Applications of laser diagnostics to thermal power plants and engines

Accepted Manuscript Applications of laser diagnostics to thermal power plants and engines Y. Deguchi , T. Kamimoto , Z.Z. Wang , J.J. Yan , J.P. Liu , Hiroaki Watanabe , Ryoichi Kurose PII:

S1359-4311(14)00438-4

DOI:

10.1016/j.applthermaleng.2014.05.063

Reference:

ATE 5665

To appear in:

Applied Thermal Engineering

Received Date: 17 January 2014 Revised Date:

26 April 2014

Accepted Date: 7 May 2014

Please cite this article as: Y. Deguchi, T. Kamimoto, Z.Z. Wang, J.J. Yan, J.P. Liu, H. Watanabe, R. Kurose, Applications of laser diagnostics to thermal power plants and engines, Applied Thermal Engineering (2014), doi: 10.1016/j.applthermaleng.2014.05.063. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Title page

Applications of laser diagnostics to thermal power plants and engines

RI PT

Y. Deguchia,*, T. Kamimotoa, Z.Z. Wanga,b, J.J. Yanb, J.P. Liub, Hiroaki Watanabec, Ryoichi Kurosed a

Graduate School of Advanced Technology and Science, The University of Tokushima, Tokushima 770-8501, Japan b

State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University, Xi’an 710049, China

Energy Engineering Research Laboratory, Central Research Institute of Electric Power Industry, Kanagawa, 240-0196, Japan d

*

Department of Mechanical Engineering and Science, Kyoto University, Nishikyo-ku, Kyoto 615-8540, Japan

Corresponding author: Yoshihiro Deguchi

SC

c

M AN U

Graduate School of Advanced Technology and Science, The University of Tokushima TEL: (+81)-88-656-7375 FAX: (+81)-88-656-9082

Email address:[email protected]

Highlights

TE D

Postal address: 2-1, Minamijyosanjima, Tokushima, 770-8506 Japan

EP

1. The applicability of advanced laser diagnostics has been demonstrated for the improvement of thermal power plants and engines.

AC C

2. Time resolved 2D temperature and NH3 concentration distributions were measured in engine exhausts using the developed method of CT-TDLAS. 3. The real time measurement method of fly ash contents according to the particle diameter was developed using LIBS for the thermal power plant application. 4. The sensitive trace species measurement method was developed for Hg using low pressure LIBS and LB-TOFMS with enhanced detection limit.

1

ACCEPTED MANUSCRIPT

Abstract

M AN U

SC

RI PT

The demands for lowering the burdens on the environment will continue to grow steadily. It is important to monitor controlling factors in order to improve the operation of industrial thermal systems. In engines, exhaust gas temperature and concentration distributions are important factors in nitrogen oxides (NOx), total hydrocarbon (THC) and particulate matter (PM) emissions. Coal and fly ash contents are parameters which can be used for the control of coalfired thermal power plants. Monitoring of heavy metals such as Hg is also important for pollution control. In this study, the improved laser measurement techniques using computed tomography-tunable diode laser absorption spectroscopy (CT-TDLAS), low pressure laserinduced breakdown spectroscopy (LIBS), and laser breakdown time-of-flight mass spectrometry (LB-TOFMS) have been developed and applied to measure 2D temperature and species concentrations in engine exhausts, coal and fly ash contents, and trace species measurement. The 2D temperature and NH3 concentration distributions in engine exhausts were successfully measured using CT-TDLAS. The elemental contents of size-segregated particles were measured and the signal stability increased using LIBS with the temperature correction method. The detection limit of trace species measurement was enhanced using low pressure LIBS and LBTOFMS. The detection limit of Hg can be enhanced to 3.5 ppb when employing N2 as the buffer gas using low pressure LIBS. Hg detection limit was about 0.82 ppb using 35 ps LB-TOFMS. Compared to conventional measurement methods laser diagnostics has high sensitivity, high response and non-contact features for actual industrial systems. With these engineering developments, transient phenomena such as start-ups in thermal systems can be evaluated to improve the efficiency of these thermal processes.

AC C

EP

TE D

Keywords: Temperature and concentration measurement; Engine; Thermal power plant; Computed tomography-tunable diode laser absorption spectroscopy; Laser-induced breakdown spectroscopy; Time-of-flight mass spectrometry

