UNIVERSITY OF CALIFORNIA, IRVINE - · UC Irvine School of Physical Sciences Faculty Endowed Fellowship, 2008-2009 Outstanding Presentation, UC Irvine Institute of Geophysics and

UNIVERSITY OF CALIFORNIA,IRVINE

Radiocarbon of Black Carbonin Marine Dissolved Organic Carbon

DISSERTATION

submitted in partial satisfaction of the requirementsfor the degree of

DOCTOR OF PHILOSOPHY

in Earth System Science

by

Lori Anne Ziolkowski

Dissertation Committee:Professor Ellen Druffel, ChairProfessor James Randerson

Professor Susan TrumboreProfessor Richard Chamberlin

Professor Caroline Masiello

2009

UMI Number: 3369198

Copyright 2009 by Ziolkowski, Lori Anne

All rights reserved

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy

submitted. Broken or indistinct print, colored or poor quality illustrations and

photographs, print bleed-through, substandard margins, and improper

alignment can adversely affect reproduction.

In the unlikely event that the author did not send a complete manuscript

and there are missing pages, these will be noted. Also, if unauthorized

copyright material had to be removed, a note will indicate the deletion.

______________________________________________________________

UMI Microform 3369198Copyright 2009 by ProQuest LLC

All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

_______________________________________________________________

ProQuest LLC 789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, MI 48106-1346

Chapter 3 c© 2009 Elsevier B.V.All other materials c© 2009 Lori Anne Ziolkowski

DEDICATION

For Oma,whom I wish could be hereto see me become Dr. Z.

ii

TABLE OF CONTENTS

Page

LIST OF FIGURES vi

LIST OF TABLES viii

ACKNOWLEDGMENTS ix

CURRICULUM VITAE xi

ABSTRACT OF THE DISSERTATION xv

1 Introduction 1Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 Oxidation of PAHs using the BPCA method 92.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

2.3.1 Sample treatment . . . . . . . . . . . . . . . . . . . . . . . . 132.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 17

2.4.1 Nitration of BPCAs . . . . . . . . . . . . . . . . . . . . . . . . 172.4.2 Carbon yields . . . . . . . . . . . . . . . . . . . . . . . . . . . 172.4.3 BPCA products of PAHs . . . . . . . . . . . . . . . . . . . . . 192.4.4 Time course and mechanistic experiments . . . . . . . . . . 222.4.5 Analysis of black carbon ring trial materials . . . . . . . . . . 28

2.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

3 Detection of fullerenes and carbon nanotubes using BPCAs 353.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

iii

3.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

3.4.1 BPCA distributions . . . . . . . . . . . . . . . . . . . . . . . . 413.4.2 Carbon yield . . . . . . . . . . . . . . . . . . . . . . . . . . . 463.4.3 Mixtures in sediments . . . . . . . . . . . . . . . . . . . . . . 48


4 Quantification of extraneous carbon during CSRA of BC 574.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

4.3.1 Chemical Oxidation . . . . . . . . . . . . . . . . . . . . . . . 604.3.2 Radiocarbon analysis of isolated samples . . . . . . . . . . . 62

4.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 634.4.1 Carbon mass balance and corrections . . . . . . . . . . . . . 634.4.2 Extraneous carbon added during PCGC isolation (CPCGC) . . 654.4.3 Extraneous carbon added during chemical oxidation and PCGC

isolation (Cchemistry+PCGC) . . . . . . . . . . . . . . . . . . . . . 694.4.4 Correcting for extraneous carbon and associated uncertainties 70


5 Black carbon in marine dissolved organic carbon 795.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 795.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 805.3 Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 815.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

6 Conclusions and thoughts on future research 966.1 BPCA method and its applicability . . . . . . . . . . . . . . . . . . . 966.2 Evaluating extraneous material added during the preparation of CSRA

samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 986.3 Black carbon in the marine DOC pool . . . . . . . . . . . . . . . . . 996.4 Future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

A Determination of Carbon Yields 105A.1 Polycyclic Aromatic Hydrocarbons . . . . . . . . . . . . . . . . . . . 105A.2 Calculating the percentage of black carbon in UDOM . . . . . . . . . 106

iv

B BPCA protocol 109B.1 Chemical extraction of BPCAs . . . . . . . . . . . . . . . . . . . . . 109

B.1.1 Cleaning the bomb . . . . . . . . . . . . . . . . . . . . . . . . 109B.1.2 Pre-treatment of samples (if required) . . . . . . . . . . . . . 110B.1.3 Filter sample . . . . . . . . . . . . . . . . . . . . . . . . . . . 110B.1.4 Cation column (if required) . . . . . . . . . . . . . . . . . . . 110B.1.5 Dry samples . . . . . . . . . . . . . . . . . . . . . . . . . . . 111B.1.6 Derivatize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111B.1.7 Solvent change . . . . . . . . . . . . . . . . . . . . . . . . . . 111

B.2 PCGC settings and parameters . . . . . . . . . . . . . . . . . . . . . 112B.2.1 Determination of sample concentration and retention times . 112B.2.2 Program the preparative fraction collector (PFC) to collect at

selected RTs . . . . . . . . . . . . . . . . . . . . . . . . . . . 113B.2.3 Prime the PCGC for collection . . . . . . . . . . . . . . . . . 113B.2.4 Collect sample(s) . . . . . . . . . . . . . . . . . . . . . . . . . 113B.2.5 Check the concentration and purity of the isolate . . . . . . . 114B.2.6 Prepare sample for combustion . . . . . . . . . . . . . . . . . 114

v

LIST OF FIGURES

Page

1.1 The black carbon continuum, adapted from Masiello {2004}. . . . . . 2

2.1 Theoretical chemical structure of black carbon . . . . . . . . . . . . 112.2 The PAHs used in this study. . . . . . . . . . . . . . . . . . . . . . . 142.3 Structures of BPCAs used as markers of aromatic carbon in this

study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162.4 Schematic of the oxidation of phenanthrene into BPCAs . . . . . . . 202.5 Oxidation products and yields for anthracene as a function of time. . 232.6 Proposed reaction scheme for the oxidation of anthracene to B2CA. 232.7 Distribution of non-, mono- and di-nitrophthalic acid as a function of

reactants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242.8 Change in B3CA and B6CA oxidation products from perylene over

time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262.9 BPCA distribution of materials used in the black carbon ring trial. . . 30

3.1 Oxidation schematic of the PAH perylene . . . . . . . . . . . . . . . 383.2 Three of the six carbon nanoparticles studied. . . . . . . . . . . . . 393.3 BPCA distribution of fullerenes, carbon lampblack and soot. . . . . . 423.4 BPCA distribution of SWCNTs with and without cation column. . . . 453.5 Standard addition of soot and SWCNTs to marine sediments. . . . . 503.6 Theoretical and measured BPCA distributions in mixtures. . . . . . . 52

4.1 The magnitude of column bleed and oven temperature as a functionof retention time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

4.2 Grass char and hexane soot before and after Cex correction. . . . . . 73

5.1 Map illustrating sample locations. . . . . . . . . . . . . . . . . . . . . 835.2 BPCA distribution and 14C of BC for the samples. . . . . . . . . . . 845.3 ∆14C of black carbon and marine DOC as a function of depth. . . . . 89

vi

6.1 Chemical structure of asphaltene . . . . . . . . . . . . . . . . . . . . 101

vii

LIST OF TABLES

Page

2.1 PAH Carbon yield and BPCA distributions . . . . . . . . . . . . . . . 182.2 Time course carbon yields . . . . . . . . . . . . . . . . . . . . . . . . 222.3 Quantification of black carbon materials (g BC / kg dry weight) . . . 292.4 Variations on converting BPCAs to BC . . . . . . . . . . . . . . . . . 30

3.1 Percent carbon yield for the carbon nanoparticles in this study . . . . 44

4.1 Materials processed and associated solvents used for CSRA of blackcarbon. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

4.2 Type and treatment of samples evaluated for mass and FM of extra-neous carbon added during sample processing. . . . . . . . . . . . . 68

4.3 Radiocarbon values (fraction modern) and associated uncertainty ofblack carbon reference materials before and after correction for Cex. 72

5.1 Sample information for UDOM samples in this study. . . . . . . . . . 825.2 Measurements of black carbon isolated from UDOM. . . . . . . . . . 85

A.1 Carbon content of BPCAs. . . . . . . . . . . . . . . . . . . . . . . . 106A.2 Example calculation of BPCAs in PAHs. . . . . . . . . . . . . . . . . 107

B.1 PCGC settings and parameters . . . . . . . . . . . . . . . . . . . . . 112

viii

ACKNOWLEDGMENTS

This work would not have possible without Ellen Druffel introducing me to the worldof black carbon, by not so naıvely assigning me a summer project isolating blackcarbon from a dissolved organic carbon sample! Her quiet guidance and boundlesssupport allowed me to have the greatest freedom in carrying out this project. I ameternally thankful to her for her patience, belief in my abilities, critical feedback andwillingness to discuss potentially frivolous matters. In the future whenever I crosspaths with a rubber chicken, I will think of you, with great fondness.

Sheila Griffin, the rock of the Druffel lab, was instrumental in my progress. Sheprovided support, structure and delicious baked goods even when I didn’t know Ineeded it. I am forever indebted to her and her willingness to turn a blind eye tomy chaotic ways. I am grateful to my academic big brother, Steve Beaupre, whowas always willing to lend a helping and educational hand, and for teaching me thevalue of returning to first principles.

I’d like to thank my thesis committee for grounding me, while still believing in myability to tackle this project. Sue Trumbore taught me the value in being challenged,while Jim Randerson taught me the value of taking a step back, to evaluate the bigpicture. Carrie Massiello provided the foundation for black carbon in the Druffel lab.Thank you for introducing me to the wonderful world of biochar. Dick Chamberlinwas an indispensable organic chemistry resource, always willing to answer eventhe simplest of my questions with great patience.

Without the guidance (and tough love) of the KECK Carbon Cycle AMS facility,these radiocarbon measurements would be meaningless. Always willing to answermy questions, from trivial to complex, John Southon was a valuable resource forthis work. I am honoured my CSRA blank evaluation work has received the Gua-ciara dos Santos seal of approval. For better or worse, never again will I be ablelook at 14C data without questioning the blank.

If it were not for Lihini Aluwihare and Matt McCarthy opening their freezers to me,I would still be searching for marine UDOM samples. I would like to thank ClaudiaCzimczik for first introducing me to the BPCA method and showing me that youcan have it all in science. I am most appreciative to Christopher Reddy for his ac-cessibility, which has brought my understanding of (and confidence in) collegialityto a new level. I am indebted to John Greaves at the UCI Mass Spec facility for hispatience and eye to detail with regards to my work.

Thank you to the UCI ESS graduate students, past and present, who have openedmy mind to different schools of thought and challenged my perceptions. With theadministrative support of Cynthia, Liz, Linda and Jeff, my paper work was alwaysin order, money was always deposited to my bank account and immigration alwayslet me back in the country. Thank you!

ix

Finally, I am indebted to my family for their support along this great journey. Thankyou to my parents, Werner and Shirley, and my sister Kathi, who always encour-aged me to follow my heart. And thank you to Cam, who patiently listens to all myideas, humours me when I riddle off scientific nonsense to him, and continues tochallenge me.

x

CURRICULUM VITAE


EDUCATIONDoctor of Philosophy in Earth System Science 2009University of California, Irvine Irvine, California

Master of Science in Earth System Science 2006University of California, Irvine Irvine, California

Master of Science in Chemical Oceanography 2000Dalhousie University Halifax, Nova Scotia

Bachelor of Science in Environmental Chemistry 1998University of Waterloo Waterloo, Ontario

RESEARCH EXPERIENCEGraduate Research Assistant 2004–2009University of California, Irvine Irvine, CaliforniaResearch Technician 2000–2004Dalhousie University Halifax, Nova ScotiaGraduate Research Assistant 1998–2000Dalhousie University Halifax, Nova ScotiaGuest Student Investigator Jan – Aug,1997Woods Hole Oceanographic Institution Woods Hole, MassachusettsResearch Technician May – Aug,1996University of Notre Dame, Radiation Laboratory South Bend, IndianaGeochemical Assistant Jan – Apr,1995National Water Research Institute Burlington, Ontario

HONORS AND AWARDSOrigins Institute CREATE Astrobiology Postdoctoral Fellowship, 2009 - 2011Isocompound Meeting, Young Investigator Award, June 2009UC Irvine Earth System Science Outstanding Departmental Contributions, 2009UC Irvine Graduate Dean’s Dissertation Quarter Fellowship, Summer 2009UC Irvine School of Physical Sciences Faculty Endowed Fellowship, 2008-2009Outstanding Presentation, UC Irvine Institute of Geophysics and Planetary Physics,2008Pedagogical Fellowship, University of California Irvine, 2006-2008Jenkins Graduate Fellowship, University of California Irvine, 2004-2006Graduate Student fellowship, Dalhousie University,1998-2000

xi

REFEREED JOURNAL PUBLICATIONSBourbonniere, R.A., S.L. Telford, L.A. Ziolkowski, J. Lee, M.S. Evans and P.A.Meyers (1997) Biogeochemical marker profiles in cores of dated sediments fromlarger North American lakes Molecular Markers in Environmental Geochemistry,ACS Symposium Series, 671, 133-150.

Ziolkowski, L., K. Vinodgopal, and P.V. Kamat (1997), Photostabilization of or-ganic dyes on poly(styrenesulfonate)-capped TiO2 nanoparticles Langmuir, 13(12):3214-3128.

Xie, H., S.S. Andrews, W.R. Martin, J. Miller, L. Ziolkowski, C.D. Taylor, and O.C.Zafiriou (2002), Validated methods for sampling and headspace analysis of carbonmonoxide in seawater, Marine Chemistry, 77(2-3), 93-108.

Clark, C.D., W.T. Hiscock, F.J. Millero, G. Hitchcock, L. Brand, W.L. Miller, L. Zi-olkowski, R.F. Chen and R.G. Zika (2004) CDOM distribution and CO2 productionon the southwest Florida shelf, Marine Chemistry, 89(1-4): 145-167.

Bouillon, R.C., W.L. Miller, M. Levasseur, M. Scarratt, A. Michaud, and L. Zi-olkowski (2006) The effects of mesoscale iron enrichment on the marine photo-chemistry of dimethylsulfide in the NE subarctic Pacific, Deep-Sea Research PartII Topical studies in Oceanography, 53(20-22): 2384-2397,doi:10.1016/j.dsr2.2006.05.024.

Ziolkowski, L. A. and W. L. Miller (2007), Marine photochemical production ofcarbon monoxide. Marine Chemistry, doi:10.1016/j.marchem.2007.02.004.

Ziolkowski, L.A. and E.R.M. Druffel (2009), Feasibility of isolating and detectingfullerenes and carbon nanotubes using the benzene polycarboxylic acid method.Marine Pollution Bulletin, doi:10.1016/j.marpolbul.2009.04.018

PAPERS IN PREPARATIONZiolkowski, L.A., E.R.M. Druffel, R.A. Chamberlin and J. Greaves, Evaluation of theoxidation of polycyclic aromatic hydrocarbons using the benzene polycarboxylicacid method.

Ziolkowski, L.A., E.R.M. Druffel and J. Southon. Microscale compound specificradiocarbon analysis.

Ziolkowski, L.A. and E.R.M. Druffel. Radiocarbon of black carbon in marine dis-solved organic carbon.

xii

ORAL PRESENTATIONS (presenter in bold)Ziolkowski, L.A., W.L. Miller, D. J. Kieber, and K. Mopper, Rapid Shipboard Deter-mination of the Efficiency Spectrum for Photochemical Carbon Monoxide Produc-tion, AGU Ocean Sciences, January 2000, Texas.

Ziolkowski, L. A., W.L. Miller, H. Xie and O.C. Zafiriou, Efficiency Spectra for theMarine Photochemical Production of Carbon Monoxide Determined Using a SolarSimulator, ASLO, February 2001, New Mexico.

Ziolkowski, L. A. and W.L. Miller, Spatial and Temporal Variation of CDOM FadingEfficiency, AGU Ocean Sciences, February 2002, Hawaii.

Ziolkowski L.A. and C. Gardiner, The Professional Utility of Teaching AssistantTraining beyond the Classroom, Lilly West, March 2008, Pamona, CA.

Ziolkowski, L.A. and E. R. M. Druffel, Radiocarbon content of benzene polycar-boxylic acids, GSA/SSA Joint Meeting, October 2008, Houston, TX (Invited pre-sentation)

Ziolkowski, L.A., Black magic: can we make carbon disappear?, UCI ESS 1/2Baked Seminar Series, February 2009 (Invited presentation)

Ziolkowski, L.A. and E.R.M. Druffel, Compound specific radiocarbon analysisof black carbon in marine dissolved organic matter, Isocompound Meeting, June2009, Potsdam, Germany.

POSTER PRESENTATIONS (presenter in bold)Ziolkowski, L.A. and W.L. Miller, U.V. Optical Properties During the Evolution of aPhytoplankton Bloom (SERIES), ASLO/TOS Ocean Sciences, 2004, Hawaii.

Ziolkowski, L.A., C.S. Law and W.L. Miller, Investigation of the Marine Photochem-ical Production of Carbon Monoxide in the Waters South-East of New Zealand,IGAC Conference, September 2004, Christchurch, N.Z.

Ziolkowski, L.A., E.R.M. Druffel and S. Griffin, Progress Towards an Estimate ofthe Radiocarbon Content of Black Carbon in Marine Organic Matter, AGU OceanSciences, February 2006, Hawaii.

Ziolkowski, L.A. and E.R.M. Druffel. Black carbon measurements using a revisedBPCA method. EGU, April 2007, Vienna, Austria.

Ziolkowski, L.A. and E.R.M. Druffel, Radiocarbon Content of Soot and CharredBlack Carbon using the Benzene Polycarboxylic Acid Method, AGU Ocean Sci-ences, March 2008, Orlando, Fl.

Ziolkowski, L.A. and E.R.M. Druffel, Radiocarbon values of black carbon usingthe Benzene Polycarboxylic Acid Method, AGU Fall Meting, December 2008, San

xiii

Francisco, CA.

FIELD EXPERIENCECruise:R/V Tangaroa, M. Harvey, ANZ-SOLAS, Mar 20-Apr 19, 2004, S. Ocean.CCGS Martha Black, M. Gosselin, C-SOLAS, July 3-28, 2003, N. Atlantic.R/V Pelican, R.T. Powell, SWISS-III, Aug 29-Sept 13, 2002, Gulf of Mexico.R/V El Puma, M. Levasseur, C-SOLAS, July 4-30, 2002, N. Pacific.R/V Pelican, W. Landing, SWISS-II, August 11-25, 2001, Gulf of Mexico.R/V Pelican, R.T. Powell, SWISS-I, April 16-28, 2001, Gulf of Mexico.R/V Endeavor, O. C. Zafiriou, March 13-April 2, 2000, Sargasso Sea.R/V Endeavor, O. C. Zafiriou, August 2-20, 1999, Sargasso Sea.R/V Endeavor, D. J. Keiber, July 8-28, 1999, Gulf of Maine.R/V New Horizon, T. Hayward, CalCOFI, April 2-30, 1997, So. Cal. coast.Other: Thompson, MB, Canada, BOREAS, 1993, 1994, 1995, collection of beaverpond water.

OTHER ACTIVITIESReviewerPublications: AGU Book Series, Environmental Science and Technology, MarineChemistryGrants: National Science FoundationService & Other Activities2008 GSA Short Course: Starting out in Undergraduate Research and Education:A Professional Development Workshop for Young Faculty2006-2008: Science Fair Judge, California State Science Fair2007-present: AGU Education and Human Resources Student Advisory Board2007-present: Earth System Science Journal Club Coordinator2005-2006: Earth System Science Departmental Graduate Student Representa-tive

TEACHING EXPERIENCEUC IrvineIntroduction to the Earth System, ESS 25, Teaching Assistant (2005, 2007)Organic Biogeochemistry, ESS 53, Teaching Assistant (2006)Teaching Assistant Professional Development Program, Instructor (2006-08)

AFFILIATIONSAmerican Geophysical Union memberEarth Science Womens Network memberThe Oceanography Society member

xiv

ABSTRACT OF THE DISSERTATION

Radiocarbon of Black Carbonin Marine Dissolved Organic Carbon

By


Doctor of Philosophy in Earth System Science

University of California, Irvine, 2009

Professor Ellen Druffel, Chair

Black carbon (BC), a bi-product of combustion, is a major long-term carbon sink

in the Earth system. Known storage pools for BC are marine sediment and soil.

Previous studies found significant 14C age differences between BC and organic

carbon in sediments, and projected that BC must reside in an intermediate pool,

such as dissolved organic carbon (DOC), before deposition to the sediment. This

research applied compound specific radiocarbon analysis (CSRA) of BC using the

benzene polycarboxylic acid (BPCA) method, to provide the first estimates of BC

cycling in marine DOC.

First, the BPCA method was adapted for CSRA of marine DOC. This method was

applied to nine polycyclic aromatic hydrocarbons (PAHs) to examine the oxidation

mechanism of the BPCA method. These experiments showed larger BPCAs are

preferentially formed for large (>4 ring) PAHs and an average C recovery of 26 ±

7 %. Quantification of nitrated BPCAs was found to be essential for accurate as-

sessment of BC. Next, I evaluated the mass and radiocarbon of extraneous carbon

(Cex) added in the processing and isolation of CSRA samples. The Cex originated

equally from column bleed and the processing steps prior to compound isolation.

xv

While constant over a few weeks, the mass and radiocarbon signature of Cex varied

over longer time periods and must be frequently re-evaluated.

Finally, the radiocarbon signatures of BC in marine high molecular weight (HMW)

DOC samples from a river and five locations in the Atlantic and Pacific Oceans

are presented. BC exported from the river was 14C modern, while ocean samples

were uniformly old (average open ocean BC ∆14C = -888 ± 25 h, n=6). The

concentration of BC in HMW DOC (also known as UDOM) ranged from 0.5 to 3.5 %

and suggests that a substantial portion of BC should be in the low molecular weight

DOC pool. The presence of 14C-depleted BC in modern HMW DOC demonstrates

that there are widely different turnover times for these two pools in the marine

carbon cycle.

xvi

1Introduction

Black carbon, (BC) is the term applied to materials that have undergone combus-

tion and are characterized by a broad spectrum of properties. As detected in soil

and sediment reservoirs, BC is operationally defined and represents a range of

combustion residues (i.e.: char and charcoal) and combustion condensates (i.e.:

as soot, Figure 1.1). BC may represent a significant sink of the global carbon cycle

{Kuhlbusch, 1998}, affect the Earth’s radiative heat balance {Crutzen and Andreae,

1990; Ramanathan and Carmichael, 2008}, influence the albedo of snow {Flanner

et al., 2008}, serve as a paleo-tracer for Earth’s fire history {Bird and Cali, 1998}

and represent a significant portion of carbon buried in soil and sediment {Masiello

and Druffel, 1998}.

