NGHIÊN CỨU PHÂN BỐ OZONE TRONG KHÍ QUYỂN TẦNG THẤP VỚI ĐỘ PHÂN GIẢI CAO TRÊN CƠ SỞ PHÁT TRIỂN VÀ ỨNG DỤNG PHƯƠNG PHÁP LIDAR HẤP THỤ VI SAI TÓM TẮT LUẬN ÁN TIẾN SĨ QUANG HỌC

BỘ GIÁO DỤC VÀ ĐÀO TẠO VIỆN HÀN LÂM KHOA HỌC VÀ CÔNG NGHỆ VIỆT NAM HỌC VIỆN KHOA HỌC VÀ CÔNG NGHỆ - Phạm Minh Tiến NGHIÊN CỨU PHÂN BỐ OZONE TRONG KHÍ QUYỂN TẦNG THẤP VỚI ĐỘ PHÂN GIẢI CAO TRÊN CƠ SỞ PHÁT TRIỂN VÀ ỨNG DỤNG PHƯƠNG PHÁP LIDAR HẤP THỤ VI SAI Chuyên ngành: Quang học Mã số: 44 01 09 TÓM TẮT LUẬN ÁN TIẾN SĨ QUANG HỌC Hà Nội, 2017 Cơng trình hồn thành Học Viện Khoa học Công nghệ – Viện Hàn lâm Khoa học Công nghệ Việt Nam Người hướng dẫn Khoa học: PGS.TS Đinh Văn Trung Phản biện 1: Phản biện 2: Luận án bảo vệ trước Hội đồng đánh giá luận án tiến sĩ cấp Học Viện, họp Học Viện Khoa học Công nghệ - Viện Hàn lâm Khoa hoc Công nghệ Việt Nam vào hồi … …, ngày … tháng… năm 201… Có thể tìm hiểu luận án : - Thư viện Học Viện Khoa học Công nghệ - Thư viện Quốc Gia Việt Nam TABLE OF CONTENTS PREFACE 1 The necessary of the thesis Objectives of the thesis The main research contents of the thesis Chapter INTRODUCTION 1.1 Ozone in the lower atmosphere 1.1.1 Formation and distribution 1.1.2 Ozone absorption cross section 1.1.3 Role and impact of atmospheric ozone 1.2 Measurement of atmospheric ozone 1.2.1 Overview 1.2.2 Measuring ozone in the atmospher 1.2.2.1 Total ozone measurements 1.2.2.2 Measurement of the vertical profile of ozone 1.3 Differential Absorption LIDAR technique for measuring atmospheric ozone dítribution 1.3.1 Physical principle of LIDAR and DIAL 1.3.2 LIDAR system and the LIDAR equation 1.3.3 Differential Absorption LIDAR technique 1.3.4 Wavelength selection for ozone measuring DIAL 1.3.5 DIAL measurement of ozone distribution in the lower atmosphere 1.3.6 Calculation of ozone concentration distribution 1.3.7 Accuracy of ozone DIAL measurement Chapter DESIGN AND SIMULATION OF A DIAL SYSTEM FOR MEASURING OZONE DISTRIBUTION IN THE LOWER ATMOSPHERE 2.1 Design of a DIAL system for measuring ozone distribution 2.1.1 Diagram of Differential Absorption LIDAR 2.1.2 Optical transmitter 2.1.3 Optical receiver 2.1.4 Opto-electronic receiver 2.1.5 Processing and calculation program 2.2 Selection of pair of wavelength 2.3 Simulation of received backscattered DIAL signal 10 2.4 Simulation results and discussion 10 Chapter DEVELOPMENT OF A DIFFERENTIAL ABSORPTIN LIDAR SYSTEM TO MEASURE ATMOSPHERIC OZONE DÍTRIBUTION 10 3.1 Configuration of DIAL system 10 3.2 Development of two DFDL 10 3.2.1 Oscillator 10 3.2.2 Optical pumping system 11 3.2.3 Optical amplifier 11 3.2.4 Active medium 12 3.2.5 Dye transfer pump 12 3.3 Development and evaluation of DIAL’s transmitter 12 3.4 Development of UV telescope and optical receiver 12 3.4.1 Development of UV telescope 12 3.4.2 Making a grinding and polishing machine 13 3.4.3 Development of optical receiver 13 3.5 Development of electronic receiver 13 3.6 Programming for signal acquirement and data processing 13 3.7 Testing of UV DIAL 14 Chapter MEASUREMENT OFF OZONE DISTRIBUTION IN THE LOWER ATMOSPHERE 14 4.1 Data processing 14 4.2 Calculation of ozone concentration distribution 14 4.3 Results of vertical ozone distribution measurement 16 4.4 Error analysis 17 CONCLUSIONS 19 NEW CONTRIBUTIONS OF THE THESIS 20 LIST OF PUBLISHED ARTICLES 21 Preface The necessary of the thesis Ozone is of particular interest in the atmospheric composition because of its presence, distribution, and properties that greatly affect the life of our planet With higher concentrations in the stratosphere, ozone contributes greatly to protecting the earth by absorbing most of the dangerous ultraviolet radiation from the sun in a wavelength range of 200 to 300 nm In the atmospheric layer close to the ground, although only a small component (about several tens of billions - ppb), but ozone is an important contributor to pollution smoke It is one of