"iii?" 0‘. ' F H. '58". "‘91:" i." .1 l‘ I L") I 1 in“; .9 M 5mi- . .0535 ......;.A ..-5‘P«‘~' LR" V1:4>';v ' Nu..3«u . ‘ ..u-, . "mm; 5.1L :VA'I:)::- “Law 9:“: '6: m. lawn“ 31.x37“‘“ fia"“€5"“’" JG)“ 4 us My , .71‘ 11's» ; , r; *4 Hum...» . arr; ruin-hug «"Ilijtx ":5..-; : <53? 4-15 . . t ,5 fairy-fa "J; ‘..: ' : 3332i 1. m. . 4"::‘h fidflm' "i: .1 ‘3‘. 3133‘.“ "W ""0"” Jar-«pan: 'n v . u 2&3}: ru'"~vn..'~fi in: u- 1 "‘1‘: '53“ r ;. ;l‘rvm "xv-r1" "W‘s ‘rn «Fir m- s ‘T-‘-« rv g} vw-r ‘ 2: J".- 1': vu-‘Iw'w! :3. '_N’1'hvr~:v' grx? ’ r , 2451‘: r: 41. r '1' . -. 133;.” run: - :3. Lt. " .., ,....'. .; .. , « r2234; ' :.. ,.. r ‘ ’Jf ‘13; ‘ f: d 9' , (I w: 1 f w. "‘ 1'3.” -".' ‘: .2112"? “"12"”, - ' :n 1: nu r:n--~ .J 33,, r.:..w.g§':.-...~ ..._.t 3:." :: :1:'1L'1"" P1 "V: J ,‘m ‘.."?.'.'":."1;’ -r r 3:; H ’. J r . . -. - n“. 3'; :37...» J. {wit 4' :23.» .F '37 .40" .1ng v‘," 00": "' 31': ~va :rv-wv: ’w—v-‘h- EDM ‘ m ":7 5%”; «mums pl R ITY LIBRARIES llljllll ill 85 2711 TATE lllllllill | 31293C‘)O This is to certify that the thesis entitled IMPLEMENTATION AND TESTING OF A LASER INDUCED FLUORESCENCE SYSTEM FOR THE CHARACTERIZATION OF A MULTIPOLAR ELECTRON CYCLOTRON RESONANCE PLASMA REACTOR presented by GREGORY LOUI S KING has been accepted towards fulfillment of the requirements for Master of Science degree in Electrical Engineering ’ gm professor Date August 9, 1991 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution IMPLEMENTATION AND TESTING OF A LASER INDUCED FLUORESCENCE SYSTEM FOR THE CHARACTERIZATION OF A MULTIPOLAR ELECTRON CYCLOTRON RESONANCE PLASMA REACTOR by Gregory Louis King A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Electrical Engineering 1991 ABSTRACT IMPLEMENTATION AND TESTING OF A LASER INDUCED FLUORESCENCE SYSTEM FOR THE CHARACTERIZATION OF A MULTIPOLAR ELECTRON CYCLOTRON RESONANCE PLASMA REACTOR by Gregory Louis King In materials processing using electron cyclotron resonance (ECR) plasmas the ion energy distribution is important in determining the processing rate and magnitude of any material damage. This thesis describes the concepts and implementation of the laser induced fluorescence technique as a non-intrusive measure of these distributions at various spatial locations and under various experimental conditions in a multipolar electron cyclotron resonance plasma. This thesis begins with a discussion of the theoretical basis of laser induced fluorescence spectroscopy and its applicability to ion energy and ion density studies in an ECR plasma. With this background the ion energy within the source and processing regions of the plasma reactor is measured. Measured Doppler shifted energy distributions indicate plasma potential variation from the source to the processing region. Finally, relative metastable ion density measurements are described at various locations Mthin the source and processing regions of the plasma reactor. ACKNOWLEDGEMENTS I would like to thank Dr. Timothy Grotjohn for his generosity. He has given his knowledge. his time and his support to guide me through the work presented here. I would also like to thank Dr. J es Asmussen for his insights and advice and Dr. Donnie K. Reinhard for his comments and suggestions. Since this thesis would not have been possible without the laser accessible baseplate, I would like to thank Peng-Un Mak for his work designing this integral piece of equipment. Finally, I would like to thank Jim Bradeen for his encouragement when it was needed most and to my family for a lifetime of support. LIST OF TABLES LIST OF FIGURES viii Chapter 1 Introduction 1.1 Motivation for an MPDR ECR Plasma Ion Study 1 1 .2 Research Goals 1 1.3 Thesis Outline 2 Chapter 2 LIF Utilized for Plasma Diagnostics: A Review 2. I Introduction 3 2.2 The Theory Of Laser Induced Fluorescence 3 2.2. 1 LIF for Ion Density Measurements 4 2.2.2 LIF for Ion Velocity Measurements 6 2.2.3 LIF for Ion Energ' Measurements 7 2.3 Recent LIF Work in Plasmas 8 TABLE OF CONTENTS iv 3. l 3.2 3.3 3.4 3.5 3.6 4. l 4.2 4.3 4.4 4.5 5. l 5.2 5.3 5.4 5.5 5.6 Chapter 3 LIF Apparatus / Multipolar ECR Plasma System Introduction 1 1 Multipolar ECR Plasma System 1 1 3.2.1 The MPDR Baseplate 1 1 3.2.2 The Microwave Subsystem 15 3.2.3 Gas [Vacuum Subsystems l 5 The Lasers 1 7 3.3. 1 Nd:YAG Pulsed Laser 17 3.3.2 Dye Laser 19 Fluoresced Light Collection 20 3.4. 1 Optics 20 3.4.2 Monochromator 20 Gated Integrator 22 Computer Control 25 Chapter 4 Ion Energies and Velocities Introduction 26 Spectral Line Broadening 29 4.2. 1 Doppler Broadening 29 4.2.2 Laser Broadening 49 4.2.3 Zeeman Splitting 32 4.2.4 Power Broadening 32 Ion Energies 33 Ion Velocities 33 Conclusions 35 Chapter 5 Relative Ion Densities Introduction 38 Relative Ion Density versus Microwave Power 38 Relative Ion Density versus Gas Flow 40 Relative Ion Density versus Pressure 40 Relative Ion Density versus Radial Position 43 Conclusions 43 Chapter 6 Summary of Results 6. 1 Implementation of an LIF System 47 6.2 Testing of the LIF System 47 6.2. 1 Ion Energy Measurements 47 48 6.2.2 Directed Ion Velocity Measurements 6.2.3 Relative Density Variation Measurements 6.3 fixture Research APPENDIX LIST OF REFERENCES 49 50 61 LIST OF TABLES Table 1. 