Geochemistry of Radiocarbon in Organic Materials

Size: px

Start display at page:

Download "Geochemistry of Radiocarbon in Organic Materials"

Adele McCoy
5 years ago
Views:

1 Geochemistry of Radiocarbon in Organic Materials Suggested Reading: General: McNichol A.P. and Aluwihare L.I. (27) The Power of Radiocarbon in Biogeochemical Studies of the Marine Carbon Cycle: Insights from Studies of Dissolved and Particulate Organic Carbon (DOC and POC). Chem. Rev. 17, Ingalls A.E. & Pearson A. (25) Ten years of compound specific radiocarbon analysis. Oceanography magazine Specific topics: Eglinton T.I., Benitez-Nelson B.C., Pearson A., McNichol A.P., Bauer J.E. and Druffel E.R.M. (1997) Variability in radiocarbon ages of individual organic compounds from marine sediments. Science 277, Pearson A., Eglinton T.I. and McNichol A.P. (2) An organic tracer for surface ocean radiocarbon. Paleoceanog. 15, Wang X.-C., Druffel E.R.M., Griffin S., Lee C. and Kashgarian M. (1998) Radiocarbon studies of organic compound classes in plankton and sediment of the northeastern Pacific Ocean. Geochim. Cosmochim. Acta 62, Raymond P.A. and Bauer J.E. (21) Riverine export of aged terrestrial organic matter to the North Atlantic Ocean. Nature 49, Blair N.E. et al. (23) The persistence of memory: The fate of ancient sedimentary organic carbon in a modern sedimentary system. Geochim. Cosmochim. Acta 67, Ingalls A.E., Shah S.R., Hansman R.L., Aluwihare L.I., Santos G.M., Druffel E.R.M. and Pearson A. (26) Quantifying archaeal community autotrophy in the mesopelagic ocean using natural radiocarbon. PNAS 13 (17)

interaction of neutrons (produced by cosmic rays) and nitrogen atoms.

2 Geochemistry of Radiocarbon in Organic Materials Natural abundance of stable and radio isotopes of carbon 12 C - 99% 13 C - 1% C - 1 part per trillion (1-12 ) in modern carbon C is a cosmogenic nuclide - continually formed in the upper atmosphere (lower stratosphere/upper troposphere) by interaction of neutrons (produced by cosmic rays) and nitrogen atoms. N + neutron C + proton After formation, C atoms rapidly combine with oxygen to form CO 2 which mixes throughout atmosphere, dissolves in the oceans, and enters the biosphere via carbon fixation by marine and terrestrial photoautotrophs. 2

3 Geochemistry of Radiocarbon in Organic Materials There is a dynamic equilibrium between C formation and decay, leading to an approximately constant level in the atmosphere. Current best estimate for half-life, T½, of C = 573 yr Conventional (Libby) half-life adopted for reporting C ages = 5568 yr (3% smaller than true half-life). The half-life is related to the meanlife,τ, by: T½ = (ln2)τ (or T½ =.693τ) Corresponding meanlife, τ, for Libby half-life is 833yr. The radiocarbon age ( C yr), t, can be determined from: t = -τ ln(a/a ) (or t = -833 ln(a/a )) where A is the number of atoms left after time t and A is the initial number of atoms. Key attribute: 573 yr half life of C is suitable for studying processes and dating carbonaceous materials over yr time-scales. Methods of C measurement 1. Conventional method - Determination of C activity of a weighed sample by counting the number of electrons (beta particles) emitted from nucleus per unit time by the decay of C. Beta-counting can be performed by samples combusted to CO 2 (gas proportional beta counting) or on samples converted to benzene and measured photometrically after addition of a scintillator (liquid scintillation counting). Sample size requirements: > 1 g C and long counting times (days). 2. Direct measurement of the proportion of C atoms (relative to 13 C or 12 C) by accelerator mass spectrometry (AMS). Measurements are typically made on graphite (CO 2 also possible). Graphite is formed by combustion of sample to CO 2 and then reduction of CO 2 to graphite. Measurement times as short as 2 min. Key attribute of AMS - Isobar rejection: Negative ions (Cs sputter source) remove ( N + ) Electron stripping (accelerator) to remove hydrides ( 13 CH - ) Sensitivity of AMS = ( 6, yr; 1 half-lives). Sample size requirements: Standard targets < 1 milligram C* (as little as 3 μg C) for full precision (+/- 4 ) As low as 5 μg C now possible at reduced precision (+/ ). Blanks become major issue at these low sample sizes. Standards: Oxalic acid (HOxI, HOxII) *Small sample size has opened up many new applications for C. 3

