MEASUREMENTS OF THE 13СО2/12CO2 ISOTOPE

COMPOSITION USING THE TDLS

TECHNIQUE

A.Berezin, S. Chernin, A. Makkaveiskii, A. Nadezhdinskii, Ya. Ponurovskii, D.

Stavrovskii

Natural Sciences Center of the A.M. Prokhorov Moscow General Physics Institute, RAS

38 Vavilov street, 119991 Moscow,Russia E-mail: jakov@nsc.gpi.ru

A.Babichev, G.Grigoriev, S.Malyugin,

Sh.Nabiev

Russian Research Center “Kurchatov Institute” 1 Kurchatov sq., 123182 Moscow, Russia

Introduction Isotopes abundance measurements are important now for variety

of applications: Deposit oil fields and gas; control of an ecological situation and man-caused accidents, it is especial in connection with recent entrance into force of the Kyoto treaty; and several medical applications (for example, “breath-test” - an efficient method of diagnostics of gastrointestinal tract diseases), etc. Goal of present paper was instrument development for accurate absolute isotopes abundance measurements for CO2 molecule. Absorption lines R4 12CO2 and Р12 13СО2 located near 6232.5 cm-1 were chosen taking into account independence of results obtained on sample temperature [1]. Fiber pigtail Anritsu DFB DL was used to cover spectral range near 1.6 µm. We’ve used our traditional DL operation mode (see separate poster). “Chernin” multipass cell with total optical path of 42 m was used in the instrument analytical channel. Absorption of main isotopomer spectral line was chosen close to 0.1 to reduce influence of DL electromagnetic field spectrum on results obtained [2]. To reduce instrument operation time observed spectral lines were fitted by Gaussian profile using HITRAN [3] data to determine isotopomers partial pressure. Special procedure was developed to removed baseline (see separate poster) influence on results measured. Intercomparison campaign between Mass-Spectrometer (MS) and the instrument developed (TDLS) was performed using several gas samples having different isotopes abundance.

Precision achieved for present measurement is equal to 0.013 % and is caused by DL quantum noise. Analysis shows that this value can be considered as accuracy if HITRAN data are accurate enough. Achieved precision corresponds to minimum detectable absorption at level of 1.3 10-7.

1. J.McManus, M.Zahniser, D.Nelson, L.Williams, C.Kolb, Spectrochimica Acta, A58, 2465-2479 (2002)

2. Nadezhdinskii A I, et al, Quantum electronics, 29, (10) 916-920 (1999). 3. L.S. Rothman, et al, The HITRAN 2004 molecular spectroscopic

database, JQSRT, vol. 96, No.2, Dec. 1 (2005) in press.

1000 2000 3000 4000 5000 6000 7000 8000 9000 100001E-4

1E-3

0.01

0.1

1

10

100

1000

cm-1/a

tm

ν, cm-1

р д

L.G

iafra

ny [4

]

M.Z

ahna

iser

[1]

Pre

sent

wor

k

CO2 cross section (cm-1/atm) [3] and spectral ranges used by different groups to measure carbon isotope ratio

4. G. Gagliardi, A. Castrillo, R. Iannone, E. R. Th. Kerstel, and L. Gianfrani, Applied Physics B, 77, 119-124 (2003).

Comparison between CO2 absorption spectrum and photo-diode NEP (Noise Equivalent Power).

1000 2000 3000 4000 5000 6000 7000 8000 90001E-4

1E-3

0.01

0.1

1

10

100

1000

1E-14

1E-13

1E-12

1E-11

1E-10

1E-9

1E-8 NEP CO2

K, c

m-1/a

tm

ν, cm-1

NE

P, W

/Hz1/

2

Experimental Setup

Experimental setup block-scheme.

Diode module incorporated an “Anritsu” DFB DL with fiber pigtail emitted at wavelength λ=1.6 µm, a thermistor and a Peltier thermo element. Chernin multipass cell with the 25 cm base length The cell design permitted evacuation of the cell and filling it with the gas to be analyzed under low pressure.

Photodetector with a preamplifier installed on the cell provided for registration of the outcoming radiation.