1: Plasma Species Studied using LIF Figure 2.1 figure 3.1 Figure 3.2a Figure 3.2b Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Figure 4.6 Figure 5.1 LIST OF FIGURES Three Level Energy Diagram Cavity. Baseplate. Vacuum and Gas Systems MPDR Baseplate - Top View MPDR Baseplate - Side View Microwave Subsystem The Laser System 11 12 13 15 17 Optics and Spatial Resolution 20 Photomultiplier Signal Amplifier 22 Gated Integrator Timing Diagram 23 Source and Processing Regions Longitudinal and Radial Laser Beam 27 28 Geometry within Source and Processing Regions Radial Velocity Distribution in Source Region Longitudinal Velocity Distribution in Source Region LIF Spectra at Various Longitudinal Positions LIF Intensity versus Microwave Power _30 __3 1 __34 —36 39 figure 5.2 Figure 5.3 Figure 5.4 Figure 5.5 ix LIF Intensity versus Gas Flow 41 LIF Intensity versus Pressure LIF Spectra at Various Radial Positions 42 MPDR Baseplate - Top View Showing ECR Volumes 45 Chapter 1 Introduction 1 . 1 Motivation for an MPDR ECR Plasma Ion Study The scope of microwave electron cyclotron resonance (ECR) pro- cessing is increasing dramatically. Applications for the high density, low species energy, clean processor are numerous. ECR plasmas are used for etching. thin film deposition and growth of oxide layers [1-3]. For any of these the energy and density of species irnpinging on the processing sur- face is important in determining the rate of processing and the magnitude of any surface damage. To understand the processing mechanisms, an understanding of the species dynamics is necessary. The charged particles having energy greater than the thermal energy are particularly energetic and worthy of study. Although the ion energies have been studied in a microwave plas- ma disk reactor (MPDR) using a multi-grid ion energy analyzer, limita- tions precluded a complete characterization [4]. 1.2 LIF Research Goals This thesis endeavors to describe a laser induced fluorescence sys- tem for the characterization of an MPDR ECR plasma with respect to its ion energy and density attributes. A preliminary characterization study to test and evaluate the LIF system includes a study of the ions created un- der various experimental parameters and at various locations. This thesis describes the method, the cautions and the preliminary results of an LIF study of singly ionized argon in an MPDR plasma. 1 1.3 Thesis Outline Chapter 2 begins with a description of the theory of LIF as it relates to the measurements of later chapters. This chapter is an overview of the theory and a complete description is found in the references cited. Chap- ter 2 concludes with a brief look at pertinent literature using LIF as a plasma diagnostic tool. Chapter 3 describes the MPDR source and supporting equipment. The bulk of the chapter describes the lasers, optics and light collection equipment used. Ion energy measurements are described in chapter 4. The chapter begins with a look at the various broadening mechanisms that affect the LIF absorption lineshape and continues with an ion energt measurement in the source of the MPDR. Chapter 4 concludes with a study of the ion velocity distribution in both the source and processing re- gions of the plasma. Chapter 5 switches to relative ion density measure- ments in the MPDR at various locations and under difl’erent experimental conditions. Chapter 6 summarizes the work of this thesis. An appendix is included which contains the computer program used to collect and aver- age the LIF signal. Chapter 2 LIF Utilized for Plasma Diagnostics: A Review 2.1 Introduction Laser induced fluorescence (LIF) is a sensitive diagnostic tool well suited for detailed studies of ECR plasmas. The power of LIF lies in its ability to probe the plasma without intrusive probes or analyzers. Also. the spatial resolution of LIF allows detailed study of each particular re- gion of the plasma. The following sections review the theory of absorption spectroscopy relative to the energy, velocity and density studies undertak- en for this thesis. The final section of this chapter reviews the pertinent literature in plasma diagnostics. Particular attention is paid to LIF used in ECR pro- cess plasmas under similar conditions studied in this thesis. 2.2 The Theory of Laser Induced Fluorescence Laser induced fluorescence involves exciting an atom or ion in a ground or metastable electronic energy level to a higher energy level through absorption of laser radiation. The excited species can spontane- ously decay to the beginning energy level or to another level through emission of a photon. The intensity of this emission or fluorescence is proportional to the excited state species density and indirectly proportion— al to the beginning state density. Also, the lineshape of the absorption ra- diation spectrum gives information about the energy of the excited species. 4 2.2. 1 LIF for Ion Density Measurements Laser induced fluorescence is particularly useful for studies of spe- cies density. The following describes the theory of LIF in terms of rate equations for transitions from one energy level to another. The fluores- cence signal resulting from the emission of radiation as the ion decays from its excited state is proportional to the density (N2 cm‘3) Of the excited state. A relationship between the number of photons per second incident on the detector (np) and this density is then: where A23 (sec'l) is the Einstein rate coefficient for spontaneous emission and 8 (cm3) is the collection coefficient based on such factors as the effi- ciency of the collection system and the solid angle subtended by the col- lection optics. The fluorescence signal is then directly proportional to the number of photons. In order to understand the relationship between this signal and the total species density. an equation for the upper level densi- ty is needed. Finding an equation for N2 begins with a look at a typical LIF tran- sition, shown in Figure 2. 