2

ACCEPTED MANUSCRIPT

1 Introduction

AC C

EP

TE D

M AN U

SC

RI PT

Recent years have seen tighter regulation of harmful substances such as NOx, CO, particulates, and heavy metals in several types of commercial plants using combustion processes, including engines, boilers, gas turbines and so on. These pollutants cause the serious problems concerning the environment and human health [1-3]. It is necessary to make efforts to protect natural ecosystems and effectively utilize fossil fuels in various fields. Therefore, the understanding of the reaction mechanism of combustion is becoming more important to minimize environmental disruption and to improve the efficiency of combustors. In particular, detailed measurement techniques for temperature and species concentrations are necessary to elucidate the overall nature of industrial combustion systems. There are a number of standard methods to detect these parameters, such as well-known “Industrial Standards”. These standard methods are well established and easily accessible, although they are limited in terms of meeting the industrial needs described above because of slow response, low sensitivity, complicated pre-concentration, etc. In contrast, laser diagnostics makes it possible to monitor these parameters due to their fast response, high sensitivity, and non-contact features [4]. For example, laser-induced fluorescence (LIF) [5-7], tunable diode laser absorption spectroscopy (TDLAS) [8-10], laser-induced breakdown spectroscopy (LIBS) [11-13], and time-of-flight mass spectrometry (TOFMS) [14-16] have been widely used in various applications to meet the practical industrial requirements noted above. Temperature and concentration distributions play an important role for the combustion structure and the combustor efficiency in engines, burners, gas turbines and so on. In engines, exhaust gas temperature distribution and species concentrations are important factors in NOx, THC and PM emissions, as Fig. 1 shows. They are also catalytically important parameters in both gasoline and diesel engines. 2D temperature distribution plays an important role for the catalytic efficiency. A thermocouple, which has been widely used as a temperature measurement device, is intrinsically a point measurement method. Non-contact 2D temperature distribution cannot be attained by thermocouples. The development of these methods is important to meet the needs of these applications. In thermal power plants, the boiler control system is important for the adjustment of combustion process, as shown in Fig. 2. 2D temperature and concentration distributions also play an important role for the efficiency. The coal and fly ash contents, especially unburned carbon in fly ash, are important factors for efficient combustion. As is well known, fly ash produced during coal combustion is one of the sources of fine particles. The compositions of fly ash, which contains Si, Al, Fe, Ca, unburned carbon and other materials, are complex and highly depending on the coal quality and combustion procedure. Recently, LIBS technique has been applied to coal combustion and other industrial fields because of the fast response, high sensitivity, real-time and non-contact features. LIBS has been applied to the detection of unburned carbon in fly ash, char and pulverized coal under high-pressure and high-temperature conditions. This automated LIBS apparatus were also applied in a boiler-control system of a power plant with the objective of achieving optimal and stable combustion [17,18], which enabled real-time measurement of unburned carbon in fly ash as shown in Fig. 2 (a) [18]. This apparatus featured a detection time capability of less than 1 minute, which showed better real-time capability compared to other methods (preparation of samples) used in other studies [19-21]. On the other hand, there is an urgent need to evaluate the heavy metal emission for the purpose of reducing and eliminating heavy metal influence. For example, coal combustion in power plant constitutes the large share of global anthropogenic Hg emission. The harmful substances, such as heavy metals, are produced during the combustion processes in engines,

3

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

Fig. 1 Engine system

RI PT

boilers, and other combustors. The rapid and precise measurement of trace species with variable compositions is imperative according to different applications.

AC C

EP

(a) Measurement requirement

(b) Advanced control system by the real-time measurement of unburned carbon in fly ash Fig. 2 System of coal-fired power plant

4

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

The aim of current research is the applications of newly developed laser diagnostics to thermal power plants and engines. In this study, the 2D temperature and concentration distributions, the contents of particles, and the trace heavy metals in exhausts were measured using computed tomography TDLAS (CT-TDLAS), low pressure LIBS and laser breakdown TOFMS (LBTOFMS) methods. CT-TDLAS is based on a computed tomography method using absorption spectra of molecules such as H2O and NH3. Using CT-TDLAS it becomes possible to measure non-contact and fast response 2D temperature and concentration distributions, which cannot be measured by conventional TDLAS methods. The method was applied to engine exhausts to measure 2D temperature and NH3 concentration distributions. NH3 is an important species for NOx removal equipment in both thermal power plants and engines. The temporal and spatial resolutions of this method have also been discussed to demonstrate its applicability to various types of combustors. An advanced method has been demonstrated with regard to further applications of particle measurement. This research focused on the quantitative composition measurement of fly ash. Particles were classified and then measured using LIBS technique with the temperature correction method. The trace species in exhausts were also measured using low pressure LIBS and LB-TOFMS methods with the features of increased sensitivity and rapid analysis compared to the normal LIBS and TOFMS methods. The detection limit of measured species using TOFMS is often ppb or less [22,23]. The applicability of these methods to thermal power plants was also discussed.

2 Theory

2.1 CT-tunable diode laser absorption spectroscopy (CT-TDLAS)

AC C

EP

TE D

Gas temperature and species concentrations can be determined by measuring molecular absorbance at multiple wavelengths. Tunable diode laser absorption spectroscopy was used in this research. The principle of TDLAS is based on Lambert Beer's law. When light permeates an absorption medium, the strength of the permeated light is related to absorber concentration according to Lambert Beer's law. TDLAS uses this basic law to measure temperature and species concentrations. The number density of the measured species is related to the amount of light absorbed as in the following formula [4].     (1) I λ / I λ 0 = exp {− Aλ } = exp −∑  n(i ) L∑ Si , j (T ) GVi , j   j    i  Here, Iλ0 is the incident light intensity, Iλ is the transmitted light intensity, Aλ is the absorbance, ni is the number density of species i, L is the path length, Si,j is the temperature dependent absorption line strength of the absorption line j, and Gvi,j is the line broadening function. Temperature can be measured to evaluate several absorption lines which have different temperature dependence. In this study, 16 optical paths were intersected to each other to form analysis grids, reconstructing the 2D temperature and concentration distributions by a computed tomography method [6,7]. A set of measured H2O and NH3 absorption spectra was compared to theoretical spectra to minimize the mean squared errors. 2D temperature and NH3 concentration were determined using a polynomial noise reduction technique [4]. 2.2 Laser-induced breakdown spectroscopy (LIBS)

