Deliverable D8.7 Mesocosm experimental · Linus Malmquist, UCPH Thorvaldsensvej 40, 1871 Frb C...

This project is supported by the European Union under the Food, Agriculture and Fisheries and Biotechnology theme of the 7th Framework Programme for Research and Technological Development under GA no. 312139

Kill•Spill

Integrated Biotechnological Solutions for Combating Marine Oil Spills

Deliverable D8.7

Mesocosm experimental protocols on (a) compound specific isotope and (b) mass balance/weathering analyses

Grant Agreement no. 312939 Deliverable D8.7 Mesocosm experimental protocols on (a) compound specific isotope and (b) mass balance/weathering analyses

ii

Work package WP8 Demonstration

Deliverable no. D8.7

Deliverable title Two mesocosm experimental protocols on: (a) compound specific isotope and (b) mass balance/weathering analyses to monitor in situ biodegradation efficacy during running tests of bioaugmentation, bioelectrochemical remediation and within modular system

Due date: 2015-06-30 (Month 30)

Actual submission date: 2015-09-20 (Month 33)

Start date of project: 2013-01-01

Deliverable Lead Beneficiary (Organisation name)

UCPH

Participant(s) (Partner short names) UCPH1, HMGU2, GEUS3.

Author(s) in alphabetic order: J. Aamand3, J. Christensen1, M. Elsner2, A. Johnsen3, L. Malmquist, S1. Marozava2, R. Meckenstock2, A. Meyer2,

Contact for queries: Linus Malmquist, UCPH Thorvaldsensvej 40, 1871 Frb C (Denmark) T: +45 35320486 E: [email protected]

Dissemination Level: (PUblic, Restricted to other Programmes Participants, REstricted to a group specified by the consortium, COnfidential only for members of the consortium)

PU

Deliverable Status: Draft v1 2015-05-21 Redited version V2: 2015-09-15, TUC contribution deleted (RH)


iii

Table of Content

1 About this deliverable ..................................................................................................................... 1

2 Introduction ..................................................................................................................................... 1

3 Mesocosm experimental protocols for compound specific isotope analysis (HMGU) ................... 1

3.1 Introduction ............................................................................................................................. 1

3.2 Procedures ............................................................................................................................... 2

3.2.1 Principles ......................................................................................................................... 2

3.2.2 Materials .......................................................................................................................... 4

3.2.3 Methods .......................................................................................................................... 5

3.2.4 Notes ............................................................................................................................. 14

3.3 References ............................................................................................................................. 15

4 Mesocosm experimental protocols for mass balance/weathering analysis (UCPH) .................... 18

4.1 Introduction ........................................................................................................................... 18

4.2 Extraction and cleanup of petroleum hydrocarbons for GC analysis .................................... 19

4.2.1 Introduction ................................................................................................................... 19

4.2.2 Extraction from sediment .............................................................................................. 21

4.2.3 Extraction and cleanup of oil from water and oil/water emulsions ............................. 25

4.3 GC-FID method for TPH analysis............................................................................................ 26

4.3.1 Introduction ................................................................................................................... 26

4.3.2 Sample preparation ....................................................................................................... 26

4.3.3 Standards for GC-FID analysis for TPH quantification ................................................... 27

4.3.4 Calibration ..................................................................................................................... 27

4.3.5 Internal standards ......................................................................................................... 27

4.3.6 Instrument settings and equipment .............................................................................. 27

4.3.7 GC-FID data analysis ...................................................................................................... 27

4.4 GC-MS method for PAC and biomarker analysis ................................................................... 28

4.4.1 Introduction ................................................................................................................... 28

4.4.2 Sample preparation ....................................................................................................... 28

4.4.3 Standards ....................................................................................................................... 28

4.4.4 Instrument settings and equipment .............................................................................. 29

4.4.5 GC-MS data analysis ...................................................................................................... 33

4.5 References ............................................................................................................................. 38


iv

List of Figures

Figure 1 A) Logarithmic plot of carbon isotope ratios according to the Rayleigh equation for an exemplary compound. From the slope of the regression the enrichment factor ɛ is derived. For practicability ɛ is given in ‰. B) Measured (blue rhombi) and modelled (black line) isotope fractionation according to Equation (2) with an enrichment factor ɛ of -5.3 ‰. ........................................................................................................................ 3

Figure 2 Schematic view of the GC-IRMS hyphenated with a purge and trap autosampler and a temperature programmable (PTV) injector. ....................................................................... 7

Figure 3 Example of a control chart for the background gas argon, analyzed in straight mode on a daily base when applying an emission current of 1.5 mA in the ion source. ............. 8

Figure 4 Exemplary chromatogram of ten CO2-monitoring gas pulses (“CO2-On/Offs”). ................ 8

Figure 5 Check for a low amount-dependency (“linearity”) of the IRMS by analysis of CO2 monitoring gas peaks. ......................................................................................................... 9

Figure 6 Schematic view of a GC-IRMS hyphenated with a purge and trap autosampler. The working scheme lists convenient parameters (purging time, temperature programs, scan times, etc.) for carbon isotope analyses of benzene, toluene, ethylbenzene, xylene and naphthalene. .................................................................................................. 10

Figure 7 Validation of amount-independency (“linearity”) exemplarily for the in-house standard benzene. ............................................................................................................ 11

Figure 8 Parameters for peak detection of BTEX, naphthalene, and 2-methylnaphthalene in the Isodat 3.0 software of the manufacturer Thermo Fisher Scientific. .......................... 12

Figure 9 Exemplary data chart for the quality control of in-house standard measurements of benzene. ........................................................................................................................... 14

Figure 10 Diagram showing the content of this protocol, covering sample preparation, chemical analysis and data treatment. The light-blue box represent the CHEMSIC and Pixel-based data treatment approach, which is still under development. ............................... 18

Figure 11 Illustration of the preparation of a small chromatographic column, packed from its based with a glass filter, activated silica and anhydrous sodium sulfate ......................... 22

Figure 12 PLE extraction cell components. ....................................................................................... 23

Figure 13 Scheme of the PLE extraction protocol ............................................................................ 25

Figure 14 GC-FID chromatograms showing examples of a weathered and an unweathered crude oil analyzed by GC-FID and the sections of the four boiling point fractions (BPFs). ............................................................................................................................... 28

Figure 15 Chromatograms of injections of a mix of DFTPP, Pentaclorophenol, Benzidine and DDT from a GC-MS operating in ideal conditions (a) and a GC-MS in bad conditions (b). ..................................................................................................................................... 30

Figure 16 PW-plot. Example of relative changes in composition of n-alkanes in soil extracts compared to original oil after 1 year of weathering. ....................................................... 34

Figure 17 Example of relative changes in composition of mono-, di- and tri cyclic aromatic compounds in soil extracts compared to original oil after 1 year of weathering. ........... 35


v

Figure 18 Example of relative changes in composition of sulfur containing di- and tricyclic aromatics in soil extracts compared to original oil after 1 year of weathering. .............. 36

Figure 19 Employment of diagnostic ratios to show the degree of weathering of an oil after one year of degradation. ......................................................................................................... 37

List of Tables

Table 1 Recommended maximum signal intensities (in backflush and straight mode) of daily checks on background gases. According to the technical notes of the manufacturer and the experience of the authors, higher signal intensities can influence the measurement performance negatively .............................................................................. 7

Table 2 List of the most appropriate internal and recovery standards for the quantification of 19 selected PACs. .............................................................................................................. 20

Table 3 Ion DFTPP key ions and ion abundance criteria. (USEPA 1998) See USEPA method 8270D, Table 3 for DFTPP acceptance criteria. ................................................................. 30

Table 4 Ions of the key compounds responsible for the control of chromatographic conditions. ........................................................................................................................ 31

Table 5 List of compounds, SICs and corresponding groups of GC-MS/SIM (Gallotta and Christensen 2012) ............................................................................................................. 32

Table 6 Biodegradation diagnostic ratios (DRs) of selected petroleum hydrocarbons................. 36


vi


1

1 About this deliverable

Work in other work packages has resulted in a set of tools and methodologies to degrade oil spilled into the marine environment and to evaluate the different strategies employed. Thereby there exists a need to test both these evaluation methodologies as well as the individual biotechnological solutions during large-scale experiments in mesocosms. This deliverable collects the protocols needed for assessing the effects of applied biotechnological solutions with the two chemical methodologies:

• Compound specific isotope analysis (by HMGU) and • detailed mass balance/weathering analysis (by UCPH),

in mesocosm experiments.

Part of the work presented here has been submitted for publication. The report is thus deemed to stay confidential until the editors have decided on acceptance. The actual journal publication may later replace this deliverable.

2 Introduction

Experimental protocols for mesocosm studies often resemble those for real data collection, depending on analytical strategy and the size of the mesocosm. The large size of the mesocosm facility in Messina and the small size of the samples needed for the chemical analysis means that the protocols employed will be the same as for sampling and extracting samples from real environmental sampling programmes.

Therefore, this deliverable will describe the sampling, extraction and analytical strategies as they were performed on real environmental samples. The setup of the mesocosms will not have an effect of how to do the analysis, but should only consider the aim of the study (i.e. testing of a selected biotechnological solution).

3 Mesocosm experimental protocols for compound specific isotope analysis (HMGU)

3.1 Introduction

In recent years compound-specific isotope analyses (CSIA) has become an established approach to obtain a better understanding of transformation reactions of organic compounds in contaminant hydrology, biochemistry and microbiology (1-5). Furthermore, CSIA enables the allocation of sources for food authentication, in pharmaceutical research as well as in doping analysis and environmental studies (6-9). By measuring the relative abundance of stable isotopes of an element within single organic compounds at natural abundance – even at very low concentrations (µg/L) – CSIA taps information that goes beyond the chemical identity of a compound or the measurement of its concentration. Changes in isotopic signatures reflect underlying kinetic isotope effects that can give evidence on the natural degradation of chemicals in complex natural systems if other approaches such as mass balances and metabolite detection fail (1,10). Biodegradation of aromatic hydrocarbons is associated to changes in their natural (13C/12C) and (2H/1H) isotope ratios. With the detection of these isotope shifts by compound specific isotope analyses, hydrocarbon biodegradation can be identified within environmental samples. Further, changes in isotope ratios during biodegradation experiments, under controlled conditions in the lab, allow calculating isotope enrichment factors, which can then be used to infer the extent of biodegradation in the field. Within the following protocol, we describe the procedure of compound specific carbon isotope analyses for mono- and


2

poly-aromatic contaminants, which are commonly released during an oil spill. With the developed methods, we aim to trace degradation processes of selected contaminants within the planned mesocosm studies. The following protocol provides recommendations on sample preparation and storage, daily quality control of the instrument, sample analyses, necessary parameters for gas chromatography and isotope ratio mass spectrometry, data analyses and quality assurance.

3.2 Procedures

3.2.1 Principles

CSIA to identify degradation

When organic compounds are degraded, bonds containing a light isotope (e.g. 12C-H) typically react faster than those containing a heavy isotope (e.g.13C-H). This phenomenon is known as kinetic isotope effect (KIE) (11). As a consequence, the molecules, which are left behind in the pool of unreacted substrate, contain increasingly more 13C as the reaction progresses (12,13). Non-degradative processes (sorption, transport, dilution), in contrast, cause smaller isotope effects (14-16). On a most fundamental level, changes in the isotope ratio of a compound compared to its original isotopic composition can therefore be a strong qualitative indicator of biotic or abiotic transformation.

CSIA to distinguish transformation pathways

In addition, observable changes in isotope ratios can give information about different (bio)chemical transformation pathways of an organic compound (17). For example, if changes in isotope ratios are particularly pronounced, this indicates that the element for which CSIA is performed is likely to be located in the reacting bond of the target compound (primary isotope effect). In addition, isotope effects can vary depending on the type of reaction mechanism (18). This line of evidence is particularly informative if CSIA is performed on more than one element (19). More detailed explanations on how observable isotope fractionation can be linked to degradation and transformation pathways can be found in Elsner et al., (18,19), Hofstetter et al.,(2) , and Schmidt et al. (9).

Conventions to express isotope values

Isotope ratios are expressed as the proportion of the heavier isotope relative to the lighter isotope (e.g. 13C/12C, 2H/1H, 15N/14N). In addition, isotope values of a sample are given as the relative difference of the isotope ratio of the sample to that of an international reference material (20). Generally, an isotope ratio is expressed with respect to the δ-notation in per mil (‰) according to

δ13Ccompound = (13C/12C)compound(13C/12C)reference

− 1 (1)

where 13C/12Ccompound and 13C/12C reference are the ratios of the heavy isotope (here 13C) to the light isotope (here 12C) in compound x and in the international reference material (21), respectively. Typically, the abundance of the lighter isotope is higher than of the heavy isotope (e.g. natural abundance of 12C = 98.89 %, abundance of 13C = 1.11 %).