1 , where N is the number density of species for that level, W12 and W21 are the laser induced rates of absorption and emission, 911 is the collisional excitation or quenching rate and A41 is the spontaneous emission rate [5]. Species are excited from one level to a higher one by absorption of laser radiation or collisions with other parti- cles. Species decay to a lower energy level spontaneously. through colli- 5 sions or through laser radiation induced emission. N2921 N2923 N3932 N2A23 N19 12 I ' Level 3 N1W12 NgA N2W21 21 N3931 N3A31 v i v v v Figure 2.1: Three Level Energy Diagram Level 1 Assuming a constant total number density (NT) for the system, the rate equations for each Of the two excited levels are [6]: 2 E = N1W12_N2(W21+Q21+A21+QZ3+A23)+N3932 st dt N2(923+A23) ‘N3(931+A31+932) NT=N,+N2+N3 In steady state (dN3/dt = O), the population ratio N3/N2 becomes: g Q23 + A23 = = B N2 Q31 “‘31 + Q32 (2.2) (2.3) (2.4) (2.5) 6 Then substituting Equation 2.4 into Equation 2.2 and assuming steady state and full saturation: W12 N (2 6) W12(1+B)+W21 T ° N2: Full saturation means the laser induced rates are larger than both spon- taneous and collisional rates (W 12, W21 >> 91], A11) [5]. Equation 2.6 shows that N2 is proportional to the total species den— sity. Earlier Equation 2. 1 showed that the fluorescence signal is propor- tional to the level 2 number density. This leads to the conclusion that the fluorescence intensity measured is a relative measure of the total species density (N 1 + N2 + N3). Since in most systems N1 >> N2 or N3, the LIF sig- nal can primarily be interpreted as a measure of N1 . This conclusion is the basis for the relative density measurements of Chapter 5. 2.2.2 LIF for Ion Velocity Measurements The Doppler effect, where the relative motion of a particle shifts the observed frequency of light emitted or absorbed by that particle, serves as the basis for the ion velocity distribution measurements using LIF. By tuning the dye laser (Section 3.3.2) through the Doppler shifted absorp- tion wavelengths the resulting lineshape of the fluorescence intensity is proportional to the ion velocity distribution along the laser beam. In other words [7]: I (A) (ll = Kf(v) dv (2.7) 7 where I is the fluorescence intensity distribution as a function of wave- length, f(v) is the velocity distribution and K is a proportionality constant The dependence of the absorption wavelength on the ion velocity is [8]: V lo = MHz) (2.8) where A is the Doppler shifted wavelength, 2.0 is the center wavelength. c is the speed of light in a vacuum and v is the velocity of the ion along the laser beam. An ion with a velocity in the direction of the laser beam re- quires light of a shorter wavelength to make the transition to the upper energ' level. 2.2.3 LIF for Ion Energy Measurements For a Maxwellian distribution of the ion velocity at a temperature, T,, the lineshape due to Doppler broadening has a Gaussian shape: where m, is the mass of the ion and c is the speed of light in vacuum. The full width at half the maximum value of g (FWHM) is found from Equation 2.9 as [9]: 2A ,2kT ln2 A}. = o i (2.10) C mi 8 The ion emery or temperature is found by measuring Al for a particular distribution and solving for T1: 2 _ mic AA 2 All ion enery estimates in this work are found using this equation. The velocity distributions presented are found using Equations 2.7 and 2.8. 2.3 Recent LIF Work in Plasmas Laser induced fluorescence has come into common usage since the advent of high powered pulsed lasers. Some of the early work in LIF plas- ma diagnostics is in regard to flames. Eckbreth, et al. used LIF as a densi- ty measurement in flames [10]. This work also described methods for absolute density calibration. J. W. Daily proposed operating in saturation to overcome quenching difficulties in Reference [6]. The work of Wright. et al., described the study of velocity distributions using LIF [8, 1 1]. Many references in the literature describe the use of LIF for density and enery measurements in a wide variety of plasmas. Various gases have been studied. Table 1.1 lists some of those gases and the emery transitions studied. Some of the species listed absorb two photons of light to make the transition to the higher enery level. The most complete LIF studies in low pressure microwave ECR plasmas have been undertaken on diverging field ECR reactors [7, 12, 13, 14]. These plasma sources use solenoidal magnetic coils to produce the ECR fields. They are characterized by magnetic field gradients which pro- duce associated ion drifts. Table 1 . 1: Plasma Species Studied using LIF Plasma Wavelength (nm) Species Absorption Emission Comment Reference Ar+ 61 1 .49 460.96 12 Ar+ 617.23 458.99 15 Ar+ 624.3 488.0 16 Ar 696.54 772.4/ 727.3 17 CF 232.9 240.0 18 CF2 261.7 271.0 18 Cl 210. 1 904. 1 Two photon 19 Cl 233.3 725-775 Three peaks 20 in emission range F 690.25 677.40 17 Ge 265.1 275.5 21 H 205. 