Interface module is intended for controlling the DL-module and the monitoring system. It incorporates a DL pumping current supply, a DL thermal stabilization system, and a PD preamplifier power supply.

To control DL pumping and thermal stabilization systems, and to record signals with their subsequent processing in a gas analyzer, use was made of I/O Card NI DAQ PCI-6052E of “National Instruments”. The control of the measurement process was executed by software running under the NI LabView 7.0 environment.

The diode laser was pumped with trapezoidal pulses. The current pulse magnitude, Ioper~80 mA, the pulse length was 0.4-1 ms, and the repetition rate was 5 kHz. DL frequency tuning (∆v~2 cm-1) took place because of changing of the refraction index of the laser active region during a pumping current pulse.

Scanning over the whole region of frequency tuning (~60 сm–1) was performed with the temperature stabilization system via smooth variation of the laser crystal temperature with a Peltier thermo element.

thermal stabilization

system

DL current supply

PA power supply

Input PD

I/O Card NI DAQ

PCI

DL-moduleS.M. Chernin multipass mirror

system

Sample input

Instrument View

General view of the 13СО2/12СО2 measuring setup

Anritsu DFB DL with fiber pigtail λ=1.6 µm

Sample feeding system

Multipath

Optical cell

Electron unit

PC (notebook)

Chernin Multipass Cell

Chernin multipass cell: baselength - 25 cm, dielectric mirrows with 99.5 % reflectivity; 182 passes, 45.5 m.

Electrical Block-Scheme

PD2

PC

Electronic board

Inputs Outputs

Interface

Thermosensor power source

Current power source

Preamp power source

Thermoelement control

DL module

DL Thermoelement Thermosensor

PD1

View of interface module

13CO2 and 12CO2 absorption spectra. Procedure of determining the isotope

composition

Fig.a shows a model CO2 transmission spectra in the vicinity of 1.6 µm (according to the HITRAN). Red curves represent the contribution of 13CO2 absorption for the natural isotope abundance. Analytical spectral lines have been chosen so that the gas temperature in the cell produced practically no effect on the measurement results. Therefore, lines with equal temperature dependence of absorption cross-section (see Fig.b) have been chosen. The bottom of Fig. a shows the chosen spectral region and analytical spectral lines of two isotopes of carbon dioxide. Analytical lines were R4 for 12CO2 and Р12 for 13CO2.

6220 6230 6240 6250 62600.96

0.97

0.98

0.99

1.00

6228 6230 6232 6234 62360.99

1.00

CO 2 17 Torr, L=35 m

Пропускание

ν, cm -1

R4

P12

Fig.a 13СО2 and 12СО2 transmission spectra [1]

6200 6220 6240 6260

0.00

0.01

0.02

0.03

d/dT

ln(S

)

ν, cm-1

12C16O2

12C16O2 hot band

13C16O2

16O12C18O

Tran

smis

sion

Fig.b. Relative temperature dependence coefficient for absorption cross-section of spectral lines.

Data Processing

1. Signal recording. 2. Spectrum normalization and frequency scale linearization. Absorption coefficient determination. 3. Spectrum filtering. Absorption spectral line second derivative is used as filter giving second derivative auto-correlation function of the line. 4. Spectrum approximation. It was made with the second derivative of the Doppler profile. 5. Calculation of the d-factor. The ratio of R integral absorption coefficients of 13СО2/12СО2 is found and then magnitude d=(R/R0-1)*1000 0/00 is calculated. R0 was obtained from [3].

0 20 40 60 80 100 120 140 160 180

0.0

0.5

1.0

1.5

2.0

S, V

t, ì êñ

6231.0 6231.2 6231.4 6231.6 6231.8 6232.0 6232.2-0.005

0.000

0.005

0.010

0.015

0.020

0.025

Ê, î

òí. å

ä.

v , ñì -1

31.2 31.4 31.6 31.8-0.004

-0.003

-0.002

-0.001

0.000

0.001

0.002

31.6 31.8-0.00010

-0.00008

-0.00006

-0.00004

-0.00002

0.00000

0.00002

0.00004

0.00006

0.00008

0.00010

d2 /dv2 (K

)