Loss processes for BC are not well understood. When Seiler and Crutzen {1980}

estimated the annual natural emissions of BC to be on the order of 120 Tg BC, they

pointed out that without a loss term, all carbon on the surface of the Earth would

be present as BC in less than 100,000 years. BC has been found in 400 million

year old sediments in various locations {Venkatesan, 1989; Killops and Killops,

1992} and terrestrial soil samples that range in age from thousands to millions of

years {Bird and Cali, 1998}. Herring {1985} found no obvious decay of BC upon

1

Figure 1.1: The black carbon continuum, adapted from Masiello {2004}.

2

visual inspection of Cenozoic sediments. Despite its inert nature, losses of BC

have been detected in some situations. For example, Bird et al. {1999} determined

that BC in African soils could be significantly degraded on centennial timescales.

Using stable carbon isotopes they found that coarse BC particles are degraded

faster than finer particles. In laboratory experiments, Winkler {1985} observed

that organic compounds within the BC undergo acidic breakdown in an anaerobic

bog and lake sediments. Measurements of BC decay in sediments were made by

Middelburg et al. {1999}, where they found a diminished amount of BC in Madeira

Abyssal Plain turbidites that were exposed to oxygen over long time periods (10-20

kyr).

There is no standardized analytical technique for measuring BC. Quantification

techniques generally fall into one of two categories, isolation techniques and iden-

tification techniques (Figure 1.1). Separation techniques isolate the BC from the

bulk sample using some form of oxidation to remove the non-BC carbon com-

pounds. Quantification techniques analyze the bulk sample for chemical signatures

indicative of black carbon, such as condensed aromatic structures. Several review

papers {Schimdt and Noack , 2000; Schmidt et al., 2001; Preston and Schmidt ,

2006} and intercomparison projects {Currie et al., 2002; Hammes et al., 2007} on

the quantification and importance of BC have recently been published. In the most

recent intercomparison project, termed the “BC ring trial” {Hammes et al., 2007},

seventeen different teams using seven different isolation methods measured the

BC content in 12 standard reference materials (SRMs), and recommended that all

future BC studies calibrate using this set of BC reference materials. The ”BC ring

trial” clearly demonstrated that different methods each have associated artifacts

and biases.

3

Outline of this thesis

The primary goal of this dissertation was to determine the role that BC plays in the

DOC cycle within the global ocean.

The method used to quantify BC is often chosen based on the type of BC of in-

terest (Figure 1.1). Since we were interested in not only quantifying the BC but

also measuring its radiocarbon content, we selected a BC method that would allow

for unambiguous isotopic measurement of BC. The benzene polycarboxlyic acid

(BPCA) method seemed well suited for CSRA of BC, as it simultaneously oxidizes

non-BC and transforms BC into BPCAs, leaving the BC-signature in an aqueous

solution. This method can provide both qualitative and quantitative information

about the BC. Before CSRA of BPCAs could be employed, two methodological

issues needed to be resolved. First, the accurate quantification of BC as BPCAs

requires a reliable, repeatable and robust conversion of BPCAs to BC, which had

not been previously demonstrated in previous BPCA studies. Secondly, our in-

terest in CSRA of BPCAs required minimizing the carbon added in derivatization

and a new derivatization method was essential for the most accurate 14C measure-

ments. These improvements, along with additional information a proposed reaction

scheme for the formation of BPCAs is discussed in Chapter 2.

The similarity of naturally produced BC to some manufactured BC-like products,

such as fullerenes and carbon nanotubes (CNTs), led to the work presented in

Chapter 3. Here we investigate the feasibility of using the BPCA method for isolat-

ing fullerenes and CNTs from BC and non-BC sedimentary material.

Marine DOC can be concentrated through either size ultratfiltration (UDOM) using

tangential flow filtration or solid phase extraction (SPE-DOM). Both methods are

costly especially for obtaining large quantities of DOC. Further, the DOC they iso-

4

late is only a fraction of the pool. UDOM concentrates the high molecular weight

DOC while SPE-DOM is composed of hydrophobic compounds within DOC. We

elected to isolated BC from UDOM in an effect to minimize potential artifacts that

may originate from the breakdown of the solid phase in SPE-DOM, which might

influence our 14C measurements.

Because BC was postulated to be a small percentage of the UDOM, large quan-

tities of UDOM are required to produce BPCAs for CSRA. Any sample handling

for radiocarbon analysis inadvertently adds extraneous carbon. Since the BPCAs

produced from the oxidation of UDOM would generate small CSRA samples (<

30 µg carbon), the extraneous carbon would contribute to the measured isotopic

composition of the CSRA samples. In Chapter 4 we discuss how we quantify the

magnitude and variability of extraneous carbon originating from the chemical oxi-

dation and subsequent isolation by PCGC.

Based the work of Masiello and Druffel {1998}, it was postulated that BC was 4

to 22 % of deep DOC. Using a suite of samples, ranging from fresh river water to

the deep Pacific, we investigate the concentration and radiocarbon of BC in marine

DOC. These results are presented in Chapter 5.

The final chapter (Chapter 6) summarizes the work presented here and suggests

some potential avenues of future research that stem from this thesis.

Two appendixes follow the conclusion. Appendix A outlines carbon recovery cal-

culations. Appendix B contains the method protocol used to generate the BPCA

data presented within this thesis.

5

Bibliography

Bird, M., and J. Cali, A million-year record of fire in sub-saharan africa, Nature,

394, 767 –769, 1998.

Bird, M., C. Moyo, E. Veenedaal, J. Lloyd, and P. Frost, Stability of elemental carbon

in savanna soil, Global Biogeochemical Cycles, 13, 923 – 932, 1999.

Crutzen, P., and M. Andreae, Biomass burning in the tropics: Impact on at-

mospheric chemistry and biogeochemical cycles, Science, 250(4988), 1669 –

1678, 1990.

Currie, L., et al., A critical evaluation of interlaboratory data on total, elemental, and

isotopic carbon in the carbonaceous particle reference material, nist srm 1649a,

Journal of Research of the National Institute of Standards and Technology, 107,

279 – 298, 2002.

Flanner, M. G., C. S. Zender, P. G. Hess, N. M. Mahowald, T. H. Painter, V. Ra-

manathan, and P. J. Rasch, Springtime warming and reduced snow cover from

carbonaceous particles, Atmospheric Chemistry and Physics Discussions, 8(6),

19,819–19,859, 2008.

Hammes, K., et al., Comparison of quantification methods to measure fire-derived

(black/elemental) carbon in soils and sediments using reference materials from

soil, water, sediment and the atmosphere, Global Biogeochem. Cycles, 21(3),

18, doi:10.1029/2006GB002914, 2007.

Herring, J., Charcoal fluxes into sediments of the north pacific ocean: the cenozoic

record of burning, in The Carbon Cycle and Atmospheric CO2: Natural Varia-

tions Archean to Present., edited by E. Sunquist and W. Broecker, pp. 419 –

442, AGU, Washington, 1985.

6

Killops, S., and V. Killops, An Introduction to Organic Geochemistry, 228 pp., John

Wiley and Sons, 1992.

Kuhlbusch, T., Black carbon and the carbon cycle, Science, 280, 1903 –1904,

1998.

Masiello, C., New directions in black carbon organic geochemistry, Marine Chem-

istry, 92, 201–213, 2004.

Masiello, C., and E. Druffel, Black Carbon in Deep-Sea Sediments, Science,

280(5371), 1911–1913, doi:10.1126/science.280.5371.1911, 1998.

Middelburg, J., J. Nieuwenhuize, and P. van Breugel, Black carbon in marine sedi-

ments, Marine Chemistry, 65, 245 –252, 1999.

Preston, C., and M. Schmidt, Black (pyrogenic) carbon: a synthesis of current

knowledge and uncertainties with special consideration of boreal regions, Bio-

geoscience, 3, 397 – 420, 2006.

Ramanathan, V., and G. Carmichael, Global and regional climate changes due to

black carbon, Nature Geoscience, 1(4), 221, doi:doi:10.1038/ngeo156, 2008.

Schimdt, M., and A. Noack, Black carbon in soils and sediments: Analysis, distri-

bution, implications, and current challenges, Global Biogeochem. Cycles, 14(3),

777 – 793, 2000.

Schmidt, M., J. Skjemstad, C. Czimczik, B. Glaser, K. Prentice, Y. Gelinas, and

T. Kuhlbusch, Comparative analysis of black carbon in soils, Global Biogeochem-

ical Cycles, 15, 163 – 167, 2001.

Seiler, W., and P. Crutzen, Estimates of gross and net fluxes of carbon between

the biosphere and the atmosphere from biomass burning, Climate Change, 2,

207 –247, 1980.

7

Venkatesan, M., Tetrahymanol, its widespread occurrence and geochemical signif-

icance, Geochim. Cosmochim. Acta, 53, 3095 – 3101, 1989.

Winkler, M., Charcoal analysis for paleoenvironmental interpretation, a chemical

assay, Quaternary Research, 23, 313 – 326, 1985.

8

2Oxidation of polycyclic aromatic hydrocarbons

and natural materials using the benzene

polycarboxylic acid method

2.1 Abstract

Quantification of black carbon (BC), carbonaceous material of pyrogenic origin, has

typically required either chemical or thermal oxidation methods for isolation from

heterogeneous matrices, such as sediment or soil. The benzene polycarboxylic

acid (BPCA) method involves chemical oxidation of aromatic structures, such as

those in BC, into BPCAs. Using a revised BPCA method, we studied the oxida-

tion of nine polycyclic aromatic hydrocarbons (PAHs). After 8 hours of oxidation

at 180 oC, the average carbon yield was 25.7 ± 6.8 % C and was not correlated

to the molecular weight of the PAH oxidized. The majority of the BPCAs observed

were nitrated, which has serious implications for the quantification of BC. Smaller

PAHs favor the formation of less substituted BPCAs, while larger PAHs, such as

coronene, favor the formation of more fully substituted BPCAs. Time course experi-

ments revealed variations of BPCA distributions over time, favoring less substituted

BPCAs with longer oxidation times, while the carbon yield remained constant. No

9

decarboxylation of fully substituted mellitic acid (B6CA) was observed during the

time course experiments. Using the model compound anthracene, a potential in-

ternal standard, we proposed a mechanism for the oxidation reaction based on

time course experiment data. Quantification of BC in reference materials revealed

that this revision of the BPCA method is significantly more efficient than previous

versions and is effective for quantifying soot BC.

2.2 Introduction

Black carbon (BC) particles are by-products of combustion processes that can be

defined by a broad range of characteristics. Sizes range from nanometer soot par-

ticles to millimeter pieces of charcoal. Chemically, BC has a condensed, highly

aromatic structure (Figure 2.1). Environmental scientists are interested in isolating

BC from organic matrices, such as soils and sediments, to address questions re-

garding the time scales of carbon storage in the Earth system. The wide range

of physical and chemical characteristics of both the BC and the heterogenous

matrices in which it is found pose challenges when isolating and quantifying BC

{Masiello, 2004}. Various techniques are used to quantify BC isolated from envi-

ronmental matrices including thermal oxidation {Gustafsson et al., 1997}, chemical

oxidation (nitric acid {Glaser et al., 1998; Brodowski et al., 2005} and acid dichro-

mate {Wolbach and Anders, 1989}) and mild chemical or photo-oxidation (NaClO

{Simpson and Hatcher , 2004} and ultra-violet {Skemstad et al., 1996}) followed by

NMR.

The benzene polycarboxylic acid (BPCA) method {Glaser et al., 1998; Brodowski

et al., 2005} converts BC to benzene rings that are substituted with various num-

bers (2-6) of carboxylic acid groups. Assuming that BC is primarily aromatic carbon

10

OC OHO

O

O

O

O OO

C=O

O

O

OO O OHO

O

O

O

O

O

O

OC

OOH

H

CH C (CH

2)

13

CH CCH

3

O

OH

OO

CH2

CH2

CH2

CH2

CH2

O

HO-CH2

CO

O

O

O

OC=O

O

O

CO

CH3

O

Figure 2.1: Theoretical black carbon type of molecule Goldberg {1985}. High-lighted are a number of PAHs used in this study to mimic edge functionalities of thetheoretical BC structure.

and additional carbon is neither added nor exchanged, the method provides both

yield and structural information about the BC. Nitric acid oxidizes the BC struc-

ture to BPCAs in which a single aromatic ring from the BC is maintained and is

substituted with carboxylic acids derived from adjacent rings or side chains. This

method retains the carbon from the molecular structure, which is essential for sub-

sequent isotopic abundance assays such as compound specific radiocarbon anal-

ysis {Eglinton et al., 1996}. While nitric acid oxidation of BC results in a significant

loss of carbon (roughly 25 % is retained), it provides both qualitative and quantita-

tive information about the original BC with negligible methodological artifacts.

A recent intercomparison of BC analysis methods {Hammes et al., 2007} revealed

that the previous version of the BPCA method {Glaser et al., 1998; Brodowski

et al., 2005}, which silylated rather than methylated the BPCAs, was more efficient

for the analysis of char-like BC than soot-like BC. Char-like BC is less condensed,

contains more functional side chains and forms BPCAs that are less substituted.

11

Soot-like BC is highly condensed with few additional side chains and forms BPCAs

with a higher degree of substitution. Generally, the chemical structure of BC is un-

known in environmental matrices, such as marine sediment. Therefore, a method

that provides both quantitative and qualitative data is needed.

To understand the distribution and yield of BPCA oxidation products, we studied

the oxidation of PAHs to model edge functionalities of BC (Figure 2.1). This is the

first study to systematically examine the oxidation products of the BPCA method

and suggest an oxidation mechanism. Polycyclic aromatic hydrocarbons (PAHs) of

various sizes and structures were digested and the resulting products were quan-

tified. A previous version of this method {Glaser et al., 1998; Brodowski et al.,

2005}, using GC-FID analysis of silylated BPCAs, was not as efficient for quanti-

fying all BPCAs and required the use of response factors. Also, previous methods

did not quantify nitrated BPCAs. Here we present results of experiments that quan-

tified the BPCAs and nitrated-BPCAs as methyl esters. No response factors were

used, as all BPCAs were methylated with the same efficiency. We examined the

reaction kinetics by evaluating the reaction products as a function of time. Using

commercially available BPCAs, we also address the possibility of decarboxylation

of BPCAs during the oxidation and subsequent steps. Furthermore, we suggest a

mechanism of oxidation for the PAH anthracene. Finally, using this revised method

the BC in reference materials was quantified. Compared to previous versions of

the BPCA method, this method produced higher estimates of BC and was able to

quantify soot BC.

12

2.3 Methods

2.3.1 Sample treatment

All glassware and quartz filters that came in contact with the samples and stan-

dards were baked at 550 oC for 2 hours prior to use in order to minimized carbon

contamination. Individual PAHs (Figure 2.2), varying in amounts from 2 to 10 mg C

were weighed into 12 mL quartz digestion tubes. Two mL concentrated nitric acid

(grade ACS) were added to each tube, then were capped and heated to 180 oC in

a high pressure digestion apparatus {Schramel et al., 1980}. Briefly, the digestion

apparatus consisted of an aluminum block with holes to fit up to six teflon sleeves

and quartz digestion tubes and caps. The block was mechanically clamped closed

to secure the quartz tubes within the teflon sleeves and then placed into an oven

at 180 oC for 0.5 to 16 hours. Post digestion, the samples were filtered through

quartz fiber filters (27 mm diameter, 0.8 µm pore diameter) and rinsed with 15 mL

Milli-Q water that was generated immediately prior to use. The filtrate was then

freeze dried.

Dried samples were redissolved in 5 mL methanol and the internal standard,

biphenyl-2,2’-dicarboxylic acid (1 mg mL-1 in methanol) was added. Samples were

derivatized by titration with 2.0 M trimethylsilyl diazomethane in ethyl ether (Sigma

Aldrich). Derivatization was considered complete when the solution retained the

yellow color of the trimethylsil-diazomethane.

The derivatized oxidation products were blown dry under ultrahigh purity nitrogen

and re-dissolved in methylene chloride, and subsequently separated and quanti-

fied on a Hewlett Packard 6890 gas chromatograph outfitted with a Gerstel cooled

injection system and a DB-XLB capillary column (30 m x 0.53 mm I.D., 1.5 µm film

13

23

1

41056

78 9

(a) Anthracene

12

3457

8

9 10

6

(b) Phenanthrene

12

3457

8

9 10

6

(c) Retene

12

34

567

8

9

11

10

12

(d) Chrysene

1

3

2

456

7

8

910

(e) Pyrene

1

3

2

456

7

8

910

NO2

(f) 1-Nitropyrene

10

1112 1

32

45

678

9

(g) Perylene

12

3

4

56

7

89

10

1112

(h) Benzo-ghi-perylene

12

3

4

567

8

9

10

1112

(i) Coronene

Figure 2.2: The PAHs used in this study.

14

thickness) and a flame ionization detector (FID). After injection, the column tem-

perature was maintained at 100 oC for 1 minute, then raised at 25 oC min-1 to 250

oC followed by a 5 oC min-1 ramp to a final temperature of 280 oC. The column was

held at the final temperature for 10 minutes. The detector temperature was 300 oC.

The split-less injection volume was between 1 and 3 µL.

Compound verification was performed using a Finnigan Trace MS+ GC/MS system

operating in electron ion (EI) mode. The GC was equipped with a J&W Scientific

DB-5 capillary column (30 m x 0.32 mm I.D., 0.25 µm film thickness). Helium was

used as the carrier gas. The temperature program used was 50 oC ramping at 10

oC min-1 to a final temperature of 290 oC. The injector temperature was 250 oC.

BPCAs were identified by comparison of their retention times with those obtained

for a commercially available mixture and were verified using the GC/MS. All methy-

lated BPCAs were quantified relative to the biphyenl-2,2’-dicarboxylic acid internal

standard. Unlike previous BPCA studies {Glaser et al., 1998; Brodowski et al.,

2005}, response factors were not required to correct for incomplete derivatization.

All methylated BPCAs exhibited the same response factor and BPCA calibration

curves were calculated relative to the internal standard peak area. Except where

otherwise noted, samples were processed and analyzed in triplicate. Detection of

BPCAs was limited to 10 ng BPCA per injection.

15

COOH

COOH

(a) phthalic acid

COOH

COOH

COOH

(b) hemimellitic acid

HOOC COOH

COOH

(c) trimellitic acid

COOH

COOH

HOOC

(d) trimesic acid

COOH

COOH

COOH

COOH

(e) prehnitic acid

COOH

COOH

COOH

HOOC

(f) mellophanic acid

HOOC

HOOC COOH

COOH

(g) pyromellitic acid

COOH

HOOC

HOOC COOH

COOH

(h) benzene pentacarboxylicacid

COOH

HOOC

HOOC

COOH

COOH

COOH

(i) mellitic acid

Figure 2.3: Structures of benzene polycarboxylic acids used as markers of aro-matic carbon in this study.

16

2.4 Results and Discussion

2.4.1 Nitration of BPCAs

Previous BPCA studies {Glaser et al., 1998; Brodowski et al., 2005} quantified only

the non-nitrated BPCAs produced during oxidation and did not consider whether

the oxidation products of the BC also included significant amounts of nitrated BP-

CAs. Early studies of the organic chemistry of electrophilic substitution found that

nitration of PAHs was important {Dewar and Mole, 1956; Watts, 1873}. We have

found that the majority of the BPCAs produced from all PAHs studied were sub-

stituted with at least one nitro-group (-NO2). Mono- and di-nitrated B2CAs were

observed. All other BPCAs (B3CA, B4CA and B5CA) were mono-nitrated. Both

3-nitrophthalic and 4-nitrophthalic acid are commercially available and were used

for calibration of the nitrated B2CAs. Larger nitrated BPCAs were not commer-

cially available. However, since the calibration curves for 3-nitrophthalic and 4-

nitrophthalic acid were the same as that for phthalic acid, we applied the non-

nitrated calibration curves to the larger BPCAs (e.g.: the calibration curve for B3CA

was applied to all nitrated B3CA isomers). The BPCA distribution and carbon yields

discussed below include nitrated BPCAs.

2.4.2 Carbon yields

For each oxidized and derivatized PAH, the BPCAs were quantified and a carbon

yield was calculated by comparing the sum of BPCA carbon to the initial carbon

(see Appendix A for an example). For all nine PAHs analyzed in this study, the

average carbon yield was 25.7 ± 6.8 % C, with values for individual PAHs ranging

17

Tabl

e2.

1:C

arbo

nyi

eld

and

perc

entB

PC

Adi

strib

utio

nof

the

nine

PAH

sox

idiz

edin

this

stud

y.A

llsa

mpl

esw

ere

oxid

ized

for8

hour

s.±

isth

est

anda

rdde

viat

ion

ofth

ree

repl

icat

es.

-ind

icat

esno

BP

CA

sw

ere

dete

cted

.

#of

C%

C%

Cre

cove

red

B2C

AB

3CA

B4C

AB

5CA

B6C

AA

nthr

acen

e14

94.4

24.2±

1.6

100.

0±

0.0

--

--

Phe

nant

hren

e14

94.4

23.7±

n.a.

72.1±

2.1

-27

.9±

2.1

--

Ret

ene

1892

.329

.3±

1.4

1.3±

2.2

62.4±

0.8

36.4±

1.7

--

Chr

ysen

e18

94.7

21.5±

2.3

64.4±

2.5

-35

.6±

2.5

--

Pyr

ene

1695

37.0±

7.2

-18

.8±

3.5

79.3±

5.2

--

1-N

itrop

yren

e16

77.4

36.1±

9.3

4.0±

3.5

17.9±

3.5

78.1±

5.4

--

Pery

lene

2095

.222

.5±

1.3

-80

.3±

1.7

--

19.7±

1.7

Ben

zo-g

hi-p

eryl

ene

2295

.522

.5±

3.9

-6.