14 656.2 TWO photon 22 N2+ 389.05 389.05 Doublet emission 14 Ne 597.55 626.65 13 O 226 845 Two photon 23 / 24 $0 248 270/314 25 Xe 605. 12 529.22 15 Zr” 595.53 838.94 8 10 Den Hartog, et a1. [14] studied the enery of a nitrogen ion in an ECR discharge. They report the transverse temperature of the ion in terms of its implications in plasma etching. Their study took place in the bulk of the plasma to determine if the ion gains transverse enery before it reaches the processing surface. A series of studies using argon and argon / helium mixtures have characterized these ECR reactors with respect to ion temperatures and ion velocity distributions [7 . 12, 13]. These distributions were measured in both the source and processing regions of the plasma and under various experimental conditions. In an argon discharge downstream from the source with microwave power of 1 .5 kWatts and at 0.35 mTorr. an ion em- ery upper bound of 0.46 eV is given [7]. The LIF ion velocity distributions measured in the processing region Show a distinct bimodal nature which is divided into a fast component and a slow component. The fast component is reported to arise from those ions which follow the magnet flux lines from the source into the reaction chamber; whereas. the slow component arises from those ions created where the source expands into the reaction chamber. The ion velocity dis- tribution functions are nearly isotropic in the source region but become strongly anisotropic in the processing region as they follow electrostatic fields [12]. Chapter 3 LIF Apparatus / Multipolar ECR Plasma System 3.1 Introduction This chapter introduces the specific equipment used to carry out the research described in this thesis. First the MPDR and its supporting equipment is described. The final sections detail the lasers, optics and signal collection equipment used for the laser induced fluorescence work. 3.2 Multipolar ECR Plasma System The multipolar ECR source consists of a seven inch microwave cav- ity, the baseplate and the quartz cavity (Figure 3. 1 and Figures 3.2a/b). The microwave cavity is a cylindrical resonant cavity which can be tuned with a sliding short. It directs intense microwave emery into the plasma source region. The baseplate contains the eight gas inlets and also a ring of eight rare earth magnets (Figure 3.2a). Each magnet measures two inches by one inch by one inch. These magnets create the ECR zones within the quartz cavity. A detailed description of the MPDR is found in reference [26]. 3.2. l The MPDR Baseplate The baseplate. shown in Figures 3.2a and 3.2b. was designed to al- low laser access to the plasma source region [27]. The plasma is created within a fused quartz chamber (Figure 3.2b) that is attached to the base- plate. The eight rare earth magnets are housed in a high permeability iron 11 12 C Cavity L___, / (Eaeplafi‘ Mass Flow Controller Processing Region To Manual Gate Valve Gas CYHHdCI’ \\\\\“{ ..\\\\' {x ‘\ \Freon B §§ Mechanical Pump _| | Diffusion Pump \__/ Figure 3.1: Cavity, Baseplate. Vacuum and Gas Systems Figure 3.2a: MPDR Baseplate Inside Diameter = 12.5 cm Laser Source Region Access Figure 3.2b: MPDR Baseplate - Side View Laser Port A is 2.5 cm high Laser Port B is 0.6 cm high Dashed Line Indicates Magnet Locations 15 keeper which focuses the magnetic fields within the quartz chamber and eliminates magnetic fields in the downstream processing region. These magnets have a low Curie temperature and therefore need cooling to pro- tect them from the relatively high temperature of the plasma. Water cool- ing is provided by that section of the baseplate which surrounds the magnet ring. The baseplate design keeps the function of the iron keeper and the water coong intact while allowing laser access to the discharge region in front of the magnets. The baseplate also serves as the mecha- nism for the distribution of the worldng gas into the chamber. Eight pin— holes are arranged around and below the inner side of the quartz chamber for gas access. Also, air cooling is available for the quartz cham- ber if necessary. 3.2.2 The Microwave Subsystem Microwave emery is supplied by a 2.45 GHz microwave power sup- ply (Micro-Now 420B 1). The experiments described here were performed with microwave power ranging from 1 50 Watts to 300 Watts. This power range refers to the power absorbed by the cavity which is found by sub- tracting the microwave power reflected by the cavity from the power inci- dent om the cavity. The nominal value used was 250 Watts with less than 0.3% reflected. The microwave circuit. shown in Figure 3.3, includes the three port circulator and dummy load to protect the power supply and the dual directional coupler for sampling both the reflected and incident pow- 61'. 16 Reflected Power Meter Dummy Load Incident 0 Power C Meter Microwave avity Power 1 3 __ Supply CD Coaxial r \ 1 TYansmission ' \\ 1 Line Directional Coupler and Attenuators Circulator Figure 3.3: Microwave Subsystem 17 3.2.3 Gas/Vacuum Subsystems The 99.99996 pure Argon gas is fed into the baseplate through a mass flow controller (Tylan FC-280) with a range of flow from 10 to 30 sccm. The nominal value used is 20 sccm. The vacuum system includes a 2500 1/ sec oil diffusion pump and a 33 m3/Sec mechanical pump both filled with a hydrocarbon-free oil to allow the use of reactive gases. To re- duce backstreaming of oil, a freon-cooled baffle separates the processing chamber from the diffusion pump. This minimized the contaminates in the chamber at the expense of pumping speed. A manual high-vacuum gate valve separates the diffusion pump from the processing chamber and allows manual throttling of the cham- ber pressure. The pressure is measured in the chamber with a capacitive manometer (MKS-390HA) down to about 1 x 10'5 Torr. Also. an ionization pressure gauge is located at the opening to the difiusion pump but the sensitivity of this instrument to non-nitrogen environments brings its ac- curacy into question. The vacuum and gas systems are represented sche- matically in Figure 3.1. 