Application of thermal analysis in radiocarbon dating

Article abstract of DOI:10.1023/A:1011509516021

Thermal analysis combined with mass spectrometry was applied to radiocarbon dating procedures (age determination of carbon containing samples) to determine the optimal temperature range for the reduction of CO₂ over metallic cobalt of various particle sizes. Experiments were carried out to show the different catalytic activities of cobalt of particle sizes 1, 1-45 and 44 μm. The morphology of the cobalt samples and the deposited carbon were investigated. The quantification of CH₄ and CO formed during the reduction of CO₂ was done by means of pulse thermal analysis.

Full text of DOI:10.1023/A:1011509516021

Journal of Thermal Analysis and Calorimetry, Vol. 64 (2001) 1219–1228
APPLICATION OF THERMAL ANALYSIS IN
RADIOCARBON DATING
D. Brandova1, M. Maciejewski2 and W. A. Keller1
1Department of Geography, University of Zurich, Winterthurerstrasse 190, 8057 Zurich,
Switzerland
2Laboratory of Technical Chemistry, Swiss Federal Institute of Technology, ETH Zentrum,
8092 Zurich, Switzerland
Abstract
Thermal analysis combined with mass spectrometry was applied to radiocarbon dating procedures
(age determination of carbon containing samples) to determine the optimal temperature range for the
reduction of CO2 over metallic cobalt of various particle sizes. Experiments were carried out to show
the different catalytic activities of cobalt of particle sizes 1, 1–45 and 44 µm. The morphology of the
cobalt samples and the deposited carbon were investigated. The quantification of CH4 and CO
formed during the reduction of CO2 was done by means of pulse thermal analysis.
Keywords: pulse thermal analysis, thermal analysis in radiocarbon dating
Introduction
Radiocarbon dating is one of the main methods used for estimating the age of carbon
containing materials [1]. It is based on the measurement of the concentration of the
radioactive isotope 14C that occurs in extremely small quantities. 14C is continuously
produced in the atmosphere by the interaction of cosmic ray neutrons with nitrogen
molecules. The half-life of this isotope is 5 600 years. 14C formed in the atmosphere
oxidises to 14CO2 and mixes with common 12CO2. The ratio of 14C to 12C is, in nature,
constant but when the continuous, dynamic cycle of carbon isotope exchange is
stopped (e.g. death) then the amount of 14C slowly reduces by radioactive decay (to
14N) at a rate governed by its half-life. Thus, the age of carbon-containing matter can
be determined by measuring its 14C content. Elemental carbon isolated from the sam-
ple is the target used in 14C-dating by accelerator mass spectrometry. Elemental car-
bon is produced by the catalytic reduction of CO2 over magnesium [2], iron [3, 4],
zinc [5] or metallic cobalt [6, 7]. In the presented paper thermal analysis was used to
investigate and optimise the processes necessary for the formation of elemental car-
bon by the catalytic reduction of CO2 over metallic cobalt.
1418–2874/2001/ $ 5.00
Akadémiai Kiadó, Budapest
© 2001 Akadémiai Kiadó, Budapest
Kluwer Academic Publishers, Dordrecht