9±

0.4

53.1±

1.2

-40

.0±

1.5

Cor

onen

e24

9619

.5±

2.2

--

67.4±

2.4

-32

.6±

2.4

18

from 19.5 ± 2.2 to 37.0 ± 7.2 % C (Table 2.1). The two smallest PAHs studied,

anthracene and phenanthrene, exhibited carbon yields of 24.2 ± 1.6 % and 23.7 %

(n=1), respectively. Retene, a three ring PAH with two small side chains, exhibited

a significantly higher carbon yield (29.3 ± 1.4 %). Chrysene, a four ring PAH with

a structure similar to phenanthrene, had a carbon yield of 21.5 ± 2.3 %. Four-

ring pyrene and 1-nitropyrene had the highest measured carbon yields, 37.0 ± 7.2

% and 36.1 ± 9.3 % respectively. The five ring perylene and six ring benzo-ghi-

perylene had the same carbon yield (22.5 ± 1.3 % and 22.5 ± 3.9 % respectively).

The largest PAH studied, seven ring coronene, had the lowest carbon yield of 19.5

± 2.2 %. No significant correlations were found between the carbon yield and the

number of aromatic rings or the percentage of carbon in the PAH; thus we are not

able to draw any conclusions how the type or size of PAH oxidized is related to the

carbon yield.

The BPCA method requires the use of a conversion factor to convert the BPCAs

formed into an estimate of BC mass. Previously activated charcoal was used to

determine the conversion factor {Glaser et al., 1998; Brodowski et al., 2005}. Since

the composition and character of this material may vary between production lots of

activated charcoal, the distribution and yield of BPCAs from this material may vary.

We recommend using materials of known chemical formulas, such as PAHs, to

calibrate the BPCA method. Our results from the oxidation of PAHs suggest using

the average carbon yield of 25.7 ± 6.8 % C to calculate the BC mass in samples.

2.4.3 BPCA products of PAHs

The BPCA method oxidizes condensed aromatic structures to produce single ben-

zene rings with carboxylic acid functional groups derived from adjacent aromatic

19

COOH

COOH

HOOC

HOOC

COOH

COOHHNO3180oC

COOH

HOOCA

A’A’A

BBor or

Figure 2.4: Schematic of the oxidation products of phenanthrene using the BPCAmethod. One molecule of phenanthrene theoretically could produce either onemolecule B2CA (A or A’) or one molecule B4CA (B). Using the observed distribu-tion of BPCAs produced and assuming no losses occurred during oxidation, thetheoretical carbon yield would be 57 %.

rings or side chains. For example, when phenanthrene is oxidized (Figure 2.4) only

two BPCAs are produced: phthalic acid (B2CA) and benzene-1,2,3,4-tetracarboxylic

acid (B4CA). If the method does not oxidize the aromatic structure preferentially

(regioselective), we would expect two molecules of B2CAs to form for every one

molecule of B4CA, based on the fact that one phenanthrene consists of two outer

rings suited to become B2CAs and one central ring suited to become B4CA. In-

deed, oxidized phenanthrene preferentially formed B2CA (72.1 ± 2.1 % of the

carbon recovered) and the remainder was B4CAs (27.9 ± 2.1 %). Oxidized an-

thracene also is expected to yield 66.7 % B2CA and 33.3 % B4CA, however, it

produced exclusively B2CAs. The four-ring PAH chrysene is expected to yield

equal proportions of B2CA and B4CA, however it yielded 64.4 ± 2.5 % B2CA and

35.6 ± 2.5 % B4CA. Anthracene and chrysene also preferentially formed smaller

BPCAs than expected.

The oxidation of retene yielded the expected distribution of BPCAs (2 B3CAs:1

B4CA, Table 2.1). Carbon on each side chain was oxidized to a carboxylic acid.

This result suggests that aliphatic side chains of BC could be oxidized to carboxylic

acids. Theoretically, perylene would yield four B3CAs for every one B6CA and

indeed, we found 80.3 ± 1.7 % B3CAs and 19.7 ± 1.7 % B6CA formed. Previously,

20

perylene oxidation was reported {Glaser et al., 1998} to yield 75 % B3CAs and 25

% B6CAs, similar to ours. In contrast, Dittmar {2008} reported 19 % B3CA and 81

% B6CA from oxidized perylene, however his quantification methods (microwave-

assisted oxidation and HPLC quantification) were different from our study. This

study did not use microwave-assisted oxidation and BPCAs were quantified as

methyl-esters by GC rather than carboxylic acids via HPLC.

Other PAHs used in this study formed more of the larger BPCAs than expected.

Pyrene formed 18.8 ± 3.5 % B3CA and 79.3 ± 5.2 % B4CA instead of the ex-

pected equal distribution. The oxidation products of nitrated pyrene, 1-nitropyrene,

was not significantly different from the non-nitrated compound, except that a small

percentage (4.0 ± 3.5 %) of B2CA was produced. The observed B2CA was a mix-

ture of two isomers of dinitro-B2CA. Benzo-ghi-perylene was expected to form 33

% B3CA, 50 % B4CA and 17 % B6CA; instead it formed 6.9 ± 0.4 % B3CA, 53.1

± 1.2 % B4CA and 40.0 ± 1.5 % B6CA. The proportion of B4CAs formed was as

expected, however more B6CAs and less B3CAs were measured than expected.

The largest PAH studied, coronene, was expected to form 86 % B4CA and 14 %

B6CA. Instead, coronene produced 67.4 ± 2.4 % B4CA and 23.6 ± 2.4 % B6CA,

again more B6CAs than expected. These results show that larger PAHs generally

formed larger BPCAs than predicted.

These BPCA distribution data illustrates the complexity of the oxidation reaction.

Since there seems to be no systematic pattern of oxidation, we cannot accurately

model the oxidation products. Nor can we, without additional data, reconstruct the

original structure of the BC using the BPCA distribution. However, these distribu-

tion data are useful when drawing qualitative distinctions between different types

of BC {Ziolkowski and Druffel, 2009}. It is possible to distinguish between material

with aliphatic side-chains and fully condensed BC material.

21

Table 2.2: Carbon yields for time course experiments of two PAHs (anthraceneand perylene) and mellitic acid (B6CA). All samples were oxidized at 180 oC forthe time listed. ± is the standard deviation of three replicates. - indicates timepoints that were not studied.

oxidation time % C recovered(hours) Anthracene Perylene Mellitic acid

0.5 16.5 ± 8.5 - -1 11.7 ± 3.4 - 95.9 ± 7.72 18.6 ± 1.7 20.4 ± 0.1 97.4 ± 134 20.5 ± 0.7 24.0 ± 1.1 97.3 ± 5.78 23.6 ± 0.2 22.6 ± 1.3 104.7 ± 1.0

16 18.9 ± 3.1 19.9 ± 3.4 95.2 ± n.a.

2.4.4 Time course and mechanistic experiments

To evaluate the optimal time of the high pressure and high temperature oxidation of

PAHs to BPCAs, we conducted time course experiments. Anthracene was chosen

as a model compound for time course and mechanistic experiments because it was

being evaluated for use as an internal standard. First, we evaluated the evolution of

BPCAs from anthracene by conducting the high temperature nitric acid oxidation

from 0.5 to 16 hours. The carbon yield of BPCAs increased from 1 to 8 hours

(Table 2.2). The 16 hour oxidations did not yield significantly different amounts or

distributions of BPCAs than 8 hour oxidations, although the standard deviations at

16 hours was much larger than the previous three time points. Therefore oxidation

of at least 8 hours were optimal, as shorter oxidations gave lower carbon yields and

longer oxidations a greater variability of carbon yield. We also examined the degree

of nitration of the B2CAs formed from anthracene as a function of time (Figure

2.5). With oxidations of 0.5 and 1.0 hour the B2CA formed were exclusively dinitro-

B2CA. Between the one and two hour oxidations, the carbon yield increased and

the quantity of dinitro-B2CA produced decreased significantly and was replaced by

B2CA and mono-nitro-B2CA (Figure 2.5).

22

0%

20%

40%

60%

80%

100%

% o

f to

tal

0.5 1 2 4 8 16duration of oxidation (hours)

B2CAmononitro-B2CAdinitro-B2CAC yield

Figure 2.5: Distribution of non-nitrated, mono-nitrated and di-nitrated dicarboxylicacid formed and carbon yield from oxidation of anthracene as a function of oxida-tion time.

HNO3180oC

O

O

O

O

NO2

NO2

O

O NO2

NO2

slow

fast

COOHCOOH

COOHCOOH

NO2

COOHCOOH

NO2O2N+ +

Figure 2.6: Proposed reaction schematic for the high pressure, high temperatureoxidation of anthracene to B2CA. The initially formed products undergo thermody-namic equilibration to primarily mononitro-B2CA (see Figure 2.7).

23

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

anthracene, 45min

3,5-dinitrophthalic acid,16 hours

phthalic acid, 16 hrs

3-nitrophthalic acid,16hrs

4-nitrophthalic acid,16hrs

% of total B2CA measured

B2CA

mononitro-B2CA

dinitro-B2CA

Figure 2.7: Distribution of non-, mono- and di-nitrophthalic acid as a function ofreactants. Four forms of phthalic acid were oxidized for 16 hours: phthalic acid,3-nitrophthalic acid, 4-nitrophthalic acid and 3,5-dinitrophthalic acid. Regardlessof the starting materials 3-nitrophthalic acid is the most abundant product after 16hours. This experiment was conducted in duplicate and the difference betweenduplicates was ≤ 5 %.

24

These results suggest that the oxidation mechanism of anthracene is a multistep

process with dinitrophthalic acid as the initial product (≤ 1 hr). With increased ox-

idation time (8 hr) the B2CA reaches an “equilibrium” state as nitro-phthalic acid.

When the shortest oxidations of anthracene were filtered, a solid remained, identi-

fied by GC/MS and NMR as 100 % anthraquinone. Oxidation of anthracene in ni-

tric acid, under milder conditions, has been found to produce anthraquinone {Cho,

1995}. We hypothesize that the anthraquinone generated was then nitrated to dini-

troanthraquinone before being oxidized to dinitrophthalic acid (Figure 2.6) which

was then denitrated to mononitro-B2CA, the thermodynamic sink under these con-

ditions.

We tested this hypothesis by oxidizing four forms of B2CA: phthalic acid, 3-nitro-

phthalic acid, 4-nitrophthalic acid and 3,5-dinitrophthalic acid for 16 hours. With

the exception of 4-nitrophthalic acid, we found that regardless of the form of ph-

thalic acid we oxidized, the primary product was 3-nitro-phthalic acid (Figure 2.7),

supporting our hypothesis. Furthermore, after 16 hours of oxidation 4-nitrophthalic

acid yielded 85 % mononitro-B2CA and 15 % dinitro-B2CA and more than half of

the mononitro-B2CA was 3-nitrophthalic acid. This demonstrated the conversion of

4-nitrophthalic acid to the more stable 3-nitrophthalic acid. These results confirm

the importance of quantifying nitrated BPCAs, because nitration occurs before the

formation of BPCAs and can comprise a significant portion of the products.

Perylene was also studied in a time course experiment. Although the carbon yield

was relatively constant over the course of the these experiments at 21.7 ± 1.9 %

(Table 2.2), the distribution of BPCAs changed significantly as a function of oxi-

dation time (Figure 2.8). The shortest perylene oxidation (2 hours) yielded more

B6CA than the 16 hour oxidation (23.3 ± 0.5 % B6CA versus 14.4 ± 0.3 %). The

mononitro-B3CA was predominantly 4-nitro-1,2,3-benzenetricarboxylic acid and a

25

0

20

40

60

80

100

0 5 10 15 20oxidation duration (hours)

% o

f to

tal BPC

As

obse

rved

(ci

rcle

s)

0

10

20

30

40

50

% c

arbon y

ield

(tr

iangle

s)

B3CAB6CAC yield

Figure 2.8: Change in B3CA (filled circles) and B6CA (open circles) oxidationproducts (% of total BPCAs observed) and carbon yield (filled triangles) from pery-lene over time (% carbon yield). Error bars represent the standard deviation ofthree replicates.

26

small proportion was 5-nitro-1,2,3-benzenetricarboyxlic acid (≤ 10 %). Relative dis-

tributions of these two acids did not vary with increased oxidation time (up to 16 hr).

Dewar and Mole {1956} reported that perylene nitrates at the #3 position, which is

consistent with our observation that 4-nitro-1,2,3-benzenetricarboyxlic acid is the

dominant B3CA formed. This provides further confirmation that nitration occurs

before the break-up of the PAHs. The decrease in B6CA with oxidation time may

indicate that decarboxylation may take place with longer oxidation times. It is im-

portant to note that, while the ratio of B3CA to B6CA doubled between 2 and 16

hours, the carbon yield did not significantly change. Thus if decarboyxlation is

occurring, in this case it did not change the carbon yield. At no point during the

perylene time course experiments did we observe equivalent amounts of oxidation

products, as reported when microwave assisted oxidation was employed {Dittmar ,

2008}. In the future, it is important for users of the BPCA to quantify nitrated BP-

CAs and calibrate the oxidation performed in each lab. Comparing the ratio of

smaller to larger BPCAs (including nitrated ones) of a known compound, such as

a PAH, should be used to calibrate the method for inter-lab comparisons of BPCA

distributions.

To test for decarboxylation, we oxidized commercially available mellitic acid (B6CA)

at 180 oC for 1 to 16 hours (Table 2.2). The average recovery of mellitic acid for

all time points was 98.1 ± 3.8 %. For all time points except 8 hours, the amount

of mellitic acid remaining after oxidation was between 95 and 97 %. The 8 hour

oxidation yielded 104.7 ± 1.0 % of the initial carbon. These results demonstrate

that mellitic acid is not decarboxylated over the course of the oxidation.

The time course results demonstrate that oxidations conducted for 4 to 16 hours

show no change in the carbon yield, whereas the relative distribution of BPCAs

changes as a function of oxidation time. Since the anthracene carbon yield (Table

27

2.2) and nitration (Figure 2.5) continued to evolve from 4 to 8 hours of oxidation

time, we elected to conduct all further oxidations at 8 hours.

2.4.5 Analysis of black carbon ring trial materials

An analytical challenge in the analysis of black carbon is its wide variety of chemical

and physical characteristics. Many methods of BC quantification focus on partic-

ular components of BC (i.e. soot or char). Recently, an inter-comparison of BC

quantification methods for BC rich materials {Hammes et al., 2007} revealed that

the previous version of the BPCA method were more well suited for the analysis of

char than for soot BC. Additionally, the inter-comparison revealed that the conver-

sion factor of BPCAs to BC was not easily reproducible.

Since many modifications of the BPCA method were made in this work, we ana-

lyzed a suite of BC rich materials to contextualize this version of the BPCA method

(Table 2.3, Figure 2.9). The PAH carbon yield data reported here were combined

with carbon yield data of soot-like BC materials in another study (Ziolkowski and

Druffel {2009}, Chapter 3) to generate a robust conversion factor, for converting

BPCAs to BC, of 4 ± 1 or the inverse of 25 ± 6 %. A wide range of BC materials

(e.g.: carbon nanotubes, soot, char, PAHs, etc) were used to generate this con-

version factor. This is much higher than the 2.27 conversion factor reported in the

original BPCA study {Glaser et al., 1998}, but lower than the highest BPCA conver-

sion factor measured (4.5) by Brodowski et al. {2005}. Activated charcoal, which

was used to generate previous BPCA conversion factors, was not used as in the

determination of the conversion factor reported here. Although, carbon yields of

oxidized activated charcoal (23.5 ± 1.0 % C yield, n=3) agree with the conversion

factor reported here.

28

Tabl

e2.

3:Q

uant

ifica

tion

ofbl

ack

carb

onm

ater

ials

(gB

C/k

gdr

yw

eigh

t)by

this

and

othe

rm

etho

ds.

Dat

afo

rC

TO-3

75,

BP

CA

,C2O

7an

dTO

R/T

OT

are

from

Ham

mes

,200

7H

amm

eset

al.{

2007}.

The

quan

tyof

BC

inth

isw

ork

was

conv

erte

dfro

mal

lBP

CA

s(in

clud

ing

thos

eni

trate

d)to

BC

usin

ga

conv

ersi

onfa

ctor

of4

(the

inve

rse

of25

%).

Ada

shin

dica

tes

that

noda

taw

asre

port

ed.

The

unce

rtai

nty,

s,is

the

prop

agat

eder

ror

ofth

eB

PC

Ato

BC

conv

ersi

onfa

ctor

(25±

6%

)or

the

stan

dard

devi

atio

nbe

twee

nre

plic

ates

,whi

chev

eris

larg

er.

BP

CA

,thi

sw

ork

CTO

-375

BP

CA

C2O

7TO

R/T

OT

mea

ns

mea

ns

mea

ns

mea

ns

mea

ns

aero

sol

32.7

8.2

14.9

7.0

14.5

4.5

63.9

20.8

66.5

20.4

mar

ine

sedi

men

t10

.78.

15.

11.

41.

71.

511

.86.

610

.96.

5IH

SS

NO

M63

.015

.81.

11.

920

.525

.1-

-16

2.3

135.

1he

xane

soot

945.

523

6.4

410.

08.

323

9.5

206.

946

9.9

97.8

887.

624

.3w

ood

char

478.

411

9.7

--

183.

296

.352

4.4

106.

765

2.7

93.5

gras

sch

ar48

8.0

122.

09.

07.

215

4.6

18.3

205.

849

.447

8.0

76.0

29

Figure 2.9: BPCA distribution of materials used in the black carbon ring trial.

Table 2.4: Wood char and hexane soot BC yield data (g BC / kg dry wt) using var-ious scenarios.

∑nBPCA is the summation of all BPCAs, including those nitrated.∑

BPCAs is the summation of all non-nitrated BPCAs. BPCA (last column) is thedata reported in the BC ring trial Hammes et al. {2007} summing the non-nitratedBPCAs. For the work presented here (

∑nBPCA and

∑BPCA), the uncertainty, s,

is the propagated error of the BPCA to BC conversion factor (25 ± 6 %).

BPCA summation∑

nBPCA∑

BPCA∑

BPCA BPCAconversion factor 4.09 4.09 2.27 2.27

mean s mean s mean s mean swood char 478.4 119.6 385.2 96.3 213.8 53.4 183.2 96.3

hexane soot 945.5 236.5 616.8 154.2 334.0 83.5 239.5 206.9

30

For all materials assessed, our BC yields were higher than both the CTO-375

and original BPCA method. Our results are similar to those generated using the

chromic acid oxidation method (C2O7) for charred materials, while soot materials

are closer to the thermal optical method (TOR/TOT). The increased BC yield using

this version of the BPCA method is due to various factors, most like to the quan-

tification of nitrated BPCAs and possibly in small part to the derivatization method.

For wood char, if the 2.27 conversion factor was applied to the sum of non-nitrated

BPCAs, the BC yield is lowered from 478 g/kg to 213 g/kg, much closer to the

previous BPCA methods of the amount of BC in wood char (Table 2.4). The BC

yield for hexane soot using only the non-nitrated BPCAs were converted to BC

with a 2.27 conversion factor is 334 g/kg and is still significantly higher than the

previous BPCA estimates for hexane soot. Thus, it appears that the BPCA method

presented here is not biased for char and can equally quantify char and soot BC.

After the conversion factor and quantification of nitrated peaks are accounted for,

the BC yield is still greater than previous versions of this method, likely due to the

increased oxidation temperature and derivatization method.

2.5 Conclusions

Using a revised BPCA method we have shown that oxidation of nine PAHs results

in nitrated BPCAs. On average, 25.7 % of the PAH carbon was recovered as BP-

CAs (including nitrated BPCAs). Although the number of acid groups is related to

the original structure, the distribution of oxidation products does not systematically

correlate with the structure of the original PAH. More highly substituted BPCAs are

preferentially formed from larger PAHs. Time course experiments revealed that the

ratio of oxidation products changed over time, favoring smaller BPCAs with longer

31

oxidation times, while the carbon yield did not change. We also found that quanti-

fying the nitrated BPCAs is essential as the PAHs were nitrated before they were

oxidized. Future work with the BPCA method should assess the degree of nitrated

BPCAs when oxidizing BC in environmental samples, as it may provide further in-

formation about the oxidation process. Measurements of BC in reference materials

reveal that this version of the BPCA method is overall more efficient at quantifying

BC and no longer is biased against the quantification of soot BC. The increased

efficiency is a function of the oxidation conversion factor, quantification of nitrated

peaks, derivatization method and increased temperature of oxidation of BPCAs to

BC.

Acknowledgements

The authors would like to thank Sheila Griffin and Claudia Czimczik for their help

with this work. We acknowledge funding from the NSF Chemical Oceanography

Program.

Bibliography

Brodowski, S., A. Rodionov, L. Haumaier, B. Glaser, and W. Amelung, Revised

black carbon assessment using benzene polycarboxylic acids, Organic Geo-

chemistry, 36(9), 1299–1310, 2005.

Cho, B., Recent progress in the synthesis of nitroarenes. A review., Organic Prepa-

rations and Procedures Int., 27(3), 243 – 272, 1995.

32

Dewar, M., and T. Mole, Electrophilic substitution. Part II. The nitration of naptha-

lene and perylene., Journal of the Chemical Society, pp. 1441 – 1443, 1956.

Dittmar, T., The molecular level determination of black carbon in marine dissolved

organic matter, Organic Geochemistry, 39(4), 396–407, 2008.

Eglinton, T., L. Aluwihare, J. Bauer, E. Druffel, and A. McNichol, Gas chromato-

graphic isolation of individual compounds from complex matrices for radiocarbon

dating, Analytical Chemistry, 68(5), 904–912, 1996.

Glaser, B., L. Haumaier, G. Guggenberger, and W. Zech, Black carbon in soils:

the use of benzenecarboxylic acids as specific markers, Organic Geochemistry,

29(4), 811–819, 1998.

Goldberg, E. D., Black carbon in the Environment, Wiley, 1985.

Gustafsson, O., F. Haghseta, C. Chan, J. Macfarlane, and P. Gschwend, Quantifi-

cation of the dilute sedimentary soot phase: Implications for PAH speciation and

bioavailability, Environmental Science and Technology, (31), 203 – 209, 1997.



soil, water, sediment and the atmosphere, Global Biogeochemical Cycles, 21(3),

18, 2007.

Masiello, C. A., New directions in black carbon organic geochemistry, Marine

Chemistry, pp. 201 – 213, 2004.

Schramel, P., A. Wolf, R. Seif, and B. Klose, Eine neue apparatur zur druck-

veraschung von biologischem material, Fresenius Zeitschrift fur Analytische

Chemie, (302), 62–64, 1980.