3.3 The Lasers A number of different lasers could be chosen for LIF work each hav- ing distinct advantages. The laser chosen, though, must have high peak power, a narrow spectral width and be frequency tunable. The work pre- sented in this thesis uses a tunable dye laser pumped by a Nd :YAG pulsed laser. The complete LIF apparatus is shown schematically in Fig- ure 3.4. 18 Dye _ NdIYAG Lens Laser Pulsed Laser f—A—l Computer ' Boxcar Integrator Monochromator Photomultiplier Figure 3.4: The Laser System 19 3.3. 1 Nd:YAG Pulsed Laser In order to adequately pump the dye laser and increase the signal to noise ratio by minimizing the loss of excited species due to quenching, saturation of the laser induced absorption transition should be assured (Section 2.2.1). One way to saturate this transition is through the use of a high power laser. The necessary high power is achieved with a Q-switched neodymium doped yttrium aluminum garnet (Nd:YAG) pulsed laser (Spec- tra-Physics DCR- 1 1). The peak power of 40 Watts with a pulse duration of 6 nsec assures adequate pumping of the lossy tunable dye laser after frequency doubling to facilitate saturation of the absorption transition. The laser output of the pump laser is frequency doubled using an harmonic generator (Spectra-Physics HG-2). The harmonic generator con- sists of a KD‘P crystal (potassium dideuterium phosphate) which inter- acts with the fumdamental 1064 mm light from the Nd :YAG laser to produce a secondary wave with half the wavelength. Since the conversion efficiency of the crystal is highly dependent on its temperature. a temper- ature controller is necessary for optimum frequency conversion. 3.3.2 Dye Laser Since many suitable transitions exist and many different gases can be used, a tunable laser is necessary to fully analyze any plasma system. The dye laser as used in this work (Spectra-Physics PDL-3) provides tun- able laser radiation from 380 nm to 960 nm.The particular range of laser radiation available depends on the dye used. In order to scan the entire velocity distribution of the particular argon species studied here (624.3 mm), Exciton DCM dye was chosen. 20 In addition to the dye, the angle of the wavelength tuning grating is adjustable to permit spectral tuning of the output radiation. Control of the grating angle is achieved through the use of a stepper motor and step- per motor comtroller. A personal computer is used to coordinate the mo- tor. the light collection stages and the gated integrator. 3.4 Fluoresced Light Collection Laser access to the plasma is achieved as described in Section 3.2. 1 . This section will describe the collection of the fluoresced light and the apparatus necessary for spatial characterization of the ion species. 3.4. l Optics Emitted light is collected by a lens of diameter 6.3 cm and focal length 5 cm. The light is focused onto a 1 mm diameter fiber cable which carries the light out of the vacuum system to the monochromator. The im- aging system just described is shown in Figure 3.5 along with the transla- tion stages necessary to move the focal point within the vacuum. Three dimensional movement is available with the three translation stages which allows precise positioning of the collection volume along the laser beam. The x and y stages allow movement along a distance of 4 inches and the 2 stage allows movement along 2 inches. Figure 3.5 also shows the actual spatial dimensions of the collec- tion volume and gives an estimate of the spatial resolution of the system as used in this work. The sample volume is 0.05 cm3 in this configura- tion. Using a calculation of the solid angle subtended by the optics and assuming fluorescence occurs isotropically. about 1% of all photons emit- ted are collected by the lens. 21 Sample Volume 0.05 cm3 1 Laser I 5 mm Beam 2 17.5 cm A 6.3 cm 4 > W V Translation A I Stages Lens f = 5 cm 7 cm W : Y 1 mm Fiber Cable Figure 3.5: Optics and Spatial Resolution 22 3.4.2 Monochromator A 1 meter, f / 9. Spex, Inc. monochromator is used to filter unwanted light from the fluorescence signal. In addition, an optical filter with a passband centered near the fluoresced light wavelength is used to further reduce the amount of unwanted light affecting the signal. The light from the monochromator is detected using an EMI, Inc. cooled photomultiplier. The monochromator entrance and exit slits are set at 1 mm to ensure good signal to noise ratio. Since the spectral resolution of the apparatus is determined by the laser linewidth and other broadening mechanisms the spectral width of the monochromator is not a factor. 