CSIA to quantify degradation of groundwater contaminants

Since remaining substrate molecules contain increasingly more 13C as the reaction progresses, changes in the isotope composition provide yet another opportunity to make use of CSIA: these


3

isotope values are indicative of the extent of degradation and can be described by the Rayleigh equation

𝑅𝑅𝑡𝑡𝑅𝑅0

=1 + δ13Ct

1 + δ13Co= 𝑓𝑓ɛ 2)

where Rt and R0 (or δ13Ct and δ13C0) describe the average isotope composition of the heavy isotope to the light isotope in a specific compound at a given time and at the beginning of the reaction, respectively (i.e., when nothing has been degraded so far). The remaining fraction f of the compound is given by the ratio Ct/C0, where Ct is the concentration of this compound at a given time and C0 at the beginning of the reaction. The enrichment factor 𝜀𝜀 links the shift in isotope ratios to the extent of degradation and can be derived from the linearized form of the Rayleigh equation (Figure Figure 1A). In good approximation the isotope shift after 50% of compound degradation is ∆δ13C ≈ 0.7•ɛ (Figure 1B). Further, ∆δ13C ≈ 2•0.7*ɛ for f = 0.25 (75 % degraded), ∆δ13C ≈ 3•0.7*ɛ for f = 0.125 (87.5 %) degraded and so on.

Figure 1 A) Logarithmic plot of carbon isotope ratios according to the Rayleigh equation for an exemplary

compound. From the slope of the regression the enrichment factor ɛ is derived. For practicability ɛ is given in ‰. B) Measured (blue rhombi) and modelled (black line) isotope fractionation according to Equation (2) with an enrichment factor ɛ of -5.3 ‰.

y = -0.0053x + 0.9997R² = 0.9812

0.999

1.004

1.009

1.014

1.019

1.024

-3 -2.5 -2 -1.5 -1 -0.5 0

ln((δ

13C t

+1)/

(δ13

C 0+1

)) A

-30

-28

-26

-24

-22

-20

-18

-16

-14

-12

00.250.50.751

δ13 C

[‰]

remaining compound f

B

∆δ13Cf=0.125 ≈ 3•0.7ɛ

∆δ13Cf=0.25 ≈ 2•0.7ɛ

∆δ13Cf=0.5 ≈ 0.7ɛ


4

On the one hand, the enrichment factors ε are informative because they reflect the magnitude of underlying kinetic isotope effects giving mechanistic information. On the other hand, they have the virtue that they can be used to quantify degradation in the field. Once values of ε have been established for a certain degradation of a specific compound – either because they are already reported in the literature or because they have been determined in controlled microcosm experiments mimicking field conditions (1) – the extent of degradation can be estimated from changes in isotope values according to

𝐵𝐵 = (1 − 𝑓𝑓) · 100 [%] = 1 − �1+δ13Ct

1+δ13Co�1𝜖𝜖 ∙ 100[%] (3)

where B is the extent of its degradation in %. In contrast to conventional methods, which are usually based on the measurement of compound concentrations, this concentration-independent approach makes it possible to estimate degradation of an organic contaminant in natural systems without the need of identifying metabolites and of establishing closed mass balances. A telling indication of the added benefit of CSIA is the fact that the method has already found its way into US-EPA guidelines (10). These guidelines give useful recommendations to environmental consultants and regulative authorities for monitoring of bioremediation at contaminated sites including aspects of quality assurance. Here, we offer a complementary perspective and provide a step-by-step guide of the analytical method which lies at the heart of the vast majority of CSIA studies to date, namely gas chromatography (GC) coupled to continuous flow isotope ratio mass spectrometry (IRMS) as briefly introduced in the following paragraph.

GC-IRMS - instrumentation and specifications.

Gas chromatography-isotope ratio mass spectrometry (GC-IRMS) makes it possible to analyze the isotope ratios of individual compounds of a complex mixture within one analytical run. Analytes are chromatographically separated and converted into a suitable measurement gas (here: CO2) in an online combustion oven. A helium carrier stream transfers the separated CO2 peaks to a dedicated isotope ratio mass spectrometer where the generated 13CO2 and 12CO2 molecules are detected simultaneously, which results in a high precision which is applied to analyze small shifts in the isotope ratio (4th decimal place). With an analytical uncertainty of ± 0.5‰ sample amounts as low as 0.8-1 nmol C (roughly 10 ng of a typical hydrocarbon) on-column can be analyzed (22,23).

3.2.2 Materials

Instrumentation

• Purge and trap unit (P&T): purge and trap concentrator Tekmar Velocity XPT with a liquid autosampler Tekmar AQUATek 70 (Teledyne-Tekmar, Mason, OH, USA; http://www.teledynetekmar.com/)

• VocarbTM 3000 Trap (Supelco, Bellefonte, PA, USA) • Programmable temperature vaporization (PTV) injector: Optic 3-SC High Power Injection

System (ATASTM GL International B.V., Veldhoven, Netherlands; http://www.atasgl.com/), equipped with a baffled liner (PAS Technology, Magdala, Germany; http://www.pas-tec.com)

• Gas chromatograph: TRACE GC Ultra gas chromatograph (Thermo Fisher Scientific, Milan, Italy; http://www.thermoscientific.com/en/home.html),

• Agilent DB-624 column (60 m × 0.25 mm, 1.4 µm film thickness, J&W Scientific, Folsom, Canada; http://www.home.agilent.com)

• 2 m Fused Silica pre- and postcolumn (ID 0.25 mm, OD 0.38mm, CS-Chromatopgraphie Service GmbH, Langerwehe, Germany, http://www.cs-chromatographie.de/

http://www.teledynetekmar.com/

http://www.atasgl.com/

http://www.pas-tec.com/

http://www.thermoscientific.com/en/home.html

http://www.home.agilent.com/

http://www.cs-chromatographie.de/


5

• Combustion unit: Finnigan GC combustion interface (Thermo Fisher Scientific, Bremen, Germany), equipped with a Pt/CuO/NiO reactor tube or a ceramic tube for δ13C analysis (Thermo Fisher Scientific, Bremen)

• Isotope ratio mass spectrometer: Finnigan MAT 253 (Thermo Fisher Scientific, Bremen, Germany)

• Software: Isodat 3.0 (Thermo Fisher Scientific, Bremen, Germany)

Chemicals

• N2 liquid gas (Linde Gas, Munich) for cooling device of PTV injector • Carrier gas: ultrapure He 5.0 (99.999 %; Linde AG, Pullach, Germany); CO2 gas (99.99 %, 10 kg

bottles, CARBO Kohlensäurewerke GmbH & Co. KG, Bad Hönningen, Germany) used as monitoring/reference gas

• Benzene, toluene, ethylbenzene, m-xylene, naphthalene, 2-methylnaphthalene (99.9 %, Sigma Aldrich, Seelze, Germany) used as in-house standards

Materials for storage

• 40 mL glass vials equipped with PTFE/silicone septa for Purge and Trap Autosampler (CS-Chromatographie Service, Düren, Germany; http://www.cs-chromatographie.de/

• Glass bottles/vials for storage of standard stock solutions and septa o 8 mL and 12 mL vials; Wheaton: Cat.# W240509 - (15-425 Cap), Phenolic, Black,. o 4 mL vials; Wheaton: Cat.# W240508 - (13-425 Cap), Phenolic, Black, o 1.5-2 mL; Wheaton: Cat.# W240506 - (8-425 Cap), Phenolic, Black,

(http://wheaton.com/)

3.2.3 Methods

Sample Storage & Preparation

It is important to stress the necessity of obtaining a representative sample for analysis. Otherwise, results may be meaningless or lead to severe misinterpretations. Additionally, good sampling practice for isotope analyses should ensure that there is no mass discrimination meaning that sampling and storage needs to be carefully checked for the absence of isotope fractionation.

1. Samples should be analyzed as soon as possible. Otherwise, they can be either stored in fully filled gas tight glass bottles at 4°C, or if possible even be frozen. In the latter case, the bottles have to be prevented from bursting by leaving a small head space, and they need to be insulated on the top while being frozen (24)

2. To avoid ongoing microbial degradation, samples should be preserved by setting the pH of the samples to greater than 10 by addition of NaOH or to smaller than 2 by adding concentrated hydrochloric acid;

3. To avoid sorption of the aromatic hydrocarbons to storage containers, samples should be stored in glass bottles which have to be closed with Teflon-coated sealing caps. Alternative strategies for sampling and storage can be found in Hunkeler et al (10).

To prepare samples for analyses the following steps have to be carried out:

1. Fill samples into 40 mL purge and trap vials. Avoid headspace volume within the vials. Vials have to be closed with gas tight teflon septa.

2. Take replicates of the samples for possible re-analyses. 3. Prepare working standards freshly every second day from stock solutions of authentic in-

house standards of benzene, toluene, ethylbenzene, o-xylene, m-xylene, naphthalene, and 2-methylnaphthalene. The stock solutions have to be prepared at least monthly from authentic

http://www.cs-chromatographie.de/

http://wheaton.com/


6

in-house standards. The in-house standards, in turn, should be characterized independently for their carbon isotopic composition. Afterwards they must be stored in tight containers without headspace in order to strictly avoid any volatilization over time. (see note 1)

4. Prepare blanks filled with the same water (tap or deionized) as used for the standards 5. Adjust the sample/standard concentration by proper dilution (either manually or by the

choice of the right split-flow within the injector) to obtain amplitude signals for the mass m/z 44 which are within the linearity range (normally between 500-25000 mV) of the isotope ratio mass spectrometer (25,26)

Instrument Preparation for δ13C GC-IRMS Analysis

A performance check should be executed on a daily basis to ensure proper functioning of the instrument. It is recommended to implement a “morning routine” for instrument checks and quality control as detailed below. In the following paragraph we describe a list of tests to be performed prior to a sample analysis.

1. Control of source vacuum; should be approximately 1.5 10-6 mbar 2. Control of source heating; check indicator lights; power supply control lamps for the ion source

light bulbs are visible from outside of the IRMS casing 3. Check electron emission current (ion beam generated by an heated filament to ionize CO2

molecules) at control panel of the IRMS software, LED lits green 4. Check column flow and check backflush flow (note 2): backflush flow has to be at least twice the

column flow plus 0.4 mL/min 5. Check intensities of the masses of CO2 (m/z=44), H2O (m/z=18), Ar (m/z=40), N2 (m/z=28) and O2

(m/z=32), on detector cup 2 in high amplification mode to record background values of air constituents as indicators of possible leaks. For a typical analysis (Figure 2) the test would, for example, be run with the following parameters:

- Column-flow: 1.4 mL/min - Split-flow: 14 ml/min - Electron-Emission: 1.5 mA - Backflush-flow: 3.2 ml/min

i) start first in backflush mode (backflush on, open split connected) to check only for the performance of the isotope ratio mass spectrometer (Figure 2)

ii) straight mode (backflush off, open split connected) enabling to monitor for the whole GC-IRMS system (constituting of gas chromatograph, interface and IRMS; Figure 2)

6. Begin with setting the instrument to mass m/z = 40 for Ar to identify leakages then continue with the other gases listed in table 1 It is useful to record the performance with automated charts as shown in Figure 3 for Ar.

7. To check stability and precision of the IRMS performance, measure five times a series of ten CO2 monitoring gas pulses (”Standard On /Offs”) (Figure 4). According to manufacturer specifications (27) the standard deviation of these CO2-isotope values must be below 0.04‰ in each series.

8. Subsequently, perform a linearity check to guarantee that carbon isotope values do not strongly change with signal size (Figure 5). To this end, analyze a series of ten monitoring gas pulses while increasing the monitoring gas pressure from 0.25 to 2.5 bar in 0.25 bar steps. Plot these ten δ13C values against the amplitudes of the mass m/z 44 (given in Volts) as illustrated in Figure 5. The slope of the regression line must not be steeper than ± 0.06‰/Volt (according to the specific amplifier configuration of the faraday cups set up by the manufacturer Thermo Fisher Scientific.; where the ion beam of the mass m/z 44 is caught on cup 4 (resistance R: 3•108 Ω), m/z 45 on cup 5 (R: 3•1010 Ω ) and m/z 46 on cup 6 (R: 1•1011 Ω)).