Creation and cooling processes of plasma in LIBS can be described as follows. In the generation of plasma, the core of plasma is firstly produced by absorption of incident laser energy, such as multi-photon ionization in solids, liquids, or gases. The creation of the plasma 5

ACCEPTED MANUSCRIPT

SC

RI PT

core induces the rapid growth of plasma through the absorption of the laser light by electrons and the electron impact ionization process in it, that is, the inverse bremsstrahlung process. After the termination of the laser pulse, the plasma continues expanding because of its high temperature and pressure gradients compared with ambient conditions. At the same time, recombination of electrons and ions proceeds due to the collision process and temperature decreases gradually compared to the plasma generation process. Therefore the continuum emission is released by bremsstrahlung and recombination processes in the optically thin plasma. LIBS signals arise in the plasma cooling period [4]. The continuum emission is considered as one of the interferences to LIBS signals. Despite the fact that the processes involved in LIBS are complex, the emission intensity from the atomized species during the cooling process is mostly examined by the following equation with the assumption of a uniform plasma temperature:   E(i ), j   (2) I (i ) = n(i ) ∑  K (i ), j g (i ), j exp  −  k T   j  B    

TE D

M AN U

In the above expression, I(i) is the emission intensity of species i, ni is the number density of species i, K(i),j is a variable that includes the Einstein A coefficient from the upper energy level j, g(i),j is the statistical weight of species i at the upper energy level j, E(i).j is the upper level energy of species i, kB is the Boltzmann constant and T is the plasma temperature. Eq. (2) is applicable under the conditions of local thermodynamic equilibrium (LTE). In this study the low pressure and short pulse LIBS technique was developed and applied to enhance the detection limit. The interference of the continuum emission from plasma itself decreases dramatically at reduced pressure. Because the collisional and plasma quenching processes are not significant under reduced pressure conditions, stable and longer existing plasma is formed through the plasma expansion, which makes the signal detection much easier with low interference. 2.3 Laser breakdown time-of-flight mass spectrometry (LB-TOFMS)

EP

In TOFMS system, a measurement sample is introduced into a vacuum chamber and it is atomized and ionized by laser irradiation. The electric field potential is simultaneously applied to acceleration of ions. The accelerated ions enter the drift region with no potential difference and undergo uniform motion. An ion detector records the signals of ionized species and ion counter takes over to digitize and display the results. Due to the law of energy conservation, the ions’ electric field potential is equivalent to their kinetic energy. The following formula is established by the energy conservation law [4]: 2

AC C

m L (3)   2t  In the above expression, z is the ionic valence, V is the acceleration electric field potential, m is the ion mass, L is the ion distance of flight, t is the time of flight. TOFMS distinguishes the ions of different atoms or molecules based on their arrival time to ion detector. The following relation between the ion mass and the time of flight can be expressed from Eq. (3). 2zV m = 2 t2 (4) L The laser breakdown process of gas phase materials consists of laser dissociation, multiphoton ionization, electron impact ionization and so on. At high pressure (atmosphere pressure), the electron diffusion and electron impact ionization processes are the major sources of the plasma generation. When reducing the pressure (a few kPa), the effect of electron diffusion and zV =

6

ACCEPTED MANUSCRIPT

electron impact ionization decreases compared with that under high pressure condition, which usually occurs in LIBS process. The multi-photon ionization process becomes the remarkable influence at low pressure, especially less than 1 Pa. For example, in the case of LB-TOFMS, the plasma mainly generates during the multi-photon ionization process. The laser dissociation and multi-photon ionization contribute to the ion signal detection.

RI PT

3 Experimental apparatus 3.1 2D temperature and concentration measurement

AC C

EP

TE D

M AN U

SC

Fig. 3 shows the outline of an experimental apparatus used in this study. DFB lasers (NTT Electronics Co., NLK1E5GAAA) at 1388 nm and 1512 nm were used to measure water vapor (temperature and H2O concentration) and NH3, respectively. The laser wavelength was scanned at 1-3 kHz and the absorption spectra were measured to calculate the instant 2D temperature and NH3 concentration using a 16 path measurement cell as shown in Fig. 3(a). The laser beam was separated by fiber splitter (OPNETI CO., SMF-28e 1310 nm SWBC 1×16) and the separated laser beams were irradiated into target gas by 16 collimators (THORLABS Co., 50-1310-APC). The transmitted light intensities were detected by photodiodes (Hamamatsu Photonics and G8370-01), and taken into the computer (HIOKI E.E. Co., 8861 Memory Highcoda HD Analog16). Temperature was also measured by chromel-alumel thermocouples with a diameter of 100 µm (KMT-100-100-120). The experiment was performed using gasoline engine (FUJI HEAVY INDUSTRIES, Inc., EX13) as shown in Fig. 3(b). The laser paths were set at the outlet of the engine exhaust pipes. The diameter of 16 path measurement cell was 70 mm. The diameter of the engine exhaust pipe was 22-40 mm. The exhaust pipe length was 160 mm and it had an inner pipe with diameter of 8 mm to input NH3 into the exhausts. Evaluation of temperature and NH3 concentration measurements was also performed using a single laser path setup. A CH4-Air flat flame burner with an inner diameter of 40 mm was used for the temperature measurement evaluation. Temperature of the flame was controlled by placing stainless steel meshes on in flat flames. A NH3 flow cell with 220 mm path length was used for the NH3 concentration measurement evaluation. NH3 concentration was controlled by mass flowmeters by mixing the 2% NH3 standard gas and N2.

