of organics
128
such as benzene and CO2, it was possible to determine when the injection was complete by a
129
small, but distinct increase in the slope of the pressure curve (e.g.,