7

Table 1 Recommended maximum signal intensities (in backflush and straight mode) of daily checks on background gases. According to the technical notes of the manufacturer and the experience of the authors, higher signal intensities can influence the measurement performance negatively

Gas m/z Tolerable Intensity

Problem and likely reason

CO2 44 300 Interfers with carbon isotope ratios during analysis Ar 40 60 Indication of system leak O2 32 400 Reducese life-time of filament in the ion source N2 28 2000 Indication of system leak H2O 18 2000 Forms protonated species H 12CO2

+ interfering with the mass m/z 45 o the ion 13CO2

Figure 2 Schematic view of the GC-IRMS hyphenated with a purge and trap autosampler and a

temperature programmable (PTV) injector. A) shows the system in straight mode, meaning that the GC-effluent (green dots) passes through the combustion oven, reduction oven, water trap and the open split into the IRMS. In the so called backflush mode (B), a helium stream flows (red line) into the combustion reactor and through the backflush valve, preventing the GC-effluent entering the combustion reactor. To reoxidize the combustion reactor oxygen (valve is opened) is transported together with the backflush stream to


8

the combustion reactor. Further the scheme indicates spots which may cause problems. Parts of the purge & trap system are indicated by roman numbers, parts of the control unit of the PTV injector by alphabetical letters and parts of the GC-IRMS by arabian numbers.

Figure 3 Example of a control chart for the background gas argon, analyzed in straight mode on a daily

base when applying an emission current of 1.5 mA in the ion source. The black line indicates the recommended maximum signal intensity of 60 mV. Points below the line indicate a tight system, whereas values above the line evidence a leak within the system.

Figure 4 Exemplary chromatogram of ten CO2-monitoring gas pulses (“CO2-On/Offs”).

Standard deviation of the δ13C values of all ten peaks should be < 0.04 ‰. δ13C values are automatically calculated by the manufacturer's used software Isodat3.0 using the ratio of ion beams of the mass m/z 44 (12CO2) and mass m/z 45 (13CO2).

020406080100120140160180200

020406080

100120140160180200

08/0

9/20

0627

/11/

2006

16/0

2/20

0727

/04/

2007

10/0

7/20

0713

/09/

2007

14/1

2/20

0707

/03/

2008

27/0

5/20

0831

/07/

2008

09/1

0/20

0811

/12/

2008

03/0

3/20

0924

/06/

2009

10/0

9/20

0917

/03/

2010

22/0

6/20

1011

/09/

2010

21/0

2/20

1128

/06/

2011

13/1

0/20

1110

/01/

2012

26/0

3/20

1218

/06/

2012

21/0

8/20

1220

/02/

2013

22/0

9/20

14

Ampl

itude

m/z

= 4

0 [m

V]

date

Ar-Background (straight mode)

100 200 300Time [s]

400 500

0

1000

Inte

nsity

[mV]

2000

3000

4000

5000


9

Figure 5 Check for a low amount-dependency (“linearity”) of the IRMS by analysis of CO2 monitoring gas

peaks. The lower graph shows how peak amplitudes (m/z 44) increase as the reference gas pressure is augmented. The upper graph gives the evaluation of the corresponding δ13C values in dependence of the increasing amplitudes (m/z 44).

Purge and Trap online δ13C GC-IRMS Analysis

The following chapter describes the different steps which have to be carried out for carbon isotope measurements via GC-IRMS connected to a purge and trap autosampler device (note 3). Analytical samples are preconcentrated by the P&T unit coupled online to the PTV-injector of the GC-IRMS system. Instrument methods and settings for all devices (P&T autosampler PTV injector GC IRMS) are given in Figure 6. For best performance and quality control the following procedure should be carried out:

100 200 300Time [s]

400 500

5000

10000

15000

20000

Inte

nsity

[mV]

0

y = 0.0408x - 0.025R² = 0.9501

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

0 5 10 15

∆δ1

3 Crm

onito

ring-

gas

[‰]

Amplitude m/z 44 [V]


10

Figure 6 Schematic view of a GC-IRMS hyphenated with a purge and trap autosampler. The working

scheme lists convenient parameters (purging time, temperature programs, scan times, etc.) for carbon isotope analyses of benzene, toluene, ethylbenzene, xylene and naphthalene.

1. Start with the analysis of a chromatographic blank (= no sample injected) to identify possible column-bleeding or matrix effects from the GC system.

2. Check for interfering peaks or matrix effects originating from the P&T system by analyzing water blanks

3. Before analyzing samples, the concentration range for precise isotope analysis (“linearity range”) of the target compounds must be validated (note 4). Use at least 4 different concentrations of isotopically characterized target compounds of the in-house standards which bracket expected concentrations of the samples. Analyze each of them three times. Subsequently, plot them as shown for the compound benzene in Figure 7 to guarantee that true and reproducible δ13C values are obtained which are independent of signal size. Accurate values are typically obtained for peak amplitudes (mass 44) between 500 mV and

Purge & Trap Desorb Preheat 250 °CPresurize Time 1 min Purge Time 11 min Desorb Time 2.5 minPressure 130 kPa Purge Temp. 45 °C Desorb Temp. 250 °CSample loop 25 mL Purge Flow 40 mL min-1 Desorb Flow 200 mL min-1

Transfer Time 3 min Dry Purge Time 1 min START Injector ProgrammPressure 130 kPa Dry Purge Temp. 40 °C Bake Time 8 min

Dry Purge Flow 200 mL min-1 Bake Temp. 300°CBake Flow 400 mL min-1

InjectorInitial ConditionsTemperature -100 °C Temperature 250 °CColumn Flow 1.4 mL min-1 Delay Time 20s Column Flow 1.4 mL min-1

Split Flow 14 mL min-1 START GC Programm Split Flow 14 mL min-1

Gas ChromatogaphColumn Ramp rate Temperature Hold timeAgilent DB 624 50 °C 4 min60 m x 0,25 mm 10 °C min-1 120 °C 1 minFilm thickness 1.4 µm 3 °C min-1 140 °C 1 min + fused silica post and pre-column each 1 m 18 °C min-1 250 °C 6 min

Combustion InterfaceOxidation Reactor (Pt/Cu/Ni) 940 °CReduction Reactor (Cu) 650 °C

IRMSEmission : 1.5 mA

Time (s) 0 30 50 80 100 130 150 600 1800 1850 1870 1900 1920 1950 1970 2000Backflush On Off OnOpen Split Open Close OpenMonitoring Gas Off On Off On Off On Off On Off On Off On Off


11

25000 mV, but can even be achieved at amplitudes as low as 150 mV (28). Before analysis of samples, such a test needs to be carried out for each target compound. Within the investigated range, precision should be within ± 0.3 ‰ (standard deviation) and the systematic difference of the δ13C instrument values from true δ13C values of in-house standards determined by elemental analyzer-IRMS (EA-IRMS) should not be greater than ± 0.5 ‰ (26).

Figure 7 Validation of amount-independency (“linearity”) exemplarily for the in-house standard benzene.

Blue diamonds are δ13C instrument values of in-house standards analyzed by GC-IRMS at different concentrations. The blue dashed line gives the average (-24.6 ‰) of all GC-IRMS measurements with a standard deviation (n=12) of 0.1‰. The black line represents the “true” δ13C value of benzene (-24.4 ‰) determined by EA-IRMS using international reference materials. The black dashed line indicates the total uncertainty of isotope analyses of ±0.5‰. δ13C values are given according to Equation 1 relative to Vienna-Pee Dee Belemnite (VPDB).

4. After validating the range for accurate isotope analysis, sample sequences can be started. At the beginning and the end of each sequence working standards, which are prepared from isotopically characterized in-house standards, are measured for two to three times. It is recommended to introduce additional standards within the sequence.

5. Analyze samples at least in duplicates (with the option of an additional measurement if the results differ by more than 1‰) or even in triplicates (29).

6. After every 9-10 sample runs (e.g.: 5 samples analyzed in duplicates or 3 samples analyzed in triplicates) the lab standard should be analyzed again (1x) (note 5).

7. If the concentration of a sample is outside of the accepted linearity range, the sample should be adjusted to the proper concentration range and be analyzed again.

8. Also add water blanks after highly concentrated samples (> 1000 µg/L) to (i) avoid carry over effects which impair the isotope values of the target compounds of the next sample, and (ii) to check for background alterations of the analytical system.

9. To avoid carry over effects, check also that the rinse water tank of the P&T is filled with enough water to rinse the sample loop and sparge vessel of the P&T unit after each sample.

-26.0

-25.5

-25.0

-24.5

-24.0

-23.5

-23.0

0 20 40 60 80 100 120 140

δ13

CVP

DB

[‰]

concentration [ug/L]


12

Data Analysis

In isotope analysis, baseline separation of target peaks, as well as the right choice of integration parameters, are crucial for two reasons:

(i) After passing the combustion oven, both target analytes and the interfering matrix are combusted to CO2 and, therefore, become indistinguishable. Hence, if a certain setting of integration parameters inadvertently includes additional peaks, this can seriously bias isotope values.

(ii) Because 13C- and 12C-containing molecules of the same compound typically have slightly different retention times, 13CO2 and 12CO2 do not exactly coelute, but appear in, what is known as, “isotope swing”: 13CO2 typically elutes first, followed by 12CO2. Hence, if inadequate integration parameters miss part of the peak this leads to systematic errors (29,30). To avoid this, the maximum peak width has to be chosen broad enough. Additionally, the peak resolution should be high enough. The peak is defined by the determination of the peak start and its end time. The software Isodat 3.0 of the manufacturer Thermo Fisher Scientific continuously analyses the rate of change of the signal using a five point linear regression for each data point. Start and end of each peak can be determined by threshold slope values chosen by the user. Peak detection parameters, which are typically used in the laboratory for the analyses of BTEX (benzene, toluene, ethylbenzene and xylenes) and naphthalene, and 2-methylnaphthalenen, are given in Figure 8

Figure 8 Parameters for peak detection of BTEX, naphthalene, and 2-methylnaphthalene in the Isodat 3.0

software of the manufacturer Thermo Fisher Scientific.

The Isodat 3.0 software automatically converts isotopologue ( isotopic composition of a molecule) ratios of analyte peaks into isotope ratios. To this end, peaks are compared to one of several monitoring gas peaks which are introduced at the beginning and at the end of each chromatographic run(23). The monitoring gas, in turn, is calibrated against international reference materials (note 6). In contrast to CO2 from analytes, the CO2 of the monitoring gas has not been generated in the combustion oven and – according to the principle of identical treatment of standard and sample – is therefore not a true standard. As shown in figure 7, the GC-IRMS instrument values that are


13

automatically generated nonetheless often agree with the “true” values of lab standards within the uncertainty of the measurement so that for many compounds, among them BTEX and naphthalene, they are reported without further correction. This referencing strategy is called “single point anchoring” strategy and is common for the application of CSIA using GC-IRMS.

We recommend to check each chromatogram individually, in order to be aware of possible systematic errors that may derive, for example, from an unstable background, strong deviations in the δ13C values of the reference gas peaks, or insufficient peak separation (29,30). Peak separation can be defined by the following expression (31),

𝑅𝑅𝑠𝑠 =(𝑡𝑡𝑅𝑅2 − 𝑡𝑡𝑅𝑅1)

𝑤𝑤𝑏𝑏1 + 𝑤𝑤𝑏𝑏2 ≈

𝑡𝑡𝑅𝑅2 − 𝑡𝑡𝑅𝑅1𝑤𝑤ℎ1 + 𝑤𝑤ℎ2

where tR2 and tR1 are the retention times of the adjacent peaks, wb1 and wb2 the corresponding widths at the bases of the peaks, wh1 and wh2 are defined as the peak half widths and Rs is the chromatographic resolution. For baseline peak separation Rs should be at least 1.5 (0.2 % peak overlap), whereas complete peak separation is achieved for Rs =2 (0% overlap) (25).

Quality assurance/control

Quality assurance of isotope measurements by GC-IRMS is based on the trueness and precision of the values from external (lab) standards that are analyzed along with the samples in daily sequences. Since these working standards are prepared from isotopically characterized in-house standards, and since they fulfill the criterion of identical standard and sample treatment(25), they can serve as a true quality indicator of measurement performance. The standard deviation (n > 3) of δ13C values should be within ±0.3 ‰ and the total uncertainty (encompassing deviations from the characterized in-house standard values) should be within ±0.5 ‰ (26). To monitor long-term instrument performance, we recommend using control charts for the evaluation of analyzed in-house standards in order to detect possible mismatches in precision and trueness as early as possible. These charts are conveniently constructed in Excel spreadsheets and should contain the following information:

a) Information to be typed in:

• Date of analysis • User • Date of the preparation of the stock-solution from which daily standards are prepared • Concentration of the standard • Column and split flow • chromatographic column • amplitude of mass m/z 44 • δ13C values of the respective in-house standard from which the daily standard has been

made

b) Information to be calculated:

• mean value • standard deviation • range of the total uncertainty of the analyzes • linearity of the instrument tested with in-house standards for the range of concentrations of

the samples that will be analyzed

Furthermore, it is helpful to express some of these data/parameters in automated quality control charts as illustrated for benzene in Figure 9.


14

Figure 9 Exemplary data chart for the quality control of in-house standard measurements of benzene.