1220
BRANDOVA et al.: THERMAL ANALYSIS IN RADIOCARBON DATING
Experimental
A sample for radiocarbon dating is treated with hydrochloric acid to remove non-or-
ganic carbon in the form of carbonates, followed by a treatment with sodium hydrox-
ide to remove humic acids formed during the rotting process. This procedure should
leave only the carbon that was incorporated into the organism during its life-time. In
order to obtain the carbon dioxide that is later used for the synthesis of elemental car-
bon a sample is heated in an evacuated sealed quartz tube with CuO as an oxygen
source. The resulting carbon dioxide is mixed with hydrogen in the ratio 2.5:1 and
catalytically reduced over cobalt powder to elemental carbon.
Cobalt powders of particle sizes 1, 1–45 and 44 µm (Johnson Matthey) were se-
lected in order to study their catalytic activities in the laboratory-designed system
used for catalytic conversion. In order to check the morphological properties of the
different cobalt samples the BET surface area (krypton adsorption) and SEM analy-
ses were undertaken. The determination of the amount of oxide in the cobalt samples
was made under a 20 vol% H2 atmosphere on a Netzsch STA 409 thermoanalyzer
coupled to a QMG 420 (Balzers) quadrupole mass spectrometer. Investigation of the
behaviour of the cobalt samples under a CO atmosphere was carried out under a 20
vol% CO, balance He, atmosphere. To study the reaction(s) that occur during the cat-
alytic reduction of CO2 the cobalt samples were heated in a CO2:H2 (1:2) atmosphere.
The temperature was raised at a rate of 3 K min-1 over a range of 30 to 700°C. The for-
mation of H2O, CO and CH4 over Co (1–45 µm) during the injection of 1 mL of CO2
was quantified by pulse thermal analysis [8]. The same method was also applied for
measuring the amount of H2O, CO2 and CH4 formed during the injection of 1 mL of
CO over Co (1–45 µm). The determination of the optimum temperature for the reduc-
tion of CO2 by hydrogen over metallic cobalt (44 µm) was investigated in the range
480 to 580°C in the laboratory-designed system used for catalytic conversion.
Results and discussion
The dependence of pressure change, during the reduction of CO2 in the labora-
tory-designed system, on the various particle sizes of the cobalt samples is presented
in Fig. 1. The changes of pressure at the beginning of the reaction, as well as the pres-
sure changes during the course of the reaction, indicate that the activity of the cobalt
samples depends on their particle sizes. The particle size also influences the forma-
tion of undesirable gaseous products, manifested as a higher pressure in the system,
which considerably delay the formation of carbon deposited as a result of CO
disproportination (Boudouard reaction).
In order to check the morphological properties of the cobalt samples under
study, the BET surface area (krypton adsorption) and SEM analyses were undertaken.
The surface areas of the cobalt samples were determined to be:
44 µm:0.54 m2 g–1 1–45 µm:0.95 m2 g–1
1 µm:1.22 m2 g–1
J. Therm. Anal. Cal., 64, 2001

BRANDOVA et al.: THERMAL ANALYSIS IN RADIOCARBON DATING
1221
Fig. 1 Change of pressure due to reduction of CO2 by hydrogen in the labora-
tory-designed system over metallic cobalt powder (1, 1–45, 44 µm) at 550°C
Fig. 2 SEM microphotographs of the cobalt samples, the horizontal side has a length of
27.0 µm
The microphotographs presented in Fig. 2 show the morphology of the samples.
The microphotographs of graphite (Fig. 3) deposited on these cobalt samples show
the different morphologies in accordance to their particle sizes. The 1–45 µm cobalt
particles catalyse the formation of filamentous carbon [9], common hexagonal graph-
ite is formed on the 44 µm cobalt.
Thermal analysis combined with mass spectrometry was applied to optimise the
formation of carbon by the catalytic reduction of CO2 over metallic cobalt to CO and
finally to carbon. Carbon formation should be: (i) fast, (ii) occur with maximum yield
and (iii) lead to optimal morphological properties so that it can be used as a target in
accelerator mass spectrometry. In the first step of the investigation, the amount of ox-
ide phase present in the cobalt samples was investigated. The results of the TA-MS
analysis (Fig. 4) of the reaction of CoO by hydrogen show that the amount of CoO
present in the samples was: 3.7 mass% in the 1 µm, 3.2 mass% in the 1–45 µm and
1.6 mass% in the 44 µm sample. The reduction of the 1 µm cobalt sample occurs in
J. Therm. Anal. Cal., 64, 2001

1222
BRANDOVA et al.: THERMAL ANALYSIS IN RADIOCARBON DATING
Fig. 3 SEM microphotographs of graphite formed on the cobalt samples, the horizontal
sides have lengths of 4.3, 2.1 and 27.0 µm respectively
Fig. 4 TA-MS investigation of the reduction of the cobalt samples by 20 vol% H2
the range 180–250°C with a maximum at 217°C. In the case of the 1–45 µm and
44 µm samples the reductions occur in two steps within the ranges 180–290 and
200–400°C, respectively.
To study the disproportionation of CO over the cobalt samples experiments were
carried out under a 20 vol% CO atmosphere. Disproportionation of carbon monoxide
takes place according to the Boudouard equilibrium, which can be presented by the
equation:
2CO↔C+CO2 ∆H= –173.4 kJ mol–1
(1)
The formation of carbon is exothermic and is favored at low temperatures. The
reduction of CoO, present in the investigated cobalt samples, followed by carbon for-
mation due to CO disproportionation, is shown in Fig. 5. The disproportionation of
CO over the 1 and 1–45 µm samples starts at ca 330°C and at ca 300°C for the 44 µm
sample and is reflected by the mass gain due to carbon formation. The interpretation
J. Therm. Anal. Cal., 64, 2001

