Microsoft Word - CET--006.docx
CHEMICAL ENGINEERING TRANSACTIONS
VOL. 59, 2017
A publication of
The Italian Association
of Chemical Engineering
Online at www.aidic.it/cet
Guest Editors: Zhuo Yang, Junjie Ba, Jing Pan
Copyright © 2017, AIDIC Servizi S.r.l.
ISBN 978-88-95608- 49-5; ISSN 2283-9216
Research on Optical Fiber Gas Sensing Based on Absorption
Spectrometry in Optical Communication
Na Zhang
Xichang College, Xichang 615000, China
zhangna2112@163.com
The Internet of things (IOT) is one of the most popular research fields at present, and the optical fiber sensing
network is an important part of the Internet of things. In the field of environmental protection and modern
industrial production, there is a need for a sensor network which can monitor the gas concentration
information. Because of the strong anti-interference ability, intrinsic safety, optical communication technology-
based, intelligent monitoring and other significant characteristics, optical fiber gas sensing equipment and
scheme has become the main direction of the field of domestic and foreign research. In this paper, we
presents a 1665nm methane sensor structure based on single fiber coupler ring, and the performance of the
sensor is analyzed, including the relationship between the optimal ratio of 1550nm coupler in the 1665nm
system and the number of rings of signal light. Compared with the traditional absorption spectrum method, the
effective absorption length of the chamber is increased 4 times by the sensor. Methane gas samples of two
groups were measured by the sensor, and the measurement error was less than 2.5%.
1. Introduction
With the rapidly development of global industrialization and the rapidly increase of global population, the
global environment is becoming worse and worse. Air pollution, greenhouse effect and other environmental
problems have become the focus of attention because of the serious threat to human health, the earth's
ecological environment, biodiversity and climate. In order to solve the problem of air pollution, it is the key to
control and monitor the emission of gaseous pollutants (Arapatsakos et al., 2015; Cuisinier et al, 2013;
Cannistraro et al., 2015). In addition, in the industrial production, it should timely and accurately monitor,
forecast, alarm and take corresponding measures to the flammable, toxic and harmful gas to ensure the order
of life, to ensure the safety of people's lives and property of the enterprise, which has become an important
problem of petroleum chemical industry, electric power, electrical and other industries (Castellanos-Gomez et
al, 2016). In the actual application of sensor, the understanding to the gas concentration of individual
monitoring points is not enough, in order to accurately forecast, it must master the gas concentration
distribution in a large range and forecast its trend over time, in order to take corresponding measures before
the disaster (Wild et al, 2014). So, no matter in the field of environmental protection or modern industrial
production, the safe, reliable and stable system that can monitor the gas concentration in many points is
needed. The Internet of things, wireless sensor networks and optical fiber sensing network have become the
focus and the main direction of the field of domestic and foreign research.
2. Theory and method
2.1 Cavity ring-down spectroscopy
Cavity ring-down spectroscopy (Long, et al, 2014) is an absorption spectroscopy appeared in 80s in last
century. The technology has an effective path length and high sensitivity, and is not affected by the instability
of light source, so it has attracted attention. A typical cavity ring-down spectroscopy system is shown in Figure
1 (Truong et al, 2013). The system consists of a laser source, a high reflectivity mirror composed of a
resonator and an oscilloscope. The light emitted by the laser enters into the cavity, and is restricted to travel
between the two high reflectivity mirror oscillation (Renganathan et al, 2014). Each oscillation will carry sense
DOI: 10.3303/CET1759110
Please cite this article as: Na Zhang, 2017, Research on optical fiber gas sensing based on absorption spectrometry in optical communication,
Chemical Engineering Transactions, 59, 655-660 DOI:10.3303/CET1759110
655
information because of the loss by the substances to be measured, and a small part of light will be transmitted
into the oscilloscope every oscillation.