Accelerator Mass Spectrometer C systematics The absolute international standard of C activity (A abs ) is defined as 95% of the C activity of the original oxalic acid standard (HOxI), in the year 195.

4 Accelerator Mass Spectrometer C systematics The absolute international standard of C activity (A abs ) is defined as 95% of the C activity of the original oxalic acid standard (HOxI), in the year 195. This is equivalent to the activity of 19 th century (189 AD) wood, and represents the C concentration of the atmosphere prior to anthropogenic influence (fossil fuel combustion, atomic weapons testing). The measured activity of HOxI (A ox ) is corrected from fractionation effects using a defined δ 13 C ox value of 19 to yield the fractionationnormalized activity (A ON ): A ON =.95A ox ( δ 13 C) This is corrected to account for radioactive decay between 195 and the year of measurement: A abs = A λ( y 195) ONe 4

5 C systematics The measured C activity of a sample (A s ) is normalized (A sn ) to a constant δ 13 C value of 25 to remove the influence of isotopic fractionation on the reported concentration: A sn 13 2(25 + δ C = A s 1 1 sample ) To a first approximation, the above equation treats the C fractionation as twice the 13 C fractionation (to account for the greater mass difference). This is based on physical-chemical derivations that suggest the C fractionation is approximately equal to the square of the 13 C fractionation. The mean age correction is about 16 years for every 1 difference from -25. This may be simplified to: A A s 13 δ C sn = 2 2 sample The above equations were first developed for C measurements from decay counting techniques. AMS yields absolute ratios of C/ 12 C (or C/ 13 C) in a sample, rather than the rate of decay (activity). The above equations are still applicable, as activity and R /12 are proportional via the decay constant, λ (based on the real half-life = ). AMS data are reported as fraction modern (f m ) values, rather than activities: Δ C = A A f m sn λ ( y 195) ONe C systematics Asn R =.1( pmc) = = AON R 1 *1 = 12 sn 12 ON When a radiocarbon age (year date) is not desired, data are reported as Δ C values in one of two forms. For samples with no age correction, where y is the year of measurement: λ ( y 195) ( f e 1) * 1 For samples of known geochronological age, where y is the year of measurement, and x is the year of sample formation: m λ (195 x) ( f e 1) * 1 λ ( y x) A sne Δ C = 1 *1 = ( 195) λ y m AONe The radiocarbon age of a sample is strictly defined as the age calculated using the Libby half-life (5568 y) for radiocarbon. - In classical radiocarbon dating applications, the calculated radiocarbon ages are converted to calendar ages using calibration curves. 5

6 Reporting of Radiocarbon Data 1 1. Percent Modern Carbon (pmc) Fraction Modern (F m ) Δ C (permil) C age (yr BP) Fm and pmc report the fractional amount of C in a sample relative to that in the standard. - A radiocarbon age ( C yr), is not a calendar or chronological age and must be calibrated. - Δ C normalizes the C content of a sample to the same δ 13 C (-25 permil) and time-point (195 AD). It is a linear quantity & can be used in mass balances. Δ C is useful for isotopic mass balance calculations Radiocarbon age versus Δ C Δ C originally defined by Broecker and Olson (1959). Am. J. Sci. Radiocarbon Suppl. 1,