Blue diamonds are the measured δ13C values of the analyzed lab standards during various measurement sequences at different days. The blue dashed line gives the average (-24.6 ‰) of all GC-IRMS measurements with the calculated standard deviation SD (orange dots) of every measurement day. The black line indicates the true δ13C value of benzene (-24.4 ‰) determined by EA-IRMS using international reference materials. The black dashed line indicates the total uncertainty of the isotope analyses of ± 0.5 ‰.

3.2.4 Notes

1. To assure quality control of isotope analyses it is important to bracket samples with external authentic in-house standards of the target compounds (these are different from the “reference gas” / “monitoring gas” CO2 peaks that do not constitute an internal standard). For these external standards the true isotope value must be determined in advance by analyzing them against international reference materials. Ideally, characterization of an in-house standard is accomplished by offline combustion (32) and subsequent analysis against CO2 in dual-inlet IRMS. In practice, excellent results are also accomplished by measurements conducted on an Elemental Analyzer hyphenated with an isotope ratio mass spectrometer. Since the combustion process is typically quantitative, true and precise values (standard deviation 0.1‰, (4) are obtained after calibration with international reference materials like IAEA-600 (caffeine), IAEA 601 (benzoic acid) or USGS24 (graphite). For more details about the referencing procedure we refer to Werner et al., 2001(32). These in-house standards should be re-characterized in regular intervals in order to minimize effects of day-to-day variation during characterization, and to ensure that isotope values of standards do not change over time. We recommend storing a sufficient amount of these standards (1-2 L) in sealed glass ampoules or in numerous small screw cap glass bottles closed with teflon-coated septa (see section materials) without any headspace to avoid subtle but possible systematic isotope fractionation due to volatilization (25)

2. The backflush system is implemented to the GC-IRMS to allow diverting the GC effluent to a waste-line. This prevents that an excess of carbon (solvent, sample matrix, column bleeding) exhausts the oxidation power of the combustion reactor and/or reduces the lifetime of the filament in the ion-source. To this end, a counter-He-gas flow is applied which should be twice the GC effluent flow plus an additional 0.4 mL/min to sustain a necessary constant flow to the IRMS) (more detailed description of the setup can be found in Jochmann and Schmidt (25) . The

0

0.2

0.4

0.6

0.8

1

-26.0

-25.5

-25.0

-24.5

-24.0

-23.5

-23.0

24/03/2006 02/07/2006 10/10/2006 18/01/2007 28/04/2007

SD

[‰]

δ13

CVP

DB

[‰]

date


15

backflush should, on the one hand, be implemented to avoid that non-target compounds enter the combustion unit and should, on the other hand, be switched on also when the system is in stand-by and when the reactor is in the process of reoxidation in order to provide maximum protection of the ion source of the IRMS.

3 With the lowest reported method detection limits to date (MDLs in the low µg/L range) purge and trap (P&T)(10) is the most effective known pre-concentration technique for on-line CSIA of volatile and semi-volatile compounds (33). It has been shown that for most substances, enrichment and application of the sample by P&T is not associated with significant carbon isotope fractionation (34). Besides 40 ml sample volumes also higher volumes (up to 100 mL) can be injected and purged. Alternative injection techniques for GC-IRMS are SPME (solid phase microextraction), headspace injection and liquid injection. An informative overview of achievable limits of detection is given in Jochmann and Schmidt, 2012, (25).

4 Besides determining the linear range of the ion source by applying different CO2-reference gas pulses, also the linear range for measurements of each target compound has to be determined with in-house standards. This test ensures that also the application of a sample into the GC-IRMS, chromatographic separation, and combustion to CO2 is size-independent for a distinct signal size range. The linear range is limited to both the side of the lowest concentration by the detection limit and on the higher end by the capacity of the column and the combustion capacity.

5 For quality control it is important to analyze working standards not only at the beginning and at the end but also in-between each sequence. It guarantees that bias in isotope values can be identified early on (and, possibly, corrected for retrospectively). In particular, isotope shifts towards lighter δ13C values can be the result of incomplete analyte conversion to CO2 due to an exhausted combustion reactor, which leads to incomplete combustion and preferential formation of 12CO2.

6 For GC-IRMS, isotope values for a target compound are given by the calculation of the relative difference of the isotopic composition to the CO2 monitoring gas which enters the IRMS at the beginning and at the end of each analysis. For the comparability of these data with other laboratories worldwide, it is recommendable that the monitoring gas be characterized against the internationally agreed zero value for δ13C. As the primary reference standard Pee Dee Belemnite is not available anymore, isotope laboratories normalize their monitoring gases and also their in-house standards to international carbon isotope reference materials which have been calibrated against the primary reference material Pee Dee Belemnite (VPDB). For the determination of the true δ13C value of the CO2 monitoring gas we use the following reference materials (all CO2 gases) RM 8562, RM 8563 and RM 8564, which can be purchased from the IAEA (International Atomic Energy Agency) or NIST (National Institute of Standards and Technology). Determination of the δ13Cvalue of the monitoring gas, in literature also termed reference gas, is commonly carried out in a dual-inlet system.

3.3 References

1. Meckenstock RU, Morasch B, Griebler C, Richnow HH. Stable isotope fractionation analysis as a tool to monitor biodegradation in contaminated acquifers. JContamHydrol 2004;75:215-55.

2. Hofstetter TB, Berg M. Assessing transformation processes of organic contaminants by compound-specific stable isotope analysis. TrAC Trends in Analytical Chemistry 2011;30:618-27.

3. Thullner M, Fischer A, Richnow HH, Wick LY. Influence of mass transfer on stable isotope fractionation. Applied Microbiology and Biotechnology 2013;97:441-52.


16

4. Elsner M, Jochmann MA, Hofstetter TB, et al. Current challenges in compound-specific stable isotope analysis of environmental organic contaminants. Anal Bioanal Chem 2012;403:2471-91.

5. Hatzinger PB, Böhlke JK, Sturchio NC. Application of stable isotope ratio analysis for biodegradation monitoring groundwater. Curr Opin Chem Biol 2013;24:542-9.

6. Meier-Augenstein W. Applied gas chromatography coupled to isotope ratio mass spectrometry. Journal of Chromatography A 1999;842:351–71.

7. Brand WA. High precision isotope ratio monitoring techniques in mass spectrometry. Journal of Mass Spectrometry 1996;31:225-35.

8. Benson S, Lennard C, Maynard P, Roux C. Forensic applications of isotope ratio mass spectrometry - A review. Forensic Science International 2006;157:1-22.

9. Schmidt TC, Zwank L, Elsner M, Berg M, Meckenstock RU, Haderlein SB. Compound-specific stable isotope analysis of organic contaminants in natural environments: a critical review of the state of the art, prospects, and future challenges. Anal Bioanal Chem 2004;378:283-300.

10. Hunkeler D, Meckenstock RU, Sherwood Lollar B, et al. A Guide for Assessing Biodegradation and Source Identification of Organic Ground Water Contaminants using Compound Specific Isotope Analysis (CSIA). Oklahoma, USA: US EPA; 2008. Report No.: PA 600/R-08/148 | December 2008 | www.epa.gov/ada.

11. Wolfsberg M, Van Hook WA, Paneth P. Isotope Effects in the Chemical, Geological and Bio Sciences. Dordrecht, Heidelberg, London, New York: Springer; 2010.

12. Mariotti A, Germon JC, Hubert P, et al. Experimental determination of nitrogen kinetic isotope fractionation: some principles; illustration for the denitrification and nitrification processes. Plant and Soil 1981;62:413-30.

13. Hoefs J. Theoretical and Experimental Principles. In: Wyllie PJ, ed. Stable isotope geochemistry. 3rd ed. Chicago: Springer-Verlag; 1987:1-25.

14. Qiu S, Eckert D, Cirpka OA, et al. Direct Experimental Evidence of Non-first Order Degradation Kinetics and Sorption-Induced Isotopic Fractionation in a Mesoscale Aquifer: 13C/12C Analysis of a Transient Toluene Pulse. Environmental Science & Technology 2013;47:6892-9.

15. Harrington RR, Poulson SR, Drever JI, Colberg PJS, Kelly EF. Carbon isotope systematics of monoaromatic hydrocarbons: vaporization and adsorption experiments. Org Geochem 1999;30:765-75.

16. Slater GF, Dempster HS, Lollar BS, Ahad J. Headspace analysis: A new application for isotopic characterization of dissolved organic contaminants. Environmental Science & Technology 1999;33:190-4.

17. Hirschorn SK, Dinglasan MJ, Elsner M, et al. Pathway dependent isotopic fractionation during aerobic biodegradation of 1,2-dichloroethane. Environ Sci Technol 2004;38:4775 - 81.

18. Elsner M, Zwank L, Hunkeler D, Schwarzenbach RP. A new concept linking observable stable isotope fractionation to transformation pathways of organic pollutants. Environ Sci Technol 2005;39:6896-916.

19. Elsner M. Stable isotope fractionation to investigate natural transformation mechanisms of organic contaminants: principles, prospects and limitations. J Environ Monit 2010;12:2005-31.

20. Gröning M. International stable isotope reference materials. In: de Groot PA, ed. Handbook of Stable Isotope Analytical Techniques, Vol 1 Amsterdam: Elsevier; 2009:874-906.

21. Coplen TB. Guidelines and recommended terms for expression of stable-isotope-ratio and gas-ratio measurement results. Rapid Communications in Mass Spectrometry 2011;25:2538-60.

22. Mosandl A. Authenticity Assessment: A Permanent Challenge in Food Flavor and Essential Oil Analysis. Journal of Chromatographic Science 2004;42:440-9.


17

23. Sessions AL. Isotope-ratio detection for gas chromatography. J SepSci 2006;29:1946-61. 24. Elsner M, Couloume GL, SherwoodLollar B. Freezing To Preserve Groundwater Samples and

Improve Headspace Quantification Limits of Water-Soluble Organic Contaminants for Carbon Isotope Analysis. Anal Chem 2006;78:7528-34.

25. Jochmann MA, Schmidt TC. Compound-specific stable isotope analysis. Cambridge: The Royal Society of Chemistry; 2012.

26. SherwoodLollar B, Hirschorn SK, Chartrand MMG, Lacrampe-Couloume G. An Approach for Assessing Total Instrumental Uncertainty in Compound-Specific Carbon Isotope Analysis: Implications for Environmental Remediation Studies. Anal Chem 2007;79:3469-75.

27. MAT 253 operating manual. 2002. 28. Schreglmann K, Hoeche M, Steinbeiss S, Reinnicke S, Elsner M. Carbon and nitrogen isotope

analysis of atrazine and desethylatrazine at sub-microgram per liter concentrations in groundwater. Anal Bioanal Chem 2013;405:2857-67.

29. Blessing M, Jochmann M, Schmidt T. Pitfalls in compound-specific isotope analysis of environmental samples. Analytical and Bioanalytical Chemistry 2008;390:591-603.

30. Meier-Augenstein W, Watt PW, Langhans CD. Influence of gas chromatographic parameters on measurement of C-13/C-12 isotope ratios by gas-liquid chromatography combustion isotope ratio mass spectrometry .1. Journal of Chromatography A 1996;752:233-41.

31. Jennings W, Mittlefehldt E, Stremple P, eds. Analytical Gas Chromatography Academic Press, 2nd edition; 1997.

32. Werner RA, Brand WA. Referencing strategies and techniques in stable isotope ratio analysis. Rapid Commun Mass Spectrom 2001;15:501-19.

33. Jochmann MA, Blessing M, Haderlein SB, Schmidt TC. A new approach to determine method detection limits for compound-specific isotope analysis of volatile organic compounds. Rapid Communications in Mass Spectrometry 2006;20:3639-48.

34. Zwank L, Berg M, Schmidt TC, Haderlein SB. Compound-specific carbon isotope analysis of volatile organic compounds in the low-microgram per liter range. Analytical Chemistry 2003;75:5575-83.


18

4 Mesocosm experimental protocols for mass balance/weathering analysis (UCPH)

4.1 Introduction

This protocol standardizes the extraction, sample preparation and analysis of Total Petroleum Hydrocarbons (TPH), Polycyclic Aromatic Compounds (PACs) and biomarkers from sediment and soil samples. With this protocol, samples evaluating remediation technologies of oil spills are standardized, allowing comparisons and summation across different remediation approaches. The protocol is based on current governmental standard methods (USEPA, NIST, CEN Bonn-OSInet, etc.) and published scientific data analysis approaches developed to take full advantage of state-of-the-art instrumentations for detailed analysis of the progress and extent of employed remediation technologies. A workflow of the protocol is depicted in Figure 10.

Figure 10 Diagram showing the content of this protocol, covering sample preparation, chemical analysis

and data treatment. The light-blue box represent the CHEMSIC and Pixel-based data treatment approach, which is still under development.