Laser Diode OscilloscopeTest
Sample
Resonant Cavity
R1 R2
Figure 1: A Principle architecture of Cavity Ring-Down Spectroscopy System
The R1 and R2 in the figure are high reflectivity mirrors. The relationship between the intensity of transmitted
light and time can be expressed by (1).
e
t
t
ItI exp, 0
(1)
I0() is the initial intensity of the optical signal with wavelength in the cavity. It(t, ) is the initial intensity of
the optical signal with wavelength in the t moments. The ring-down time of optical signal with wavelength
is defined as the time required in the attenuation of the transmission light intensity to the initial intensity. It is
related to all the loss in the cavity, including the measured medium absorption loss and other losses (medium
scattering loss, mirror transmission, diffraction and scattering loss etc.). RDTe() can be obtained from (1).
cA
nL
e
(2)
n is the refractive index of the cavity, L is the cavity length of the resonant cavity, C is the velocity of light in
vacuum, and the total loss of the cavity is A, e() is the time of the optical signal, which is related to the loss
of the cavity. According to (3), the down time is only related to the physical parameters of the resonator, and it
is not affected by the incident light intensity I0(). When the sample in the cavity introduces the loss such as
absorption, the total loss of A increases which can make the ring down time become small. The loss of the
cavity can be obtained by measuring the attenuation variation down time after inducing the additional loss, and
then the additional value of the loss can be obtained. The principle of measurement can be described as
follows:
When the sample is not present in the cavity, the formula (2) can be written as:
Rc
nL
-1
0
(3)
When the concentration C was measured in the ring cavity, the refractive index of the sample was ignored
when the absorption coefficient of the optical signal was measured. At this point, (3) can be written as:
RCLc
nL
sample
1
(4)
Therefore, the concentration C of the medium to be measured can be obtained.
0
11
sampleabsorptionc
n
C
(5)
It can be seen that the concentration C of the sample to be measured is only related to the transmission rate
c/n of light in the cavity and the absorption coefficient of the sample. From (3) and (5) we can see that sample
and 0 are related to the transmission speed c/n of light, the cavity length L and the reflectivity of the
endoscopic. The medium of n with high refractive index is filled with the resonant cavity, or the cavity length
and the the reflectivity of the endoscopic are increased, so the corresponding splitting ratio of the coupling in
FLCRD is reduced or the loss in ring is decreased, which can increase the ringdown time, so as to improve
the measurement accuracy.
656
2.2 Fiber ring methane gas sensor
Gas absorption chamber (Renganathan, et al, 2017) is one of the key components in the absorption spectrum
gas sensor. In order to improve the sensitivity of detection, it often needs a long absorption chamber. The gas
chamber in the laboratory has a longer effective length, which needs to reach about 50cm. Based on the
absorption spectrum of the sensor, the signal light source must be aligned with the methane absorption peak.
In the past, most of the methane sensors use 1310nm wavelength light source, and the absorption coefficient
is small. In this paper, a 1650nm wavelength light source is developed, which uses the Super-luminescent
Light Emitting Diode with 1653nm center wavelength and 54nm Full Maxmium Half Width (Mishra et al, 2015)
as the light source, measuring the peak position and depth of the methane gas in the range of 1600-1700nm
by using the AQ6317C spectrometer of ANDO company. At the same time, in order to realize the
measurement of the ring down cavity, the light source must be a pulse laser which can generate the pulse,
and the width and the repetition period can be adjusted.
Figure 2 shows the connection of the Distributed Feed-Back Laser Diode which the working wavelength is
1665nm laser and oscillograph LeCory WavePro 7000 to the 2 and 4 port of the coupling device, respectively.
(DFB LD: Distributed Feed-Back Laser distributed feedback Diode LeCory WavePro 7000), the oscilloscope
(OSC: oscillograph) respectively and the coupling device 2 and 4 port connection. The utility model is
characterized in that a pair of collimators with a working distance of 50mm is fixed in the V groove to be
loaded into the air chamber. The two ends of the collimator are respectively connected with the 1 and the 3
ports of the coupling device to form a closed fiber ring.
50mm collimators
Coupler
Gas chamber
DEB
LD
OSC
1 3
2 4
Figure 2: The experimental setup of the fiber loop gas sensor
OFB LD can give out a pulse of light into the fiber ring every certain period, and the output pulse sequence is
shown by the receiver. The down time RDTe of the output signal light is related to the loss in the ring, which
can be expressed as:
e
T
e lg10
(6)
T is the time that the signal light travels a round in the ring. The length of the fiber is 20 meters in the
experiment, then the corresponding single ring time T is 100ns. Α is the total loss of optical signals of a round,
then =fibre loss + collimator loss + coupler loss + absorption loss, and the unit of the loss is dB.