the main factors affecting human health, the life of organisms, and contributing to the greenhouse effect Therefore, the determination of the concentration and distribution of ozone in the atmosphere is essential, especially the atmosphere surrounding the ground According to the report of the National Hydro-Meteorological Services of Vietnam (May, 2012), our country has about 20 AeroMeteorological Observatories However, there are no annual atmospheric ozone monitoring data The continuous monitoring of ozone concentration distribution will help predict and warn the air pollution to protect human health, increase our understanding of space weather and climate change, and build future development plans Objectives of the thesis Development of a UV Differential Absorption LIDAR ( DIAL) system for high resolution study of the distribution of ozone in the lower atmosphere The main research contents of the thesis The main content of the thesis is to develop an DIAL system with two UV pulses emitted at wavelengths on (282.9 nm) and off (286.4 nm) into the atmosphere The elastic backscattering LIDAR signals of these radiations are acquired and their intensity is used to calculate the vertical ozone concentration The DIAL system includes: + UV transmitter + UV receiver + Single photon counter, software program to process data and calculate ozone concentration distribution Chapter Introduction 1.1 Ozone in the lower atmosphere Ozone (O3) is a pale blue gas and a powerful oxidant It has a distinctively pungent smell and strongly absorbs UV light [2,5] There is very little ozone in the earth's atmosphere, with an average of 10 million molecules of air per molecules of ozone 1.1.1 Formation and distribution Tropospheric ozone is produced from photochemical reactions with oxides of nitrogen (NOx) and volatile organic compound (VOC) molecules, in the presence of sunlight The highest ozone concentration tends to be concentrated in and around urban areas, where generate precursors necessary for ozone production, and often have peaks at noon and lowest at night Ozone concentration also vary from day to day depending on weather conditions, temperature, humidity, wind speed, etc 1.1.2 Ozone absorption cross section The absorption cross section of ozone in the wavelength range from 200 to 1100 nm includes four absorption bands: Hartley, Huggins, Chappuis and Wulf Hartley Huggins are intense bands in UV region They are particularly important in atmospheric ozone monitoring using remote sensing techniques (Differential Optical Absorption Spectroscopy and Differential Absorption LIDAR) 1.1.3 Role and impact of atmospheric ozone Stratospheric ozone filters out sunlight harmful UV wavelegths and protects the life on Earth In contrast, ozone in the lower atmosphere is a major component of photochemical smog in urban environments, an atmospheric pollutant, harmful to human health and a greenhouse gas 1.2 Measurement of atmospheric ozone 1.2.1 Overview Ozone measuring devices can be placed on the ground or flying objects Atmospheric ozone is measured both by remote sensing and by in situ techniques 1.2.2 Measuring ozone in the atmosphere 1.2.2.1 Total ozone measurements Total ozone is measured by remote‑sensing techniques using ground‑based and satellite instruments that measure irradiances in the UV absorption spectrum of ozone between 300 and 340 nm 1.2.2.2 Measurement of the vertical profile of ozone The vertical profile of ozone expresses ozone concentration as a function of height or ambient pressure It is measured with ozonesondes, LIDARs, Umkehr technique with ground-based spectrometers and various satellite-borne instruments [19] 1.3 Differential Absorption LIDAR technique for measuring atmospheric ozone dítribution 1.3.1 Physical principle of LIDAR and DIAL The main components of a LIDAR system consists of laser transmitter, optical