(a) 16 path measurement cell

(b) Engine experiment

Fig. 3 Experimental apparatus of CT-TDLAS

7

ACCEPTED MANUSCRIPT

3.2 Content measurement of particles

TE D

M AN U

SC

RI PT

The objective of this experiment is to detect the classified particles according to the diameter. Fig. 4 illustrates the experimental apparatus used in this study. The experimental set-up composed of laser, beam focusing system, detection system, separator and auxiliary device. The beam from a Q-switched Nd:YAG laser (LOTIS TII, LS-2137U, energy stability: 2.5%, beam diameter: 8 mm) operating at 1064 nm with 6-8 ns pulse width was focused into the measurement area using the lens with focus length of 200 mm. The measurement chamber was a vacuum cell with four quartz windows and its internal volume was about 200 cm3. The emission signal was collected using an optical fiber and detected using a spectrometer (JASCOCT-10S) and an ICCD camera (iStar 334T Series, Andor). The particles of fly ash and coal (Newlands Coal) in powder form with a range of elemental composition were provided by Central Research Institute of Electric Power Industry (CRIEPI, Japan), which were measured here as practical samples.

EP

Fig. 4 Measurement of particles using LIBS

3.3 Trace species measurement

AC C

The trace species of Hg and HgCl2 were measured using low pressure LIBS and LB-TOFMS in this study. The schematic diagram of experimental apparatus including input and detection systems is shown in Fig. 5(a). The gaseous mixtures of trace species with air or N2 were fed into the chamber from the constant-temperature bath. The vaporizing temperature of Hg and HgCl2 was 370 K according to their vaporizing pressure. Hg in combustion gas from the burner was also measured. The concentration of Hg was determined using a mercury gas detector tube (Komyo Rikagaku Kogyo K.K. No. 142S) by sampling the outlet gas. The three lasers with different pulse width operated at 1064 nm were used in this study, including nanosecond laser (Quantel Brilliant b, 6 ns, 10 Hz, beam diameter: 9 mm), picosecond laser 1 (EKSPLA SL312, 150 ps, 10 Hz, beam diameter: 10 mm) and picosecond laser 2 (Quantel YG901C-10, 35 ps, 10 Hz, beam diameter: 9.5 mm). The low pressure LIBS detection system for trace species measurement is illustrated in Fig. 5(b). The apparatus fundamentally consisted of lasers, a vacuum chamber, lens, a spectrometer,

8

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

an ICCD camera and auxiliary equipment. The LB-TOFMS detection system is shown in Fig. 5(c), consisting of lasers, a jet chamber, a test chamber, a flight chamber (R.M. Jordan Co., D850 AREF; drift tube length: 500 mm), an ion detector (R.M. Jordan Co., 40 mm MCP Z-gap detector), an ion counter (SRS, Model SR430 multi-channel scaler) and auxiliary equipment. For the measurement of sample with heavy mass, the jet chamber was used to ensure heavy materials go straight ahead to the test chamber. The temperature of gas inlet and outlet pipes was controlled at 423 K. The test and the flight chambers were equipped with two turbo pumps (Pfeiffer vacuum, MVP 055-3). The pressure in the test chamber was used as an indicator under different experimental conditions.

AC C

EP

TE D

(a) Schematic diagram of experimental system

(b) Detection system of low pressure LIBS

(c) Detection system of LB-TOFMS

Fig. 5 Experimental system of trace species measurement

9

ACCEPTED MANUSCRIPT

4 Results and discussion 4.1 2D temperature and concentration measurement results in engine using CT-TDLAS

TE D

M AN U

SC

RI PT

2D temperature and concentration measurement is important for both thermal power plant and engine systems. The required areas for these applications are different and they are approximately 1-20 m for thermal power plants and 50-200 mm for engines. TDLAS uses the absorption phenomena and the large scale measurement is also possible for CT-TDLAS. In this study CT-TDLAS was applied to engine exhausts to measure 2D temperature and NH3 concentration. Fig. 6(a) shows the temperature measurement evaluation result using the flat flame burner. Temperatures measured by TDLAS and the thermocouple show good agreement with each other. Exhaust gas temperature distribution for 5 seconds was measured using the 16 path measurement cell. Fig. 6(b) shows the temperature history measured by thermocouple at the center point of the CT measurement cell. Fig. 6 (c) and (d) show 2D temperature measurement results at 0 s and 4 s in engine exhausts using the 16 path measurement cell, respectively. The

(b) Engine revolving speed and temperature history measured by thermocouple

AC C

EP

(a) Temperature measurement evaluation result using flat flame burner

(c) Temperature distribution at t = 0 s

(d) Temperature distribution at t = 4 s

10

ACCEPTED MANUSCRIPT

Fig. 6 2D Temperature measurement results in engine exhausts using 16 path measurement cell

