Subscriber access provided by United Arab Emirates University | Libraries Deanship
Article
Experimental determination of isotope enrichment factors – bias from mass removal by repetitive sampling Daniel Buchner, Biao Jin, Karin Ebert, Massimo Rolle, Martin Elsner, and Stefan B. Haderlein Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b03689 • Publication Date (Web): 22 Nov 2016 Downloaded from http://pubs.acs.org on December 1, 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 23
Environmental Science & Technology
Experimental determination of isotope enrichment factors – bias from mass removal by repetitive sampling Daniel Buchner1, Biao Jin1,2, Karin Ebert1,3, Massimo Rolle1,2, Martin Elsner4, Stefan B. Haderlein1* 1
Department of Geosciences, Center for Applied Geosciences, University of Tübingen, Hölderlinstraße 12, 72074 Tübingen, Germany 2 present address: Department of Environmental Engineering, Technical University of Denmark, Miljøvej Building 113, DK-2800, Kgs. Lyngby, Denmark 3 present address: BoSS Consult GmbH, Lotterbergstraße 16, 70499 Stuttgart 4 Institute of Groundwater Ecology, Helmholtz Zentrum München, Ingolstädter Landstraße 1, 85764 Neuherberg, Germany *Corresponding author: Stefan B. Haderlein University of Tübingen Department of Geosciences Center for Applied Geosciences Hölderlinstr. 12 D-72070 Tübingen Phone +49-7071-2973148 e-mail:
[email protected] Abstract Art
ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 23
1
Abstract
2
Application of compound-specific stable isotope approaches often involves comparisons of
3
isotope enrichment factors (ε). Experimental determination of ε-values is based on the
4
Rayleigh equation, which relates the change in measured isotope ratios to the decreasing
5
substrate fractions and is valid for closed systems. Even in well-controlled batch experiments,
6
however, this requirement is not necessarily fulfilled, since repetitive sampling can remove a
7
significant fraction of the analyte. For volatile compounds the need for appropriate corrections
8
is most evident and various methods have been proposed to account for mass removal and for
9
volatilization into the headspace. In this study we use both synthetic and experimental data to
10
demonstrate that the determination of ε-values according to current correction methods is
11
prone to considerable systematic errors even in well-designed experimental setups.
12
Application of inappropriate methods may lead to incorrect and inconsistent ε-values entailing
13
misinterpretations regarding the processes underlying isotope fractionation. In fact, our results
14
suggest that artifacts arising from inappropriate data evaluation might contribute to the
15
variability of published ε-values. In response, we present novel, adequate methods to
16
eliminate systematic errors in data evaluation. A model-based sensitivity analysis serves to
17
reveal the most crucial experimental parameters and can be used for future experimental
18
design to obtain correct ε-values allowing mechanistic interpretations.
19 20 21
Introduction
22
Compound specific isotope analysis (CSIA) is widely applied in various fields of
23
environmental sciences (i) as forensic tool for contaminant source identification
24
detect and quantify in situ (bio)-degradation 3–5 and (iii) for process characterization including
25
identification of transformation mechanisms in (bio)-degradation studies
26
applications involve isotope enrichment factors (ε), which relate quantitatively measured
27
isotope ratios to substrate fractions of an analyte.
28
ε-values are usually determined using laboratory batch experiments and may be lumped
29
parameters, not only reflecting the underlying kinetic isotope effect of a reaction but also
30
physical constraints or experimental artifacts. Thus, reported ε-values for a given process may
31
span a wide range; for instance, the published εcarbon-values for reductive dehalogenation of
32
trichloroethene (TCE) range from -2.5‰ to -19.6‰
7,12–14
1,2
, (ii) to
6–11
. The latter two
. Numerous experimental studies
3 ACS Paragon Plus Environment
Page 3 of 23
Environmental Science & Technology
33
tried to unravel the processes and mechanisms that determine the extent of isotope
34
fractionation for a given reaction, e.g., by investigating the transport of the substrate across
35
the cell wall
7,8
, the efficacy of the involved enzymes
15
, the cell density
9,16
or the nutrient
13,17
36
supply
37
For closed systems, isotope enrichment factors can be obtained using a simplified form of the
38
Rayleigh distillation equation
39
between the natural logarithms of the measured substrate fraction, ln fsubstrate(t) and the
40
measured isotope ratios, ln (R(t)/R(0)). While it is obvious that field sites often cannot be
41
treated as closed systems and require modified forms of the Rayleigh equation
42
laboratory batch reactors the applicability of the linearized form of the Rayleigh equation
43
seems to be straightforward as such systems appear to fulfill all requirements of closed
44
systems. These conditions are strictly met in experiments with numerous parallel replicates
45
which are sacrificed sequentially for sampling
46
uncertainty and variability due to different dynamics in such parallel assays.