33

Simpson, M., and P. Hatcher, Determination of black carbon in natural organic

matter by chemical oxidation and solid-state 13C nuclear magnetic resonance

spectroscopy, Organic Geochemistry, (358), 923 – 935, 2004.

Skemstad, J., P. Clarke, J. Taylor, J. Oades, and S. McClure, The chemistry and

nature of protected carbon in soil, Australian Journal of Soil Research, (34), 251

– 271, 1996.

Watts, H., A Dictionary of Chemistry and the Allied Branches of Other Sciences,

vol. IV, Longmans, Green and Co., 1873.

Wolbach, W., and E. Anders, Elemental carbon in sediments: Determination and

isotopic analysis in presence of kerogen, Geochimica Cosmochimica Acta, (53),

1637 – 1647, 1989.

Ziolkowski, L., and E. Druffel, The feasibility of isolation and detection of fullerenes

and carbon nanotubes using the benzene polycarboxylic acid method, Marine

Pollution Bulletin, 59, 213 – 218, 2009.

34

3The feasibility of isolation and detection of

fullerenes and carbon nanotubes using the

benzene polycarboxylic acid method

Presented here with minor editing:

Ziolkowski and Druffel (2009), The feasibility of isolation and detection of fullerenes

and carbon nanotubes using the benzene polycarboxylic acid method, Marine Pol-

lution Bulletin, 59 (4-7) 213-218, doi: 10.1016/j.marpolbul.2009.04.018.

3.1 Abstract

The incorporation of fullerenes and carbon nanotubes into electronic, optical and

consumer products will inevitably lead to the presence of these anthropogenic com-

pounds in the environment. To date, there have been few studies isolating these

materials from environmental matrices. Here we report a method commonly used

to quantify black carbon (BC) in soils, the benzene polycarboxylic acid (BPCA)

method, for measurement of two types of single-walled carbon nanotubes (SWC-

NTs), two types of fullerenes and two forms of soot. The distribution of BC prod-

35

ucts (BPCAs) from the high pressure and high temperature oxidation illustrates

the condensed nature of these compounds because they form predominantly fully

substituted mellitic acid (B6CA). The conversion of carbon nanoparticles to BP-

CAs was highest for fullerenes (average of 23.2 ± 4.0 % C recovered for both C60

and C70) and lowest for non-functionalized SWCNTs (0.5 ± 0.1 % C). The recov-

ery of SWCNTs was 10 times higher when processed through a cation-exchange

column, indicating the presence of metals in SWCNTs compromises the oxidation

chemistry. While mixtures of SWCNTs, soot and sediment revealed small losses

of black carbon during sample processing, the method is suitable for quantifying

total BC. The BPCA distribution of mixtures did not agree with theoretical mixtures

using model polyaromatic hydrocarbons, suggesting the presence of a matrix ef-

fect. Future work is required to quantify different types of black carbon within the

same sample.

3.2 Introduction

As the industrial application for carbon nanotubes (CNTs) and fullerene production

increases, their presence in the environment is an eventuality. While these com-

pounds have much biomedical promise {Bianco et al., 2005}, there is conflicting

eco-toxic evidence {Oberdorster , 2004; Tong et al., 2007} about their impact on

organisms in nature. The analytical methods used to isolate these materials from

environmental matrices are limited, inhibiting our ability to directly quantify these

compounds.

Fullerenes, also known as buckyballs, consists of twelve pentagonal rings sur-

rounded by an appropriate number of aromatic hexagonal rings. C60, first reported

by Kroto et al. {1985} has 20 hexagon rings and is spherical, while C70 has 25

36

hexagons and an elongated shape. Used in industrial polymer products, such

as thin films, electro-optical devices {Prato, 1999} and drug delivery agents {Bosi

et al., 2003; Bianco et al., 2005}, fullerene production has increased annually.

Numerous techniques have been used to characterize fullerenes, such as mass

spectrometry and UV-Vis spectroscopy {Isaacson et al., 2007; Andrievsky et al.,

2002}, but few methods have successfully isolated fullerenes from environmental

matrices. Carbon nanotubes exhibit different chemical and physical properties de-

pending on the method of production, removal of amorphous carbon and function-

alization {Dai, 2002; Niyogi et al., 2002; Plata et al., 2008}. Commercially available

SWCNTs are typically produced on a metal catalyst and can be up to one-third

metal by weight {Plata et al., 2008}. To date, CNTs have been studied mostly by

size exclusion chromatography {Bauer et al., 2007}, electron microscopy {Rasheed

et al., 2007} and chemo-thermal oxidation {Sobek and Bucheli, 2009}.

Recently the benzene polycarboxylic acid (BPCA) method has been employed to

study BC in soils {Glaser et al., 1998; Brodowski et al., 2005}, marine sediment

{Sanchez-Garcia et al., 2007} and marine dissolved organic matter {Dittmar , 2008;

Ziolkowski and Druffel, 2008}. Using a high-temperature and high-pressure oxi-

dation, the BC is chemically oxidized with concentrated nitric acid and converted

to BPCAs. The number of carboxylic acid groups on each BPCA is a function

of the number of aromatic carbons attached to it prior to oxidization (e.g.: Fig-

ure 3.1). Currently, the mechanism of this reaction is unknown. Fully substituted

BPCAs (B6CA) are formed from aromatic rings surrounded on all sides by other

aromatic rings, while less substituted BPCAs (i.e. B3CA) are formed from aro-

matic rings with only two adjacent aromatic rings. Thus by examining the relative

distribution of BPCAs, information about the source BC material may be obtained.

Post-oxidation, if a BPCA distribution is predominantly B6CA the original BC mate-

rial was likely a condensed aromatic, while BC material that is less condensed and

37

Figure 3.1: Oxidation of the PAH perylene yields two BPCAs: 20 % mellitic acid(B6CA) and 80 % trimellitic acid (B3CA), suggesting that the oxidation productsare reflective of the original structure of the condensed aromatic material. Thereported percentages are based on mg C recovered.

more oxidized will form fewer B6CAs with a greater proportion of smaller BPCAs.

Since BC and CNTs are similar in their condensed aromatic structure, it is likely

that similar extraction techniques could be used. Here we test the feasibility us-

ing the BPCA method to isolate and quantify carbon nanoparticles in the marine

environment. In this paper we present the distribution of BPCAs and the percent

carbon recovered for two fullerenes, two carbon nanotubes and two other carbon

nanoparticles, hexane soot and carbon lampblack. Using mixtures of these carbon

nanopoarticles in marine sediment, we quantify CNTs and evaluate possible matrix

effects.

3.3 Methods

We obtained polycyclic aromatic hydrocarbons fullerenes (C60 and C70) and two

single walled carbon nanotubes (SWCNT) from Sigma Aldrich (Figure 3.2). The

first SWCNT was 1 - 2 nm O.D. x 0.5 - 2 µm in length. The second was function-

alized (SWCNT-F) with 3-6 % carboxylic acid groups and was 4 - 5 nm O.D. by 0.5

-1.5 µm in length. Hexane soot obtained from D.M. Smith (University of Denver)

was analyzed previously by Akhter et al. {1985} using spectroscopic techniques

and quantified for BC by Hammes et al. {2007}. Commercially available carbon

38

(a) Buckminsterfullerene, C60 (b) Buckminsterfullerene, C70, afterMckenzie et al. {1992}

(c) SWCNT, nonfunctionalized

Figure 3.2: Three of the six carbon nanoparticles studied.

lampblack (Fisher) was also analyzed. Perylene (Sigma Aldrich) was used as a

model compound to investigate the oxidation process (Figure 3.1). Marine sedi-

ment (NIST SRM 1941b) was used as an environmental matrix for mixed samples.

Two to seven mg carbon were digested in 2 mL 65 % HNO3 at 180oC for 8 hours

(unless otherwise noted) as reported by Glaser et al. {1998}; Brodowski et al.

{2005} and in Chapter 2. During the digestion BC is chemically oxidized to form

BPCAs. The solution was passed though a 0.8 µm pore size quartz fiber filter into

a filtration flask and washed with 30 mL of deionoized water. To remove polyvalent

cations from the filtrate that interfere with sample analysis, a number of samples

received additional treatment. This subset of samples was pretreated with 10 mL

39

of 4 M trifluoroacetic acid for 4 hours at 104 oC. Following nitric acid oxidation and

filtration, samples were passed through a cation column (H+ form, Dowex 50W-X8,

200-400 mesh, packed 18 cm ID x 10 cm high) that was subsequently rinsed with

an additional 30 mL of deionized water and combined with the filtrate. Samples

were then freeze-dried for 24 hours.

Five mL of methanol was added to the dried BPCAs along with 500 µL of a 1 mg

mL−1 solution of biphenyl-2,2-dicarboyxlic acid (Sigma Aldrich) in methanol that

was used as a derivatization standard. Samples were then methylated by titra-

tion with (trimethylsilyl)diazomethane in diethyl ether (Sigma Aldrich) until the sam-

ple solution remained yellow, indicating the presence of un-reacted diazomethane.

Samples were then dried under a stream of purified ultra-high purity nitrogen. A

fixed volume of dichloromethane was then added as a solvent. All samples were

separated on a Hewlett Packard 6890 GC outfitted with a Gerstel cooled injec-

tion system and a DB-XLB capillary column (30 m x 0.53 mm I.D., 1.5 µm film

thickness) and a flame ionization detector (FID). After injection, the column tem-

perature was maintained at 100 oC for 1 minute, then raised 25 oC min−1 to 250 oC,

then raised 5 oC min−1 to a final temperature of 280 oC. The detector temperature

was 300 oC. The split-less injection volume was between 1 and 3 µL. Benzenepoly-

carboxylic acids were identified by comparison of their retention times with those

obtained by a commercially available mixture, verified by GC-mass spectrometry

and quantified by GC-FID. All methylated BPCAs were quantified relative to the

biphyenl-2,2-dicarboxylic acid internal standard. No additional response factors

were applied, as all methylated BPCAs exhibited an equal response to detection.

Initial work studying the oxidation products of PAHs with the BPCA method yielded

BPCAs substituted with -NO2 groups, due to the nitric acid oxidation (Chapter 2).

This study quantifies these nitrated BPCAs. Omitting these nitrated peaks from

40

quantification would lead to an underestimate of the BPCAs formed in oxidation.

Non-nitrated BPCAs are used as reference materials, as nitrated BPCAs are not

commercially available. Phthalic acid and 3-nitro and 4-nitrophthalic acid exhibited

nearly identical calibration curves, therefore we assumed the same relationship

would hold true for larger nitrated BPCAs. All measurements were performed in

triplicate, unless otherwise noted.

3.4 Results

3.4.1 BPCA distributions

The BPCA method forms BPCAs only from condensed aromatic materials, such

as char, soot or polycyclic aromatic hydrocarbons (PAHs) {Glaser et al., 1998;

Brodowski et al., 2005}. Although some PAHs and BC materials form BPCAs with

two carboxylic acids, only those compounds with three or more acids groups were

quantified in this work. This assumption is employed to avoid erroneously quantify-

ing BC as BPCAs from non-BC material, such as lignin. Initially, the PAH perylene

was studied to understand the mechanism of high temperature and high pressure

nitric acid oxidation (Chapter 2). Upon oxidation of perylene we measured only two

BPCAs: the tri-substituted hemimelltic acid (B3CA) and the fully substituted mellitic

acid (B6CA) with the measured mole ratio of 4:1 (Figure 3.1). These results sug-

gest that the quantitative distribution of BPCAs can provide structural information

about the material being oxidized.

Our results show that most of the oxidation products of the fullerenes (C60 and C70)

and soots are the fully substituted mellitic acid (B6CA) with small portions of less

41

0

10

20

30

40

50

60

70

80

90

100

B3CA B4CA B5CA B6CA

% o

f tot

al B

PCAs

form

ed

hexane soothexane soot + cationcarbon lampblackC70C60C60 + cation

Figure 3.3: Distribution of BPCAs formed upon high temperature and high pressureacid oxidation relative to total BPCAs formed from fullerenes, carbon lampblackand soot with and without cation column processing.

42

substituted BPCAs (Figure 3.3). Oxidation of both C60 and C70, processed with-

out cation removal, produces the greatest yield of B6CAs (94.4 ± 0.7 and 92.2 ±

2.8 % of total BPCAs, respectively). The BPCA distribution of C60 did not change

significantly when processed through the cation column (Table 3.1). Based on the

structure of C60 (Figure 3.2a), only B6CA should be formed, which is confirmed by

these results. Carbon lampblack, processed without the cation column, had almost

equal proportions of B3CA, B4CA and B5CA (about 10 % each) and 70.8 ± 10.0 %

B6CA, suggesting that the structure of carbon lampblack is predominantly aromatic

rings surrounded by other rings. Without the cation column, hexane soot produced

the smallest proportion of B6CA of the materials in this study (46.5 ± 4.6 %) with

10.3 ± 1.6 % of B5CA, 21.5 ± 0.6 % B4CA and 21.7 ± 5.3 % B3CA. Process-

ing soot through the cation column (n=1) drastically shifted the BPCA distribution.

There was an increase in the proportion of B6CA and B5CA and a corresponding

decrease in the proportion of B4CA and B3CA. If one theorized about the structure

of hexane soot using the BPCA distribution using the cation processed samples,

a more condensed aromatic structure would be proposed than that using the non-

cation processed samples.

The distribution of oxidation products formed from SWCNTs varied markedly de-

pending on sample treatment (Figure 3.4). The non-carboxylic acid functionalized

SWCNTs exhibited significantly different distributions when processed with and

without the cation column. Oxidation products of SWCNTs processed through the

cation column were all B6CA (98.4 ± 2.2 %) and a trace of B4CA, while SWCNT

samples not processed through the cation column contained a mixture of mostly

B3CA and B4CA (37 ± 13 % B3CA and 43 ± 8 % B4CA) with the remainder as

B6CA (20 ± 17). B5CA were not observed in either sample. The SWCNT not

processed through the cation exchange resin had the greatest uncertainty asso-

ciated with the distribution of oxidation products, due to the low carbon yield (as

43

Table 3.1: Percent carbon yield for the carbon nanoparticles in this study. Unlessotherwise noted, all samples were treated for 8 hr.

No cation column processing

n B3CA B4CA B5CA B6CA % C YieldC70 3 0.9± 1.5 3.0±1.2 4.0±0.2 92.2± 2.8 26.0± 3.2C60 3 0.0± 0.0 2.6±0.7 3.0±0.0 94.4± 0.7 20.3± 2.1Carbon lamp black 3 11.3± 8.9 9.1±1.8 8.8±0.7 70.8±10.0 20.0± 4.1Hexane soot 3 21.7± 5.3 21.5±0.6 10.3±1.6 46.5± 4.6 25.3± 4.0SWCNT 3 37.1±13.3 43.3±8.1 0.0±0.0 19.6±17.1 0.8± 0.2SWCNT-F 3 18.4± 2.7 9.2±1.5 4.1±0.7 68.3± 3.5 8.2± 1.5SWCNT 16 hr 3 7.0± 6.5 5.3±2.2 0.0±0.0 87.7± 5.2 7.8± 1.2SWCNT-F 16 hr 3 27.1± 7.7 11.0±1.9 0.0±0.0 61.9± 9.6 11.1± 2.7

Processed through cation

n B3CA B4CA B5CA B6CA % C YieldC60 1 -± - 1.1± - 8.8± - 90.1± - 17.0± -Hexane soot 1 -± - 5.0± - 26.4± - 68.6± - 15.7± -SWCNT 2 -± - 1.2±1.8 0.0±0.0 98.4± 2.2 7.2± 1.6SWCNT-F 2 -± - 2.1±0.3 10.9±3.7 87.0± 3.4 7.2± 1.5SWCNT 16 hr 2 -± - 11.0±4.7 5.9±8.4 83.0±13.1 6.0± 1.9SWCNT-F 16 hr 2 -± - 13.7±3.4 5.2±3.8 81.1± 7.2 8.8± 0.3

discussed in the next section). Theoretically, SWCNT should form predominantly

B6CA because the structure consists of benzene rings surrounded on all sides by

other benzene rings. In contrast, oxidation of SWCNT-F produced mostly B6CA

regardless of cation processing (87.0 ± 3.4 % with and 68.3 ± 3.5 % without cation

processing). The other oxidation products (B4CA and B5CA) varied with cation

processing, such that fewer B5CA and more B4CAs and B3CAs were formed when

cation processing was not done. These results indicate a greater proportion of in-

terfering cations, likely the metal catalyst, in SWCNTs that was not observed with

the SWCNT-F samples.

We also investigated the effect of oxidation duration on the SWCNT BPCA distri-

bution. After 16 hours of oxidation (8 additional hours), the SWCNT-F oxidation

products were predominantly B6CA (81.1 ± 7.2 % with and 61.9 ± 9.6 % without

cation processing). The 16 hour oxidation of SWCNT did not yield a significantly

44

0

10

20

30

40

50

60

70

80

90

100

B3CA B4CA B5CA B6CA

% o

f tot

al B

PC

As

form

ed

SWCNT , 8 hrs

SWCNT, 16 hrsSWCNT-F, 8 hrsSWCNT-F, 16 hrs

0

10

20

30

40

50

60

70

80

90

100

B3CA B4CA B5CA B6CA

% o

f tot

al B

PCAs

form

ed

SWCNT , 8 hrsSWCNT, 16 hrs

SWCNT-F, 8 hrsSWCNT-F, 16 hrs

Figure 3.4: Distribution of BPCAs formed upon high temperature and high pressureacid oxidation relative to total BPCAs formed for two types of single walled carbonnanotubes for 8 hour oxidations and 16 hour oxidations. (a) samples processedwithout cation column and (b) samples processed through cation column.

45

different distribution of oxidation products for samples processed with and without

cation column processing. These distributions are not different from those obtained

from shorter oxidations of SWCNT.

3.4.2 Carbon yield

Carbon recoveries were calculated as a percentage of mg BPCA C formed relative

to the mg C used in each experiment (Table 3.1). The BPCA carbon yields of the

compounds in this study ranged from 0.8 to 26 % and 6.0 to 17 % for samples

processed with and without the cation column. The range of the carbon yields

decreases with cation processing and the overall recoveries were lower. These

losses, due to additional sample handling, may be accounted for in the future by

incorporating an additional recovery standard that could be added before chemical

oxidation. For fullerenes not cation processed, the carbon yield was equal; C70

exhibited a carbon yield of 26.0 ± 3.2 % and C60 had a carbon yield of 20.3 ± 2.1

%. Processed through the cation, there was a small loss of C60 (17 % recovery).

A sample of C70 was not processed through the cation column. By definition, this

method cannot recover 100 % of the carbon from fullerenes because carbon is lost

due to the breakup of adjacent rings. The maximum number of B6CA molecules

that could form from one C70 is three (i.e.: 36 carbons), corresponding to only

51.4 % carbon yield. If the carbon yield is adjusted to account for only the carbon

available to form BPCAs, then the C yield for C70 would be 47.0 ± 6.4 % within this

study. Similarly, if only the carbon available to form BPCAs was used to calculate

the carbon yield for C60 then the maximum possible carbon yield would be 33.5

± 3.5 %. Adjusting the carbon yield to reflect only the available BPCAs formed is

not always practical or possible, because the correct structures of the compounds

studied are not always known.

46

Although carbon lampblack and hexane soot exhibited significantly different BPCA

distributions (Figure 3.3), the C yields of these materials (20.0 ± 4.1 % and 25.3

± 4.0 % respectively) are statistically equivalent and equal to the C60 yield. The C

yield for perylene was also similar to these values (22.6 ± 1.3 %, Chapter 2), sug-

gesting an average C yield of 20.7 ± 4.2 % for the non-SWCNT materials in this

study. The conversion of BPCAs to BC was previously been made using activated

charcoal as a model BC material {Glaser et al., 1998; Brodowski et al., 2005}.

This average C yield is in agreement with the BPCA conversion factor reported by

Brodowski et al. {2005} for activated charcoal but about half of the original conver-

sion factor reported by Glaser et al. {1998}.

After oxidation of SWCNTs, there was always black particulate material left in the

flask that was most likely undissolved SWCNTs. When processed without the

cation column, the nonfunctionalized SWCNT had the lowest carbon yield at 0.5

± 0.1 % and the functionalized SWCNT had a carbon yield of 6.8 ± 1.4 %. If the

oxidation duration was increased to 16 hours, the SWCNT carbon yield increased

to 7.3 ± 0.7 % with no change for the SWCNT-F. When processed with the cation

column both SWCNTs had the same carbon yield (7.2 ± 1.6 and 7.2 ± 1.5 %) after

8 hours and longer oxidations did not show a significant change in carbon yield

(6.0 ± 1.9 % SWCNT and 8.8 ± 0.3 % SWCNT-F) from the shorter oxidations.

Carboxylic acid functionalized SWCNTs are produced as a bi-product of metal cat-

alyst removal using HNO3 alone or in combination with H2SO4 {Liu et al., 1998}.

Since the carbon yield for the 16-hour functionalized SWCNT was not significantly

different from that of the non-functionalized SWCNT, it is plausible that the oxida-

tion procedure that functionalized the SWCNT initially made it more susceptible to

BPCA formation. Future experiments should include longer oxidation times (i.e.:

32 hours) to assess the effect on the SWCNT C yield. Acid treatment has also

been shown to shorten the length of the CNTs {Chen et al., 2001; Liu et al., 1998}

47

and to form carbonaceous impurities {Hu et al., 2003}.

When pure compounds are assessed using the BPCA method (e.g.: perylene),

cation column processing is not required. However, environmental samples most

often contain a significant concentration of metals and other polyvalent cations that

requires removal by cation column. Internal standard, biphenyl-2,2-dicarboxylic

acid, recoveries are lower when polyvalent cations are present, suggesting that the

presence of polyvalent cations compromises the dervitization reaction. Since the

BPCA distributions of samples processed without the cation column were distinctly

different from those processed with the cation column, polyvalent cations appar-

ently change the mechanism of the oxidation process. Hexane soot did not exhibit

a loss of the internal standard when processed without the cation column, yet the

BPCA distribution was significantly different under the two processing regimes.

Further study is required to determine under what conditions cation column pro-

cessing affects BPCA distributions. Samples containing polyvalent cations not pro-

cessed through the cation column form a smaller proportion of B6CA and a larger

proportion of smaller BPCAs. Therefore estimated structures of BC for samples

processed without the cation column would be a less condensed aromatic struc-

ture than for those samples processed through the cation column, underestimating

the aromaticity of the original structure.