7 Factors influencing radiocarbon abundances: 1. Atmospheric C variations Variations in solar (cosmic ray flux) activity (long-term, ~ 1 3 yr variations in production rate). Variations in Earth s geomagnetic field strength (short-term, < 1 2 yr variation in production rate) Climate induced variations - solubility of CO 2 in water a function of temperature. Volcanic activity Anthropogenic activity. - Fossil fuel burning ( Suess effect ). - Nuclear weapons testing ( Bomb spike ). 2. Fractionation effects The fractionation effect for C is assumed to be double that for 13 C (reflecting mass difference relative to 12 C). Conventional radiocarbon ages are corrected to a single δ 13 C value (-25 = approximate value for wood). 3. Source or reservoir effects. There is rapid global mixing between the atmosphere and the terrestrial biosphere. However, mixing rates in deep ocean are slow. Mixing between surface mixed layer (high C) and deeper layers (lower C) gives rise to an offset between mixed layer and atmosphere. This offset ( reservoir effect ) for the pre-bomb era is on average ca. 4 yr, but varies spatially and temporally. Thus organic matter synthesized in the oceans will have an apparent age which is 4 yr older than terrestrial biomass synthesized at the same time. C variations and radiocarbon calibration High precision C calibration curve for the past 7 yr (from Irish Oak tree ring chronology). Straight line is the 1:1 correspondence between C age and dendrochronological age. Short-term (1 2 yr) variations are due to geomagnetic field variations. Long-term (1 3 yr) variations are due to variations in cosmic ray flux. 7

Recent variations of C in atmospheric CO 2 2 1 Fraction modern (f M ) 1.8 1.6 1.4 1.

8 Recent variations of C in atmospheric CO Fraction modern (f M ) Maunder Minimum The Bomb Spike Δ C ( ) 1 The Suess Effect? Calender Year 2 8

Wiggle-matching, dendrochronologies and varve chronologies C variations vs varve age for Cariaco basin sediments, compared to those from tree

9 Wiggle-matching, dendrochronologies and varve chronologies C variations vs varve age for Cariaco basin sediments, compared to those from tree rings Cariaco Basin Hughen et al., 1998 Nature, v391 Tree-rings Δ C in the atmosphere as a function of calendar year. Data obtained from INTCAL4 9

10 Potential limitations in assigning Calendar ages from C data Geochemical Applications of Radiocarbon Development of sediment chronologies Tracer studies (e.g., bomb-spike) Isotopic mass balance 1

11 The Global Organic Carbon Cycle (ca. 195) ATMOSPHERIC CO 2 (75) fossil fuel utilization SEDIMENTARY ROCKS (KEROGEN) (15,,) -2 to -35 t = infinite uplift physical weathering (.2) LAND PLANTS (57) -25 to -3 (C3) - to -2 (C4) t = yr humification (5) SOIL HUMUS (1,6) -25 to -3 (C3) - to -2 (C4) t =1-1, yr net terrestrial primary production (5) river or eolian transport (POC=.15) OC preservation (.15) net marine primary production (5) OCEAN (BIOTA) (3) -2 to -3 (C3) t = 4 yr sediment redistribution (units=1 15 gc) Modified after Hedges (1992) burial RECENT SEDIMENTS (1) -2 to -25 t =?? yr C age of bulk OC in surface-most (-2cm) marine sediments C age (yr) Water Depth (m) river-dominated shelf abyssal anoxic shelf basin OMZ open shelf continental slope

12 Radiocarbon age of bulk OC in Long Island Sound sediments C age (yr BP) Depth (cm) (Benoit et al., 1979) Variations in stable carbon isotopic and radiocarbon composition of organic matter as a function of particle size Coastal N. Sea surface sediment Percent modern carbon δ 13 C Megens et al. (22) Org. Geochem

Beaufort Sea - River-dominated (Mackenzie) - Old organic carbon inputs Molecular Multi-Isotopic Studies of Marine Sediments Box

Systematic gradients in: - Organic matter input (terrestrial vs marine) - Oxygen exposure time Bulk Bulk-vapor >25 25-1 1-63 63-38