The protocol first describes two methods for extraction and cleanup of petroleum hydrocarbons from sediment and one for extraction of petroleum hydrocarbons from water. Furthermore, the protocol includes two analytical approaches, GC-FID and GC-MS, and data analysis approaches to reach to the assessment of the efficiency of the applied cleanup and remediation technologies. These are:

I. Total petroleum hydrocarbon (TPH) concentrations and the concentrations of petroleum hydrocarbons in four boiling point intervals: C6H6-nC10, nC10-nC15, nC15-nC20 and nC20-nC35. Data are obtained from GC-FID analysis.

II. Calculation of the total weathering degree. Data are obtained from quantifications of C30-α,β-hopane in the oil.

III. Chemical fingerprints of C0-C4 alkyl-substituted polycyclic aromatic compounds (PACs). IV. Calculation of process-specific diagnostic ratios to assess the contribution of different

weathering and degradation processes to the total removal. V. Integrated multivariate oil hydrocarbon fingerprinting using the CHEMSIC method. In this

method section of GC-FID or GC-MS chromatograms are treated as pixels in images in an automated procedure. First, chromatographic and detection artifacts such as retention time


19

shifts and changes in sensitivity are removed. Data are then scaled and normalized prior to multivariate analysis (e.g., principal component analysis) of the processed chromatograms.

The protocol lists quantification standards, certified reference materials and quality control procedures that would be beneficial to employ in standard analyses, when assessing cleanup and remediation technologies applied to oil spills. Furthermore, data analysis specific excel-templates are enclosed with the protocol.

4.2 Extraction and cleanup of petroleum hydrocarbons for GC analysis

4.2.1 Introduction

This chapter standardizes the extraction of petroleum hydrocarbons from sediment and water samples to be used for the assessment of applied remediation technologies.

Efficient extraction, preconcentration and cleanup are indispensable prior to the determination of PACs in sediments. There exists wide varieties of solvent extraction techniques commonly used for extracting hydrocarbons from sediments. Soxhlet extraction, which has high extraction efficiency, has been the standard method for preparing a solvent extract of solid matrices containing PACs. Other traditional extraction procedures include ultrasonication (Luque-Garcia and de Castro, 2003; Banjoo and Nelson, 2005), mechanical shaking (Berset et al., 1999), reflux with methanolic KOH and steam distillation. Modern instrumental techniques include supercritical fluid extraction (SFE) (Barnabas et al., 1995; Benner, 1998), pressurized liquid extraction (PLE; Dionex trade name ASE for accelerated solvent extraction) (Sporring and Bjorklund, 2004;Sporring et al., 2005) and microwave assisted extraction (MAE) (Ericsson and Colmsjo, 2000). Each technique has its own merits and the choice of extraction depends on several factors including capital cost, operating cost, simplicity of operation, amount of organic solvent required, sample throughput and availability of appropriate equipment and a standardized method. For sediment extraction, we display a simple technique with stepwise extraction and cleanup viz. ultrasonication, and a more sophisticated approach with simultaneous extraction and cleanup using pressurized liquid extraction (PLE). For oil in water, we describe the simple liquid/liquid extraction.

Internal Standards used in sample preparation

The employed analytical platform and the purpose of the analysis defines the type of standard (internal and external) that should be used. This protocol distinguish between GC-FID analysis for TPH quantification, GC-MS analysis for PAC and biomarker quantification and GC-MS analysis for weathering specific semiquantitative diagnostic ratio calculations.

Internal standards for GC-FID analysis for TPH quantification

LMW internal standards chlorofluorobenzene, bromobenzene or chlorobenzene (250µg/mL in pentane giving 10µg/mL in GC-vial when 200µl is added to sample)

HMW internal standards Ortho-terphenyl, squalane or squalene (250µg/mL in pentane giving 10µg/mL in GC-vial when 200µl is added to sample)

The solutions are prepared in pre-cleaned 5 mL volumetric flasks, transferred to 8 ml amber glass vials, and stored in the freezer at -20 °C until use.

Internal standards for biomarker quantification (GC-MS)

Since weathering of oil only rarely affect the content of C30-α,β-hopane, this compound is often used as an internal preserved reference compound for the assessment of specific oil compound removal within the oil. If exact quantifications of C30-α,β-hopane and other biomarkers is needed (ex. for total


20

weathering degree calculation), we suggest to use the thermally unstable (and therefore rarely seen in oil) isomer C30-α,α-hopane as an internal standard, referring to the suggestion given by Nytoft and Bojesen-Koefoed (2001).

Internal standards for semiquantitative diagnostic isomer ratio calculations

Since this method is based on changes in selected C0-C4-alkylated PACs relative to internal preserved reference compounds (such as C30-α,β-hopane or highly alkylated PACs) and the original undegraded oil, internal standards as such are not needed in the sample preparation. However, if a need for quantification of PACs exists, we provide a list of recommended standards and concentrations below (Table 2). It is recommended to always use internal and recovery standards, if a need for quantification should arise, or if a need for evaluation of extraction efficiency, evaporative losses during sample prep or discrimination within the analytical instrument.

Internal standards for GC-MS analysis for PAC quantification

Internal standards: Naphthalene-d8, acenaphthene-d10, fluorene-d10, dibenzothiophene-d8, phenanthrene-d10, pyrene-d10, chrysene-d12, benzo(k)fluoranthene-d12 and benzo(g,h,i)perylene-d12 (8µg/ml in isooctane giving 0.32µg/mL in final sample). Label the solution (RED 3) Recovery standards: acenaphthylene-d8, anthracene-d10, fluoranthene-d10, benz(a)anthracene-d12, benzo(a)pyrene-d12 and indeno(1,2,3-c,d)pyrene-d12 (8µg/ml in isooctane giving 0.32µg/mL in final sample). Label the solution (RED 4)

The solutions are prepared in pre-cleaned 5 mL volumetric flasks, transferred to 8 ml amber glass vials, and stored in the freezer at -20 °C until use.

Table 2 List of the most appropriate internal and recovery standards for the quantification of 19 selected PACs.

Internal standard PACs Recovery standard Naphthalene-d8 Naphthalene, Acenaphthylene Acenaphthylene-d8 Acenaphthene-d10 Acenaphthene Anthracene-d10 Fluorene-d10 Fluorene Anthracene-d10 Dibenzothiophene-d8 Dibenzothiophene Anthracene-d10 Phenanthrene-d10 Anthracene, phenanthrene Anthracene-d10 Pyrene-d10 Pyrene, fluoranthene Fluoranthene-d10 Chrysene-d12 Chrysene, benzo(a)anthracene Benzo(a)anthracene-d12 Benzo(k)fluoranthene-d12

Benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(e)pyrene, benzo(a)pyrene, perylene

Benzo(a)pyrene-d12

Benzo(g,h,i)perylene-d12 Dibenzo(a,h)anthracene, Benzo(g,h,i)perylene, indeno(1,2,3-c,d)pyrene

Indeno(1,2,3-c,d)pyrene-d12

CHEMSIC

CHEMSIC is based on image analysis of selected chromatograms relative to the original undegraded oil, and therefore internal standards are not needed.


21

4.2.2 Extraction from sediment

Sampling

The importance of obtaining a representative sample for analysis cannot be overemphasized. Without it, results may be meaningless or even grossly misleading. Sampling is particular crucial where a heterogeneous material is to be analyzed. It is vital that the aims of the analysis are understood and an appropriate sampling procedure adopted. In some situations, a sampling plan or strategy may need to be devised to optimize the value of the analytical samples such as sediment samples. It is strongly recommended to examine relevant literature for standardized protocols for the specific compartment and analytes. For shipment of samples, at least three times the amount needed for an analysis should be sampled, i.e. dryweight and PAC extraction= 3x(3x5 g+5 g)=60 g

Determination of dry weight:

1. Representative sampled sediment samples are homogenized thoroughly and large fragments are removed.

2. 3 × 5 gram of material is weighed into alumina pans or ceramic crucibles (the exact weight is noted) and placed in an oven at 105 °C overnight. The cooled samples are weighed and the exact weight noted.

Extraction model 1: Stepwise ultrasonication extraction and silica gel cleanup

Ultrasonication procedures have been compared with the Soxhlet method for extraction of PACs in samples, with short time ultrasonic extractions producing equivalent or better recoveries than 6-8 hours of Soxhlet extraction (Lundstedt et al., 2000). Since the equipment for ultrasonication is simple, easy to operate and relatively fast, ultrasonication is often a preferred extraction method. To separate the analytes from the sample matrix that could interfere with the analysis, a column cleanup procedure (with silica and sodium sulfate) is applied.

Sample extraction and cleanup

Extractions are performed with 2 blank extractions (procedure blanks), 1 extraction of an appropriate reference sample (e.g. NIST SRM 1941b - Organics in sediment) used for proficiency testing and sample extractions where 1 of these are extracted in duplicate.

1. Accurately weighed known amounts (≈5 gram to reach a final concentration of about 2000 µg oil/mL solvent) of wet sediment/soil is ground thoroughly with 5 gram of hydromatrix.

2. Each sample is subsequently transferred quantitatively to a 100 ml Erlenmeyer flask and internal standard solution is added on to the top of the sample. The sample is left to stand for sorption of the added standards, capped for at least 20 min.

3. 30 ml solvent (n-hexane and acetone, 1:1, v/v) is added to the Erlenmeyer flask and the suspension is ultrasonicated at room temperature for 10 min, after which the solution from the settled mixture is decanted. The remaining suspension is washed with approximately 2 ml solvent and withdrawn using a Pasteur pipette.

4. The extraction process is subsequently repeated twice (a total of three extractions per sample) and the extracts obtained from the ultrasonication are filtered (in-between each extraction) to remove sediment particles by use of a glass funnel with filter paper directly into a 250 ml round bottom flasks.

5. 1 ml of isooctane is added to the combined extract as keeper and the combined extracts are evaporated slowly (maximum 5 ml/min) to approximately 1 mL on a rotary evaporator


22

Chromatographic column cleanup of extracts for GC analysis

1. A small glass column (1.4 id × 7 cm) is prepared per extract (Figure 11). Each column is dry packed from its base with a glass filter, 1 gram of activated silica (180°C for 20 hours) and 0.2 cm of anhydrous sodium sulfate. The column is covered with alumina foil until use.

Figure 11 Illustration of the preparation of a small chromatographic column, packed from its based with a

glass filter, activated silica and anhydrous sodium sulfate

2. The column is conditioned with 5 ml of n-hexane. 3. Before the column runs dry, the extract is transferred quantitatively to the top of the column

using a Pasteur pipette. Then ½ ml of n-hexane is transferred to the Erlenmeyer flask using a clean pipette and the first pipette is used to wash the flask and transfer the wash to the top of the column. The wash procedure is repeated three times to obtain a quantitative transfer of the extract.

4. The column is washed with 4 ml of n-hexane, which is discarded. 5. The column is subsequently eluted with 4 ml of a 1:1 mixture of n-hexane:dichloromethane

into a 5 ml volumetric flask. 6. If needed, recovery standard solution is added. 7. The volume of the flask is adjusted to 5 ml with n-hexane:dichloromethane (1:1), and stored

in amber glass vials at -20 °C until analysis.

Extraction model 2: Simultaneous extraction and cleanup using pressurized liquid extraction (PLE)

Pressurized liquid extraction (PLE) is a fast and efficient alternative to traditional extraction methods for many types of contaminants. The technique, which is based on extraction at elevated pressure and temperature, can be used with the same solvents as those used for Soxhlet extraction. The method used in this protocol is based on the development by Lundstedt, Haglund et al. (2006), and Boll, Christensen et al. (2008).

Preparations of extraction components

1) PLE extraction cells (as depicted in Figure 12)

Silica

Condition with 5 ml n-hexane Transfer the extract quantitatively Elute with 4 ml n-hexane (discard) Elute with 4 ml 1:1 n-hexane : dichloromethane

to 5 ml volumetric flask Add 200 µL spike standard to the volumetric flask Adjust the flask to 5 ml

Sodium sulfate

Glass filter


23

Figure 12 PLE extraction cell components.

a) Separate the extraction cell completely and remove old sample material form the cylinder (A) b) Clean the cylinder inside with a tissue c) Wash all parts (A, B, C, D and E) in detergent d) Rinse all parts (A, B, C, D and E) in tap water e) Rinse all parts (A, B, C, D and E) in deionized water f) Sonicate small parts (D and E) for 5 minutes in methanol, the larger parts (A, B and C) are

rinsed with methanol g) Sonicate small parts (D and E) for 5 minutes in pentane:acetone (1:1), the larger parts (A, B

and C) are rinsed with pentane:acetone (1:1) h) Let all parts dry on alumina foil

2) Cellulose filters a) Sonicate cellulose filters two times 5 minutes with dichloromethane, and leave to dry on

alumina foil 3) Hydromatrix:

a) Hydromatrix (e.g. Agilent Technologies) is placed in a thin layer (max 1 cm) on alumina foil and burned in a muffle oven at 400°C overnight. Cooled hydromatrix is stored in red cap bottles in a dry environment (e.g. desiccator).

b) Sodium sulfate (pro analysis, Merck) is dried in an oven at 85°C overnight. Dry sodium sulfate is stored in red cap bottles in a dry environment (e.g. desiccator).