BRANDOVA et al.: THERMAL ANALYSIS IN RADIOCARBON DATING
1223
of the TG and MS curves is difficult due to the simultaneous occurrence of both pro-
cesses i.e. reduction of CoO and CO disproportionation.
Fig. 5 TA-MS investigation of the reduction of the cobalt samples by 20 vol% CO and
of carbon formation due to CO disproportionation
Fig. 6 TA-MS analysis of the 1 µm cobalt sample heated at a rate of 3 K min–1 under a
CO2:H2 (1:2) atmosphere
J. Therm. Anal. Cal., 64, 2001

1224
BRANDOVA et al.: THERMAL ANALYSIS IN RADIOCARBON DATING
To study the catalytic reduction of CO2 over metallic cobalt the samples were
heated in a CO2:H2 (1:2) atmosphere. The results are depicted in Figs 6–8. The 1 µm
sample catalyses the formation of CH4 with a maximum at 413°C. The results of
TA-MS analysis show that the formation of carbon, due to CO disproportionation,
occurs in the temperature interval ca 380–600°C (Fig. 6).
Fig. 7 TA-MS analysis of the 1–45 µm cobalt sample heated at a rate of 3 K min–1
under a CO2:H2 (1:2) atmosphere
The 1–45 µm sample also produces significant quantities of CH4 with maximum
production at 370°C. The carbon forms in the temperature interval 383 to 600°C
(Fig. 7). The 44 µm cobalt sample does not catalyse CH4 formation, CO begins to
form at temperatures above 510°C (Fig. 8). The comparison of the change of mass
and the formation of methane and carbon monoxide during the heating of each of the
cobalt samples in a CO2 : H2 atmosphere shows the different catalytic activities of the
cobalt samples, depending on their morphologies. The most active samples, 1 and
1–45 µm, catalyse the formation of CH4 which leads to a decrease of carbon yield,
moreover, above 600°C, carbon that has been already formed in the range 380–600°C
reacts with gaseous products. This process distinctly decreases the yield of carbon
formation by the reduction of CO2 by hydrogen.
The quantification of the CO, CH4 and CO2 formed over the 1–45 µm cobalt
sample during the injection of either 1 mL of CO2 or 1 mL of CO into a 20 vol% H2 at-
mosphere was carried out by pulse thermal analysis. The results shown in Figs 9 (CO2
pulses) and 10 (CO pulses) allow the determination of the range of methane forma-
tion and the temperature of the beginning of CO disproportionation. In both cases, the
J. Therm. Anal. Cal., 64, 2001

BRANDOVA et al.: THERMAL ANALYSIS IN RADIOCARBON DATING
1225
Fig. 8 TA-MS analysis of the 44 µm cobalt sample heated at a rate of 3 K min–1
under a CO2:H2 (1:2) atmosphere
Fig. 9 Formation of H2O, CO and CH4 during the injection of 1 mL of CO2 into a
20 vol% H2, balance He, carrier gas stream flowing over the 1–45 µm cobalt
sample heated at a rate of 3 K min–1
J. Therm. Anal. Cal., 64, 2001

1226
BRANDOVA et al.: THERMAL ANALYSIS IN RADIOCARBON DATING
amount of methane produced is negligible above 550°C but at temperatures below
400°C it is the main gaseous product. The reduction of CO2 to carbon occurs in two
steps; in the first, CO is formed in a reversed water-gas shift reaction:
H2+CO2→H2O+CO
(2)
and in the second, CO disproportionation to carbon occurs according to reaction (1).
In order to check the process of CO disproportionation in a hydrogen atmosphere, as
carried out in the laboratory procedure, pulses of 1 mL CO were injected over the Co
1–45 µm sample into a 20 vol% H2 atmosphere. The mass spectrometric signals
m/z=15, 18 and 44, presented in Fig. 10, show the temperature range of possible pro-
cesses such as methane formation, water-gas shift or Boudouard reactions. The for-
mation of CH4 occurs in the range 250–450°C, the visible mass gain due to carbon de-
position begins at ca 350°C. Note the mass loss at temperatures above 500°C due to
secondary reactions of the formed carbon with gaseous reactants and products. Simi-
lar phenomenon were observed in the experiments presented in Figs 6 and 7, how-
ever, direct comparison of all results can only be made qualitatively due to the varied
experimental conditions. The yields of formation of CO and CH4 during the injection
of CO2 into 20 vol% H2 are presented in Fig. 11 A, the yields of formation of CH4 and
CO2 during the injection of CO into 20 vol% H2 are presented in Fig. 11 B.
Fig. 10 Formation of H2O, CO2 and CH4 during the injection of 1 mL of CO into a
20 vol% H2, balance He, carrier gas stream flowing over the 1–45 µm cobalt
sample heated at a rate of 3 K min–1
The optimal temperature for the reduction of CO2 over the 44 µm cobalt sample
was investigated in the system used for the preparation of elemental carbon for radio-
J. Therm. Anal. Cal., 64, 2001