3.4.3 Mixtures in sediments

A key challenge remains for applying this methodology to environmental matrices,

such as coastal sediments. In environmental samples it will be important to not only

quantify the amount of BC present, but also to determine the relative contributions

of different BC sources. Two methods were used to evaluate matrix effects that

48

may occur during oxidation. First, using standard addition of SWCNT we quan-

tified BC to known mixtures of marine sediment, soot and SWCNT (Figure 3.5).

Marine sediment (NIST 1941b), without any additional BC, was found to contain

4.4 ± 0.4 g/kg BC, which is greater than previous BPCA estimates and not signifi-

cantly different than chemo-thermal oxidation (Hammes et al. {2007}). While these

measurements were converted from BPCAs to BC using the 20.7 % conversion

factor determined in this study, other loss processes must be present. Since we

do not know the types of BC materials being quantified we must use this average

value. In some cases this may lead to an over estimate of BC abundance. The

slope of the data from the standard addition experiments indicates recovery of 95

± 20 % of the added BC when the samples were oxidized for 8 hours, when using

the 20.7 % conversion factor. Prolonged oxidation, of 16 hours, also fell on the

observed trend, indicating that the duration of oxidation did not affect the observed

matrix effect. SWCNTs oxidized in isolation were found to have a low carbon yield

(7.2 %) and in mixture samples a conversion factor of 20.7 % was applied, yet

the BC yield is lower than expected. The lower than expected BC yield in mixture

samples demonstrates that a small matrix effect is present. In other words, the

presence of sedimentary material, mainly clay, causes the nitric acid oxidation of

BC to proceed less efficiently than when no sedimentary material is present.

A second method used to evaluate matrix effects was to examine the BPCA dis-

tributions of the mixtures, comparing the theoretical and measured BPCA distribu-

tions. When oxidized together, soot and SWCNTs were not as fully substituted as

predicted by a theoretical mixture of these compounds (Figure 3.6a). This shift to-

wards the production of less substituted BPCAs was more dramatic when soot and

SWCNTs were oxidized with marine sediments present (Figure 3.6b). Although the

matrix effect generates BPCA distributions similar to those not processed through

the cation column (Figure 3.4), these observed results are not due to residual

49

0

50

100

150

200

250

0 50 100 150 200 250

BC q

uant

ified

(µg)

BC added (µg)

8 hour16 hourLinear (8hour)

y = 0.95±0.20x - 3.4±28.4r2=0.92

Figure 3.5: Standard addition of soot and SWCNTs to marine sediment (NIST1941b) after 8 hours (filled squares) and 16 hours (grey circle). The trendline,generated using only 8 hour oxidation samples, indicates the recovery of SWCNT.The y-intercept corresponds to the BC content in the marine sediment and thenegative value indicates loss of BC. Error bars represent propagated errors.

50

cations. If cations were present, the internal standard recovery would have dimin-

ished, whereas this was not observed in the mixture samples. Thus, this shift could

be due to interactions with the cation column.

The presence of this matrix affect limits the applicability of identifying the types

of BC present in mixtures using the BPCA method alone. Compound specific

isotopic analysis of BPCAs, for the purpose of isolating the source of BC, is not

likely to be feasible when mixtures of BC are present. Stable carbon isotopes are

not suitable as SWCNTs show a wide range of δ13C values (-53.2 to -23.5 h,

Plata et al. {2008}), which may be from the carbon source material or fractionation

during fabrication and post-production treatments. Since carbon source materials

for SWCNTs, fullerenes and most soots are mostly fossil in origin, radiocarbon

(∆14C) analysis of BPCAs would not garner information about the source of BC.

BC oxidization techniques will not be able to parse the source of BC when mixtures

of soot and SWCNTs are present due to their structural similarity. Therefore, for

BC source appointment, additional analytical techniques must be employed.

3.5 Conclusions

This paper investigated the suitability of using the BPCA method to isolate two

SWCNTs, two fullerenes and two types of soot from natural samples. The materials

studied exhibit distinct BPCA distributions, favoring the production of larger BPCAs.

Mixtures of BC do not exhibit BPCA distributions predicted by oxidation of single

compounds. Although the BPCA method is suitable for isolating and quantifying

BC mixtures in environmental samples, matrix effects complicate the feasibility of

identifying the relative contributions of different types of BC using this method.

51

0

10

20

30

40

50

60

70

80

B4CA B5CA B6CA

% o

f tot

al B

PCA

dist

ribut

ion

theoretical 1.3soot:SWCNT

theoretical 2.2soot:SWCNTmeasured 1.3soot:SWCNT

measured 2.2soot:SWCNT

0

10

20

30

40

50

60

70

80

B4CA B5CA B6CA

% o

f tot

al B

PCA

dist

ribut

ion

theoretical sed, 1.3soot:SWCNT

theoretical sed, 3.3soot:SWCNTmeasured sed, 1.3soot:SWCNT

measured sed, 3.3soot:SWCNT

Figure 3.6: Theoretical and measured BPCA distributions in mixtures of (a) sootand SWCNT and (b) marine sediment, soot and SWCNT. Error bars on theoreticalBPCA distributions are 5 % while measured BPCA distribution errors are propa-gated errors.

52

Acknowledgments

The authors thank Sheila Griffin, John Greaves, Richard Chamberlin and Dachun

Zhang for their technical expertise and advice. We acknowledge support of Na-

tional Science Foundation EAR-04473232 and EAR-0502519 (to E.R.M.D.).

Bibliography

Akhter, M., A. Chughtai, and D. Smith, The structure of hexane soot: Spectroscopic

studies, Applied Spectroscopy, 39(1), 143–153, 1985.

Andrievsky, G., V. Klochkov, A. Bordyuh, and G. Dovbeshko, Comparative analysis

of two aqueous-colloidal solutions of C60 fullerene with help of FTIR reflectance

and UV-Vis spectroscopy, Chemical Physical Letters, 364(1-2), 8–17, 2002.

Bauer, B., M. Becker, V. Bajpai, J. Fagan, and E. Hobbie, Measurement of single-

wall nanotube dispersion by size exclusion chromatography, Journal of Physical

Chemistry C, 111, 17,914–17,918, 2007.

Bianco, A., K. Kostarelos, and M. Prato, Applications of carbon nanotubes in drug

delivery, Current Opinion in Chemical Biology, 9(6), 674–679, 2005.

Bosi, S., T. D. Ros, G. Spalluto, J. Balzarini, and M. Prato, Synthesis and anti-

HIV properties of new water-soluble bis-functionalized[60]fullerene derivatives,

Bioorganic Medical Chemical Letters, 13(24), 4437–4440, 2003.



chemistry, 36(9), 1299–1310, 2005.

53

Chen, J., et al., Dissolution of full-length single-walled carbon nanotubes, Journal

of Physical Chemistry B, 105(13), 2525–2528, 2001.

Dai, H., Carbon nanotubes: Synthesis, integration, and properties, Accounts of

Chemical Research, 35, 1035–1044, 2002.

Dittmar, T., The molecular level determination of black carbon in marine dissolved

organic matter, Organic Geochemistry, 39(4), 396–407, 2008.



29(4), 811–819, 1998.




18, 2007.

Hu, H., B. Zhao, M. Itkis, and R. Haddon, Nitric acid purification of single-walled

carbon nanotubes, Journal of Physical Chemistry B, 107(50), 13,838–13,842,

2003.

Isaacson, C., C. Usenko, R. Tanguay, and J. Field, Quantification of fullerenes by

LC/ESI-MS and its application to in vivo toxicity assays, Analytical Chemistry,

79(23), 9091–9097, 2007.

Kroto, H., J. Heath, S. O’brien, R. Curl, and R. Smalley, C60: Buckminsterfullerene,

Nature, 318(6042), 162–163, 1985.

Liu, J., et al., Fullerene pipes, Science, 280(5367), 1253, 1998.

Mckenzie, D., C. Davis, D. Cockayne, D. Muller, and A. Vassallo, The structure of

the C70 molecule, Nature, 355(6361), 622–624, 1992.

54

Niyogi, S., M. Hamon, H. Hu, B. Zhao, P. Bhowmik, R. Sen, M. Itkis, and R. Haddon,

Chemistry of single-walled carbon nanotubes, Accounts of Chemical Research,

35(12), 1105–1113, 2002.

Oberdorster, E., Manufactured nanomaterials (fullerenes, C60) induce oxidative

stress in the brain of juvenile largemouth bass, Environmental Health Perspec-

tives, 112(10), 1058–1062, 2004.

Plata, D., P. Gschwend, and C. Reddy, Industrially synthesized single-walled car-

bon nanotubes: compositional data for users, environmental risk assessments,

and source apportionment, Nanotechnology, 19(18), 185,706, 2008.

Prato, M., Fullerene materials, Topics in Current Chemistry, 199, 173 – 187, 1999.

Rasheed, A., J. Howe, M. Dadmun, and P. Britt, The efficiency of the oxidation

of carbon nanofibers with various oxidizing agents, Carbon, 45(5), 1072–1080,

2007.

Sanchez-Garcia, L., J. de Andres, J. Martin-Rubi, F. Gonzalez-Vila, J. de la Rosa,

M. Schmidt, and K. Hammes, Comparative analysis of black carbon in marine

sediments from a mediterranean river influenced coastal area (SW Iberia)., in

EGU Meeting, Vienna, Austria, 2007.

Sobek, A., and T. Bucheli, Testing the resistance of single- and multi-walled carbon

nanotubes to chemothermal oxidation used to isolate soots from environmental

samples, Environmental Pollution, 157(4), 1065 – 1071, the Behaviour and Ef-

fects of Nanoparticles in the Environment, 2009.

Tong, Z., M. Bischoff, L. Nies, B. Applegate, and R. Turco, Impact of fullerene

(C60) on a soil microbial community, Environmental Science & Technology, 41(8),

2985–2991, 2007.

55

Ziolkowski, L., and E. Druffel, Radiocarbon values of black carbon using the ben-

zene polycarboxylic acid method, in AGU Fall Meeting, San Francisco, CA, 2008.

56

4Quantification of extraneous carbon during

compound specific radiocarbon analysis of black

carbon

4.1 Abstract

Radiocarbon (14C) is a radioactive isotope that is useful for determining the age

and cycling of carbon-based materials in the Earth system. Compound specific ra-

diocarbon analysis (CSRA) provides powerful insight into the cycling of individual

components that make up the carbon cycle. Extraneous, or non-specific back-

ground carbon (Cex), is added during sample processing and subsequent isolation

of CSRA samples. Here, we evaluate the quantity and radiocarbon signature of Cex

added from two sources: during preparative capillary gas chromatography (PCGC,

CPCGC) and sample processing of CSRA for black carbon samples (Cchemistry). To

normalize our Cex estimates to samples with different isolation durations on the

GC, we report the amount of Cex as µg C per minute of GC collection over 50

injections. Using both a direct and indirect method of assessment, we determine

that the added CPCGC was 0.1 ± 0.05 and 0.5 ± 0.3 µg C min-1 50 injections-1 with

a fraction modern ranging from 0.15 to 0.2. We found that the direct and indirect

57

assessment of Cchemistry+PCGC agreed, both in magnitude and radiocarbon value

(1.1 ± 0.5 µg C, fraction modern = 0.2). Half of the Cex was introduced before

PCGC isolation, likely from solvents used in the extraction method. The magni-

tude of propagated uncertainties of CSRA samples were found to be a function of

sample size and collection duration. Small samples collected for a brief amount

of time have less propagated 14C uncertainty than larger samples collected for a

longer period of time. CSRA users are cautioned to consider the magnitude of

uncertainty they require for their system of interest and to frequently evaluate the

magnitude of Cex added during sampling processing and isolation.

4.2 Introduction

Radiocarbon dating of bulk organic and inorganic carbon reservoirs has allowed

the average residence time of carbon in each of the respective pools to be cal-

culated. However, these reservoirs comprise of complex heterogenous mixtures

whose components have different residence times that may not be well repre-

sented by bulk radiocarbon measurements. Initially, the heterogenous mixtures

were studied via compound class specific radiocarbon analysis (CCSRA). The sub-

sequent introduction of compound specific radiocarbon analysis (CSRA) allowed

measurement of 14C signature in a single compound {Eglinton et al., 1996}. CSRA

usually involves a multiple-step purification procedure that culminates in the collec-

tion of a single compound (or group of compounds) of high purity. The applications

of CCSRA and CSRA range from source apportionment of atmospheric particles

{Reddy et al., 2002; Sheesley et al., 2009}, biomarkers with paleoclimatic implica-

tions {Prahl and Wakeham, 1987; Sachs and Lehman, 1999; Mollenhauer et al.,

2005}, microbial incorporation of fossil material {Petsch et al., 2001; Slater et al.,

58

2005} and compound class studies in marine sediments {Hwang and Druffel, 2005}

and marine dissolved organic carbon {Aluwihare et al., 2002; Loh et al., 2004}.

New developments in accelerator mass spectrometry (AMS) have decreased the

sample size requirements for CSRA. Ultra-small samples{Santos et al., 2007} and

online 14C measurements {von Reden et al., 2008} enable CSRA as small as 2 µg

C. Preparation of CSRA samples requires two-distinct and rigorous sets of labo-

ratory protocols (sample isolation and 14C analysis), each step inadvertently and

unavoidably introducing Cex. Thus a CSRA sample of 2 µg C may be ≥50% Cex. To

date few studies have quantified Cex{Shah and Pearson, 2007}. Accounting for Cex

has largely been avoided by processing samples large enough so as to overwhelm

the Cex. However, all environmental CSRA techniques allow for the preparation of

large sample sizes, because the compound of interest might be in low abundance.

Constraining the uncertainty of 14C measurements is done by evaluating the mass

and variability of Cex added during sample preparation. Here we assess the mass

and radiocarbon signatures of Cex specific to the chemical oxidation of organic

matter for quantifying black carbon using PCGC. We employed the benzene poly-

carboxylic acid method that chemically oxidizes black carbon to benzene rings sub-

stitued with three to six carboxylic acid groups.

4.3 Methods

Natural and synthetic vanillin (4-hydroxy-3-methoxybenzaldehyde, Table 4.1) were

used as standards to assess the extraneous carbon added during PCGC isolation.

Black carbon (BC) reference materials were used as process standards to quantify

Cex added throughout the entire isolation procedure (Table 4.1){Hammes et al.,

59

2007, 2008}.

4.3.1 Chemical Oxidation

To minimize carbon contamination, all glassware and quartz filters that came in

contact with the samples and standards were baked at 550oC for 2 hours prior to

use. Samples were processed using a modification of the benzene polycarboxylic

acid (BPCA) method ({Ziolkowski and Druffel, 2009}, Chapter 2). Process materi-

als, wood char and hexane soot (Table 4.1), were oxidized in 2 mL of concentrated

nitric acid (grade ACS) in quartz tubes inside a high pressure digestion apparatus

at 180oC for 8 hours. Post digestion, the samples were filtered through quartz fiber

filters (27 mm diameter, 0.8 µm pore diameter) and 15 mL Milli-Q water was used

to rinse any remaining BPCAs from the filter. The filtrate was collected and freeze

dried overnight.

Dried samples were redissolved in 5 mL methanol and the internal standard, biphenyl-

2,2’-dicarboxylic acid (1 mg mL-1 in methanol) was added. Samples were deriva-

tized by titration with 2.0 M trimethylsilyl diazomethane in ethyl ether (Sigma Aldrich).

Derivatization was considered complete when the solution retained the yellow color

of the trimethylsil-diazomethane. Methanol was dried with a purified stream of UHP

nitrogen. A fixed volume of dichloromethane was added.

The derivatized oxidation products were separated and quantified on a Hewlett

Packard 6890N outfitted with a Gerstel cooled injection system, a DB-XLB capil-

lary column (30 m x 0.53 mm I.D., 1.5 µm film thickness), and a flame ionization

detector (FID). After injection, the column temperature was maintained at 100oC

for 1 minute, then raised at 25oC min-1 to 250oC followed by a 5oC min-1 ramp to

280oC for 10 minutes and then raised to 320oC for 5 minutes of bake out (Figure

60

Tabl

e4.

1:M

ater

ials

proc

esse

dan

das

soci

ated

solv

ents

used

for

CS

RA

ofbl

ack

carb

on.

The

bulk

14C

was

mea

sure

din

dupl

icat

e.m

ater

ial

use

sour

cebu

lk14

C(F

M)

mat

eria

lspr

oces

sed

mod

ern

vani

llin

GC

proc

ess

stan

dard

Sig

ma

Ald

rich

1.05

2±

0.00

2sy

nthe

ticva

nilli

nG

Cpr

oces

sst

anda

rdS

igm

aA

ldric

h0.

002±

0.00

1gr

ass

char

met

hod

proc

ess

stan

dard

Uni

.of

Zuric

h1.

056±

0.00

2he

xane

soot

met

hod

proc

ess

stan

dard

Uni

.of

Den

ver

0.00

5±

0.00

1so

lven

tsan

dm

ater

ials

met

hano

lso

lven

t0.

0001

dich

loro

met

hane

solv

ent

0.00

01

biph

yenl

-2,2

’-dic

arbo

xylic

acid

inte

rnal

stan

dard

Sig

ma

Ald

rich

0.00

0±

0.00

1TM

S-d

iazo

met

hane

deriv

atiz

atio

nag

ent

Sig

ma

Ald

rich

0.00

01

DB

-XLB

GC

colu

mn

Agi

lent

0.00

0±

0.00

11 A

ssum

edra

dioc

arbo

nva

lues

.

61

4.1). The FID temperature was 300oC. The splitless injection volume was 1 µL

for all samples in this study. Approximately 1 % of the flow eluting from the capil-

lary column was diverted to the FID and 99 % was sent to the preparative fraction

collector (PFC), which consists of a zero-dead-volume valve in a heated interface

(320oC) and seven 200 µL glass U-tube traps (six sample traps and a waste trap).

The PFC transfer was kept constant at 320oC for all samples processed. U-tubes

were supported in isopropyl alcohol cooled units (-10oC). The auto-injector, CIS

and trapping device are programmable and computer controlled, and FID data was

acquired using Chemstation software.

BPCAs were identified by comparison of their retention times with those obtained

for a commercially available mixture and were verified using GC/MS. All methy-

lated BPCAs were quantified relative to the biphyenl-2,2’-dicarboxylic acid internal

standard.

4.3.2 Radiocarbon analysis of isolated samples

To avoid cross contamination from previously injected samples (e.g.: memory), the

compounds collected from the first 10 injections were disposed of and the U-tube

was replaced with a clean, baked tube. Unless otherwise noted, trapped samples

were collected from 50 injections of each sample. To avoid possible isotope frac-

tionation of isolates {Zencak et al., 2007}, care was taken to trap the entire peak.

After PCGC isolation, the U-tubes containing trapped samples were rinsed with

700 µL of CH2Cl2 into pre-baked GC autosampler vials. Samples were evaluated

by GC-FID for purity and yield. Samples were then transfered to 6 mm quartz tubes

using an additional 700 µL of CH2Cl2 and the solvent was removed in a stream of

UHP nitrogen. CuO and silver wire were added and the sample tube was evac-

62

uated to 10-6 Torr and flame-sealed under vacuum. Tubes were then heated to

850oC for 2 hours. The resulting CO2 was purified, quantified and reduced to

graphite according to standard procedures. Measurements of 14C were made at

the Keck Carbon Cycle Accelerator Mass Spectrometry Laboratory at University of

California Irvine. In all cases, radiocarbon analysis are reported as fraction mod-

ern, which is the deviation of a sample from 95 % of the activity in 1950 AD, of

National Bureau of Standards (NBS) oxalic acid 1 normalized to δ13C = -25 h with

respect to Pee Dee Belemnite {Olsson, 1970; Stuvier and Polach, 1977}. All frac-

tion modern values reported within this paper have been corrected for combustion

and graphitization and mass dependent isotope fractionation by reporting all data

to a common δ13C value of -25 h {Stuvier and Polach, 1977}.

4.4 Results

4.4.1 Carbon mass balance and corrections

The mass of carbon graphitized in CSRA samples (Creported) isolated via PCGC that

have been corrected for graphitization and combustion, originate from at least four

sources:

Creported = Csample + Cderivative + Cchemistry + CPCGC (4.1)

the mass of carbon in the compound of interest isolated from the sample (Csample),

the mass of added derivative carbon (Cderivative) and the mass of extraneous carbon

added during chemical extraction (Cchemistry) and subsequent isolation via PGCG

(CPCGC).

The compounds of interest in this study, BPCAs, contain functional groups that

63

require derivatization to adjust their polarity and volatility to enable separation by

PCGC. The derivatization adds a methyl group (-CH3) to each carboxylic acid group

and this additional carbon affects the 14C of the sample. Since the isotopic com-

position of the derivative carbon is assumed to be 14C free (FMderivative = 0), the

reported isotopic signature is known and the amount of added derivative carbon is

known, the radiocarbon composition of the parent BPCA compound can be calcu-

lated:

FMsample+chemistry+PCGC =FMreported − (FMderivative ∗ fderivative)

fsample+chemistry+PCGC(4.2)

where FMsample+chemistry+PCGC is the FM of the underivatized BPCA, FMreported is

the FM of the BPCA methyl ester corrected for graphitization and combustion,

FMderivative is the FM of the derivative carbon, fsample+chemistry+PCGC is the fraction of

underivatized BPCA in Creported and fderivative is the fraction of derivative carbon in

Creported. Tests with process standards of known 14C signatures, confirmed that the

derivative carbon is radiocarbon dead.

When samples are corrected for Cderivative (Equation 4.2), Equation 4.1 is simplified

to:

Csample+chemistry+PCGC = Csample + Cchemistry+PCGC = Csample + Cex (4.3)

To provide accurate isotopic values of Csample+chemistry+PCGC, the mass and isotopic

composition (FM) of Cex must be determined. Here we evaluated two sources

of Cex: added during chemical extraction (Cchemistry) and during PCGC isolation

(CPCGC). Reported values of CSRA samples (Creported) need to be corrected for

Cex. For the purposes of estimating the Cex via process materials, samples had

to be corrected for derivative carbon before estimating Cex, which assumes that all

Cex has been derivatized.