47
Alternatively, several samples can be withdrawn from the same batch reactor over time to
48
monitor isotope ratios at different values of fsubstrate(t). Repetitive withdrawal of water and/or
49
gas samples from batch reactors violates the closed system requirement and calls for adequate
50
consideration. Unfortunately, ε-values reported in the literature often lack a clear description
51
whether corrections of measured concentration data were applied. Repetitive sampling from a
52
batch reactor not only leads to mass removal of the target compound(s) from the system but
53
also, if water samples are withdrawn, changes the phase ratio by decreasing the water phase
54
and increasing the headspace volume. Progressive change of the phase ratio, however, needs
55
to be taken into account for volatile organic contaminants such as chlorinated ethenes. As
56
either gas or water volume is sampled, the measured data represent the concentration in only
57
one of these phases and a mass balance on the system is required using Henry’s law and the
58
gas/water ratio to account for the entire amount of substrate or products
59
corrections were suggested to account for the cumulative amount of substrate removed by
60
repetitive sampling (CSR) 13,14,23.
. 18,19
where ε is derived from the slope of the linear correlation
4,20
, for
7,8,15
, however, at the expense of increased
21,22
. Additional
61
The present study systematically investigates how such correction methods affect ε-
62
values determined in batch experiments with repetitive sample removal. Using both
63
experimental case studies as well as a synthetic dataset calculated for typical experimental
64
conditions, we evaluate the performance of various correction methods, identify pitfalls and
4
ACS Paragon Plus Environment
Environmental Science & Technology
65
propose novel methods to eliminate systematic errors in data evaluation. A model-based
66
sensitivity analysis was performed to identify which experimental parameters require the most
67
stringent control and should be used to optimize the experimental design in order to obtain
68
correct isotope enrichment factors that are suitable for mechanistic interpretation and thus
69
indicative of the underlying isotope fractionation processes.
70 71
Methods to determine isotope enrichment factors
72
We describe current and newly proposed methods to account for mass removal by repetitive
73
sampling in batch reactors considering volatile organic compounds. Based on the substrate
74
fraction and stable isotope data the isotope enrichment factor (ε) can be calculated using a
75
simplified Rayleigh distillation equation 18,19 :
() ≈ ∙ () (1) (0)
76
R(t) and R(0) are the stable isotope ratios of the substrate at a time t and the initial time and
77
fsubstrate(t) denotes the remaining substrate fraction (often approximated by the substrate
78
concentration according to c(t)/c(0)) at a time t. ε-values are derived by linear regression of
79
the data without forcing the regression through the origin as suggested by Scott, et al. 24.
80
The substrate fractions, fsubstrate(t), at different time points come from measured concentration
81
data of water samples subject to corrections accounting for volatilization into the increasing
82
headspace and repetitive removal of water samples from the batch reactors. Figure 1 illustrates
83
the effect of sampling and the challenges to adequately consider the effects of mass removal
84
and volatilization.
85
Below, we outline current (I-III) and newly proposed methods (IV and V) to calculate the
86
substrate fraction:
87
5 ACS Paragon Plus Environment
Page 4 of 23
Page 5 of 23
Environmental Science & Technology
88
Figure 1 Schematic illustration of the time course of the substrate fraction, f, during biodegradation in a repetitively sampled batch reactor considering an ideal case in which neither compound’s mass nor water volume are removed from the system and a realistic case which requires correction schemes. The Figure illustrates substrate removal due to sampling by lower water levels (“slices” withdrawn) and removal due to biodegradation by concentration changes (color shades). For example, method III (Henry + CSR) adequately corrects for mass removal by sampling (nrem), but ignores that the removed mass cannot undergo further biodegradation as it would occur in an undisturbed reactor. When all amounts are summed up, this scheme therefore overestimates the remaining substance fraction fsubstrate(t) (red curve).
89 90
(I) The substrate fraction in water phase (water) without further corrections is calculated as
91
follows: () =
() ∙ () () = (2) (0) ∙ (0) (0)
6
ACS Paragon Plus Environment
Environmental Science & Technology
Page 6 of 23
92
nw(t) and nw(0) correspond to the amount (moles) of the substrate in the water phase at times t
93
and t=0, respectively. Measured aqueous analyte concentrations, cw(t) are multiplied by the
94
water volume present at the respective time, Vw(t), to obtain nw(t) thus accounting for the
95
changes of the water volume.
96
(II)
97
substrate in the system by accounting for both the water phase and the gas phase of the batch
98
reactor. The substrate fraction, fHenry (t), is calculated as:
The correction for gas-water-partitioning (Henry) considers the total amount of
() =
() ∙ () + () ∙ () () + () = (3) (0) ∙ (0) + (0) ∙ (0) (0) + (0)
99
ng(t) and ng(0) correspond to the amount of substrate in the gas phase at time t and the initial
100
time. The substrate concentration in the gas phase, cg(t) is calculated from cw(t) by applying
101
Henry’s law. KH is the dimensionless Henry-constant (KH=cw/cg) of the analyte at the
102
appropriate temperature, calculated for 21°C according to Sander
103
Vg(t) is the volume of the gas phase in the batch reactor at time t.