receiver, electronic controller and software for processing and analyzing data In LIDAR technique, laser radiation will interact with atmospheric components including molecules, atoms, aerosols and steam Then, the range of physical processes amenable to laser remote sensing includes Rayleigh scattering, Mie scattering, Raman scattering, resonance scattering, fluorescence, absorption, and differential absorption and scattering (DAS) These processes are responsible for the extenction of laser radiation beams emitted by LIDAR system The absorbtion cross section of ozone in the ultraviolet region is much larger than the fluorescent cross section and Raman scattering cross section Therefore, the extinction of an appropriate laser beam caused by ozone will be a highly sensitive method to determine the concentration of ozone in the atmosphere 1.3.2 LIDAR system and the LIDAR equation The functional elements and manner of operation of most lidar systems are schematically illustrated in Fig 1.17 An intense pulse of optical energy emitted by a laser is directed through some appropriate output optics toward the target of interest A small fraction of this pulse is sampled to provide a zero-time marker (trigger) The radiation gathered by the optical receiver and the photodetection system The spectrum analyzer serves to select the observation wavelength interval and thereby discriminate against background ratiation at other wavelengths The Newtonian and Cassegrainian telescope are the main components in the optical receiver Fig 1.17 The essential elements of a LIDAR system [3] The detected LIDAR signal received from a distance R can be written as an equation, called the LIDAR equation: 𝑃(𝑅, 𝜆) = 𝑃0 𝑅 𝑐𝜏 𝑂(𝑅) 𝐴𝜂 𝛽(𝑅, 𝜆) 𝑒𝑥𝑝 𝛼(𝑟, 𝜆)𝑑𝑟] [−2 ∫ 2 𝑅 (1.21) P0 is the average power of a single laser pulse, τ is the temporal pulse length The factor 1/2 appears because of an apparent “folding” of the laser pulse through the backscatter process, c is the speed of light A is the area of the primary receiver optics responsible for the collection of backscattered light, and η is the overall system efficiency O(R) is the laser-beam receiver-field-of-view overlap function β(R,λ) is backscatter coefficient and α(R,λ) is the extinction coefficient The factor stands for the two-way transmission path 1.3.3 Differential Absorption LIDAR technique Differential Absorption LIDAR technique allows the detection of atmospheric gases with high sensitivity With this technique, two Chapter Design and simulation of a DIAL system for measuring ozone distribution in the lower atmosphere 2.1 Design of a DIAL system for measuring ozone distribution 2.1.1 Diagram of Differential Absorption LIDAR Fig 2.1 Diagram of Differential Absorption LIDAR system 2.1.2 Optical transmitter Distributed Feedback Dye Laser (DFDL) was successfully developed at the Institute of Physics [67 – 72] With emitted power strong enough to be able to acquire LIDAR signals, DFDLs have a number of advantages: simple structure; large range of wavelength corrections (10-20 nm depending on the dye) and linewidth of ~ps So DFDL is convenient to select the pairs of wavelengths for the DIAL system, reduce the effect of interfering gas on the measurement results and give a high frequency doubling performance Therefore, the DFDL has been selected to develop DIAL’s transmitter 2.1.3 Optical receiver The main part of the DIAL’s receiver is a telescope The telescope is designed and developed with a minimum diameter of 40 cm to increase the gain of LIDAR signals In addition, the aluminum must be deposited on the surface of the telescope’s primary mirror so that the receiver has a high performance in the ultraviolet region 2.1.4 Opto-electronic receiver Opto-electronic receiver of DIAL system consists of three parts: photomultiplier tube (PMT), preamplifier and single photon counter The quantum efficiency of PMT must be