BRANDOVA et al.: THERMAL ANALYSIS IN RADIOCARBON DATING
1227
carbon dating. Changes of pressure, due to the reduction of CO2 by hydrogen at vari-
ous temperatures, are presented in Fig. 12.
An increase of temperature increases the reaction rates but formation of carbon,
due to the Boudouard reaction, is favored at low temperatures. At higher tempera-
Fig. 11 (A) The yield of conversion of CO2 into CO and CH4 and (B) the yield of con-
version of CO into CO2 and CH4 during the injection of 1 mL of CO2, respec-
tively CO, into a 20 vol% H2, balance He, carrier gas stream flowing over the
1–45 µm cobalt sample heated at a rate of 3 K min–1
Fig. 12 Change of pressure due to the reduction of CO2 by hydrogen in the labora-
tory-designed system over the 44 µm cobalt powder. The reduction tempera-
tures (°C) are noted on the curves
J. Therm. Anal. Cal., 64, 2001

1228
BRANDOVA et al.: THERMAL ANALYSIS IN RADIOCARBON DATING
tures the pressure drop in the system is smaller because, above 560°C, the carbon al-
ready formed reacts with the gaseous products. This process is undesirable because it
decreases the amount of carbon produced.
Conclusions
The presented results indicate the necessity of choosing optimal temperature condi-
tions and morphological properties of cobalt when it is used to catalytically reduce
CO2 to carbon. The cobalt samples exhibit different catalytic behaviors depending on
their morphology.
The most active samples (1 µm) and (1–45 µm) catalyse the formation of CH4
which decreases the carbon yield, moreover, above 600°C, carbon that has already
been formed reacts with gaseous by-products and decreases the yield even further. In
addition, the physical properties of these carbon–cobalt mixtures makes them unsuit-
able for accelerator mass spectrometry targets.
The 44 µm cobalt sample, as used in the laboratory system, does not catalyse
CH4 formation. The maximum reaction rate for this sample is reached at temperatures
between 540 and 560°C but above 560°C, carbon that has already been deposited re-
acts with gaseous by-products, hence decreasing the yield. For optimal conditions the
reaction temperature should not be set higher than 560°C. The 44 µm cobalt sample
catalyses the formation of hexagonal graphite which, together with the cobalt, have
physical properties that make it a very good target material.
* * *
The authors would like to thank Prof. M. Müller from the ETH Zürich for making the SEM photo-
graphs and I. Woodhatch for his technical assistance.
References
1 K. M. Goh, Carbon Isotope in Techniques, Ed. D. D. Coleman, Academic Press, (USA) San
Diego 1991, p. 125.
2 G. A. Klouda and L. A. Currie, Nucl. Instr. Meth., B5 (1984) 265.
3 J. S. Vogel, J. R. Southon, D. E. Nelson and T. A. Brown, Nucl. Instr. Meth., B5 (1984) 289.
4 D. E. Nelson, J. S. Vogel, J. R. Southon and T. A. Brown, Radiocarbon, 26 (1986) 215.
5 G. Bonani, R. Balzer, H. J. Hofmann, E. Morenzoni, M. Nessi, M. Suter and W. Wölfli, Nucl.
Instr. Meth., B5 (1984) 284.
6 D. Brandova, M. Maciejewski and W. A. Keller, J. Therm. Anal. Cal., 56 (1999) 893.
7 J. S. Vogel, J. R. Southon and D. E. Nelson, Nucl. Instr. Meth., B29 (1987) 50.
8 M. Maciejewski, C. A. Müller, R. Tschan, W.-D. Emmerich and A. Baiker, Thermochim. Acta,
295 (1997) 167.
9 R. T. K. Baker and P. S. Haris, Chemistry and Physics of Carbon, Ed. Ph. L. Walker
and P. A. Trower, Dekker, New York 1965, p. 86.
J. Therm. Anal. Cal., 64, 2001