64

4.4.2 Extraneous carbon added during PCGC isolation (CPCGC)

Two methods were used to evaluate the mass and FM of Cex originating from

PCGC isolation. First, the direct approach was used to collected a sample over

a seven minute retention time window (Figure 4.1) from dry 400 injections (direct

CPCGC). No solvent was injected during the dry injections, that is, there was no

needle in the autosampler, and all other GC parameters (i.e. carrier gas, oven

temperature) were maintained. This sample yielded 7.6 ± 0.4 µg C and had a

FMPCGC of 0.125 ± 0.034. Because the sample collection window varies with sam-

ple type (Figure 4.1), we normalized the amount of Cex (µg C) to collection duration

(in minutes) and number of injections using the equation:

normalized Cex =Cex

(collection duration)(number of injections)(4.4)

Normalizing the Cex to time assumes the majority of Cex is due to column bleed

(sample history and/or breakdown of the GC-column stationary phase) and that

the bleed does not change over time. To standardize this non-specific background

correction, all subsequent collections maintained the same injection volume and

number of injections; only the collection time and injected materials varied for the

samples reported here. While 50 injections was the standard number used in this

study, periodically fewer injections were made due to technical difficulties. Normal-

izing the Cex to both the time and number of injections enables one to apply this

corrections to samples that were collected for different durations and or different

number of total injections. We normalized all samples that evaluated Cex, even

samples that included the Cchemistry. The Cex was normalized to µg C per minute

of collection for 50 injections. Thus, evaluated directly the CPCGC added in the dry

injections was 0.1 ± 0.05 µg C min-1 50 injections-1.

65

Figure 4.1: The magnitude of column bleed (indicated by the magnitude of thebaseline signal) and oven temperature as a function of retention time. The retentiontime windows for the isolation of vanillin and BPCAs are marked.

The second method of evaluating the mass and FM of CPCGC used various sizes of

isolated process standards of known FM values. It was assumed that the sample

was diluted with a constant mass and isotopic signature of Cex and the presence

of Cex would cause a deviation in the consensus 14C value. Vanillin, the process

standard used to estimate CPCGC added during PCGC processing, does not contain

carboxylic acid groups and is not derivatized; thus it thus does not require correc-

tion for derivative carbon (Cderivative). The FM values of samples can be expressed

by the following equation:

FMsample =FMreportedCreported − FMPCGCCPCGC

Csample(4.5)

where FMsample is the radiocarbon value of the sample corrected for CPCGC, FMreported

is the measured radiocarbon value of the sample uncorrected for CPCGC, and FMPCGC

66

the radiocarbon value of the extraneous carbon added during PCGC isolation. Typ-

ically, Cex is assessed as a combination of both dead and modern material. Thus,

we would expect the FMex to be between 0 and 1. Therefore, small samples of

modern isotopic composition isolated by PCGC will become more depleted and

samples of 14C-depleted composition isolated by PCGC will become more enriched

in radiocarbon (e.g.:Figure 4.2).

The mass and FM of Cex added during PCGC processing was assessed indirectly

using a two-component approach. The PCGC isolation size-series of modern

vanillin (FMsample = 1.052, Table 4.1) samples revealed that the amount of CPCGC

added (FMPCGC=0.0) was 0.4 ± 0.2 µg C min-1 50 injections-1. The PCGC isola-

tion of a series of different sized samples of 14C-free vanillin (FMsample = 0.002) re-

vealed an additional 0.2 ± 0.1 µg C min-1 50injection-1 was added with an assumed

FMPCGC=1.0. These two blanks were added to obtain the total indirect CPCGC of 0.6

± 0.3 µg C with an average FMPCGC=0.2 (Table 4.2).

The difference of 0.5 µg C of Cex added to isolated vanillin samples calculated

using standard materials (0.6 ± 0.3 µg C) as compared to the dry injections (0.1

µg C) may be due to several factors. First, no solvent was injected into the GC

column during dry injections. It is possible that when solvent is present in the GC

column more Cex is mobilized than during the absence of solvent. The FMex value

for vanillin (FMex = 0.2 ± 0.1) and that for the dry injections (FMex = 0.125 ± 0.034)

was similar suggesting the same source of Cex. Another possible explanations are

that CPCGC and its isotopic signature may vary with time, sample memory and / or

contamination of the injector port. Therefore, we estimate that for each minute of

collection on the PCGC, 0.6 µg C with a FM = 0.2 is being added to samples due

to contamination from the PCGC.

67

Tabl

e4.

2:Ty

pean

dtre

atm

ent

ofsa

mpl

esth

atw

ere

eval

uate

dfo

rC

exan

dFM

exdu

ring

chem

ical

oxid

atio

nan

dP

CG

Cis

olat

ion.

The

unce

rtai

nty

ofth

em

ass

ofex

trane

ous

carb

onw

ases

timat

edto

be50

%of

the

sam

ple

mas

s.Th

eun

cert

aint

yof

FMex

was

estim

ated

tobe

50%

ofth

eFM

valu

e.

eval

uatio

nm

ater

ial

BP

CA

1C

H2C

l 2P

CG

Cex

trane

ous

carb

on,C

ex

µg

CFM

Dire

ctdr

yin

ject

ion

xx

X0.

10.

125±

0.03

4In

dire

ctm

oder

nva

nilli

nx

XX

0.4±

0.2

0.0

Indi

rect

dead

vani

llin

xX

X0.

2±

0.1

1.0

tota

lind

irect

PC

GC

0.6±

0.3

0.2±

0.1

Dire

ctpr

oces

sbl

ank

XX

X1.

1±

0.2

0.20

0±

0.05

4In

dire

ctgr

ass

char

XX

X0.

80±

0.40

0.0

Indi

rect

hexa

neso

otX

XX

0.15±

0.08

1.0

tota

lind

irect

chem

istr

y1.

0±

0.5

0.15±

0.08

1 BP

CA

incl

udes

the

chem

ical

oxid

atio

nof

BC

into

BP

CA

san

dth

eirs

ubse

quen

tder

ivat

izat

ion,

see

text

ford

etai

ls.

68

4.4.3 Extraneous carbon added during chemical oxidation and

PCGC isolation (Cchemistry+PCGC)

CSRA samples are typically subjected to extensive chemical extraction procedures

prior to isolation by PCGC and consequently it is likely that extraneous carbon is

added during these procedures. Similar to the evaluation of Cex added during

PCGC isolation, we evaluated the mass and FM of Cex added during the chemical

methods and PCGC isolation using both an indirect and direct approach. To eval-

uate Cex directly, the chemical oxidation and PCGC isolation steps were carried

out but no sample material was added. Direct analysis of the Cex added during

chemical oxidation, derivatization and PCGC isolation with no sample added was

1.1 ± 0.2 µg C min-1 50 injections-1 and FM = 0.200 ± 0.054 (Table 4.2).

The Cex was evaluated indirectly by quantifying the deviation in FMs+ex from the un-

processed material for radiocarbon dead (hexane soot) and modern (grass char)

of different sizes. Samples of modern grass char were chemically oxidized, deriva-

tized and isolated by PCGC. The samples isolated by PCGC (e.g.: CCSRA) ranged

from 2 to 16 µg C. We found that 0.80 ± 0.40 µg C min-1 50 injections-1 of an

assumed FMex=0.0 was added in chemical oxidation and PCGC isolation. Fossil

hexane soot revealed 0.15 ± 0.08 of an assumed FMex = 1.0 was added in sample

processing. The total indirect method Cex was then calculated to be 1.0 ± 0.5 µg

C min-1 50 injections-1 and FMex = 0.15.

When evaluated directly and indirectly, the mass and isotopic composition of the

Cex added during sample processing and isolation was the same. If the Cex for

indirect assessment was much larger than the direct method, the source of the Cex

may be a matrix effect of the oxidation process. The agreement of the two methods

suggests that the Cex is not associated with any matrix effects in the processing of

69

a sample.

The magnitude of the Cex added during chemical oxidation (Cchemistry = 0.5 µg C) is

approximately equal to that added during PCGC isolation (CPCGC = 0.6 µg C). This

suggests that only half of the non-specific background is originating from within the

PCGC, supposedly column bleed. The remainder is likely from the reagents and

solvents used in the oxidation and derivatization processes. Because reagents

and solvents can become contaminated over time and with use, it is essential to

frequently evaluate the Cex (e.g. every 2 to 5 samples).

4.4.4 Correcting for extraneous carbon and associated uncer-

tainties

Radiocarbon measurements are typically reported with an uncertainty of the AMS

measurement alone. As we have shown above, the corrected radiocarbon value of

a CSRA sample is dependent on of the mass and FM of the Cex. If the sample is

large enough (≥ 50 µg C), the Cex will be insignificant. However, the FM of small

CSRA samples will require a correction for the presence of Cex. The uncertainties

of all these terms must be considered when reporting the uncertainty of the CSRA

FM value. To determine the propagated total mathematical uncertainty of FMsample

(e.g. Equation 4.5), we applied the following equation:

σ2FMsample

=

(∂FMsample

∂FMreported

)2

σ2FMreported

+

(∂FMsample

∂FMex

)2

σ2FMex

+

(∂FMsample

∂mreported

)2

σ2mreported

+

(∂FMsample

∂mex

)2

σ2mex

(4.6)

where σFMreported is the uncertainty of FMreported measured on the AMS (machine un-

certainty), σFMex is the uncertainty for FMex, σmreported is the uncertainty for Creported

70

(uncertainty in graphitization) and σmex is the uncertainty for Cex. The total uncer-

tainty of the direct process blank (Cchemistry+PCGC in Table 4.2) was used for FMex

and Cex.

For grass char, a modern BC standard {Hammes et al., 2007}, the measured

FMreported values for 7 small samples without Cex correction (average FMreported =

0.824 ± 0.128, Table 4.3) are significantly lower than the FM value of the unpro-

cessed material (FM = 1.058 ± 0.002, Figure 4.2). After correction for Cchemistry+PCGC,

the FMsample (average 1.098 ± 0.221) agrees with that of the unprocessed material.

For hexane soot, a dead BC standard, the measured FM values without correction

for Cex (average FMreported = 0.061 ± 0.55, Table 4.3) are significantly enriched in

14C in comparison to the unprocessed material (FM = 0.005 ± 0.001). After correc-

tion for Cchemistry+PCGC, the FMreported (average 0.036 ± 0.056) is more comparable

to the FM of the unprocessed material.

These results demonstrate that the uncertainties associated with the preparation

and isolation of samples by CSRA are significantly larger than the machine error.

Propagated total uncertainty of processed 14C modern materials is much higher

than processed 14C depleted materials, due to the nature of radioactive decay of

14C and that in our system the FMex was more 14C depleted than modern. Not all

systems will have the same FMex and each user needs to evaluate the Cex and

FMex values specific for their system.

Thus, when considering CSRA applications, one must consider the magnitude of

uncertainty requirement to provide useful information about the system being stud-

ied. For example, our interest in CSRA of BPCAs is to examine the BC in marine

dissolved organic carbon (DOC). Bulk DOC, which is comprised of a wide range of

organic molecules of varying 14C ages, typically ranges from FM = 0.8 to 0.5 {Loh

et al., 2004}. The BC in marine DOC has been postulated to be more depleted in

71

Tabl

e4.

3:R

adio

carb

onva

lues

(frac

tion

mod

ern)

and

asso

ciat

edun

cert

aint

yof

blac

kca

rbon

refe

renc

em

ater

ials

befo

rean

daf

ter

corr

ectio

nfo

rC

ex.

Dur

atio

n(m

inut

es)

isth

etim

eth

eco

llect

ion

win

dow

isle

ftop

en.

The

Cex

(Cch

emis

try+

PC

GC)

isas

sum

edto

be1.

1±

0.2µ

gC

per

min

ute

ofco

llect

ion

for

a50

inje

ctio

nru

nw

itha

FM=

0.2±

0.05

4(s

eeTa

ble

4.2)

.Th

eun

cert

aint

yas

soci

ated

with

the

FMre

port

edis

the

AM

Sm

achi

neun

cert

aint

yan

dth

eun

cert

aint

yas

soci

ated

with

FMsa

mpl

eis

the

prop

agat

edun

cert

aint

y.Th

eco

llect

ion

win

dow

dura

tion

was

varie

dto

colle

ctin

divi

dual

BP

CA

or∑ B

PC

A.

type

UC

IDdu

ratio

n(m

in)

Cre

port

ed(µ

gC

)FM

repo

rted

1C

ex(µ

gC

)FM

sam

ple2

1178

21.

216

.30.

90±

0.02

1.32±

0.36

0.96±

0.03

1180

13.

712

.00.

91±

0.02

4.07±

1.11

1.27±

0.15

1177

91.

28.

70.

86±

0.04

1.32±

0.36

0.98±

0.06

gras

sch

ar11

777

2.1

5.2

0.78±

0.07

2.31±

0.53

1.24±

0.27

1178

02.

14.

60.

82±

0.09

2.31±

0.63

1.45±

0.39

1177

80.

42.

40.

81±

0.24

0.44±

0.12

0.95±

0.30

1178

10.

41.

90.

69±

0.43

0.44±

0.12

0.84±

0.56

isol

ate

aver

age±

std

dev

0.82

4±

0.12

81.

098±

0.22

1bu

lkva

lue

1.05

2±

0.00

211

711

0.6

9.4

0.00

4±

0.01

00.

660±

0.18

00.

000±

0.01

211

723

0.6

6.6

0.04

9±

0.04

40.

660±

0.18

00.

032±

0.04

9he

xane

soot

1171

30.

94.

10.

054±

0.07

70.

990±

0.27

00.

007±

0.10

311

710

0.9

3.5

0.13

9±

0.09

00.

990±

0.27

00.

115±

0.12

511

712

0.9

2.4

0.34

9±

0.10

00.

990±

0.27

00.

454±

0.17

8is

olat

eav

erag

e±

std

dev

0.06

1±

0.05

50.

036±

0.05

6bu

lkva

lue

0.00

5±

0.00

11 a

fter d

iazo

met

hane

corr

ectio

n2

dete

rmin

edus

ing

Equ

atio

n4.

5

72

(a) Modern grass char

(b) 14C dead hexane soot

Figure 4.2: (a) Grass char and (b) hexane soot before (open symbols) and after(filled symbols) Cex correction. The radiocarbon value of the unprocessed materialis indicated by the bolded line: grass char FM=1.056 ± 0.002 and hexane sootFM=0.005 ± 0.001.

73

radiocarbon. Provided that BC extracted from marine DOC has a propagated total

uncertainty less than FM = 0.10, the results should provide valuable information

about this pool of recalcitrant carbon. However, if we were interested in studying

the removal process of BC from soils over a few centuries, we would require much

larger samples than those presented here in order to ensure that the contribution

of Cex to the FMreported is insignificant, which would in turn minimize the propagated

total uncertainty. Regardless of the application, it is equally important that CSRA

users assess their ability to duplicate CSRA measurements, as in some cases the

duplication of CSRA samples may be larger than the propagated total uncertainty.

The mass and isotopic composition of Cex should ideally be evaluated with each

batch of samples, as we found the mass of Cex to vary by over 50 % over the

course of six months {Ziolkowski, 2009}.

4.5 Conclusions

Extraneous carbon added during PCGC isolation of CSRA samples was found to

be a function of collection duration on the GC. Half of the Cex was added during

PCGC isolation and half was added during the chemical oxidation and derivatiza-

tion. The estimates of extraneous or non-specific background carbon presented

here are specific to this chemical isolation technique. Another facility using the

same chemical extraction technique would need to determine the extraneous car-

bon introduced to samples that they process. Different GC columns, solvents and

users may produce more or less Cex carbon, with unique FMex values.

74

Acknolwedgements

The authors would like to thank John Southon, Guacaria dos Santos, Sheila Griffin,

Dachun Zhang and Xiaomei Xu for their technical assistance and comments. This

work was funded by the National Science Foundation Chemical Oceanography

program.

Bibliography

Aluwihare, L., D. Repeta, and R. Chen, Chemical composition and cycling of dis-

solved organic matter in the Mid-Atlantic Bight, Deep Sea Research Part II: Top-

ical Studies in Oceanography, 49(20), 4421–4437, doi:10.1016/S0967-0645(02)

00124-8, 2002.

Eglinton, T., L. Aluwihare, J. Bauer, E. Druffel, and A. McNichol, Gas chromato-

graphic isolation of individual compounds from complex matrices for radiocarbon

dating, Analytical Chemistry, 68(5), 904–912, 1996.

Hammes, K., R. Smernik, J. Skjemstad, and M. Schmidt, Characterization and

evaluation of reference materials for black carbon analysis using elemental com-

position, colour, BET surface area and 13C NMR spectroscopy, Applied Geo-

chemistry, 23(8), 2113–2122, doi:10.1016/j.apgeochem.2008.04.023, 2008.




18, 2007.

Hwang, J., and E. Druffel, Blank correction for Delta C-14 measurements in organic

75

compound classes of oceanic particulate matter, Radiocarbon, 47(1), 75–87,

2005.

Loh, A., J. Bauer, and E. Druffel, Variable ageing and storage of dissolved organic

components in the open ocean, Nature, 430(7002), 877–881, 2004.

Mollenhauer, G., M. Kienast, F. Lamy, H. Meggers, R. Schneider, J. Hayes,

and T. Eglinton, An evaluation of C-14 age relationships between co-occurring

foraminifera, alkenones, and total organic carbon in continental margin sedi-

ments, Paleoceanography, 20(1), PA1016, 2005.

Olsson, I., The use of oxalic acid as a standard., in Radiocarbon Variations and

Absolute Chronology, Nobel Symposium, 12th Proc., edited by E. IU Olsson,

p. 17, John Wiley and Sons, New York, 1970.

Petsch, S., T. Eglinton, and K. Edwards, C-14-dead living biomass: Evidence for

microbial assimilation of ancient organic carbon during share weathering, Sci-

ence, 292(5519), 1127–1131, 2001.

Prahl, F., and S. Wakeham, Calibration of unsaturation patterns in long-chain ke-

tone compositions for paleotemperature assessment, Nature, 330(6146), 367–

369, 1987.

Reddy, C., A. Pearson, L. Xu, A. McNichol, B. Benner, S. Wise, G. Klouda, L. Cur-

rie, and T. Eglinton, Radiocarbon as a tool to apportion the sources of polycyclic

aromatic hydrocarbons and black carbon in environmental samples, Environ-

mental Science & Technology, 36(8), 1774–1782, 2002.

Sachs, J., and S. Lehman, Subtropical north atlantic temperatures 60,000 to

30,000 years ago, Science, 286(5440), 756–759, 1999.

76

Santos, G., J. Southon, S. Griffin, S. Beaupre, and E. Druffel, Ultra small-mass

AMS C-14 sample preparation and analyses at KCCAMS/UCI Facility, Nuclear

Instrumental Methods Part B, 259(1), 293–302, doi:10.1016/j.nimb.2007.01.172,

2007.

Shah, S., and A. Pearson, Ultra-microscale (5-25 mu g C) analysis of individual

lipids by C-14 AMS: Assessment and correction for sample processing blanks,

Radiocarbon, 49(1), 69–82, 2007.

Sheesley, R., M. Krusa, P. Krecl, C. Johansson, and O. Gustafsson, Source appor-

tionment of elevated wintertime PAHs by compound-specific radiocarbon analy-

sis, Atmospheric Chemistry and Physics, 9(10), 3347–3356, 2009.

Slater, G., H. White, T. Eglinton, and C. Reddy, Determination of microbial carbon

sources in petroleum contaminated sediments using molecular C-14 analysis,

Environmental Science & Technology, 39(8), 2552–2558, 2005.

Stuvier, M., and H. Polach, Discussion: Reporting of 14C data., Radiocarbon, 19,

355 – 363, 1977.

von Reden, K., M. Roberts, W. Jenkins, B. Rosenheim, A. McNichol, and

R. Schneider, Software development for continuous-gas-flow ams, Nuclear In-

strumental Methods Part B, 266(10), 2233–2237, doi:10.1016/j.nimb.2008.03.

001, 2008.

Zencak, Z., C. Reddy, E. Teuten, L. Xu, A. McNichol, and O. Gustafsson, Eval-

uation of gas chromatographic isotope fractionation and process contamination

by carbon in compound-specific radiocarbon analysis, Analytical Chemistry, 79,

2042–2049, doi:10.1021/ac061821a, 2007.

Ziolkowski, L., Radiocarbon of black carbon in marine dissolved organic carbon,

77

PhD in Earth System Science, University of California Irvine (UCI), Croul Hall,

Irvine, CA, 92697, 2009.

Ziolkowski, L., and E. Druffel, The feasibility of isolation and detection of fullerenes

and carbon nanotubes using the benzene polycarboxylic acid method, Marine

Pollution Bulletin, doi:10.1016/j.marpolbul.2009.04.018, 2009.

78

5Black carbon in marine dissolved organic carbon

5.1 Abstract

Black carbon (BC) enters the ocean through aerosol and river deposition before its

eventual incorporation into the sediment. It has been postulated that BC resides

in the marine dissolved organic carbon (DOC) pool before sedimentary deposition.

Here we report the concentration and radiocarbon content of BC in high molecular

weight DOC (UDOM). BC exported from rivers has modern levels of 14C, while

open ocean samples contained BC with an average 14C age of 20,000 ± 3,000

years. BC represents 0.5 to 3.5 % of UDOM. If marine DOC contains 4 to 22

% BC, as suggested from studies of BC in the sediments {Masiello and Druffel,

1998}, the low molecular weight DOC is rich in BC and possibly the repository of

the missing BC.

79

5.2 Introduction

Black carbon (BC) particles encompass a large range of chemical and physi-

cal properties that are produced during biomass burning and fossil fuel combus-

tion. In the atmosphere, BC can lead to increased atmospheric temperatures

and decreased precipitation {Ramanathan and Carmichael, 2008}. BC is stored

in soils where it may be slowly degraded over time before being transported to

rivers {Czimczik and Masiello, 2005; Hockaday et al., 2007}. BC enters the ocean

through aerosol and river deposition {Flores-Cervantes et al., 2009; Masiello and

Druffel, 2001; Dickens et al., 2004}. BC isolated from open ocean sediments is up

to 14,000 14C years older than non-BC sedimentary organic carbon, suggesting

that BC resides in an intermediate pool, such as marine dissolved organic carbon

(DOC), before sedimentary deposition {Masiello and Druffel, 1998}.

Marine DOC, operationally defined as the material that passes through a 0.2 - 1.0

µm filter, is the largest exchangeable pool of organic carbon in the ocean. Over

80 % of the marine DOC cannot be characterized at the molecular level {Benner ,

2002}. High molecular weight ultrafiltered DOM (UDOM, ≥1000 Da) has been

found to contain a small portion of aged lipid like material {Loh et al., 2004}, which

may include some BC. Studies directly assessing the BC and BC-like material

using traditional BC (CTO-375) and FTIR-ICR-MS techniques, in riverine, coastal

and open ocean DOC estimate that BC could be up to 5 % of DOC {Mannino and

Harvey, 2004; Kim et al., 2004; Dittmar and Paeng, 2009}.