M AN U

SC

RI PT

spatial resolution of CT-TDLAS can be easily improved to 2-3 mm by adding laser paths to the measurement area. Fig. 7(a) shows the measurement evaluation result of NH3 concentration using the NH3 flow cell. Measured NH3 concentration using TDLAS shows the linear relation to NH3 concentration controlled by mass flowmeters. Fig. 7(b)-(j) show the 2D NH3 measurement results in engine exhausts. NH3 was mixed into the exhaust gas during t= 0.6-5.9 s and 6.2-10 s. It is demonstrated that the NH3 distribution has been successfully reconstructed and the time resolved 2D concentration measurement becomes possible using the developed CT-TDLAS measurement method. The temporal resolution can reach to 1 ms or less by controlling the scanning rate of laser wavelength.

(b) Time history of NH3 absorption intensity at the center laser path

AC C

EP

TE D

(a) NH3 concentration measurement evaluation result using flat flame burner

(c) 0 s

(d) 1 s

(e) 2 s

11

(f) 3 s

ACCEPTED MANUSCRIPT

(g) 4 s

(h) 5 s

(i) 6 s

(j) 7 s

Fig. 7 Time history of NH3 concentration between 0-7 s 4.2 Measurement results of fly ash and coal

TE D

M AN U

SC

RI PT

In the thermal power plant, the contents of coal and fly ash are the important factors. Therefore, the coal and fly ash were measured using LIBS in this study. In order to acquire the precise quantitative results, the temperature correction method was employed [17,24]. The fly ash and coal samples were measured to set up the plasma temperature correction factors under different experimental conditions, such as detection delay time. Fig. 8 shows LIBS spectra of fly ash. These temperature correction factors were measured in each experimental condition such as fly ash and coal. The typical temperature correction curve of IC/ISi using the unclassified fly ash sample is shown in Fig. 9. It is clear from Fig. 9 the ratio is influenced by the plasma temperature, which is directly related to the ratio IMg1/IMg2. Fig. 10 shows the comparison of uncorrected and corrected IC/ISi. By applying the temperature correction scheme, the fluctuation of IC/ISi became small and the standard deviation of IC/ISi (σ=14%) became half compared to that of uncorrected IC/ISi i (σ=28%), as shown in Fig. 10(b). Using this set of experimentally determined correction factors, the fluctuation of the concentration ratios such as Fe2O3/SiO2,

Fig. 9 IC/ISi correction curve of plasma temperature

AC C

EP

Fig. 8 LIBS spectra of fly ash

(a) Uncorrected IC/ISi

(b) Corrected IC/ISi

12

ACCEPTED MANUSCRIPT

Fig. 10 Comparison of uncorrected and corrected IC/ISi

TE D

M AN U

SC

RI PT

Al2O3/SiO2, CaO/SiO2, C/SiO2, was greatly reduced. Unburned carbon in fly ash (C content) can be evaluated down to less than 1%, which can satisfy the requirements in most pulverized coal fired boilers. The detection limit (S/N= 1) of Fe, Al, and Ca can be estimated to be 0.1%, 0.3%, and 0.02%, respectively. Consistent results have also been acquired in the case of coal measurement. With temperature correction, signals became much more stable than those without plasma temperature correction. Fig. 11 shows real-time measurement results of fly ash from stage 0 to stage 3 concerning the particle diameter. The diameter of particles reduced from stage 0 to stage 3. The particles from the pipe in stage 0 were fly ash samples that had not been separated yet. At different stages, concentration ratio of Fe2O3/SiO2 unchanged and was stable. On the other hand, concentration ratios of C/SiO2, Al2O3/SiO2 and CaO/SiO2 decreased slightly by 2.7%, 4.2% and 0.3% according to the stage number, i.e. the decrease of particle diameter. The smaller particles combust much more adequately, resulting in lower carbon concentration in smaller particles. Because the boiling points of Ca and Al are lower than that of Fe, concentration ratios of Al2O3/SiO2 and CaO/SiO2 also decreased during coal combustion, which were consistent with

(b) Fe2O3/SiO2

AC C

EP

(a) C/SiO2

(c) Al2O3/SiO2

(d) CaO/SiO2

13

ACCEPTED MANUSCRIPT

Fig. 11 Real-time LIBS results of concentration ratio in fly ash in different stages

4.3 Measurement results of trace species

SC

RI PT

chemical analysis results. The standard deviations at stage 0 for C/SiO2, Fe2O3/SiO2, Al2O3/SiO2 and CaO/SiO2 were 9.9%, 10%, 5.9% and 29% respectively, and 17%, 6.7%, 7.7% and 10% at stage 2. The inhomogeneity of fly ash samples mainly caused the fluctuation of each measurement. Variation of carbon content was apparent according to particle diameter. Furthermore, carbon content fluctuated broadly at the same stage and different measurement time because the content of carbon in fly ash was different from particle to particle. It is also demonstrated the dependence of elemental concentration on the diameter can be measured even if elemental distribution in fly ash is very complex. From this perspective, the experimental system and method can be applied to online measurement in real power plant. Variation of calcium content was also measured though the low calcium concentration in fly ash. The coal was also measured using the same method, which were consistent with chemical analysis results.