3.5 Gated Integrator A gated or boxcar integrator (EG&G PARC 4 12 1B) is used to repeti- tively sample the fluorescence signal emanating from the photomultiplier. The integrator samples the photomultiplier signal when triggered by the pump laser. The samples are averaged to improve the signal to noise ratio (SNR). The averaged signal is converted to a digital signal to allow data collection by computer using an A/ D converter (EG&G PARC 4 161A). A preamplifier is in place between the photomultiplier tube (PMT) and the gated integrator. The charge sensitive amplifier shown schemati- cally in Figure 3.6 is DC coupled to the gated integrator [28]. The important aspect of the gated integrator set-up is the timing of the trigger with respect to the laser pulse. In order to minimize any stray laser light from artificially enhancing the fluorescence signal, the integra- tor is timed to collect fluoresced light immediately after the end of the la- ser pulse. Figure 3.7 shows a timing diagram where the laser pulse, 23 820 _L 1 DF 22 g 0.0114 10K 36 - 4.7K [K N W f'\ 1 .2K Figure 3.6: Photomultiplier Signal Amplifier 24 6 nsec A Laser Pulse I ‘ Typical Fluorescence Signal 5 msec 4| 6 Laser adjustable Gate Delay I I Sample Gate H 60 nsec Figure 3.7 : Gated Integrator Timing Diagram 25 trigger pulse, fluorescence pulse and the Open gate of the integrator are shown. Two timing adjustments are possible. The delay between the actu- al laser pulse and the start of the trigger pulse is adjustable (shown in Figure 3.7). Also, a more sensitive adjustment is available on the gated in- tegrator which allows control of the delay between the time the gated inte- grator receives the trigger and the start of the gate (Gate Delay in Figure 3.7). In this work, a gate of 60 nsec with 30 samples averaged using an in- put sigmal sensitivity of 200 mV is found to give the best sigial strength without serious SNR problems and is used for most of the measurements. 3.6 Computer Control The entire experiment is controlled through the use of an IBM PC computer with various interface boards. LIF sigials are determined by first pulsing the laser for a prescribed number of seconds and simulta- neously collecting the gated integrator sigials. Then the measuring pro- cess is repeated for the same length of time with no laser pulses. By comparing the gated integrator sigials when the laser is pulsing to the background sigials when the laser is off, the LIF sigial strength is deter- mined. Generally, at each wavelength point the intensity shown is an av- erage of 7 to 13 on-off cycles where each cycle is about 18 sec each. The computer also controls the dye laser gating through the step- per motor and stepper motor controller. Additionally. movement of the collection Optics is achieved through control of the translation stages within the vacuum chamber. The complete QuickBasic program used for computer control is included in the Appendix. Chapter 4 Ion Emerges and Velocities 4.1 Introduction The implemented LIF system has been tested by applying it to the study of ion energies and velocities in an MPDR plasma. The MPDR sys- tem has been split into two distinct regions. The source is that region within the discharge chamber where the plasma is generated and electro- magietic enery is coupled to the electron gas by the electric fields of the microwave cavity and the ECR static magnetic fields. The processing re- gion lies below the source as in Figure 4.1 and is where the ion emery and velocity distributions are dominated by plasma potential gradients and diffusion processes from the source to the processing regions. Since the ion energies along the laser direction are measured, changing the direction of the laser provides information on the distribu- tions both horizontally, with a radial laser beam, and vertically, with a longitudinal laser beam as shown in Figure 4.2. Ion enery distributions are measured in both regions and, within the source region, in both direc- tions. Ion velocity distributions are particularly interesting when compar- ing one position with another and therefore they are presented in that context. The study described here concentrated on demonstrating the imple- mentation of an LIF system to study a singly ionized argon metastable in an ECR plasma. Specifically the absorption transition is the 3d4F7 ,2- 4p2D5/2 transition at 624.3 nm. The emission is the 4p2D5,2 - 482P3/2 transition at 488.0 mm. 26 27 Sliding / Short Microwave Input 2] ‘ AcLaser cess * Port \ 1 I (for radial laser) To Vacuum FIDCI' PumP Cable Figure 4.1: Source and Processing Regions 28 Longitudinal Laser Beam Radial Laser Beam Magiets Fiber Optic Cable Figure 4.2: Longitudinal and Radial Laser Beam 29 4.2 Spectral Line Broadening A typical ion enery distribution, measured at the center of the source region (r=0, z=1: Figure 4.3), is shown in Figure 4.4 where fluores- cence intensity in arbitrary units (arbs) is measured versus ion velocity. An obvious feature of this distribution is that it has a certain non-zero width. A number of factors contribute to the broadening of the distribu- tion, including Doppler shifts, high laser power, laser spectral width, magnetic fields and electric fields. Therefore, care must be taken in inter- preting the results. 4.2.1 Doppler Broadening Doppler shifted broadening (Sections 2.2.2 and 2.2.3) occurs due to the relative motion of the ion with the observer. From an assumed Gauss- ian distribution a full width at half maximum is found which relates the average ion enery or temperature to the observed distribution (Equation 2.5). This is an actual temperature only in the absence of all other broad- ening mechanisms. 