13 Beaufort Sea - River-dominated (Mackenzie) - Old organic carbon inputs Molecular Multi-Isotopic Studies of Marine Sediments Box Core Surface Grab Particulate Washington Margin - River-dominated (Columbia R) Washington Margin Cascadia Basin transect Coppola et al. in prep. Organic carbon (%) 3 WM WM C/N - Sediment Corg grain-size distribution WM Grain size (μm) Systematic gradients in: - Organic matter input (terrestrial vs marine) - Oxygen exposure time Bulk Bulk-vapor > <38 <38 >1m d-1 <38 <1m d Corg/Ntot δ 13 Corg (permil) Grain s Uchida et al., in prep WM4 δ 13 C Δ C Bulk Bulk-vapor > <38 <38 >1m d-1 <38 <1m d-1 Grain size (μm) Grain s Δ Corg (permil) 13

14 Variations in stable carbon isotopic and radiocarbon composition of organic matter as a function of chemical compound class δ 13 C Percent modern carbon Megens et al. Variations in stable carbon isotopic and radiocarbon composition of organic matter as a function of chemical compound class Height above bottom (m) TOC THAA TCHO Lipid Acid-insol BC alkenone aggregates flocc plankton zooplankton sinking POM Depth in sediment (cm) Station M Wang, Druffel et al a δ 13 C ( ) Δ C ( ) b

15 Isotopic ( 13 C, C) evidence for a lipid-like source for acid insoluble macromolecular organic matter in sinking POM Hwang & Druffel 23, Science 299,882 15

Molecular-level Radiocarbon Analysis The Problem: Many samples contain heterogeneous mixtures of organic compounds of diverse origin (and age).

16 Molecular-level Radiocarbon Analysis The Problem: Many samples contain heterogeneous mixtures of organic compounds of diverse origin (and age). Age variability can be a source of interference, or information. The Approach: Structurally diverse organic compounds are preserved in sediments and carry a wealth of biogeochemical information. Measure the stable- and radio- carbon isotopic composition of individual organic compounds in order to constrain the origin of OC buried in sediments. Isotopic mass balance using both C and 13 C allows for three OC source inputs (phytoplankton, vascular plant, relict organic matter) to be defined. Select compounds for C and 13 C analysis using biochemical criteria, rather than characterizing OC pools based on operational definitions. Molecular C contents also provide apparent ages for assessment of the residence times and cycling rates within (and between) carbon reservoirs. The Challenge: To measure the natural abundance of C in individual organic compounds in complex mixtures. Greater than 25 μg C required for reliable C measurement (by AMS). Isolation of target analytes in very high purity. Conventional capillary GC resolves < 5ng compound. The Approach: Automated Preparative Capillary Gas Chromatography (PCGC). 16

17 Typical isolation scheme for CSRA Sediment Solvent extraction FID Intensity C 26 Δ 5,22 + C 26 Δ μgc C 27 Δ 5 86 μgc 5α-C μgc C 28 Δ 5, μgc C 29 Δ μgc 5α-C 29 (sample lost) Extract SiO 2 chromatography Polarity fraction Derivatisation? Target compound mixture PCGC irm-gc-ms/gc-ms Trap 1 Trap 2 Trap 3 Trap 4 Trap 5 Trap 6 Trap Total Combustion irm-gc-ms/gc-ms Retention Time (min.) Carbon dioxide Reduction Graphite AMS Correction for derivative carbon N.B. Derivatives are typically derived from petrochemicals (i.e., Δ C = -1 ) 17

18 Δ 5 sterols GDGT- Smittenberg et al 22 J. Chromatog

$Alkenone purification methodology O Sediment Solvent extraction Total Lipid Extract Saponification Neutral fraction SiO 2 chromatography Ketone fraction Urea adduction$