M AN U

The exhausts releasing from the thermal power plants and some engines usually contain various harmful substances, such as heavy metals. The purpose of this study is to measure the trace species of Hg, As, Cd, Se using low pressure LIBS and LB-TOFMS methods in real applications. The specific measured items of these trace species including wavelength and mass are list in Table 1. In this paper, the measurement results of Hg and HgCl2 using low pressure LIBS and LB-TOFMS were presented. Table 1 Specific measurement conditions of LIBS and LB-TOFMS LIBS Wavelength (nm) 253.7 278.2 228.8 204.0

TE D

Element Hg As Cd Se

LB-TOFMS Mass (amu) 200 75 112 79

AC C

EP

The emission line of Hg atom at 253.7 nm was difficult to distinguish from the spectra under the high pressure condition (atmosphere pressure) because of serious background interference of continuum emission from plasma itself. Low pressure laser-induced plasma was employed to minimize the background interference and enhance the detection limit. Fig. 12 shows the measurement results of Hg and HgCl2 in buffer gas of N2 at reduced pressure of 2600 Pa. The background interference of continuum emission reduced dramatically at lower buffer gas pressure in both cases. Hg and HgCl2 were also measured at low pressure when employing air as the buffer gas. The Hg signal intensity became worse compared with that in buffer gas of N2. Hg signal submerged in the background due to the quenching of Hg signal in O2. Considering the laser-induced plasma process, another strategy for the enhancement of detection limit is the laser pulse width. The electron impact ionization process can be controlled when employing the short pulse width lasers. Different pulse width lasers have been employed to measure Hg under low pressure conditions. Hg atom signal became clearly distinguishable in the case of picosecond breakdown when employing air as the buffer gas.

14

RI PT

ACCEPTED MANUSCRIPT

(a) Hg measurement result

(b) HgCl2 measurement result

SC

Fig. 12 LIBS spectra of Hg and HgCl2 in buffer gas of N2 at 2600 Pa

AC C

EP

TE D

M AN U

In order to apply this method for the real applications, Hg was measured in combustion gas sampling from a burner [25], as shown in Fig. 5(a). Fig. 13 illustrates the comparison of Hg in air and combustion gas at lower pressure of 700 Pa. the interferences from coexisting molecular and atomic emissions and the continuum emission were reduced in combustion gas. This may have occurred because of the different air and combustion gas breakdown processes. The particles existing in combustion gas absorbed the laser light to break down samples much more efficiently compared with that employing air as coexisting material. These findings suggest that the detection ability is relevant to the coexistent atoms and molecules. It is demonstrated that the measurement of trace Hg in combustion gas is practicable using low pressure laser-induced plasma, which can be applied in real applications.

(a) Hg in air at pressure of 700 Pa

(b) Hg in combustion gas at pressure of 700 Pa

Fig. 13 Comparison of Hg in air and combustion gas

LIBS detection limit can be enhanced by reducing the pressure and using short pulse width laser. The detailed discussion was reported in elsewhere [26]. The gaseous mixture of Hg and air with different Hg concentration was measured at pressure of 2000 Pa. The measurement result of concentration dependence shows the linear growth, as shown in Figure 14. Employing air as the buffer gas, Hg detection limit of 600 shots (1 min) was estimated by subtracting the interference signal from the measured LIBS spectra and it was calculated by evaluating the ratio of the slope of the Hg calibration curve (ms) to the background noise (standard deviation: σ) around 253.7 nm. Hg detection limit of nanosecond breakdown was 450 ppb (3σ/ms) at pressure of 700 Pa. According to the enhancement of picosecond breakdown at low pressure, the detection limit was

15

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

evaluated to be 30 ppb (3σ/ms) in picosecond (35 ps) breakdown at pressure of 700 Pa. In buffer gas of N2, the detection limit cannot be enhanced using short pulse width laser because of the different interference. Therefore, the detection limit of Hg in N2 was 3.5 ppb (3σ/ms) at 6600 Pa employing nanosecond breakdown, which is much better than that of Hg in air employing picosecond breakdown.

Fig. 14 Concentration dependence of Hg emission signal using LIBS

AC C

EP

TE D

The detection limit of measured species using TOFMS is often ppb or less. In order to enhance the detection limit of trace species measurement in some applications, LB-TOFMS has also been employed to measure Hg and HgCl2. In practice, the samples of heavy metals are usually enclosed with complex materials such as hydrocarbons or as a form of compounds. It is necessary to distinguish the target signal from mixture without any interference. The method of laser breakdown time-of-flight mass spectrometry (LB-TOFMS) was employed to eliminate or minimize the fragmentation interference of the mixture. Fig. 15 shows the measurement results of hydrocarbon mixture of p-C7H6Cl2, C7H8, C6H5C2H3, p-C8H10, p-C6H4(C2H5)2 and C6H3(CH3)3 using 266 nm and 1064 nm breakdowns. LB-TOFMS using 266 nm breakdown can produce partial fragmentation and daughter ions, especially for large fragile molecules which will be pre-dissociated using 266 nm breakdown. Compared to the result of 266 nm breakdown with a lot of partial fragmentation and daughter ions, the result of 1064 nm breakdown is very distinct without any interference in the mass region of 30-300 m/z because of the complete fragmentation, such as H, C, CH, C2 and N2. The target signals, such as As, Hg and so on, can be distinguished without any interference using 1064 nm breakdown, as well as 532 nm breakdown.