high in UV region Because the LIDAR signal is low intensity signals and the elastic backscattered photons are discrete pulses, single photon counting method will be used in receiver In addition, the photon counting method is more advantageous than the analog method because of stability, high detection efficiency and high signal to noise ratio (SNR) [73] The electronic receiver is designed with fast-responding electronic components 2.1.5 Processing and calculation program The function of the designed software is LIDAR signal acquisition, data storage, data processing and calculation of vertical ozone distribution 2 Selection of pair of wavelength The differential pair of wavelengths selected for the DIAL system are two UV wavelengths 282.9 nm (λon) and 286.4 nm (λoff) The absorption cross section of ozone at λon is of 29.7.10-23 m2 and the differential absorption cross section 𝜎(𝜆𝑜𝑛 ) − 𝜎(𝜆𝑜𝑓𝑓 ) is of 8,9.10-23 m2 [3] The selection of these wavelengths results from the balance of the following the considerations: fluorescence efficiency of laser dyes, altitude range to make retrievals, reducing the impact of the solar background, reducing the impact of aerosol and SO2 interference upon the ozone retrieval [3] 2.3 Simulation of received backscattered DIAL signal The return LIDAR signals are simulated as a function of range by to estimate the expected LIDAR signals of the designed DIAL system (altitude and counting duration) In the simulation, the number of backscattered photons is calculated by the LIDAR equation (1.21) 2.4 Simulation results and discussion In this simulation, the LIDAR equation has been computed with the change of the laser pulse energy emitted, the diameter of telescope (40 cm and 60 cm), the photon counting duration Our simulations of backscattered signals indicate that transmitter is appropriate to a DIAL system which can be used to measure the vertical ozone distribution to an altitude of over 5000 m, counting duration of 10 minutes Chapter Development of a Differential Absortion LIDAR system to measure atmospheric ozone distribution 3.1 Configuration of DIAL system The configuration of DIAL system is developed according to the selections described in Chapter 3.2 Development of two DFDL DFDLs have been developed for the DIAL’s transmitter and their laser radiation is emitted at 565.8 nm and 572.8 nm (Fig 3.2) 3.2.1 Oscillator In the DFDL’s oscilator, a double-faced aluminum mirror CM plays the role of a beam splitter, which divides the pumping beam into two parts After the pumping beam propagates through a cylindrical quartz lens, both parts of the beam are reflected by two rotating 10 dielectric mirrors m1 and m2 The beam is then focused onto the dye cell C1 that contains the sample, and forms an interference pattern Fig 3.2: Diagram of DFDL system 3.2.2 Optical pumping system Two DFDLs is pumped by a frequency-doubled Nd:YAG laser (5ns, 10 Hz, 532 nm) The optical pumping system consists of mirrors M1, M2, M3 and M7; two beam splitter Rm1, Rm2; two prisms P2 and P3 to project the pumping beam to the dye cells of the oscillator and amplifier The optical pumping system extends the journey of the pumping pulse to the power amplifier in order to ensure the amplifier performance when the DFDL’s laser pulse passes through the cuvette C3 3.2.3 Optical amplifier The optical amplifier consists of a 6-pass amplifier (lens L2, mirrors from m3 to m14 and cuvette C2) and output power amplifier (cylindrical quartz lens L3, cuvette C3) 11 3.2.4 Active medium The dye of Rhodamine 6G dissolved in ethanol is the active medium for each DFDL 3.2.5 Dye transfer pump The dye transfer pumps have been designed and manufactured with glass material The transfer pump use the magnetic paddle to push the dye through the cuvettes 3.3 Development and evaluation of DIAL’s transmitter The two DFDLs have the same design There is only