In order to make the relationship between the loss parameter and the concentration more obvious and easier
to solve, the relationship between the concentration of the gas to be measured and the fitting coefficient is
derived directly from the Beer-Lambert law and after combined with the diffusion coefficient:
LC
I
Ii
explg10lg10
0
(7)
The Ii is the size of input light intensity, I0 is the size of the output light intensity, and the is the absorption
coefficient when the wave length is , L is the length of the chamber, C is the gas concentration. So it can be
defined that on the left side of the equation is the gas absorption loss of the single circle, and the unit is dB,
then we can get:
eLCg lg10
(8)
657
The total loss of the single ring is the sum of the gas absorption loss g and the insertion loss s of the
system:
sLC 34.4 (9)
L is 50mm, and can be obtained after combined with data in HITRAN database, and the relationship
between the concentration C and total loss can be obtained according to the above formula. It can be seen
from the experimental results that the diffusion time and the residual gas concentration in the gas diffusion are
in accordance with the following exponential decay:
cbtaC exp
(10)
When the boundary condition t=0, the concentration of C is 100%; when the t approaches infinity, the gas has
been diffused out of the gas chamber, and the concentration is 0%. The parameter can be determined as a=1,
C=0. The relationship between concentration and time can be expressed as:
btC exp
(11)
The relationship between the diffusion time and total loss can be obtained according (10):
sbtL exp34.4 (12)
In order to simplify the calculation, in this paper, we proposed a single loop attenuation coefficient B. The
single loop attenuation coefficient B is defined as the ratio of the single ring time e to the attenuation time:
eTB / (13)
3. Result and analysis
3.1 System calibration
The gas path is the key part of the gas sensor, and an adjustable and controllable gas path is the base of the
gas sensing experiment, including the gas source, gas path control and gas path detection. In order to ensure
the accuracy of the measurement of the concentration of the gas chamber in the experiment period, the
change of the indoor concentration of the gas should be known, so the the gas diffusion experiment is carried
out. Experiment shows that the gas concentration is still stable in the order of ms the extreme case, and when
the valve is closed, the air leakage is less than 1% in 24 hours. So it is accurate and effective to use the gas
chamber to measure the gas concentration.
The experimental system is built according to figure 2. The signal pulse length is 1665.538nm, the pulse width
is 50ns, and the cycle is 1us. In order to test the calibration of the system, the pure methane (purity is >99.9%)
is charged into the gas chamber and the valve is opened to record the output pulse sequence after 14 time
diffusion. The figure 3 is the pulse sequence single exponential fitting diagram at the 8th, 16th and 108th
minutes.
Figure 3: Single exponential fitting curve of output pulse measured at first, 8, 16 and 108 minutes after valve
opening
658
As can be seen in Figure 3, with the change of time, the fitting curve gradually slowed down which reflects the
diffusion of gas is slow from the gas chamber to the outside. The fitting coefficients of the 14 records are
calculated respectively, and the total loss of each time is obtained and represented by the x in figure 4. After
fitting according to (12), it can obtain that p=4631, q=0.07376, r=5.408.
408.5631.4
07376.0
t
e
(16)
When s=5.408db, the absorption coefficient is =0.209/cm, then the (16) can be:
631.4/408.5631.4/408.534.4 BC
(17)
After the methane gas is filled into the gas chamber, only the output pulse sequence is recorded by OSC, and
the coefficient of B in (17) is obtained according to the single exponential fitting, and the corresponding gas
concentration C can be obtained.
3.2 Sample determination
In order to evaluate the error of the scheme, the methane samples A and B of two groups were measured
respectively. The concentrations of the samples were 30.7% and 69.8%, respectively. The output pulse
sequence is shown in Figure 4. The real curve is the output pulse sequence data recorded by OSC. The ×
represents the fitting data, and the dashed line represents a single exponential fitting curve.