80

5.3 Approach

BC is a heterogenous material that is aromatic in nature. Soot BC, formed during

high temperature combustion, is characterized by a condensed aromatic structure

(e.g. Figure 2.9, BPCA distributions of BC references materials). In contrast char

BC, formed at lower combustion temperatures, is more oxidized and has a less

condensed aromatic structure. These characteristics can be determined using

NMR {Czimczik et al., 2003}, elemental analysis {Hammes et al., 2008} and the

benzene polycarboxylic acid (BPCA) method {Glaser et al., 1998}. We used the

BPCA method to quantify and characterize BC in marine DOM {Ziolkowski, 2009}.

The abundance of radiocarbon (14C) in BC will be indicative of its source. Fossil fuel

produced BC contains no radiocarbon (is isotopically “dead”) and thus has a 14C

age of greater than 50,000 years (the detection limit). In contrast BC produced from

biomass burning has a 14C value of the contemporary biosphere C. Here we use

radiocarbon measurements of BC markers using the BPCA method to determine

the cycling and residence time of BC in the marine DOC pool.

BC was extracted from a series of UDOM samples (Figure 5.1, Table 5.1) and was

analyzed using the BPCA method {Glaser et al., 1998; Brodowski et al., 2005} as

described in Chapter 4. These samples represent a wide range of ocean locations,

sources and ages of DOC and one river sample. UDOM samples were digested in

concentrated nitric acid to oxidize BC to BPCAs (marker molecules) and were sub-

sequently isolated and purified via pcGC before radiocarbon analysis (see Chapter

2). Individual and nitrated BPCAs were pooled for radiocarbon analysis. No B6CA

were collected for radiocarbon analysis.

81

Tabl

e5.

1:S

ampl

ein

form

atio

nfo

rUD

OM

sam

ples

inth

isst

udy.

Isot

opic

valu

esw

ere

dete

rmin

edon

UD

OM

.

Sam

ple

UC

IDs

lat

long

dept

hsa

linity

appr

ox.

tem

p.TO

Cδ1

3 CU

DO

MC

:N∆

14C

UD

OM

NW

mo C

µM

hh

Suw

anee

Riv

er1

1178

3,11

784

30.7

82.5

13,

270

47.7

152.

7±

1.7

Am

azon

influ

ence

d211

921,

1192

48.

455

.62

30.9

7227

78-2

4.6

16.2

-99±

3M

id-A

tlant

icB

ight

310

879,

1088

037

.973

.72

33.6

1419

74-2

2.2

16.3

2.9±

2.6

SE

Atla

ntic

411

923,

1192

4-1

2.0

-9.4

236

.071

2691

-20.

014

.9-9

0±

3N

.Cen

tralP

acifi

c611

925,

1192

919

.6-1

56.0

2068

-21.

613

.7-3

11±

3N

EP

acifi

c511

919,

1192

032

.512

3.5

1000

34.1

806

38-4

45±

2.8

1 NO

Mob

tain

edfro

mth

eIn

tern

atio

nalH

umic

Soc

iety

cata

log

#1R

101N

,May

1999

,not

acid

ified

2M

P08

-56,

May

2003

3 DF2

0606

,Jun

e20

06,n

otac

idifi

ed4 S

EA

tlant

ic,s

tn95

,Apr

il16

,200

35 C

alco

fi06

10,s

tn80

.100

,Nov

2006

,[TO

C]b

ased

onB

eaup

reet

al.{

2007},

usin

gsa

land

tem

pfro

m50

0m6 N

EH

LA20

mpi

pe,

[TO

C]

base

don

D.

Han

sell’s

web

site

,ht

tp://

ww

w.rs

mas

.mia

mi.e

du/g

roup

s/bi

ogeo

chem

/inde

x.ht

ml.

Acc

esse

don

July

3,20

09.

2,3,

4,5 C

olle

cted

byL.

Alu

wih

are.

6 Col

lect

edby

M.M

cCar

thy.

82

Figure 5.1: Map illustrating sample locations. Filled dots indicate surface (1 or 2m) or near surface (20 m) depth. Open dot indicates deep sample (1000 m).

5.4 Results

The high proportion of B5CA and B6CA products formed from BC in DOM con-

centration from the Suwannee River illustrates that terrigenous BC is condensed

in its aromatic structure. In contrast, BPCAs formed from open ocean UDOM-

BC has a uniformly smaller and less condensed aromatic structure because of

the higher proportion of B3CA and B4CAs formed and absence of B6CAs (Fig-

ure 5.2). Suwannee River DOC exhibited the largest BC structure (average BPCA

size = 4.71 acids) and the distribution of BPCAs formed resembled the distribution

of BPCAs formed from charred BC (see Figure 2.9). In the Amazon influenced

sample, that contains a mixture of marine and riverine DOC, the average BC struc-

ture (average BPCA size = 4.04 acids) is less condensed than the Suwanee River

structure. The structure of the BC in the open ocean samples (average BPCA size

= 3.5 - 3.92 acids) is less condensed and unvarying in composition, regardless of

depth or ocean location.

83

Figure 5.2: BPCA distribution and 14C of BC for the samples. For each sample,the distribution of BPCAs is calculated by relating the total carbon of an individualBPCA (including nitrated peaks) to the total BPCA carbon.

The ∆14C values of bulk UDOM ranged from +152 h in the Suwannee River to

-445 h in the the deep NE Pacific Ocean, respectively (Table 5.1). BPCAs formed

from UDOM-BC in Suwannee River were significantly more depleted in 14C. The

Suwannee River DOM is mostly bomb carbon (due to nuclear weapons testing),

while BPCAs formed from BC are pre-bomb, though mean averages are likely

less than a century old (Table 5.2). In contrast, the BPCAs formed from oceanic

UDOM-BC were 14C depleted. The ∆14C values of collected BPCAs correlate with

the BPCA distributions (Figure 5.2). That is, the younger precursor of BPCAs, BC,

is a more condensed aromatic than the older precursor of BPCAs. If the BC in the

Amazon influenced sample is conservative with salinity, a mass balance calculation

reveals that the 14C age of BC exported from the river is also modern (∆14C ∼ 0

h).

84

Tabl

e5.

2:M

easu

rem

ents

ofbl

ack

carb

onis

olat

edfro

mU

DO

M.δ

13C

are

AM

Sm

easu

red

valu

es,r

epor

ted

assu

min

gde

riva-

tive

carb

onha

saδ1

3 Cof

-12

h.

The

estim

atio

nis

anap

prox

imat

em

inim

umva

lue

ofB

Cin

DO

Cbe

caus

eul

tra-fi

ltere

dm

ater

ialw

asan

alyz

ed,

whi

chco

nstit

utes

only

apo

rtio

nof

the

DO

Cpo

ol.

Ave

rage

BP

CA

isth

eav

erag

enu

mbe

rof

acid

func

tiona

lgro

ups

prod

uced

inth

eox

idat

ion

ofB

C.

UC

IDav

erag

eB

PC

Aδ1

3 C∆

14C

BC

14C

age

BC

inD

OC

∆∆

14C

1

Suw

anne

eR

iver

211

803,

1180

44.

71-3

0±

4-4

9±

3341

0±

280

29.4µ

M-2

02A

maz

onin

fluen

ced3

1195

64.

04-3

3±

4-7

27±

4410

,400±

1300

300

nM-6

29M

id-A

tlant

icB

ight

210

878,

1172

13.

75-2

3±

4-8

58±

1915

,680±

1100

560

nM-8

61S

EA

tlant

ic3

1197

13.

68-3

3±

4-8

97±

5518

,300±

4300

330

nM-8

07N

.Cen

tralP

acifi

c311

958

3.92

-30±

4-8

80±

3817

,000±

2500

90nM

-569

NE

Pac

ific3

1195

53.

50-2

8±

4-9

18±

3120

,100±

3000

330

nM-4

73

1 ∆∆

14C

is∆

14C

UD

OM

-∆14

CB

C2

Bla

nkco

rrec

ted

with

the

aver

age

ofpr

oces

sbl

anks

UC

ID11

701,

1170

2,11

703,

1170

4,11

705

and

1170

6.3

Bla

nkco

rrec

ted

with

proc

ess

blan

kU

CID

1195

4.

85

Contemporary 14C ages of riverine and riverine-influenced BC suggests a non-

fossil fuel source of BC exported from these rivers, but is difficult to reconcile with

a biomass combustion source. The BPCAs distribution of riverine and riverine-

influenced BC resembles that of charred BC, which agrees with the findings that

char is mobilized in watersheds {Hockaday et al., 2007}. When this material reaches

the ocean, it appears that the UDOM quickly loses its aromatic character. This loss

of aromaticity could be due to photochemical degradation. In estuaries the abun-

dance of aromatic compounds exposed to ultraviolet radiation has been observed

to decrease {Tremblay et al., 2007; Gonsior et al., 2009} and / or microbial utiliza-

tion {Carlson, 2002}. Since only aromatic materials form BPCAs, it is unlikely that

this decreased aromatic character of UDOM is due to dilution with non-aromatic

material.

The BPCA distributions of the marine samples suggest that BC cycling in the open

ocean is distinct from the BC that is exported from the Suwannee and Amazon

Rivers. If the BC exported from rivers remained unaltered in the UDOM pool,

one would expect the chemical composition of BC from the Atlantic Ocean to be

more similar to terrestrial BC. However, this is not the case. The ∆14C value and

BPCA distribution of BC isolated from the Mid-Atlantic Bight and SE Atlantic do

not appear to resemble the aromatically condensed modern 14C exported from the

rivers. Additionally, the ∆14C values of BPCAs isolated from UDOM-BC from the

Pacific were not significantly different from the Atlantic samples.

Atmospheric inputs of BC reveal a wide variety ∆14C values (-220 to -600 h), in-

dicating a variety of sources {Eglinton et al., 2002; Gustafsson et al., 2009}. Ama-

zon Basin atmospheric BC had a mean particle size of 0.175 µm {Echalar et al.,

1998}, smaller than the size cutoff for DOC, indicating that more than half of the

86

BC aerosols could be incorporated into the UDOM. Unless soot, originating from

fossil fuel combustion, is chemically solubilized via atmospheric oxidation, dras-

tically changing its chemical composition, before deposition to the surface ocean

{Decesari et al., 2002} it is unlikely that the isolated BC originated from recent fossil

fuel emissions. The distribution of BC oxidation products in marine samples do not

resemble the condensed aromatic character of soot particles (see Figure 2.9).

BPCAs extracted from UDOM-BC had more depleted 14C values than bulk UDOM,

suggesting that the BC is more recalcitrant or has other sources compared to other

components of the bulk UDOM. The ∆14C off-set between UDOM-BC and bulk

UDOM is not consistent ( ∆∆14C ranged from -202 to -861 h, Table 5.2), suggest-

ing that UDOM-BC cycles on longer times scales than any chemical components

of DOC identified to date. If BC was produced in situ (e.g. mid-water column pro-

duction {Yamashita and Tanoue, 2008} or from bacterial production {Ogawa et al.,

2001}), the ∆14C of newly produced “BC” would reflect that of bulk DOC, that is

being consumed. However, little variation of the 14C age of UDOM-BC is observed

and it is consistently older than bulk DOC (Figure 5.3). While the two river systems

studied, Suwannee and Amazon, contain modern 14C levels and are condensed

in aromatic character, it is possible that other river systems could export older

graphitic material (e.g.: Dickens et al. {2004}). However, graphitic black carbon is

of a condensed aromatic structure and upon oxidation would likely produce more

substituted BPCAs (B6CA, B5CA), such as that observed for soot.

These data represent BC in UDOM, which typically represents 25 to 35 % of the

bulk DOC {Benner , 2002}. Assuming a conversion factor of 4 (see Chapter 2.4.5),

BC in my samples ranged from 0.5 to 3.5 % (on a carbon basis). This corresponds

to minimum marine BC concentrations of 85 – 500 nM. It is likely that the low

87

molecular weight (LMW) fraction of marine DOM contains a higher proportion of

BC than that in the UDOM, as suggested by an isotopic mass balance study {Loh

et al., 2004} and the decreasing size of BPCAs with ∆14C value (Figure 5.2). The

concentrations presented here are in agreement with BC concentrations (600 to

800 nM) in solid-phase extracted DOC determined in the S. Atlantic and S. Ocean

{Dittmar and Paeng, 2009}. The LWM DOM would contain aromatic molecules

from polycyclic aromatic hydrocarbons to larger molecules such as fullerenes (e.g.:

C60). The UDOM-BC studied here are likely macromolecules or more likely, smaller

molecules that are complexed thus making them larger than the 1000 Da size cut-

off. If the size of BC in the DOC pool, as inferred by the observed BPCA distribu-

tion, is a function of 14C age, it is likely that the BC in the LMW fraction of DOC is

older than the values presented here.

From the age differences between BC and non-BC sedimentary organic carbon

(SOC), Masiello and Druffel {1998} suggested that BC resides in the DOC pool

from 2400 to 13,900 14C years before deposition. Assuming the average annual

flux of pre-industrial BC to the world oceans is 10 Tg per year {Suman et al., 1997}

and a marine DOC pool of 685 x 1015 g C {Hansell and Carlson, 1998}, Masiello

and Druffel {1998} calculated that BC could be up to 4 to 22 % of the total deep

ocean DOC. Therefore 27 – 151 x 1015 g could be BC, corresponding to a con-

centration of BC in marine DOC between 1.7 and 9.4 µM. In the deep NE Pacific,

we found that the UDOM was 3.5 % BC or 0.29 µM (see Chapter A.2 for calcula-

tion details). Thus, if DOC is 4 to 22 % BC, then the lower molecular weight DOC

contains a substantial proportion of BC (1.4 – 9.1 µM).

Our measurements of the radiocarbon of BPCAs extracted from UDOM-BC from

the deep NE Pacific Ocean had a 14C age of BC of 20,000 ± 3,000 years. This

residence time calculation assumed that the source(s) of BC to the ocean are 14C

88

Figure 5.3: ∆14C of black carbon and marine DOC as a function of depth. Thedepth profiles of DOC are from the Sargasso Sea (SS) and North Central Pacific(NCP) from Druffel et al. {1992}.

89

modern. However, with industrialization fossil fuel and coal combustion (soot) has

increased the amount of 14C depleted BC {Bond et al., 2004}. If BC from fossil fuel

or coal combustion was incorporated into the UDOM-BC, it would decrease the

14C signature. Stable carbon isotope (δ13C) measurements of BPCAs extracted

from UDOM-BC may help pinpoint the source of the BC. However δ13C measure-

ments of BC would only distinguish a marine or terrestrial (forest fire and fossil

fuel combustion) source. Since the BPCA distribution of UDOM-BC does not re-

semble soot, it is unlikely that 14C depleted soot is present in the deep NE Pacific

Ocean sample. This does not exclude the presence of other yet to be discovered

source(s) of 14C depleted aromatic compounds, resembling BC, in the UDOM.

BC has a much lower ∆14C value than DOC (-900 h versus -500 h for the deep

Pacific Ocean). If BC is up to 22 % of the deep DOC (i.e. 9.4 µM), then the

remaining 78 % (28.6 µM) would have an average 14C of -388 h (4000 14C years),

which is significantly younger than the deep bulk DOC value (6000 14C years).

Our hypothesis that a large fraction of old BC is in the LMW fraction of DOC is

at odds with this calculation and needs to be tested. It is not unlikely that old

smaller BC is in the LMW fraction, as the LMW fraction of DOC has been found

to be significantly older than the HMW fraction {Santschi et al., 1995}. Molecular

analysis of BC mobilized within a fire-impact watershed had a peak mass to charge

ratio of 400 {Hockaday et al., 2006}, which is equivalent to the size of a five ring

PAH and would be in the LMW fraction of DOC. Should it be proven that there

is a large pool of aged BC in the LMW fraction of DOC, then this would help to

explain the enigma that has existed in our understanding of the BC cycle. That is,

the sources of BC far outweigh the known sinks. However, marine DOC may be a

temporary reservoir for BC, and processes that are responsible for its breakdown,

(e.g. photochemical oxidation, bacterial remineralization or physical removal to the

sediments) warrant investigation.

90

Bibliography

Beaupre, S., E. Druffel, and S. Griffin, A low-blank photochemical extraction sys-

tem for concentration and isotopic analyses of marine dissolved organic carbon.,

Limnology and Oceanography: Methods, 5, 174 – 184, 2007.

Benner, R., Biogeochemistry of Marine Dissolved Organic Matter, chap. 3, pp. 59–

90, Academic Press, 2002.

Bond, T., D. Streets, K. Yarber, S. Nelson, J. Woo, and Z. Kilmont, A technology-

based global inventory of black and organic carbon emissions from combustion,

Journal of Geophysical Research, 109, doi:10.1029/2003JD003697, 2004.



chemistry, 36(9), 1299–1310, doi:10.1016/j.orggeochem.2005.03.011, 2005.

Carlson, C., Biogeochemistry of Marine Dissolved Organic Matter, chap. 4, pp. 91

– 139, Academic Press, 2002.

Czimczik, C., and C. Masiello, Controls on black carbon storage in soils, Global

Biogeochem. Cycles, 21(3), GB3005, doi:10.1029/2006GB002798, 2005.

Czimczik, C. I., C. A. Preston, M. Schmidt, and E.-D. Schulze, How surface fire in

siberian scots pine forests affects soil organic carbon in the forest floor: Stocks,

molecular structure, and conversion to black carbon (charcoal), Global Biogeo-

chemical Cycles, 17(1), 1020, doi:doi.10.1029/2002GB001956, 2003.

Decesari, S., M. Facchini, E. Matta, M. Mircea, S. Fuzzi, A. Chughtai, and D. Smith,

Water soluble organic compounds formed by oxidation of soot, Atmospheric En-

vironment, 36(11), 1827–1832, doi:10.1016/S1352-2310(02)00141-3, 2002.

91

Dickens, A., Y. Gelinas, C. Masiello, S. Wakeham, and J. Hedges, Reburial of

fossil organic carbon in marine sediments, Nature, 427(6972), 336–339, doi:

10.1038/nature02299, 2004.

Dittmar, T., and J. Paeng, A heat-induced molecular signature in marine dissolved

organic matter, Nature Geoscience, 2(3), 175–179, doi:10.1038/NGEO440,

2009.

Druffel, E. R. M., P. M. Williams, J. E. Bauer, and J. R. Ertel, Cycling of dissolved

and particulate organic matter in the open ocean, Journal of Geophysical Re-

search, 97(C10), 15,639–15,659, doi:10.1029/92JC01511, 1992.

Echalar, F., P. Artaxo, J. V. Martins, M. Yamasoe, F. Gerab, W. Maenhaut, and

B. Holben, Long-term monitoring of atmospheric aerosols in the amazon basin:

Source identification and apportionment, Journal of Geophysical Research,

103(D24), 31,849 – 31,864, 1998.

Eglinton, T. I., G. Eglinton, L. Dupont, E. R. Sholkovitz, D. Montlucon, and C. M.

Reddy, Composition, age, and provenance of organic matter in nw african dust

over the atlantic ocean, Geochemistry Geophysics Geosystems, 3(8), 1050, doi:

10.1029/2001GC000269, 2002.

Flores-Cervantes, D., D. Plata, J. MacFarlane, C. Reddy, and P. Gschwend, Black

carbon in marine particulate organic carbon: Inputs and cycling of highly recalci-

trant organic carbon in the gulf of maine, Marine Chemistry, 113(3-4), 172–181,

doi:10.1016/j.marchem.2009.01.012, 2009.



29(4), 811–819, 1998.

92

Gonsior, M., B. Peake, W. Cooper, D. Podgorski, J. D’andrilli, and W. Cooper, Pho-

tochemically induced changes in dissolved organic matter identified by ultrahigh

resolution fourier transform ion cyclotron resonance mass spectrometry, Envi-

ronmental Science and Technology, 43, 698 – 703, doi:10.1021/es8022804, 2009.

Gustafsson, O., et al., Brown clouds over south asia: Biomass or fossil fuel com-

bustion?, Science, 323(5913), 495–498, doi:10.1126/science.1164857, 2009.

Hammes, K., R. Smernik, J. Skjemstad, and M. Schmidt, Characterisation and

evaluation of reference materials for black carbon analysis using elemental com-

position, colour, BET surface area and 13C NMR spectroscopy, Applied Geo-

chemistry, 23(8), 2113–2122, doi:10.1016/j.apgeochem.2008.04.023, 2008.

Hansell, D., and C. Carlson, Deep-ocean gradients in the concentration of dis-

solved organic carbon, Nature, 395(6699), 263–266, 1998.

Hockaday, W., A. Grannas, S. Kim, and P. Hatcher, The transformation and mobility

of charcoal in a fire-impacted watershed, Geochimica et Cosmochimica Acta,

71(14), 3432–3445, doi:10.1016/j.gca.2007.02.023, 2007.

Hockaday, W. C., A. M. Grannas, S. Kim, and P. G. Hatcher, Direct molecu-

lar evidence for the degradation and mobility of black carbon in soils from

ultrahigh-resolution mass spectral analysis of dissolved organic matter from a

fire-impacted forest soil, Organic Geochemistry, 37(4), 501 – 510, doi:DOI:

10.1016/j.orggeochem.2005.11.003, 2006.

Kim, S., L. Kaplan, R. Benner, and P. Hatcher, Hydrogen-deficient molecules in nat-

ural riverine water samples?evidence for the existence of black carbon in dom,

Marine Chemistry, 92(1-4), 225–234, doi:10.1016/j.marchem.2004.06.042, 2004.



93

Mannino, A., and H. Harvey, Black carbon in estuarine and coastal ocean dissolved

organic matter, Limnology and Oceanography, 49(3), 735–740, 2004.

Masiello, C., and E. Druffel, Black carbon in deep-sea sediments, Science,

280(5371), 1911–1913, doi:10.1126/science.280.5371.1911, 1998.

Masiello, C., and E. Druffel, Carbon isotope geochemistry of the santa clara river,

Global Biogeochem. Cycles, 15(2), 407–416, 2001.

Ogawa, H., Y. Amagai, I. Koike, K. Kaiser, and R. Benner, Production of refractory

dissolved organic matter by bacteria, Science, 292(5518), 917–920, 2001.