4.2.2 Laser Broadening Laser broadening occurs due to the non-zero linewidth of the laser light. If the laser light is Gaussian and a Doppler-broadened line takes a Gaussian shape then the two can be decomvolved [7]. 2 2 A)‘(iop 3 J (Axobs) — (Atlas) (4'1) where Mdop is the Doppler broadened FWHM, Mobs is the FWHM of the (‘— A Magnet it up HN llll oo Procsinesg Region Figure 4.3: Geometry within Source and Processing Regions 31 Power = 240 W z = 1 cm Pressure = 0.75 mTorr 000 r = 0 cm " Gas Flow = 20 sccm o 0 9° 9 0 o - 00 E o °° .33. Q 0 ° 2 - o d) *5 o '-' o o o 0 o 90% 0 0° °o l l 1 i i 1 l -8 -6 -4 -2 0 2 4 6 Velocity (km/s) Figure 4.4: Radial Velocity Distribution in Source Region 32 observed lime and Mm is the linewidth of the laser. Equation 4. 1 is used to extract a Doppler broadened linewidth from the observed line when all other broadening mechanisms are minimized. 4.2.3 Zeeman Splitting Zeeman splitting is the effect of magietic fields on the observed dis- tribution. It is not broadening in the sense of Doppler broadening but ac- tually arises as the splitting of the spectral absorption enery into two or more distinct energies. For most of the locations inside the plasma region measured in this thesis. the magietic fields are negligible and Zeeman splitting can be ignored. Reference [27] shows that the magietic fields in most of the sample areas are negligible, particularly at the center of the source and processing regions. 4.2.4 Power Broadening Lime broadening due to the power density of the laser is termed power broadening. The pulsed laser used in these experiments operates at a power density such that power broadening is observed. In order to re- duce the effect of power broadening, the laser intensity is attenuated us- ing a series of optical density filters. The power broadening effect is determined by taking a series of measurements with increasing attenuation of the beam. As the attenua- tion increases the broadening of the line decreases until the power broad- ening is negligible compared to the Doppler broadened line. The beam was attenuated using a series of optical density filters placed in the path of the laser. The crucial ion enery measurements were taken at this inci- 33 dent laser power. In the interest of sigial to noise certain relative mea- surements were taken without attenuation of the laser. 4.3 Ion Energies The ion enery distribution in Figure 4.4 is found at the center of the source region (r=0, z=1 on Figrre 4.3) with a radial laser beam. The resolved ion emery is found to be about 0.5 eV after reducing the laser power to minimize power broadening, deconvolving the laser broadening and verifying negligible magnetic fields. Due to residual power broadening this is set as an upper limit until further research refines the measure- ment. Figure 4.5 shows a similar distribution where the sample space is located at the same point (r=0, Z: 1). This distribution is taken with a lon- gitudinal laser beam and the light collection optics set at an angle greater than ninety degrees (see Figure 4.2 for laser beam direction and optics set-up). The upper limit ion enery in this case is also 0.5 eV with the same experimental parameters as Figure 4.4. 4.4 Ion Velocities Ion velocity distributions are found using the techniques described in Section 2.2.2. The velocity distributions are particularly interesting when comparing one location with another. Since the measurement is taken at the center of the source region the distribution of velocities is as- sumed to be random, therefore the peak of the distribution is chosen as the zero velocity point. The geometry used to describe the MPDR system is shown in Figure Intensity (arbs) I I I I I I 1 T I Power = 240 W z = 1 cm Pressure = 0.75 mTorr o r = 0 cm I- Gas Flow = 20 sccm 0° _, 000 0 0 o 0 - 0° ° . 6’ 00% o F 0 00$ 4 ° 9 o a» °°° .. 0° 6° ‘. S°°° o o I I I l I i I I I -10 -8 -6 -4 -2 0 2 4 6 8 10 34 Velocity (km/s) Figure 4.5: Longitudinal Velocity Distribution in Source Region 35 4.3 where the center point of the source region is chosen as the origin of polar coordinates (r = 0, z = 0). Longitudinally, this point lies at the base of the magnets and radially. at the center of the quartz chamber. The pos- itive longitudinal direction is from the source to the processing region. The positive radial direction is from the origin out to the magiets. Figure 4.6 shows a series of velocity distribution measurements taking at three different longitudinal positions. The important point to no- tice is the shift of peak ion velocity from source to processing region. This indicates the ions have picked up a directed enery component as they leave the source. If the peak velocity of an ion at z = 1 cm is taken as the reference then the peak velocity at z = 3 cm is 1.24 km/sec and at z = 5 cm is 2.02 krn/sec. This shows an increase of 2 km/sec over 4 cm. 4.5 Conclusions The operation of the LIF system for determining ion velocities has been demonstrated. In some initial test measurements, the ion emery within the source region is given an upper limit of 0.5 eV regardless of la- ser direction at a pressure of 0.7 5 mTorr with 240 Watts of microwave power and an argon gas flow of 20 sccm. This is only an upper limit since residual line broadening of those distributions shown in Figure 4.4 and Figure 4.5 may still be a factor. The ion velocity distributions given indicate a directed enery com- ponent from source to processing region. This indicates a change in plas- ma potential between the points of the measurement. The data indicates an increase in directed ion emery of about one electron volt. This corre- sponds to a potential variation over the 4 cm measured from one cm to 36 T I I I I I I I z = 1 cm 0 Power = 240 W 4 Pressure = 0.75 mTorr ° °° z = 3 cm + 1 Gas Flow = 20 sccm 9 o z = 5 cm El 0 I— 0 0° —: 4. + . ° * + o .8 - _ 9 El 0 3. S n n n 0 ° 0 Li? '1' u + co m a O a " o + %l] + .