Figure 4: The output waveform and fitting waveform when measuring 69.8% methane gas samples
Table 1: The regression results of the sustainable development of chemical industry (�̂�2 is taken as the
example)
Experiment
Sample
1st 2nd Average
B /dB C% B /dB C% Cave% Error/%
A 1.575 6.836 30.8 1.567 6.801 30.1 30.45 0.81
B 1.976 8.576 68.4 2.040 8.854 74.4 71.4 2.29
As can be seen from the table that when the gas concentration is larger, because the fitting series is less that
makes the error larger, thus the number of rings is increased by decreasing the loss in the ring, and it is very
necessary to increase the ring-down time and series.
4. Conclusions
In 1820s, people began to study the gas detection equipment and methods, there have been a variety types of
gas sensors, such as contact combustion type, electrochemical type, semi conductor type, optical fiber type
and so on. The optical fiber gas sensing equipment and the scheme have become the focus of attention and
research because of its strong anti-interference, intrinsically safe, optical communication technology-based,
simple networking, intelligent monitoring and other significant characteristics. In this paper, a single fiber ring
1665nm methane sensor is studied, and the methane 1665nm absorption peak is measured by the sensor.
The relationship between the optimal splitting ratio and the number of rings in the sensor is analyzed
659
theoretically, which is instructive for the selection of the experimental equipment and the improvement of the
detection sensitivity. The gas samples of two groups were measured, and the measurement error was less
than 2.5% when the effective absorption length of the chamber was increased by a factor of 4 by the sensor.
Reference
Arapatsakos C., Karkanis A., Anastasiadou C., 2015, The load and the gas emissions measurement of
outboard engine, International Journal of Heat and Technology, 33(4), 221-228, DOI: 10.18280/ijht.330430
Cuisinier M., Cabelguen P.E., Evers S., He G., Kolbeck M., Garsuch A., Bolin T., Balasubramanian M., Nazar
L.F., 2013. Sulfur speciation in Li–S batteries determined by operando X-ray absorption spectroscopy, The
Journal of Physical Chemistry Letters, 19(4), 3227-3232.
Cannistraro G., Cannistraro M., Cannistraro A., Galvagno A., Trovato G., 2016, Reducing the demand of
energy cooling in the CED, “centers of processing data”, with use of free-cooling systems, International
Journal of Heat and Technology, 34(3), 498-502, DOI: 10.18280/ijht.340321.
Castellanos-Gomez A., Quereda J., van der Meulen H.P., Agraït N. and Rubio-Bollinger G., 2016. Spatially
resolved optical absorption spectroscopy of single-and few-layer MoS2 by hyperspectral imaging.
Nanotechnology, 27(11), 115705.
Wild R.J., Edwards P.M., Dubé W.P., Baumann K., Edgerton E.S., Quinn P.K., Roberts J.M., Rollins A.W.,
Veres P.R., Warneke C. and Williams E.J., 2014. A Measurement of Total Reactive Nitrogen, NO y,
together with NO2, NO, and O3 via Cavity Ring-down Spectroscopy. Environmental science & technology,
48, 16, pp.9609-9615.
Long D.A., Fleisher A.J., Wójtewicz S. and Hodges J.T., 2014. Quantum-noise-limited cavity ring-down
spectroscopy. Applied Physics B, 115(2), 149-153.
Truong G.W., Long D.A., Cygan A., Lisak D., van Zee R.D. and Hodges J.T., 2013. Comb-linked, cavity ring-
down spectroscopy for measurements of molecular transition frequencies at the kHz-level, The Journal of
chemical physics, 138(9), 094201, DOI: 10.1063/1.4792372.
Renganathan B., Sastikumar D., Raj S.G. and Ganesan A.R., 2014. Fiber optic gas sensors with vanadium
oxide and tungsten oxide nanoparticle coated claddings, Optics Communications, 315, 74-78.
Mishra S.K., Tripathi S.N., Choudhary V. and Gupta B.D., 2015. Surface Plasmon Resonance-Based Fiber
Optic Methane Gas Sensor Utilizing Graphene-Carbon Nanotubes-Poly (Methyl Methacrylate) Hybrid
Nanocomposite, Plasmonics, 10(5), 1147-1157.
660