one difference in the incident angle of the pumping laser beam to the cuvette C1 The energy of DFDL and UV pulses are measured and presented in Table 3.2 Table 3.2: The energy of DFDL and UV pulses Laser beam Wavelength (nm) Energy DFDL 565,8 0,62 mJ/pulse 572,8 1,8 mJ/ pulse UV 282,9 30 J/ pulse 286,4 60 J/ pulse 3.4 Development of UV telescope and optical receiver 3.4.1 Development of UV telescope The configuration of the telescope is Newtonian It is possible to install spherical mirrors with a maximum diameter of 40 cm and a maximum focal length of 210 cm The telescope frame will be covered with a thick black fabric to prevent near-field scattered radiation The optical axis of the system is calibrated with a semiconductor laser 12 3.4.2 Making a grinding and polishing machine A grinding and polishing machine has been designed to make spherical mirrors (primary mirrors of telescope) with a diameter from 20 cm to 80 cm The optical spherical mirrors were examined and evaluated by the combination of Ronchi and Foucault methods [80] The results show that the optical mirror has a regular spherical concave and a focal length of 1.8 m after grinding and polishing process 3.4.3 Development of optical receiver The optical receiver of Differential absorption LIDAR system includes a telescope, a wavelength filter (F), two lenses L1 and L2 3.5 Development of electronic receiver The electronic receiver of UV DIAL system is developed from the one of multi-wavelength LIDAR system developed at the Institute of Physics This receiver consists of a signal amplifier, a PC oscilloscope Picoscope 5204 with a diagram as shown in Figure 3.22 Amplifier PMT Digital Oscilloscope PC: program of photon counting (Labview) Fig 3.22: Diagram of electronic receiver 3.6 Programming for signal acquirement and data processing A software has been written using Labview programming to control Picoscope oscilloscope, acquire LIDAR signals, save data, and display measurement results This software was developed at the Institute of Physics for the study of atmospheric aerosols The signal processing software was built using Matlab code to smooth the measured data, remove the dark and offset curent of the 13 electronic receiver module, and change the LIDAR signal depending to the distance P(R,) into undepending R2P(R,) The software for calculating atmospheric ozone concentration distribution is also built using Matlab code 3.7 Testing of UV DIAL The Differential Absorption LIDAR system is calibrated to acquire the LIDAR signal from the highest altitude possible After calibration, the DIAL system has been tested and it can acquire the LIDAR signals to a height of over km at both on and off wavelengths With the sampling rate of 125 MSamples/s and using data filtering technique by averaging on some measurement points, the spatial resolution of DIAL measurements is 480 m with a statistical error of ~ 18% at the altitude of km The spatial resolution may be smaller, but the statistical errors will be high The DIAL system does not acquire LIDAR signals from the height above km as the simulation calculation, this can be explained by the average thickness of aerosols of km over Hanoi [81] This aerosol scatters the radiations emitted by the laser transmitter, reducing of backscattering signals and limiting of measurement height Chapter Measurement of ozone distribution in the lower atmosphere 4.1 Data processing In order to improve the accuracy, the data (in the * txt file) will be calibrated over time, adjusted the background and averaged 4.2 Calculation of ozone concentration distribution The concentration of ozone 𝑁𝑂3 (𝑅) between the height R and R+R is calculated according to the expression (1.34), and it is the 14 sum of three terms: Ns(R) : signal term (s – signal), Nb(R) : correction term of differential backscattering (b – backscattering), Ne(R) : correction term of