(a) 266 nm breakdown

(b) 1064 nm breakdown

Fig. 15 Laser breakdown mass spectra of hydrocarbons using different wavelengths

16

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

RI PT

The trace species of Hg and HgCl2 were also measured using LB-TOFMS method. Fig. 16 shows the measurement result of Hg in hydrocarbon using 1064 nm breakdown. Hg+ signal can be detected without partial fragmentation and daughter ions of hydrocarbon because of the laser breakdown process. In the particles of fly ash, the clear Hg+ signal can also be measured. In both cases of 1064 nm and 532 nm breakdowns, Hg+ signal from mixtures was revealed without the interference of fragmentation in the mass region of 30-300 m/z. HgCl2 was also measured using 1064 nm and 532 nm breakdowns under different conditions. The clear Hg+ signal can be detected, which shows the consistent results with Hg measurement. There are several Hg+ signal lines, such as 198, 199, 200, 201, 202 and 204 amu, representing the isotopes of Hg, as shown in Fig. 17. The element can be measured precisely, as the method depends on the mass of atom and molecule with high sensitivity. According to the measurement results under different conditions, the detection ability can also be enhanced when employing short pulse width laser in LB-TOFMS, such as 35 ps laser. Fig. 18 shows the concentration dependence of Hg+ signal intensity, which was directly proportional to the concentration. The detection limit of Hg+ signal using 35 ps laser breakdown was about 0.82 ppb (3σ/ms), which was enhanced 4 times compared with LIBS results.

Fig. 17 Measured result of Hg isotopes

AC C

EP

Fig. 16 Measured result of Hg in hydrocarbon

Fig. 18 Concentration dependence of Hg+ mass signal using LB-TOFMS

5 Conclusions In this study, CT-tunable diode laser absorption spectroscopy (CT-TDLAS), improved laserinduced breakdown spectroscopy (LIBS), and laser breakdown time-of-flight mass spectrometry

17

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

(LB-TOFMS) have been applied to measure 2D temperature and concentrations in engine exhausts, coal and fly ash contents, and trace species of Hg and HgCl2. (1) The 2D temperature and concentration measurement method using CT-TDLAS was developed and successfully demonstrated to measure 2D temperature and NH3 concentration distributions in engine exhausts using the 16 path measurement cell. CT-TDLAS has a potential of the kHz response time and the method enables the real-time 2D temperature and species concentration measurement to be applicable in various fields such as engines and gas turbine combustors. (2) LIBS technique was applied to measure contents of size-segregated particles depending on particle diameter. The plasma temperature correction method was introduced to the sizesegregated fly ash and pulverized coal to detect their quantitative content information. With temperature correction, signal stability has been significantly improved. Acquired results successfully clarified the content dependence on particle diameter. (3) The trace species of Hg and HgCl2 were measured using low pressure LIBS and LBTOFMS. Employing low pressure LIBS technique, Hg detection limit was evaluated to be 30 ppb (3σ/ms) in picosecond (35 ps) breakdown in buffer gas of air. The detection limit of Hg can be enhanced to 3.5 ppb (3σ/ms) when employing N2 as the buffer gas. LB-TOFMS method was also developed and applied to detect Hg in mixture without the interference of partial fragmentations. The detection limit of Hg+ signal using 35 ps laser breakdown were about 0.82 ppb (3σ/ms). References

AC C

EP

TE D

[1] W.P. Linak, J.I. Yoo, S.J. Wasson, W.Y. Zhu, J.O.L. Wendt, F.E. Huggins, Y.Z. Chen, N. Shah, G.P. Huffman, M.I. Gilmour, Ultrafine ash aerosols from coal combustion: Characterization and health effects, P. Combust. Inst. 31 (2007) 1929-1937. [2] D.J. Swaine, Why trace elements are important, Fuel Processing Technology Fuel Process. Technol. 65-66 (2000) 21-33. [3] G. Oberdorster, E. Oberdorster, J. Oberdorster, Nanotoxicology: an emerging discipline evolving from studies of ultrafine particles, Environ. Health Persp. 113 (2005) 823-839. [4] Y. Deguchi, Industrial Applications of Laser Diagnostics, CRS Press, Taylor & Francis, New York, 2011. [5] S. Einecke, C. Schultz, V. Sick, Measurement of temperature, fuel concentration and equivalence ratio fields using tracer LIF in IC engine combustion, Appl. Phys. B: Lasers Opt. 71 (2000) 717-723. [6] T. Kim, J.B. Ghandhi, Investigation of light load HCCI combustion using formaldehyde planar laser-induced fluorescence, Proc. Combust. Inst. 30 (2005) 2675-2682. [7] M. Löffler, F. Beyrau, A. Leipertz, Acetone laser-induced fluorescence behavior for the simultaneous quantification of temperature and residual gas distribution in fired spark-ignition engines, Appl. Opt. 49 (2010), 37-49. [8] M. Yamakage, K. Muta, Y. Deguchi, S. Fukada, T. Iwase, T. Yoshida, Development of direct and fast response exhaust gas measurement, SAE Paper 20081298, 2008. [9] P. Wright, N. Terzijaa, J.L. Davidsona, S. Garcia-Castillo, C. Garcia-Stewart, S. Pegrumb, S. Colbourneb, P. Turnerb, S.D. Crossleyc, T. Litt, S. Murrayc, K.B. Ozanyana, H. McCanna, High-speed chemical species tomography in a multi-cylinder automotive engine, Chem. Eng. J. 158 (2010) 2-10.