6 200 -v, ñì -1

d2 /dv2 (K

)

6200 -v , ñì -1

1

Spectru m n or mal izatio n

2

Spectru m fi ltering

3

31.20 31.25 31.30 31.35 31.40 31.45 31.50 31.55

-6

-4

-2

0

2

4

31.60 31.65 31.70 31.75 31.80 31.85-0 .15

-0 .10

-0 .05

0.00

0.05

0.10

Y A

xis

Title

X Axis T itle

fi t_B fi t_C

12C O 213C O 2

Y A

xis

Title

X Axis T itle

fit_E fit_F

4

Spectru m ap prox imatio n5

Calcu latio n of the d-factord=(R/R0-1)*1000 0/00

Software

Software control panel view. Software running under the LabVIEW graphical programming environment permits real-time control of the measurement and data processing. Full time of 100 pulses accumulation, averaging and data processing took 0.2 s.

Device Calibration Intercalibration campaign was performed with the Institute of Molecular Physics of RRC “Kurchatov Institute”. Five calibrated carbon dioxide gas samples of various isotope composition were prepared. Isotopes ratio in these samples was measured with a mass-spectrometer and TDLS.

0 1 2 3 4 5 6

3.95

3.96

3.97

13C

/12C

, % 0 1 2 3 4 5 63.46

3.47

3.48

0 1 2 3 4 5 62.45

2.46

2.47

0 1 2 3 4 5 62.96

2.97

2.98

n

Results obtained using instrument developed.

Multipass cell was filled with five probes of each gas sample, and isotopes ratio was measured (five points on each picture). Precision achieved - 0.13 0/00 was limited by DL intensity quantum noise. This precision corresponds to minimum detectable absorption at level of 1.3 10-7.

MS and TDLS Inter-Comparison Results of intercalibration campaign are presented in Table 1.

Table 1

Sample No 15 5 25 3 1 (13С/12C)DLS, % 3.97 3.47 2.46 2.97 1.08 (13С/12C)MS, % 3.97 3.48 2.54 2.94 1.10

Inter-comparison of isotopes abundance measurements by MS and TDLS. Difference between MS and TDLS results is due to MS memory with respect to previously measured samples. Analysis shows that precision achieved - 0.13 0/00 can be considered as accuracy if HITRAN data are accurate enough.

0 1 2 3 4

-30

-20

-10

0

10

20

δ13C

, ‰

(13C/12C)MS, %

Breath Test “Breath test” is well-known non invasive medical technique to detect Helicobacter pylori responsible for different diseases. Presence of Н. pylori in this test is followed by changing of 13C/12C ratio in infected patient breath.

“Breath test” instruction and accessories

Usually MS instruments are using in this diagnostics to measure isotopes ratio in breath samples.

Breath Test “Breath test” is well-known non invasive medical technique to detect Helicobacter pylori responsible for different diseases. Presence of Н. pylori in this test is followed by changing of 13C/12C ratio in infected patient breath. Usually MS instruments are using in this diagnostics.

0 10 20 30 40-10

0

10

20

30

40

50

d, 0 / 00

t, m in

Example of real time clinical “breath test” results obtained using instrument developed for infected (red) and non-infected (blue) patients. Results of clinical tests using TDLS technique and instrument developed showed that they are quite acceptable for clinical practice.

Conclusion The paper reports on the development of

equipment and software for rapid measurements of 13СО2 concentration from the СО2 absorption spectrum in the near-infrared range. The relative error of the d-factor definition in the breath-test on 100-fold accumulation of single measurements was within 0.7 0/00. With the aim to determine the absolute error of isotopic ratio measurements, we took independent calibrating measurements of 5 СО2 samples with different isotope composition using both DLS equipment and mass-spectrometers.

Precision achieved for present measurement is equal to 0.013 % and is caused by DL quantum noise. Analysis shows that this value can be considered as accuracy if HITRAN data are accurate enough. Achieved precision corresponds to minimum detectable absorption at level of 1.3 10-7.

Results of clinical tests using TDLS technique and instrument developed showed that they are quite acceptable for clinical practice.