4) Silica gel: a) Silica gel 60 (0.063-0.200 mm, Merck) is activated at 180°C overnight. Activated silica gel is

stored in red cap bottles in a dry environment (e.g. desiccator). 5) Copper powder (used for removal of elemental sulfur in sediments)

a) Fine copper powder (Fluka, 99% pro analysis) is activated with concentrated hydrochloric acid immediately before use and rinsed with milliQ water followed by pentane:acetone (1:1) to remove the hydrochloric acid. Residual hydrochloric acid will lead to degradation of PACs.

6) Ottawa sand:


24

a) Ottawa sand is placed on alumina foil, and burned in a muffle oven at 400°C overnight. The layer of Ottawa sand on alumina foil should be no more than 2 cm in thickness. The Ottawa sand is cooled with alumina cover and when cold, stored in red cap bottles in a dry environment (e.g. desiccator).

Sediment and soil extraction procedures with PLE

Extractions are performed batch-wise: A batch of samples consists of 15 extractions: 2 blank extractions (procedure blanks), 1 extraction of an appropriate reference sample (e.g. NIST SRM 1941b - Organics in sediment) used for proficiency testing, 11 sample extractions where 1 of these are extracted in duplicate.

Integrated extraction and cleanup is performed by pressurized liquid extraction (PLE) in this case with a Dionex ASE 200 accelerated solvent extractor.

Packing of PLE cells

i) A 33 mL stainless steel extraction cell is packed from the bottom with two cellulose filters, 4 gram of activated silica gel and 4 gram of wet activated (acid treated) cupper (Note: activated cupper is added in order to remove elemental sulfur – and is only necessary for anoxic sediment samples. Thus, activated cupper is not added in the analysis of soil samples).

ii) Accurately weighed known amounts (≈5 gram) of wet sediment/soil is ground thoroughly with 5 gram of hydromatrix. The homogenized and dried sample is transferred quantitatively to the extraction cell.

iii) Internal standard is added directly on top of the sample. iv) The remaining cell volume is filled with an inert matrix (burned Ottawa sand) and the

extraction cells are closed and inserted in the PLE-autosampler. v) The PLE-program elaborated in Figure 13 is used for the extractions. vi) Each PLE cell is extracted two times (into two separate glass vials) vii) The two extracts for each sample are concentrated to 15 mL under a gentle stream of

nitrogen and heat (40°C) and the second extract is transferred quantitatively to the first extract. This extract (added 2 ml isooctane as keeper) is then concentrated to 2 mL under a gentle stream of nitrogen and transferred quantitatively to a 5 mL volumetric flask.

viii) QS with n-pentane:dichloromethane (90:10, v/v) and store in amber glass vials at -20°C until analysis.


25

Figure 13 Scheme of the PLE extraction protocol

4.2.3 Extraction and cleanup of oil from water and oil/water emulsions

4.2.3.1 USEPA method 3510C/8270D

Referring to USEPA method 3510C/8270D, organic compounds can be extracted in a simple manner by three successive liquid/liquid extractions in dichloromethane. For TPH analysis measures should be taken to reach oil concentrations of 2-2000 mg/L in the final dichloromethane extract. Adding 10 ml dichloromethane to 1 liter of water is appropriate. Internal standard is added to the water, before adding dichloromethane.

The extracts obtained from the liquid/liquid extraction are filtered to remove particles by use of a glass funnel with filter paper directly into a 250 ml round bottom flasks. 1 ml of isooctane is added to the combined extract as keeper and the combined extracts are evaporated slowly (maximum 5 ml/min) to approximately 1 mL on a rotary evaporator.

Chromatographic column cleanup of water extracts for GC analysis

1. A small glass column (1.4 id × 7 cm) is prepared per extract. Each column is dry packed from its base with a glass filter, 1 gram of activated silica (180°C for 20 hours) and 0.2 cm of anhydrous sodium sulfate. The column is covered with alumina foil until use.

2. The column is conditioned with 5 ml of n-hexane. 3. Before the column runs dry, the extract is transferred quantitatively to the top of the column

using a Pasteur pipette. Then ½ ml of n-hexane is transferred to the flask using a clean pipette and the first pipette is used to wash the flask and transfer the wash to the top of the column. The wash procedure is repeated three times to obtain a quantitative transfer of the extract.

Pressure 1500 psi. Preheat time 2 min Static time 5 min Flush volume: 70% Purge time: 60 sec

-1- 4g activated Silica

and if needed 4g wet acidic Cu

-2- 5g sediment or soil

5g hydromatrix Ground and mixed

-3- Internal standard is

added on top of sample

-4- Cell volume filled with inert matrix

(Ottawa sand)

Analysis From PLE after 2 full extractions 2×40 ml evaporated to <2ml Transfer to 5 ml volumetric flask Add recovery standard, QS with solvent


26

4. The column is washed with 4 ml of n-hexane, which is discarded. 5. The column is subsequently eluted with 4 ml of a 1:1 mixture of n-hexane:dichloromethane

into a 5 ml volumetric flask. 6. Recovery standard is added. 7. The volume of the flask is adjusted to 5 ml with n-hexane:dichloromethane (1:1), and stored

in amber glass vials at -20 °C until analysis.

4.2.3.2 Alternative water extractions

Pentane as organic phase and increasing ionic strength of water-phase

Oil in water is extracted as per the USEPA method 3510C/8270D, however, salt is added to the water. This increases the ionic strength of the water and thereby more strongly force analytes into the organic phase. Adding pentane instead of dichloromethane gives an organic phase in the top instead of in the bottom of the two-layer system. The procedure is otherwise the same as the above described.

Extraction with solid phase extraction

Oil in water can be extracted using solid phase extraction. We have successfully used Waters HLB columns 6cc, 500mg sorbent for extracting PACs from samples up to 500 ml. The SPE cartridges are preconditioned by washing 3 times with 3 ml methanol and flushing 3 times with 5ml deionized water. Water (added internal standard in methanol) is extracted with a flow of approximately 10 ml per minute. After extraction, PACs are eluted 3 times with 3ml of methanol. The combined eluate is then concentrated to <5ml under a gentle stream of nitrogen. The extract is added recovery standard (in methanol) and adjusted to 5ml.

4.3 GC-FID method for TPH analysis

4.3.1 Introduction

Total Petroleum Hydrocarbon (TPH) analysis by GC-FID, offers evaluation of the total content of petroleum hydrocarbons in the boiling point range of 80°C - 490°C (from benzene to nC35). Analysis with GC-FID aid to distinguish different types of oil (i.e. crude oil, gasoline, diesel, HFO etc.) in the sample matrix. GC-FID analysis can be used to quantify all pure hydrocarbons in a sample since they, in theory, have same response factors in the FID detector. The analysis can be performed to estimate concentration of the oil, before further more detailed analysis by e.g. GC-MS. Because of the low resolution of the GC-FID, separation of compounds is difficult, and quantification of individual compounds can be associated with high uncertainty or even impossible. Results from GC-FID will be recorded as either: Fractions of C6H6-nC10, nC10-nC15, nC15-nC20 and nC20-nC35; or as TPH (mg/kg DW).

4.3.2 Sample preparation

Samples should be prepared according to sample preparation protocol, and preferably be in n-pentane:dichlormethane (90:10, v/v), alternatively in another water-free low boiling point solvent (e.g DCM, pentane or hexane). To avoid overestimation of sample content, internal standards should be limited to only the ones listed in the extraction protocol for GC-FID analysis for TPH quantification. If internal and recovery standards are added to the sample, response from a standard containing internal and recovery standards should be subtracted the sample response in order to not overestimate the TPH content.


27

4.3.3 Standards for GC-FID analysis for TPH quantification

For changes on oil composition, where the start oil is not available or for weathered samples, results should be given as the change in content of four boiling point fractions (BPF).

1. Standards used for quantification of fractions. a. BPF1: C6H6-nC10: Toluene b. BPF2: nC10-nC15: Average response of nC12 and nC14. c. BPF3: nC15-nC20: Average response of nC16 and nC18. d. BPF4: nC20-nC35: Average response of nC24, nC28, nC32 and nC32.

2. Total Petroleum Hydrocarbons (TPH) a. TPH is calculated as the sum of the individual four fractions, quantified by the above-

mentioned method.

4.3.4 Calibration

1. Five dilutions in the range from 0.1 to 20 mg/L of each of the n-alkanes in pentane should be returning signals in a linear fashion, through origin. Once obtained, linearity can be checked daily with a single control sample (acceptance ±10%). Remember internal standard in quantification standards.

4.3.5 Internal standards

1. LMW internal standards chlorofluorobenzene, bromobenzene or chlorobenzene (250µg/mL in pentane giving 10µg/mL in GC-vial when 200µl is added to sample)

2. HMW internal standards Ortho-terphenyl, squalane or squalene (250µg/mL in pentane giving 10µg/mL in GC-vial when 200µl is added to sample)

4.3.6 Instrument settings and equipment

Standard performance criteria

1. Solvent blanks should be run at least for every ten samples. 2. Standard reference material (sediment and oil) should be run at least for every ten samples 3. Injected standards should give baseline separation of n-alkanes. 4. Discrimination of heavy analytes should be avoided. Check that standard response of nC35 is

at least 80% of that of nC20.

Inlet

1µL injected in splitless mode, injector temperature: 325°C

Column

40 m ZB-5 (0.18 mm ID, 0.25 µm film thickness) capillary column. Hydrogen should be used as carrier gas at 3.0 ml/s. An example of a temperature program is: Initial temperature: 45°C hold for 2.5 min, 20°C/min to 320°C (hold for 10 min), total analysis time 26.5 min.

4.3.7 GC-FID data analysis

Boiling point intervals

Since the unresolved complex mixture (the compounds responsible for the rise in baseline after 6 minutes as seen in Figure 14) contains hydrocarbons that should be quantified in relation to


28

weathering, data are quantified by integrating the entire area under the curve in the specific BPFs until baseline at zero.

Figure 14 GC-FID chromatograms showing examples of a weathered and an unweathered crude oil

analyzed by GC-FID and the sections of the four boiling point fractions (BPFs).

4.4 GC-MS method for PAC and biomarker analysis

4.4.1 Introduction

GC-MS analysis is performed to assess the relative quantity of PACs and biomarkers in oil, water and sediment samples. The method is sensitive and can be used for full quantification of single PACs in the concentration range from 2-2,500 µg/kg (by addition of appropriate internal and recovery standards, see table 3.3.10) as well as for diagnostic purposes using semi-quantitative approaches.

The CHEMSIC (Chemometric analysis of sections of selected ion chromatograms) data treatment methodology is based on the publication by Christensen, Hansen et al. (2005) and Gallotta and Christensen (2012), run on an Agilent 6890N/5975 GC-MS.

4.4.2 Sample preparation

Samples should be prepared according to sample preparation protocol, and preferably be in n-pentane:dichloromethane (90:10, v/v), alternatively in another water-free low boiling point solvent (e.g dichloromethane, pentane or hexane).

4.4.3 Standards

4.4.3.1 GC-MS analysis for PAC quantification and diagnostic ratio calculations

1) Quantification of 19 parent PACs including the 16 US-EPA PACs a) Exhaustive list of PACs with authentic standards: Naphthalene, acenaphthene,

acenaphthylene, dibenzothiophene, anthracene, phenanthrene, fluorene, fluoranthene,


29

pyrene, chrysene, benzo(a)anthracene, benzo(a)pyrene, perylene, benzo(g,h,i)perylene, benzo(k)fluoranthene, dibenzo(a,h)anthracene, benzo(e)pyrene, benzo(b)fluoranthene and indeno(1,2,3-c,d)pyrene.

b) Single PAC concentration range: 2.5 – 2000 µg/kg sample 2) 11 point standard curves are made (see excel sheet): 3-6-12-25-50-100-300-500-1000-1500-2000

ng/mL; Internal standards (5 – 200 ng/mL); and Recovery standards (150 ng/mL). Naphthalene-d8, acenaphthene-d10, fluorene-d10, dibenzothiophene-d8, phenanthrene-d10, pyrene-d10, chrysene-d12, benzo(k)fluoranthene-d12 and benzo(g,h,i)perylene-d12 are used as internal standards, while acenaphthylene-d8, anthracene-d10, fluoranthene-d10, benz(a)anthracene-d12, benzo(a)pyrene-d12 and indeno(1,2,3-c,d)pyrene-d12 are used as recovery standards.