18

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[10] Y. Deguchi, D. Yasui, A. Adachi, Development of 2D temperature and concentration measurement method using tunable diode laser absorption spectroscopy, Journal of Mechanics Engineering and Automation, 2 (2012) 543-549. [11] R. Noll, I. Mönch, O. Klein, A. Lamott, Concept and operating performance of inspection machines for industrial use based on laser-induced breakdown spectroscopy, Spectrochim. Acta Part B 60 (2005) 1070-1075. [12] Gaft M., Sapir-Sofer I., Modiano H., Stana R., Laser induced breakdown spectroscopy for bulk minerals online analyses, Spectrochim. Acta Part B 62 (2007) 1496-1503. [13] F. Boué-Bigne, Laser-induced breakdown spectroscopy applications in the steel industry: rapid analysis of segregation and decarburization, Spectrochim. Acta Part B 63 (2008) 11221129. [14] C.M. Gittins, M.J. Castaldi, S.M. Senkan, E.A. Rohlfing, Real-time quantitative analysis of combustion-generated polycyclic aromatic hydrocarbons by resonance-enhanced multiphoton ionization time-of-flight mass spectrometry, Anal. Chem. 69 (1997) 286-293. [15] H.J. Heger, R. Zimmermann, R. Dorfner, M. Beckmann, H. Griebel, A. Kettrup, U. Boesl, On-line emission analysis of polycyclic aromatic hydrocarbons down to pptv concentration levels in the flue gas of an incineration pilot plant with a mobile resonance-enhanced multiphoton ionization time-of-flight mass spectrometer, Anal. Chem. 71 (1999) 46-57. [16] M. Bente, M. Sklorz, T. Streibel, R. Zimmermann, Thermal desorption-multiphoton ionization time-of-flight mass spectrometry of individual aerosol particles: a simplified approach for online single-particle analysis of polycyclic aromatic hydrocarbons and their derivatives, Anal. Chem. 81 (2009) 2525-2536. [17] M. Noda, Y. Deguchi, S. Iwasaki, N. Yoshikawa, Detection of carbon content in a hightemperature and high-pressure environment using laser-induced breakdown spectroscopy, Spectrochim. Acta Part B 57 (2002) 701-709. [18] M. Kurihara, K. Ikeda, Y. Izawa, Y. Deguchi, H. Tarui, Optimal boiler control through realtime monitoring of unburned carbon in fly ash by laser-induced breakdown spectroscopy, Appl. Optics 42 (2003) 6159-6165. [19] M.P. Mateo, G. Nicolas, A. Yañez, Characterization of inorganic species in coal by laserinduced breakdown spectroscopy using UV and IR radiations, Appl. Surf. Sci. 254 (2007) 868-872. [20] T.B. Yuan, Z. Wang, L.Z. Li, Z.Y. Hou, Z. Li, W.D. Ni, Quantitative carbon measurement in anthracite using laser-induced breakdown spectroscopy with binder, Appl. Optics 51 (2012) B22-B29. [21] T. Ctvrtnickova, M.P. Mateo, A. Yañez, G. Nicolas, Characterization of coal fly ash components by laser-induced breakdown spectroscopy, Spectrochim. Acta Part B 64 (2009) 1093-1097. [22] B.K. Gullett, A. Touati, L. Oudejans, S.P. Ryan, Real-time emission characterization of organic air toxic pollutants during steady state and transient operation of a medium duty diesel engine, Atmos. Environ. 40 (2006) 4037-4047. [23] Y. Deguchi, N. Tanaka, M. Tsuzaki, A. Fushimi, S. Kobayashi, K. Tanabe, Detection of components in nanoparticles by resonant ionization and laser breakdown time-of-flight mass spectrometry, Environ. Chem. 5 (2008) 402-412. [24] Z.Z. Wang, Y. Deguchi, M. Kuwahara, T. Taira, X.B. Zhang, J.J. Yan, J.P. Liu, H. Watanabe, R. Kurose, Quantitative elemental detection of size-segregated particles using laser-induced breakdown spectroscopy, Spectrochim. Acta Part B 87 (2013) 130-138.

19

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[25] Z.Z. Wang, Y. Deguchi, M. Kuwahara, X.B. Zhang, J.J. Yan, J.P. Liu, Sensitive measurement of trace mercury using low pressure laser-induced plasma, Jpn. J. Appl. Phys. 52 (2013) 11NC05. [26] Z.Z. Wang, Y. Deguchi, M. Kuwahara, J.J. Yan, J.P. Liu, Enhancement of laser-induced breakdown spectroscopy(LIBS) detection limit by low pressure and short pulse laser-induced plasma process, Appl. Spectrosc. 67 (2013) 1242-1251.

20