25
and references therein.
104 105
(III)
106
(Henry+CSR) is an extension of the Henry-correction taking into account the amount of
107
substrate removed by repetitive sampling of the reactor. The substrate fraction, fHenry+CSR(t), is
108
calculated according to:
The correction for gas-water partitioning and cumulative sample removal
!"# () =
()
!"#
=
() ∙ () + () ∙ () + ∑*+,% ( = & − 1) ∙ ( ( = & − 1)) (0) ∙ (0) + (0) ∙ (0)
() + () + ∑*+, ( ( = & − 1) (4) (0) + (0)
109
nrem(t) corresponds to the amount of the substrate removed by sampling at a given time t and is
110
calculated by multiplying the concentration in the water phase (cw(t)) with the removed
111
volume (Vrem(t)) at time t. The summation term denotes the cumulative amount of substrate
112
removed from the system by sampling except for the last sampling. Even though this method
7 ACS Paragon Plus Environment
Page 7 of 23
Environmental Science & Technology
113
adequately calculates the mass of substrate that is removed by sampling, it considers this mass
114
as part of the substrate remaining fraction, as illustrated by the color shade in Figure 1. Even
115
so method III accounts for the removed mass it treats this mass as being part of the system that
116
is subject to further
117
fraction of substrate remaining fsubstrate(t).
(bio)-transformation. Consequently, the method overestimates the
118 119
(IV)
The stepwise correction (SW) method also accounts for gas-water partitioning and
120
sample removal. Contrary to method III (Henry+CSR) this method applies an iterative
121
correction scheme and avoids systematic over- or underestimation of substrate fractions which
122
is conceptually inherent to methods I to III.
123
First, for two consecutive sampling times individual substrate fractions, f(t-1)→t, are calculated.
124
By comparing the amounts of substrate present in the system at times t and at time (t-1) after
125
sample withdrawal, this fraction represents exclusively changes due to (bio)-degradation:
126 (.,)→ = (.,)→ =
() + () ( − 1) + ( − 1) − ( ( − 1)
() ∙ () + () ∙ () (5) ( − 1) ∙ ( − 1) + ( − 1) ∙ ( − 1) − ( − 1) ∙ ( ( − 1)
127 128
Using the dimensionless Henry’s constant (KH) equation 5 can be rearranged to: (.,)→
() ∙ () 1 = ( − 1) ∙ ( − 1) ( − 1) ∙ ( − 1) + − ( − 1) ∙ ( ( − 1) 1 () ∙ () +
(.,)→
() () ∙ 2 () + 3 1 = (6) ( − 1) ( ( − 1) ∙ 2 ( − 1) + − − 1)3 ( 1
129 130
cw(t) and cw(t-1) or Vrem(t) and Vrem(t-1) denote the measured concentrations or sample
131
volumes taken at time t and the preceding time point t-1.
132
Second, the overall substrate fraction at time t is calculated considering f(0)=1 (i.e., 100%
133
substrate fraction) and the results of equation 6 for all preceding time steps: "5 () = (0) ∙ 6→, ∙ ,→7 ∙ … ∙ (.,)→ (7)
134
8
ACS Paragon Plus Environment
Environmental Science & Technology
135
As an example, for two sampling steps this method leads to the following expression: "5 (2) = (0) ∙ 6→, ∙ ,→7
(1) (2) (1) ∙ 2 1 + (1)3 (2) ∙ 2 1 + (2)3 "5 (2) = 1 ∙ ∙ (8) (0) (1) (0) ∙ 2 (0) + 1 − ( (0)3 (1) ∙ 2 (1) + 1 − ( (1)3
136
As can be seen, the terms representing the substrate concentration of the intermediate time
137
point t=1 cancel out and equation 8 simplifies to: (2) (1) + (1)3 (2) ∙ 2 + (2)3 2 1 1 "5 (2) = 1 ∙ ∙ (9) (0) (1) (0) ∙ 2 (0) + − ( (0)3 2 (1) + − ( (1)3 1 1
138
Note that this stepwise correction method requires the phase volume ratios for each time step
139
but concentration measurements only for the initial and the final time step.
140
Thereby the propagation of measurement errors (e.g., detector drift) from intermediate
141
sampling times are minimized
142 143
(V)
The mass balance correction (MB) method determines the substrate fractionation,
144
fMB(t), from a mass balance of reactant and products for each time point, i.e., by dividing the
145
amount of the substrate at time t by the sum of the amounts of substrate and all degradation
146
products at that time:
0.96 and CI