Preparation of standards

3) Prepare Quantification stock solution Label RED1: consisting of 19 PACs. The stock solution contain approximately 4 µg/mL of each individual PAC (see excel preparation sheet)

4) Prepare a 10 time dilution of the Quantification stock solution - Label RED2: The solution contain approximately 0.4 µg/mL of each individual PAC (see excel preparation sheet)

5) Prepare internal stock solution: Naphthalene-d8, acenaphthene-d10, fluorene-d10, dibenzothiophene-d8, phenanthrene-d10, pyrene-d10, chrysene-d12, benzo(k)fluoranthene-d12 and benzo(g,h,i)perylene-d12: Label RED3: 8 µg/mL (see excel preparation sheet)

6) Prepare recovery stock solution: acenaphthylene-d8, anthracene-d10, fluoranthene-d10, benz(a)anthracene-d12, benzo(a)pyrene-d12 and indeno(1,2,3-c,d)pyrene-d12: Label RED4: Approx. 8 µg/mL (see excel preparation sheet)

7) Prepare the 11 standard solutions from the four Red solutions: RED1, RED2, RED3 and RED4 using the “Preparation of Quant std” excel sheet (see excel preparation sheet). The solutions are prepared in pre-cleaned 5 mL volumetric flasks, transferred to 8 ml amber glass vials, and stored in the freezer at -20 °C until use.

4.4.3.2 Biomarker quantification (GC-MS)

C30-α,β-hopane is quantified in oil samples where C30-α,α-hopane is used as an internal standard.

Preparation of standards

Five-point calibration curve in desired concentration levels is constructed with varying concentrations of C30-α,β-hopane and constant concentrations of the internal standard C30-α,α-hopane.

4.4.4 Instrument settings and equipment

Tuning and performance of the GC-MS system is monitored through daily analyses of a standard tuning sample (following instructions in EPA method 8270D). The tuning sample comprise 50 ng of each of DFTPP, DDT, benzidine and pentachlorophenol in DCM. Tuning and performance samples can be purchased at e.g. http://www.accustandard.com (M-TS-625 or M-TS-625-20X).

Tuning

The mass spectrometer should be able to produce the mass spectrum for the compound DFTPP that meets the criteria of Table 3 where 1 µl of the standard GC-MS tuning sample is injected.


30

Table 3 Ion DFTPP key ions and ion abundance criteria. (USEPA 1998) See USEPA method 8270D, Table 3 for DFTPP acceptance criteria.

Mass Ion abundance criteria 51 10-80% of base peak 68 <2% of mass 69 70 <2% of mass 69 127 10-80% of base peak 197 <1% of mass 198 198 base peak, or >50% of mass 442 199 5-9% of mass 198 275 10-60% of base peak 365 >1% of mass 198 441 Present but <24% of mass 443 442 base peak, or >50% of mass 198 443 15-24% of mass 442

Performance

Inlet liner and column performance is monitored with DDT, benzidine and pentachlorophenol. Benzidine and pentachlorophenol should be present at their normal responses (based on the experience of the laboratory) and should not show peak tailing. The degradation of DDT to DDE and DDD should not be greater than 20%. Figure 15 illustrates a chromatogram injected to a GC-MS operating in ideal conditions (a) and a chromatogram of a sample run with GC-MS in bad conditions (b). Table 4 shows the key ions of compounds that are used for monitoring the chromatographic conditions.

Figure 15 Chromatograms of injections of a mix of DFTPP, Pentaclorophenol, Benzidine and DDT from a

GC-MS operating in ideal conditions (a) and a GC-MS in bad conditions (b).

10.00 15.00 20.00 25.00 30.00 35.00 40.00

500000

1000000

1500000

2000000

2500000

3000000

3500000

4000000

4500000

5000000

5500000

6000000

6500000

7000000

7500000

Time-->

Abundance

TIC: DFTPPStand.D\data.ms

10.00 15.00 20.00 25.00 30.00 35.00 40.00

500000

1000000

1500000

2000000

2500000

3000000

3500000

4000000

4500000

5000000

5500000

6000000

6500000

7000000

7500000

8000000

8500000

9000000

9500000

1e+07

1.05e+07

1.1e+07

1.15e+07

1.2e+07

Time-->

Abundance

TIC: DFTPPBad.D\data.ms

Pent

achl

orop

heno

l

DFTP

P

Benz

idin

e DD

T

DFTP

P

Benz

idin

e

DDT


31

Table 4 Ions of the key compounds responsible for the control of chromatographic conditions.

Compounds Primary ion Secondary ions

Benzidine 184 92, 185 Pentaclorophenol 266 264, 268 4,4’-DDT 235 237, 165 4,4’-DDE 246 148, 176 4,4’-DDD 235 237, 165

After injection of many samples, the peaks of Benzidine and Pentachlorophenol may show tailing or disappear completely. The Benzidine is a reactive compound and its peak will tail if the liner is not inert. The DTT will decompose in the injector, if this contains any reactive sites (e.g. silanol interactions).

If these symptoms occur, replace the injector liner and cut about 15-30 cm of the start of the column. At the same time, the injector septum can be changed. While the injector is reheated to the original temperature, set the temperature to 10°C below maximum temperature of the column and hold it for 5 minutes in order to remove residual contaminants.

Reference materials and standards

Quantification of analytes should meet recommendations associated with standard reference materials such as NIST SRM 1582 (Crude oil) and NIST SRM 1941 (Organics in sediment).

Inlet

1µL injected in pulsed splitless mode, injector temperature: 315°C

Column

60 m ZB-5 (0.25 mm ID, 0.25 µm film thickness) capillary column. Helium should be used as carrier gas at 1.1 ml/s. Initial temperature: 40°C hold for 2 min, 25°C/min to 100°C, and 5°C/min to 315 (hold for 13.4 min), total analysis time 60.8 min. Transfer line should be 315°C.

MS

Ion source and quadropole temperatures should be 230°C and 150°C respectively. The ion source should be operating in electron ionization mode. For PAC and diagnostic ratio analysis, twelve groups of ions with 13 ions in each should be monitored in SIM mode, with a dwell time of 25 ms, giving 2.81 scans/s. Number of ions should be consistent between groups, to avoid differences in scanning frequency, see Table 5 for ion and group details.


32

Table 5 List of compounds, SICs and corresponding groups of GC-MS/SIM (Gallotta and Christensen 2012)

Group number 1 2 3 4 5 6 7 8 9 10 11 12 nCx 0 13 14 15 16 18 19 21 23 24 27 32 n-alkane (RT of n-alkane to be used for group division) SIC Compound name 83 x x x x x x x x x x x x n-Alkyl cyclo hexanes 85 x x x x x x x x x x x x Alkanes 105 x x x x x x x x x x x x Alkyl toluenes 123 x x x x x x Sesquiterpanes 128 x Naphthalene 134 x Benzo(b)thiophene 136 x d8-Naphthalene 138 x Decalin 142 x C1-Naphthalenes 148 x x C1-Benzo(b)thiophenes 152 x x x C1-Decalins, Acenaphthylene 154 x x x Acenaphthene 156 x C2-Naphthalenes 160 x d8-Acenaphthylene 162 x x C2-Benzo(b)thiophenes 164 x x d10-Acenaphthene 166 x x x C2-Decalins, Fluorene 168 x x x Dibenzofuran 170 x x C3-Naphthalenes 176 x x C3-Benzo(b)thiophenes, d10-Fluorene 178 x Phenanthrene, Anthracene 180 x x x x C3-Decalins, C1-Fluorenes 182 x x x C1-Dibenzofurans 184 x x x C4-Naphthalenes, Dibenzothiophene 188 x d10-Phenanthrene, d10-Anthracene 190 x x C4-Benzo(b)thiophenes 191 x x x x Tricyclic terpanes, Hopanes 192 x x x C1-Phenanthrenes/anthracenes, d8-Dibenzothiophene 194 x x x x x x C4-Decalins, C2-fluorenes 196 x x x C2-Dibenzofurans 198 x x C1-Dibenzothiophenes 202 x x x Fluoranthene, Pyrene 206 x x x C2-Phenanthrenes/anthracenes 208 x x C3-Fluorenes 212 x x C2-Dibenzothiophenes, d10-Fluoranthene, d10-Pyrene 216 x x C1-Fluoranthenes/pyrenes 217 x x x x x Steranes 218 x x x Steranes 220 x x x C3-Phenanthrenes/anthracenes 226 x x x C3-Dibenzothiophenes 228 x Benzo(a)anthracene, Chrysene 230 x x C2-Fluoranthenes/pyrenes 231 x x x Triaromatic steranes 234 x x x C4-Phenanthrenes/anthracenes, Retene, Benzonaphtothioph 240 x x x C4-Dibenzothiophen., d12-Benzo(a)anthr., d12-Chrysene 242 x x C1-Chrysenes 244 x d14-p-Terphenyl 248 x x C1-Benzonaphtothiophenes 252 x x 5 Rings PACs 256 x C2-Chrysenes 264 x x d12-Benzo(k)fluoranth., d12-Benzo(a)pyrene, d12-Perylene 270 x x C3-Chrysenes 276 x 6 Rings PACs 278 x 6 Rings PACs 288 x d12-Indeno(1,2,3-cd)pyrene, d12-Benzo(g,h,i)perylene Total 13 13 13 13 13 13 13 13 13 13 13 13


33

GC-MS sequence

This section describes the samples that are included in the GC-MS analysis of one extraction batch. The samples include those from the PLE extraction as well as quantification standards and quality control samples to control instrumental, procedural and other variations. The extended sequence is needed in order to conduct automated chemometric data analysis (CHEMSIC). If manual integration and data processing is used, the wide extent of control samples can be reduced.

2) The sequence for analysis of one extraction batch (See extraction protocol) includes a) Instrument quality control sample (DFTPP mix). Analyzed as first and last sample in the

sequence. b) Two procedure blank samples (Quality control, limit of detection (LOD) and limit of

quantification (LOQ)) c) Reference samples d) Sample extracts e) Duplicate of one of the sample extracts f) Quantification standards (1-11). Standard6 (0.1 µg/mL) is analyzed in-between every five

extracts from the sample extraction batch g) Total petroleum hydrocarbon mixture (TPH) (500 ng/mL for each individual compound) is

analyzed in-between every five extracts from the sample extraction batch. The mixture is used for quality control during the analysis in order to detect discrimination in the inlet liner (the relative ratios between alkanes should remain constant in the RT window relevant for the analytes of interest)

h) Solvents. These samples are analyzed in-between every five extracts from the sample extraction batch and used for quality control (system contamination)

i) An example of a sequence: 1: DFTPP, 2: DCM, 3: Standard6, 4: TPH, 5: Extract2, 6:Blank1, 7: Extract8, 8:Extract3, 9: Standard1, 10: DCM, 11: Standard6, 12: TPH, 13: Standard6, 14: ReferenceMaterial, 15: Extract2, 16: Standard8, 17: Standard11, 18: DCM, 19: Standard6, 20: TPH, etc., etc., etc., final: DFTPP.

4.4.5 GC-MS data analysis

Manual quantification

1) The relevant peaks are integrated in ChemStation, standard curves and quantifications are made in Microsoft Excel.

2) The concentrations of PACs in extracts are calculated batch-wise (one batch at a time) using the internal standard approach and the replicate analysis of one specific standard solution (standard6)

3) The internal standard method is used and the deuterated PACs added directly to the sample are used as internal standards. A list of the most appropriate internal standards for quantification of the 19 PACs are listed in table 1.

Total weathering degree data analysis

The total weathering degree is calculated as the ratio of biomarker concentration in the source oil relative to that of the weathered oil.

Equation 1. 𝑃𝑃(%) = �1 −𝐶𝐶𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 𝑖𝑖𝐻𝐻 𝑠𝑠𝐻𝐻𝑠𝑠𝑠𝑠𝑠𝑠𝐻𝐻 𝐻𝐻𝑖𝑖𝑜𝑜

𝐶𝐶𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 𝑖𝑖𝐻𝐻 𝑤𝑤𝐻𝐻𝐻𝐻𝑤𝑤ℎ𝐻𝐻𝑠𝑠𝐻𝐻𝑒𝑒 𝐻𝐻𝑖𝑖𝑜𝑜� ∙ 100%


34

Chemical fingerprints

The relevant peaks are integrated in ChemStation, and the results are exported to excel. To aid for the identification of peaks and groups of peaks, please refer to the CEN guideline (CEN/TR 2012).

Alkane fingerprints

Semiquantitative results of alkanes are reported as areas (A) of alkanes normalized to the area of a sufficiently high-recovered high boiling point compound that is robust to weathering. In some cases linear alkanes, such as nC23 will be sufficient, but in cases of severe weathering, compounds proven robust to weathering such as C30-α,β-hopane should be used. All n-alkanes-areas are normalized to the areas of the selected normalization compound, and the result can hereafter be normalized to the same ratio in the original oil, whereby percent removal from the original oil is calculated (see Equation 2).