differential attenuation (e – extinction) Ns(R) is calculated directly from the measured data, the correction terms Nb(R) Ne(R) are calculated using the expressions (1.41) and (1.42) correspondingly The backscattering coefficients 𝛽𝑎𝑒𝑟 (𝜆𝑜𝑓𝑓 , 𝑅), the aerosol extinction coefficient 𝛼𝑎𝑒𝑟 (𝜆𝑜𝑓𝑓 , 𝑅) and the ozone concentration 𝑁𝑂3 (𝑅) are determined by the iterative method presented in Section 1.3.6 Based on aerosol studies in Hanoi [81], in urban and polluted environments [63,76], the LIDAR ratio S was assumed to be 30 sr-1 The Ångström exponent η is often seen as an indicator of aerosol particle size The Ångström exponent was investigated in many published reports [82,83], its value for the troposphere aerosol varies from to around the wavelength of 300 nm Considering that η could be relatively small when it is applied in the UV region, we assume that η =0 at our DIAL wavelengths for urban aerosols [39] The iterative procedure can be summarized as follows:  Step 1: Calculate the first ozone concentration from [1.35]  Step 2: Substitute the first ozone concentration into (1.46) to derive the aerosol backscatter profile 𝛽𝑎𝑒𝑟 (𝜆𝑜𝑓𝑓 , 𝑅) for the off wavelength, and iterate to obtain a stable solution with (1.48)  Step 3: Calculate the differential aerosol backscatter and extinction corrections Nb(R) Ne(R) to obtain a second ozone concentration from (1.34)  Step 4: With the second ozone concentration, go back to step This loop ends when the condition 𝜉𝑂𝑘3 < 0,001 is satisfied 15 This program is written in Matlab code 4.3 Results of vertical ozone distribution measurement The UV DIAL, with two DFDL installed in the transmitter, has been used to measure the ozone concentration profile on cloudless nights Figure 4.2 presents the results of continuous ozone concentration distribution in January 2017, from about 1.2 km to a height of over km, with a spatial resolution of 480 m and an counting duảtion of 10 minutes The LIDAR signals at a height of less than 1.2 km are removed due to the small overlap between emitted laser beam and the field of view of the telescope receiver Fig 4.2: Distribution of ozone concentration measured in Jan 2017 over Hanoi 16 From the ozone profiles shown in Figure 4.2, we find that average ozone concentrations over Hanoi from a height of about 1200 m to 4,000 m varies from 2.1012 to 5,1011 molecules/cm3, respectively from 80 to 20 ppbv This trend is consistent with characteristic ozone distribution in the troposphere Because there is no data of ozonesonde at the same time, Figure 4.2 shows an ozone profile over Hanoi, measured by balloon ozonesonde with the spatial resolution of km (published at the meteorological conference in South Korea in 2007 [4]) to illustrate the trend of reduction of ozone concentration in the lower atmosphere and the equivalence of the data 4.4 Error analysis The error of DIAL measurement of ozone concentration is devide into the following four categories: Statistical uncertainties 1 arising from signal and background noise fluctuations Errors 2 associated with differential backscatter and extinction of otherwise gases (O2, NO2, SO2, etc) and aerosols Errors 3 due to uncertainties in the ozone absorption cross section Errors 4 related to instrumentation and electronics 1 is a random error 2, 3 and 4 are systematic errors With the assumption of a Poisson distribution governing the photon counting, 1 is calculated by the expression (1.51) [66] A summary of the errors in the DIAL measurements is shown in Table 4.2 with vertical resolution of 480 m, altitudes below 4km and counting duration of 10 17 Table 4.2: Summary of the errors in the DIAL measurements TT Error % 1 – Statistical error 2 – interference by non-ozone species < 18 Aerosol < 20 Non-ozone absorption gases < 0,3 Rayleigh < 0,6 3 due to uncertainty in differential cross < 2,5 section of ozone 4 due to SIB and dead-time