Ramanathan, V., and G. Carmichael, Global and regional climate changes due to

black carbon, Nature Geoscience, 1(4), 221, doi:doi:10.1038/ngeo156, 2008.

Santschi, P. H., L. Guo, M. Baskaran, S. Trumbore, J. Southon, T. S. Bianchi,

B. Honeyman, and L. Cifuentes, Isotopic evidence for the contemporary origin

of high-molecular weight organic matter in oceanic environments, Geochimica et

Cosmochimica Acta, 59(3), 625 – 631, doi:10.1016/0016-7037(94)00378-Y, 1995.

Suman, D., T. Kuhlbusch, and B. Lim, Marine sediments: A reservoir for black

carbon and their use as spatial and temporal records of combustion., in Sedi-

ment records of biomass burning and global change, vol. 51, edited by J. Clark,

H. Cachier, J. Goldammer, and B. Stocks, pp. 271 – 293, NATO ASI series, 1997.

Tremblay, L., T. Dittmar, A. Marshall, W. Cooper, and W. Cooper, Molecular char-

acterization of dissolved organic matter in a north brazilian mangrove porewa-

ter and mangrove-fringed estuaries by ultrahigh resolution fourier transform-ion

cyclotron resonance mass spectrometry and excitation/emission spectroscopy,

Marine Chemistry, 105(1-2), 15–29, doi:10.1016/j.marchem.2006.12.015, 2007.

94

Yamashita, Y., and E. Tanoue, Production of bio-refractory fluorescent dissolved

organic matter in the ocean interior, Nature Geoscience, 1(9), 579, doi:doi:10.

1038/ngeo279, 2008.

Ziolkowski, L., Radiocarbon of black carbon in marine dissolved organic carbon,

PhD in Earth System Science, University of California Irvine (UCI), Croul Hall,

Irvine, CA, 92697, 2009.

95

6Conclusions and thoughts on future research

This thesis helped improve the utility of molecular-level compound specific radio-

carbon analysis (CSRA) as an approach to understanding the source and cycling

of recalcitrant carbon in marine dissolved organic carbon. The following topics

were addressed.

6.1 BPCA method and its applicability

Measurements of black carbon (BC) in environmental matrices, such as sediment

or soil, require the isolation of BC from non-BC material. This separation is typically

done using chemical or thermal methods. The benzenepolycarboxlic acid (BPCA)

method, a chemical oxidation technique, oxidizes the BC to BPCAs that are mark-

ers of BC. Information about the temperature of combustion can be inferred from

the distribution of BPCAs. This method seemed ideal for studying BC in marine

DOC because it provides both quantitative and qualitative characterization of BC.

At the outset of this work, little was known about the chemistry of the BPCA

method. A suite of polycyclic aromatic hydrocarbons (PAHs), compounds that are

96

structurally similar to BC, were studied to identify any systematic oxidation pat-

terns, which may have proven useful for modeling the structure of oxidized BC.

PAHs larger than four rings typically favored the formation of the largest possible

BPCAs than that expected from stoichiometry. Smaller PAHs favored the formation

of smaller BPCAs than that expected from stoichiometry. The pattern of oxidation

products was not systematic enough to model the parent structure of the oxidized

BC, but was predictable enough to provide generalized information about the struc-

ture of the BC oxidized.

Preliminary experiments revealed that some of the BPCAs produced from the ox-

idation of BC were nitrated (contained one or more -NO2 functional groups). Ni-

trated BPCAs were not quantified in previous versions of the BPCA method. The

oxidation of PAHs produced exclusively nitrated BPCAs. I concluded that the quan-

tification of nitrated BPCAs was essential for the accurate estimation of BC; thus

nitrated BPCAs were quantified for all samples presented within this thesis.

During the nitric acid oxidation process, sample handling and PCGC isolation

losses of sample material occur. The distribution and recovery of the oxidation

products of the PAH anthracene was studied in great detail, as it seemed well

suited to be an internal standard that could be added before oxidation to account

for these losses. Quantitatively producing predominantly B2CA, which is not typi-

cally included as oxidation products of BC, anthracene could be used as an internal

standard to correct for losses in sample handling. However, anthracene was only

used in those samples that were passed through the cation column (e.g.: sedi-

ments and CNTs in Chapter 3).

Integrating the carbon yield data for the range of BC or BC-like materials studied,

a conversion factor from BPCAs to BC was determined. Generated BPCAs were

converted to BC using the conversion factor of 4 ± 1 (the inverse of 25 ± 6 %).

97

BC yields from reference materials presented here illustrate that this version of

the BPCA method is effective at quantifying both soot-like and char-like BC and

more satisfactory for quantifying BC than either the chemo-thermal oxidation (e.g.:

CTO-375) or previous versions of the BPCA method.

6.2 Evaluating extraneous material added during the

preparation of CSRA samples

Measurements of natural abundance 14C of individual compounds requires two-

distinct, rigorous sets of laboratory protocols. These steps consist of (i) the careful

separation of pure compounds from a complex mixture by PCGC and (ii) the prepa-

ration and analysis of this sample by 14C-AMS. Uncertainty of 14C measurements

of individual compounds is a function of both steps in sample preparation. Com-

monly, the uncertainty of the second step is the only 14C uncertainty reported and

the uncertainty of the first step is grossly overlooked.

In an attempt to constrain the uncertainty of CSRA measurements of BC, I exam-

ined the mass, sources and variability of extraneous carbon added in the first step

of sample preparation. The extraneous carbon added in the separation of pure

compounds by PCGC originate from two discrete sources: the chemical prepara-

tion of samples (e.g.: chemical oxidation and solvents) and from within the PCGC

(i.e.: column bleed). When evaluated both indirectly and directly, the mass of ex-

traneous carbon (Cex) appeared to be a function of collection time. It also became

apparent that only half of the Cex, which could be attributed to the PGCG (e.g.:

column bleed). The other half of Cex likely originates from solvents and reagents

and thus, it could change over time as new batches of chemicals are used. Discon-

98

certingly, I observed that the mass and ∆14C value of the Cex changed over time,

which means that the blank must be evaluated frequently when processing small

PCGC samples (≤20 µC).

6.3 Black carbon in the marine DOC pool

Based on radiocarbon measurements of BC in marine sediments, Masiello and

Druffel {1998} postulated that BC was 4 to 22 % of deep dissolved organic carbon

(DOC). Using the BC extraction and CSRA methods developed within this thesis

and a suite of UDOM samples from a river and five ocean locations, I tested this

hypothesis.

The BPCA distribution of the Suwannee River resembled char and was the most

condensed aromatic structure of BC measured in these UDOM samples. BC from

open ocean samples produced smaller BPCAs, indicating that the structure of BC

becomes smaller the longer it is in the ocean. The radiocarbon signature of BC

from the Suwannee River was modern, while the BC extracted from open ocean

samples was uniformly depleted in 14C. BC in the deep Pacific had a ∆14C = -918

± 31 h, and this value represents the oldest reported compound isolated from

DOC. It is likely that the lipids extracted from UDOM by Loh et al. {2004} contained

depleted BC, similar to the material quantified here. The fraction of carbon UDOM

that is ranged from 0.5 – 5 %. Since UDOM typically only represents about 25 % of

the DOC pool, the proportion of BC in the low molecular weight (LMW) pool must

be much higher if there is 4 – 22 % BC in marine DOC.

99

6.4 Future work

The conclusions of this thesis, as well as the unanswered questions, lead to many

potential avenues of new research. Some of these potential avenues are listed

below.

What is the “black carbon” in marine DOC?

BC cycling in marine DOC is complex. The source(s) of this material were not

well quantified by the CSRA presented within this thesis. The δ13C of BC ex-

tracted from UDOM values were not well defined (-24 to -33 h, UCI Keck AMS

values). Thus, the source of radiocarbon depleted BC in marine UDOM-BC is not

well constrained. Using HPLC to isolate the BPCAs, thereby eliminating the need

to derivatize of BPCAs and thus eliminate potential isotope fractionation associ-

ated with incomplete derivatization, could be valuable for determining the δ13C of

BC in marine DOC.

Chemically, what is BC? Could this material be graphitic BC {Dickens et al., 2004}

or resuspended BC from sediments? Or is it simply old terrestrial material? Can

its composition be described using other techniques or isotopes?

Since the BPCA method detects all aromatic structures and the BC isolated from

UDOM was ubiquitously old, could the material detected originate from a non-

combustion source? Could asphaltenes, which are remnants of fossil kerogen

found in suspension with crude oil, be dissolved in DOC? Asphaltenes from oil

have been found to be larger than 1000 Daltons and are theorized to have a molec-

ular structure that would produce the BPCA distribution observed in open ocean

100

SO

O=

O

OH

S

S

S

S

N

S

S

S

S

N N

S

S

S

S

OH

H2N

Figure 6.1: Proposed chemical structure of asphaltene, adapted from Artok et al.{1999}.

UDOM-BC (Figure 6.1). When oil undergoes biodegradation, the asphaltene frac-

tion increases {Peressutti et al., 2003} indicating that asphaltenes are resistant to

microbial degradation. Previously, asphaltenes have not been considered to be in

the DOC pool as they may not be soluble in DOC. An oversight of this dissertation

was not studying any kerogen materials, which may have provided more insight

into the asphaltene hypothesis. Analysis of asphaltenes extracted from crude oil

and / or BC in DOC extracted from locations close to oil seeps (i.e.: deep water in

Santa Monica basin) may provide more insight into this hypothesis.

101

Where is the BC?

The loss of highly condensed, B6CA producing, BC exported from rivers suggests

that BC is removed from HMW DOC. But where does the riverine BC go? Are

the smaller BPCAs measured in the ocean, remnants of the condensed aromatic

material exported by rivers? There are at least three possibilities of what happens

to the BC. It could:

1. undergo microbial consumption, removing some or all of the aromatic mate-

rial.

2. undergo photochemical oxidation BC in the surface ocean, quickly removing

the condensed aromatic character of BC exported from the rivers.

3. flocculate and / or sorb onto particles ultimately leading to its burial in sedi-

ments.

The most direct way to quantify the fate of the condensed aromatic BC would be

to quantify the BPCAs in marine DOC isolated by two techniques: ultrafiltration

(UF, size of particles) and reverse osmosis/electrodialysis (RO/ED) {Gurtler et al.,

2008}. If one were to analyze the BC in these two types of isolates of marine

DOC extracted at the same location and time, one could provide a more accurate

estimation of the fate of the BC in marine DOC. Is the concentration of BC higher

in the LMW DOC? Its also very likely that “BC” in LMW DOC is actually a mixture

of PAHs with alkyl side-chains and not complex macromolecules. While chemically

solid phase extraction (SPE) may be well suited for BC extraction, the range of BC

molecules extracted by SPE will not be the sample as those extracted by UF and

RO/ED thereby limiting our ability to compare the extracted BC with the total DOC

composition.

102

Studying the loss of terrestrial BC within an estuarine system may provide insight

into the mechanism(s) that remove BC from the water column. Any such study

should include quantification of the BC in sediment, DOC and particulate organic

carbon (POC). Flores-Cervantes et al. {2009} demonstrated the important role of

POC in the removal of BC from the surface ocean. The major limitation of using

CSRA to study the loss BC from riverine DOC is extracting sufficient amounts of

material for CSRA. Within this study, at least one liter of seawater was required to

provide enough UDOM-BC for one CSRA sample. The volume of water required

to extract enough material for CSRA limits the types of process studies possible

(i.e. photochemical incubation).

Since each BC method quantifies a different type of BC, it is difficult to compare

ages of BC isolated by different BC methods. This version of the BPCA method

appears to quantify the same BC as the chromic acid method used by Masiello

and Druffel {1998}, as indicated by BC standard materials. Would BC isolated from

sediments using the BPCA method have the same radiocarbon age offset between

the BC and non-BC material? Quantifying the BC in sedimentary material would

provide insight into the variability of the time BC spends in an intermediate pool

before sedimentary deposition.

This is the first work quantifying and radiocarbon of BC in marine DOC, therefore

this work leaves an open pathway for future studies on the spatial and temporal

variability of BC in the marine DOC pool. How do these measured concentrations

vary? How variable is the isotopic composition of BC in the ocean? Is there any

part of the ocean where the BC is significantly older (i.e.: where BC is accumulat-

ing)?

103

Bibliography

Artok, L., Y. Su, Y. Hirose, M. Hosokowa, S. Murata, and M. Nomura, Structure and

reactivity of petroleum-derived asphaltene, Energy Fuels, 13, 287 – 296, 1999.

Dickens, A., Y. Gelinas, C. Masiello, S. Wakeham, and J. Hedges, Reburial of fossil

organic carbon in marine sediments, Nature, 427(6972), 336–339, 2004.

Flores-Cervantes, D., D. Plata, J. MacFarlane, C. Reddy, and P. Gschwend, Black

carbon in marine particulate organic carbon: Inputs and cycling of highly recalci-

trant organic carbon in the gulf of maine, Marine Chemistry, 113(3-4), 172–181,

doi:10.1016/j.marchem.2009.01.012, 2009.

Gurtler, B., T. Vetter, E. Perdue, E. Ingall, J. F. Koprivnjak, and P. Pfromm, Com-

bining reverse osmosis and pulsed electrical current electrodialysis for improved

recovery of dissolved organic matter from seawater, Journal of Membrane Sci-

ence, 323(2), 328 – 336, doi:10.1016/j.memsci.2008.06.025, 2008.



Masiello, C., and E. Druffel, Black carbon in deep-sea sediments, Science,

280(5371), 1911–1913, doi:10.1126/science.280.5371.1911, 1998.

Peressutti, S. R., H. M. Alvarez, and O. H. Pucci, Dynamics of hydrocarbon-

degrading bacteriocenosis of an experimental oil pollution in patagonian soil,

International Biodeterioration and Biodegradation, 52(1), 21 – 30, doi:10.1016/

S0964-8305(02)00102-6, 2003.

104

ADetermination of Carbon Yields

A.1 Polycyclic Aromatic Hydrocarbons

How the carbon recovery was calculated for polycyclic aromatic hydrocarbon (PAH)

samples.

The following steps were taken:

1. The peak area relative to internal standard was determined for each BPCA.

[relative peak area]

2. The relative calibration curves were applied to each BPCA. Nitrated BPCAs

were assumed to have the same calibration as non-nitrated BPCAs of the

same number of functional groups (i.e.: B4CA-N1 was calibrated using the

B4CA data). [mg BPCA]

3. The mg BPCA from step (2) were converted to mg C as BPCA using the

percent carbon for each BPCA (see Table A.1). [mg C BPCA for individual

BPCAs].

105

Table A.1: Carbon content of BPCAs.

B#CA N# % CarbonB3CA N0 51.4B3CA N1 41.7B4CA N0 47.2B4CA N1 40.0B5CA N0 44.3B5CA N1 38.5B6CA 42.1

4. The mg C BPCAs of individual BPCAs (including those nitrated) were summed

to get a carbon yield [sum of mg C for all BPCAs].

5. If the relative carbon yield is to be reported, divide (4) by % C in starting

material [mg C as “BC”].

Results from the formation of BPCAs from anthracene are shown in Table A.2 as

an example of how the carbon yields were calculated.

A.2 Calculating the percentage of black carbon in

UDOM

How the carbon recovery was calculated for UDOM-BC samples.

The following steps were taken:

1. The peak area relative to internal standard was determined for each BPCA.

[relative peak area].

106

Tabl

eA

.2:

Exa

mpl

eca

lcul

atio

nof

BP

CA

sin

PAH

s.

UC

IZA

nthr

acen

eB

PC

Are

lativ

epe

akar

eam

gC

BP

CA

1%

Cyi

eld

mg

CB

2CA

-N1

B2C

A-N

2B

2CA

-N1

B2C

A-N

2su

m94

.3%

C45

.1%

C36

.9%

C94

4.70

2.46

0.42

1.03

0.14

1.17

24.9

955.

963.

100.

521.

300.

181.

4824

.810

73.

091.

700.

310.

710.

100.

8126

.210

82.

351.

070.

190.

450.

060.

5121

.710

94.

812.

440.

191.

030.

061.

0922

.711

05.

012.

610.

351.

090.

121.

2124

.211

25.

282.

700.

341.

130.

111.

2423

.511

37.

903.

270.

261.

810.

092.

0025

.3av

erag

e24

.2±

1.6

1 Cal

ibra

tion

of3-

nitro

phth

alic

acid

rela

tive

toin

tern

alst

anda

rd.

mg

CB

PC

A=

(nor

mal

ized

peak

area−

0.01

495)

1.06

72∗

%C.

107

2. The relative calibration curves were applied to each BPCA. Nitrated BPCAs

were assumed to have the same calibration as non-nitrated BPCAs of the

same number of functional groups (i.e.: B4CA-N1 was calibrated using the

B4CA data). [mg of individual BPCAs].

3. The mg of individual BPCAs was converted to mg C as BPCA using the

percent carbon for each BPCA (see Table A.1). [mg C of individual BPCAs].

4. The mg C of individual BPCAs are summed [mg C BPCA].

5. The BPCAs are converted to “BC” by dividing by 25 % (conversion factor)

[mg C BC].

6. The initial carbon used was calculated by multiplying the weight UDOM by

the percentage of carbon in UDOM [mg C UDOM].

7. Then, divide mg C BC by mg C UDOM to get the carbon yield [carbon yield].

108

BBPCA protocol

This is the generalized protocol used to process samples within this thesis. A more

detailed protocol is available upon request. Each step of the protocol is followed

by an explanation of the logic behind the step.

B.1 Chemical extraction of BPCAs

B.1.1 Cleaning the bomb

Add 2 mL of concentrated HNO3 directly into the Teflon sleeves within the bomb.

Put the bomb in the oven for 4 hours at 180 oC. At the same time, bake all glassware

for 2 hours at 550 oC that will be used for processing the samples.

Cleaning the bomb with HNO3 will oxidize any carbon that may have accumulated

in the Teflon over time. This step should be conducted every few weeks. Baking

the glassware will also remove any residual carbon.

109

B.1.2 Pre-treatment of samples (if required)

This step is required for soil, mineral containing aerosols and sediments containing

cations (i.e.: Fe3+). There are two additional steps in the processing of samples

that contain cations: TFA pre-treatment and cation column processing. Both steps

are required to remove the cations. It is recommended to add the anthracene

internal standard at the beginning of this step to account for any losses of sample

through the processing.

Weigh out samples into pre-baked quartz bomb tubes. Add 10 mL of 4 M trifu-

loroacetic acid to each tube. Cook in the bomb for 4 hours at 104 oC. Filter these

samples and retain the material on the filter. Dry the samples at 45 oC.

If the cations are not removed from the sample, they will interfere with the oxidation

and derivatization of the BPCAs.

B.1.3 Filter sample

Filter the samples after bombing. Rinse with at least 15 mL Milli-Q water. Retain

the filtrate.

Some solid materials are not oxidized and remain. These particles need to be

removed from the BPCA solution.

B.1.4 Cation column (if required)

For soil, mineral aerosol and sediment samples drop the filtrate onto the cation

column when it is in the H+ state.

110

This is the second step specific to process samples that contain cations. The resin

retains the cations.

B.1.5 Dry samples

Freeze samples in the freezer. Then dry for 24 hours in the freezedryer.

Water interferes with the derivatization and can damage the GC. Therefore sam-

ples are dried to remove the water.

B.1.6 Derivatize

Add 5 mL of methanol to each dried sample. Add a fixed volume of the deriva-

tization standard (biphenyl-2,2’-dicarboxylic acid). Titrate the sample with dia-

zomethane until the solution remains yellow.

Derivatization converts the acid groups into methyl esters, which increases the

separation of compounds on the GC and makes them more responsive to the de-

tector. Upon GC analysis if the yield of the derivatization standard is low, it is likely

that cations are present in the solution.

B.1.7 Solvent change

Dry the derivatized BPCA solutions under a stream of N2. Add a fixed volume of

dichloromethane. Transfer the sample to a GC autosampler vial.

Polar solvents, like methanol, cannot be injected onto a GC column. Therefore we

111

must change to a solvent more amenable for use on a GC column. The methyl-

esters of BPCAs (created in the previous step) are most soluble in dichloromethane.

B.2 PCGC settings and parameters

General parameters of the PCGC are given in Table B.1. More detailed information

is available upon request.

Table B.1: PCGC settings and parameters

Column DB-XLB column30 m x 53 mm x 1.5 µm film thickness

Oven temperature program initial temperature 100 oCramp to 280 oChold, collect all samplesramp to 320 oCbake out column (5 minutes)

B.2.1 Determination of sample concentration and retention times

Inject the samples and assess the retention times (RTs) of the compounds of in-

terest. Assess if the concentration of the compounds of interest are sufficient for

collection. If not, concentrate the solution by reducing the volume.

112

B.2.2 Program the preparative fraction collector (PFC) to col-

lect at selected RTs

Take care to trap the entire peak, as isotopic fraction occurs if the whole peak is

not captured.

B.2.3 Prime the PCGC for collection

Inject and collect the sample for 10 injections. Discard the collected material and

replace the U-tubes with clean, baked out, U-tubes.

We have found that the BPCA methyl esters from the previous sample remain in

the sample for a few injections. If this step is not done, the 14C of the collected

samples will contain an unknown portion of the previous sample.

B.2.4 Collect sample(s)

Inject and collect the sample for 50 injections.

Inject the sample many times to concentrate the compound(s) of interest into an

isolate. We try limiting the number of injections to 50 because it may minimize the

contribution of extraneous carbon added from the PCGC. Our blank corrections for

compound specific radiocarbon analysis are normalized to 50 injections.

113

B.2.5 Check the concentration and purity of the isolate

After the collection has finished, transfer the isolate to the GC autosampler vial

using a known volume of dichloromethane. Run the sample, as in Appendix B.2.1,

to assess the concentration and purity of the sample.

It is important to know if your isolated sample actually contains the only the com-

pounds of interest that you programmed the PFC to collect. It is important to con-

firm the purity of the isolate. The concentration of this solution will provide yield

information to assess the recovery of the injected samples.

B.2.6 Prepare sample for combustion

Transfer the solution (dichloromethane and isolate) to a clean 6 mm quartz tube.

Dry the dichloromethane under a stream of N2. Add the combustion chemicals,

evacuate to 10-6 Torr and flame off.

114

UNIVERSITY OF CALIFORNIA, IRVINE - · UC Irvine School of Physical Sciences Faculty Endowed Fellowship, 2008-2009 Outstanding Presentation, UC Irvine Institute of Geophysics and

Documents