Equation 2.

𝐴𝐴𝐻𝐻𝑛𝑛𝑥𝑥(𝑤𝑤𝐻𝐻𝐻𝐻𝑤𝑤ℎ𝐻𝐻𝑠𝑠𝐻𝐻𝑒𝑒 𝑠𝑠𝐻𝐻𝑠𝑠𝐻𝐻𝑜𝑜𝐻𝐻) 𝐴𝐴𝐻𝐻𝑛𝑛23(𝑤𝑤𝐻𝐻𝐻𝐻𝑤𝑤ℎ𝐻𝐻𝑠𝑠𝐻𝐻𝑒𝑒 𝑠𝑠𝐻𝐻𝑠𝑠𝐻𝐻𝑜𝑜𝐻𝐻)�

𝐴𝐴𝐻𝐻𝑛𝑛𝑥𝑥(𝑠𝑠𝐻𝐻𝑤𝑤𝐻𝐻𝐻𝐻𝑤𝑤ℎ𝐻𝐻𝑠𝑠𝐻𝐻𝑒𝑒 𝑠𝑠𝐻𝐻𝑠𝑠𝐻𝐻𝑜𝑜𝐻𝐻)𝐴𝐴𝐻𝐻𝑛𝑛23(𝑠𝑠𝐻𝐻𝑤𝑤𝐻𝐻𝐻𝐻𝑤𝑤ℎ𝐻𝐻𝑠𝑠𝐻𝐻𝑒𝑒 𝑠𝑠𝐻𝐻𝑠𝑠𝐻𝐻𝑜𝑜𝐻𝐻)�

∙ 100% = % 𝑙𝑙𝑙𝑙𝑓𝑓𝑡𝑡 𝑟𝑟𝑙𝑙𝑙𝑙𝑟𝑟𝑡𝑡𝑙𝑙𝑟𝑟 𝑡𝑡𝑡𝑡 𝑢𝑢𝑢𝑢𝑤𝑤𝑙𝑙𝑟𝑟𝑡𝑡ℎ𝑙𝑙𝑟𝑟𝑙𝑙𝑟𝑟 𝑡𝑡𝑜𝑜𝑙𝑙

Equation 2 will, when calculated for all alkanes in the sample give rise to the PW-plot, as seen in Figure 16.

Figure 16 PW-plot. Example of relative changes in composition of n-alkanes in soil extracts compared to

original oil after 1 year of weathering.

The PW-plot is sometimes referred to as an “evaporation curve”, however this calculation cannot distinguish between evaporation and biodegradation, since these processes will have the same effect on the alkanes (the shorter the alkane, the faster the removal). Therefore, the PW plot shows the combined effects of evaporation and biodegradation.


35

PAC fingerprints

Corresponding to the PW-plot, PAC removal can be assessed by choosing appropriate normalization compounds. As an example, biodegradation efficiency of a crude oil can be assessed by normalizing content of individual PACs to the sum of C3- and C4-dibenzothiophenes (DBT, m/z 226 and 240, Figure 17 and Figure 18). Due to the high molecular weight and high complexity of C3- and C4-DBTs have high boiling point and are relatively recalcitrant to biodegradation. They can therefore serve as internal markers for PAC removal as a combined function of biodegradation and short-term physical processes, such evaporation and dissolution. Likewise, other HMW PACs, such as C3- and C4-chrysenes can be used as internal standards.

Equation 3.

𝐴𝐴𝑃𝑃𝑃𝑃𝑛𝑛(𝑤𝑤𝐻𝐻𝐻𝐻𝑤𝑤ℎ𝐻𝐻𝑠𝑠𝐻𝐻𝑒𝑒 𝑠𝑠𝐻𝐻𝑠𝑠𝐻𝐻𝑜𝑜𝐻𝐻)Σ(𝐴𝐴𝑛𝑛3𝐷𝐷𝐷𝐷𝐷𝐷+𝐴𝐴𝑛𝑛4𝐷𝐷𝐷𝐷𝐷𝐷)𝑤𝑤𝐻𝐻𝐻𝐻𝑤𝑤ℎ𝐻𝐻𝑠𝑠𝐻𝐻𝑒𝑒 𝑠𝑠𝐻𝐻𝑠𝑠𝐻𝐻𝑜𝑜𝐻𝐻�

𝐴𝐴𝑃𝑃𝑃𝑃𝑛𝑛(𝑠𝑠𝐻𝐻𝑤𝑤𝐻𝐻𝐻𝐻𝑤𝑤ℎ𝐻𝐻𝑠𝑠𝐻𝐻𝑒𝑒 𝑠𝑠𝐻𝐻𝑠𝑠𝐻𝐻𝑜𝑜𝐻𝐻)Σ(𝐴𝐴𝑛𝑛3𝐷𝐷𝐷𝐷𝐷𝐷+𝐴𝐴𝑛𝑛4𝐷𝐷𝐷𝐷𝐷𝐷)𝑠𝑠𝐻𝐻𝑤𝑤𝐻𝐻𝐻𝐻𝑤𝑤ℎ𝐻𝐻𝑠𝑠𝐻𝐻𝑒𝑒 𝑠𝑠𝐻𝐻𝑠𝑠𝐻𝐻𝑜𝑜𝐻𝐻�

∙ 100% =

% 𝑙𝑙𝑙𝑙𝑓𝑓𝑡𝑡 𝑟𝑟𝑙𝑙𝑙𝑙𝑟𝑟𝑡𝑡𝑙𝑙𝑟𝑟 𝑡𝑡𝑡𝑡 𝑢𝑢𝑢𝑢𝑤𝑤𝑙𝑙𝑟𝑟𝑡𝑡ℎ𝑙𝑙𝑟𝑟𝑙𝑙𝑟𝑟 𝑡𝑡𝑜𝑜𝑙𝑙

Figure 17 Example of relative changes in composition of mono-, di- and tri cyclic aromatic compounds in

soil extracts compared to original oil after 1 year of weathering.


36

Figure 18 Example of relative changes in composition of sulfur containing di- and tricyclic aromatics in soil

extracts compared to original oil after 1 year of weathering.

Diagnostic ratios

Certain weathering processes involve highly selective chemical reactions, and therefore large differences in degradation of closely related compounds can appear. This is apparent for photo oxidation and bacterial degradation of certain alkylated PACs. Based on empirical evaluations and understanding of the reactions, several relationships have been constructed to distinguish between short-term physical weathering processes (evaporation and dissolution), photo oxidation and bacterial degradation. The relationships result in semiquantitative diagnostic ratios (DRs) that can indicate the progress of the individual weathering processes. Table 6 lists DRs for different processes, again with references to the CEN guideline for identification help (CEN/TR 2012). Selected ratios are plotted in Figure 19.

Table 6 Biodegradation diagnostic ratios (DRs) of selected petroleum hydrocarbons.

Diagnostic ratio Process Degradation severity

Reference

nC16/norpristane Biodegradation Initial nC17/pristane Biodegradation Initial nC18/phytane Biodegradation Initial Pristane/phytane Biodegradation Initial 2mN/1mN Biodegradation Initial 1,3+1,6dmN/C2N Biodegradation Initial (Wang, Fingas et al. 1998) 2,6+2,7dmN/C2N Biodegradation Initial (Bayona, Albaiges et al. 1986) 2+3mF/4mF Biodegradation Initial (Christensen, Hansen et al.

2005) 2+3mF/1+2+3+4mF Biodegradation Initial (Christensen, Hansen et al.


37

Diagnostic ratio Process Degradation severity

Reference

2005) 1mPy/2mPy Photooxidation Initial (CEN/TR 2012) 2+3mDBT/4mDBT Biodegradation Moderate (Wang, Fingas et al. 1998) 1mDBT/4mDBT Biodegradation Moderate (Lamberts, Johnsen et al.

2008) 1mPhe/9+4mPhe Biodegradation Moderate (Wang, Fingas et al. 1998,

Lamberts, Johnsen et al. 2008) 2+3mPhe/1+9+4mPhe Biodegradation Moderate (Wang, Fingas et al. 1998 2,4+2,6+3,7+1,3C2DBT/4,6+3,6+1,4+1,6+1,8C2DBT

Biodegradation Moderate (Douglas, Bence et al. 1996)

C2DBT/C2DBT+C2Phe Biodegradation Moderate (Douglas, Bence et al. 1996) C3DBT/C3DBT+C3Phe Biodegradation Moderate (Douglas, Bence et al. 1996) C2DBT/C2DBT+C2Chry Biodegradation Moderate (Douglas, Bence et al. 1996) C3DBT/C3DBT+C3Chry Biodegradation Moderate (Douglas, Bence et al. 1996) C2Chry/C2Chry+C2Py Biodegradation High (Douglas, Bence et al. 1996) C3Chry/C3Chry+C3Py Biodegradation High (Douglas, Bence et al. 1996) 2mChry/1mChry Biodegradation High (Bayona, Albaiges et al. 1986,

Diez, Sabate et al. 2005) 2mPy/1mPy+4mPy Biodegradation High (Bayona, Albaiges et al. 1986,

Diez, Sabate et al. 2005)

Figure 19 Employment of diagnostic ratios to show the degree of weathering of an oil after one year of

degradation.


38

4.5 References

Bayona, J. M., J. Albaiges, A. M. Solanas, R. Pares, P. Garrigues and M. Ewald (1986). "Selective Aerobic Degradation of Methyl-Substituted Polycyclic Aromatic-Hydrocarbons in Petroleum by Pure Microbial Cultures." International Journal of Environmental Analytical Chemistry 23(4): 289-303.

Boll, E. S., J. H. Christensen and P. E. Holm (2008). "Quantification and source identification of polycyclic aromatic hydrocarbons in sediment, soil, and water spinach from Hanoi, Vietnam." Journal of Environmental Monitoring 10(2): 261-269.

CEN/TR (2012). CEN/TR 15522-2:2012 Oil spill identification - Waterborne petroleum and petroleum products - Part 2: Analytical methodology and interpretation of results based on GC-FID and GC-MS low resolution analyses, European Committee for Standardization.

Christensen, J. H., A. B. Hansen, U. Karlson, J. Mortensen and O. Andersen (2005). "Multivariate statistical methods for evaluating biodegradation of mineral oil." Journal of Chromatography A 1090(1-2): 133-145.

Christensen, J. H., A. B. Hansen, J. Mortensen and O. Andersen (2005). "Characterization and matching of oil samples using fluorescence spectroscopy and parallel factor analysis." Analytical Chemistry 77(7): 2210-2217.

Diez, S., J. Sabate, M. Vinas, J. M. Bayona, A. M. Solanas and J. Albaiges (2005). "The Prestige oil spill. I. Biodegradation of a heavy fuel oil under simulated conditions." Environmental Toxicology and Chemistry 24(9): 2203-2217.

Douglas, G. S., A. E. Bence, R. C. Prince, S. J. McMillen and E. L. Butler (1996). "Environmental Stability of Selected Petroleum Hydrocarbon Source and Weathering Ratios." Environmental Science & Technology 30(7): 2332-2339.

Gallotta, F. D. and J. H. Christensen (2012). "Source identification of petroleum hydrocarbons in soil and sediments from Iguaçu River Watershed, Paraná, Brazil using the CHEMSIC method (CHEMometric analysis of Selected Ion Chromatograms)." Journal of Chromatography A 1235: 149-158.

Lamberts, R. F., A. R. Johnsen, O. Andersen and J. H. Christensen (2008). "Univariate and multivariate characterization of heavy fuel oil weathering and biodegradation in soil." Environmental Pollution 156(2): 297-305.

Lundstedt, S., P. Haglund and L. Oberg (2006). "Simultaneous extraction and fractionation of polycyclic aromatic hydrocarbons and their oxygenated derivatives in soil using selective pressurized liquid extraction." Analytical Chemistry 78(9): 2993-3000.

Nytoft, H. P. and J. A. Bojesen-Koefoed (2001). "17α,21α(H)-hopanes: natural and synthetic." Organic Geochemistry 32(6): 841-856.

USEPA (1998). METHOD 8270D: SEMIVOLATILE ORGANIC COMPOUNDS BY GAS CHROMATOGRAPHY/MASS SPECTROMETRY (GC/MS) SW-846 Ch 4.3.2 Update IVA, United States Environmental Protection Agency.

Wang, Z. D., M. Fingas, S. Blenkinsopp, G. Sergy, M. Landriault, L. Sigouin, J. Foght, K. Semple and D. W. S. Westlake (1998). "Comparison of oil composition changes due to biodegradation and physical weathering in different oils." Journal of Chromatography A 809(1-2): 89-107.

Date post:	19-May-2020
Category:	Documents
Upload:	others
View:	3 times
Download:	0 times

Deliverable D8.7 Mesocosm experimental · Linus Malmquist, UCPH Thorvaldsensvej 40, 1871 Frb C...

Documents