19 Alkenone purification methodology O Sediment Solvent extraction Total Lipid Extract Saponification Neutral fraction SiO 2 chromatography Ketone fraction Urea adduction Straight-chain ketones O Ag + SiO 2 chromatography Unsaturated ketones Al 2 O 3 chromatography Purified alkenones Combustion Carbon dioxide irms C 37 alkadienone Graphite Reduction AMS 19

20 Proof of Concept: Bulk vs molecular C of isotopically homogeneous samples 2

21 21

22 TOC, foram and sterol Δ C in Santa Monica Basin sediments Depth in sediment core z (cm) Δ C Benthic forams Post-Bomb interval DIC TOC Planktonic forams Pre-Bomb interval Δ C ± ± 19 SMB, TOC SMB, TOC Postbomb DIC SMB -1 SBB -1 SMB 5-6 Prebomb DIC δ 13 C Data from Pearson et al. (2) 22

23 C contents of lipid biomarkers in CBB sediments Data from Pearson et al. (21) Quantifying archaeal community autotrophy in the mesopelagic ocean using natural radiocarbon 23

Developing sediment chronologies for the Ross Sea, Antarctica Ross Sea Surface Sedimentary OC composition Site Fairy Water Depth (m) 671 TOC (%).24 δ 13 C -25.65 C age 999 Gentoo 623.34-27.

24 Developing sediment chronologies for the Ross Sea, Antarctica Ross Sea Surface Sedimentary OC composition Site Fairy Water Depth (m) 671 TOC (%).24 δ 13 C C age 999 Gentoo Emperor Chinstrap Antarctica Data source: Ohkouchi et al. (23) Fatty acid vs TOC C ages of Ross Sea sediments Chinstrap Site C fatty acid C16 fatty acid C18 fatty acid TOC Data source: Ohkouchi et al. (23) Depth in core (cm) cm/kyr Post-bomb DIC 7.5 cm/kyr Pre-bomb DIC Calendar year (yr BP) (OxCal3 calibration) 24

25 Radiocarbon ages of plant wax fatty acids in size-fractionated sediments Terrestrial (long-chain) WM3 fatty acids older WM4in WM3 WM4 WM4 than WM3 - Aging during lateral transport? Terrestrial fatty acid age increases with increasing grain-size - Eolian input of plant waxes (on fine particles/aerosols)? - More rapid nepheloid layer transport of fine particles? Terrestrial fatty acid older in bulk sediment than any measured fraction? - Age in > 25 μm fraction? Uchida et al., unpublished results C variations as a function of carbon number in sedimentary fatty acids 1 Plant wax fatty acids -1-2 Δ C (permil) Bermuda Rise Gulf of Mexico Mackenzie 35m Mackenzie 2 m Ross Sea Emperor Ross Sea Fairy New England shelf Washington margin TOC Fatty Acid Carbon # 25

26 C age variability in Bermuda Rise surface (-3 cm) sediment Δ C (permil) G. ruber G. inflata Tracer of DIC C in overlying surface waters planktonic forams Labile algal biomarkers total (minor organic advected carbon component?) 1 alkenones fatty acids 2 hydrocarbons 3 Refractory algal Pre-aged biomarkers refractory (large plant advected waxes component) 4 (advected + eolian component)? 5 TOC (-1cm) TOC (1-2cm) Alkenones (-1cm) Alkenones (1-2cm) C16 fatty acid C18 fatty acid C24 fatty acid C26 fatty acid C28 fatty acid Bulk OC = advected marine OM? C21+C23+C25 alkanes C22+C24+C26 alkanes C27 alkane C29+C31 alkanes C28+C3+C32 alkanes UCM C age (yr BP) Petrogenic hydrocarbons (fossil inputs) Ohkouchi et al., submitted 26

Geochemistry of Radiocarbon in Organic Materials

Geochemistry of Radiocarbon in Organic Materials Suggested Reading: Eglinton T.I., Benitez-Nelson B.C., Pearson A., McNichol A.P., Bauer J.E. and Druffel E.R.M. (1997) Variability in radiocarbon ages of