- g + +3 ++++¢ .. o o o D U I .. 0 +00 Ch + .. 0 [1 El 3 on + _ DE _ I I I I I I I I -10 -8 -6 -4 -2 0 2 4 6 8 10 Velocity (km/ s) Figure 4.6: LIF Spectra at Various Longitudinal Positions 37 five cm below the magrets of less than 25 volts/ meter at a pressure of 0.75 mTorr with 240 Watts of microwave power and an argon gas flow of 20 sccm. Chapter 5 Relative Ion Densities 5.1 Introduction Another interesting use of laser induced fluorescence is with regard to ion density measurements. LIF can give both relative and with proper calibration absolute density measurements. This section describes rela- tive ion density variation measurements within the source region of the MPDR plasma. To evaluate the operation of the LIF system, selective com- parisons are performed between the LIF measurements and double Lang- muir probe measurements. The density measurements are with respect to microwave input power, chamber pressure and gas flow rate. 5.2 Relative Ion Density versus Microwave Power Figure 5. 1 shows the variation of the intensity of the laser induced fluorescence with increasing microwave input power from 150 -. 290 Watts. The data shown are at two different locations within the plasma processing system as indicated in the figure. A longitudinal laser beam is used (Figure 4.2). The upper solid curve shows that the metastable spe- cies density increased as the microwave power increased. The lower solid curve shows a similar increase but at a slower rate. The data presented in Figure 5. 1 correlate very well with total ion density variation measured using a double Langmuir probe [27]. The in- crease in microwave power leads to more energetic particles in the ECR volumes and consequently more ionizations farther from these volumes 38 39 Intensity (arbs) I T I I I I I 2.0 Pressure = 0.7 5 mTorr 1 ._ z = cm - GasFlow-20sccm r=0cm 1 ..---°' ......... fl _ 1.5 ............. Probe m, ”we... - <— LIF —— ........ - """"""""" a 1.0 ..-"‘ """" "‘ - 0.5 I I l I I I I 0.0 140 160 180 200 220 240 260 280 300 Microwave Input Power (Watts) Figure 5.1: LIF Intensity versus Microwave Power Ion Density (x 101 1 cm'3) 40 which is measured as a relative increase in metastable ion density in the case of LIF and an increase in total ion density in the case of the double Langnuir probe. 5.3 Relative Ion Density versus Gas Flow Figure 5.2 shows the variation in LIF intensity with a change in gas flow rate from 10 seem to 40 sccm. The pressure in the chamber is al- lowed to vary from 0.43 mTorr up to 1.31 mTorr as the gas flow increased. The ion density remains fairly constant throughout the range of gas flow tested indicating that a simple increase in neutral species does not neces- sarfly increase ionizations. The likely affect is that the excess neutrals are simply removed from the region through the vacuum pumps. As expected, though. the relative ion density drops for measurements downstream from the ECR volumes. 5.4 Relative Ion Density versus Pressure Figure 5.3 shows the relative change in metastable ion density as pressure is increased. The gas flow and the microwave power are kept constant throughout these measurements. The drop in LIF intensity as the pressure is increased can be attributed to the decrease in volume size of the neutral ionization region as the pressure is increased. At the higher pressures the majority of the ionizations are taking place in a smaller re- gion which is close to the ECR volume. Also at the higher pressures, in- creased quenching may reduce the excited species density at locations away from the ionization regions. Therefore, at the center of the source (r=0) the LIF intensity drops since there are fewer excited species at this Intensity (arbs) 41 Microwave Power = 250 W r = 0 cm Pressure (mTorr) o z=lcm + z=3cm 0 +0 0.43 0.56 0.76 0.88 1.04 1.17 1.31 lllvii l l l 5 10 15 20 25 30 Gas Flow (sccm) 35 40 45 Figure 5.2: LIF Intensity versus Gas Flow 50 Intensity (arbs) 42 Microwave Power = 250 W Gas Flow = 20 sccm r = 0 cm I l l l l l 0.5 1 1.5 2 2.5 3 3.5 Pressure (mTorr) Figure 5.3: LIF Intensity versus Pressure 43 point. This is particularly evident at the higher pressures above 1.5 milli- Torr. 5.5 Relative Ion Density versus Radial Position ~ Figure 5.4 examines the change in relative ion density as the mea- surement volume moves closer to the ECR excitation volumes. These lines can not be used for enery measurements since the sample volume moved closer to the magnets and higher magietic fields which produce Zeeman splitting. Interestingly, the measurement at r = 3 cm is taken just at the edge of a visibly bright region within the plasma source (shown in Figure 5.5 as the ECR volumes). The measurement indicates a substantial in- crease in metastable ion density which would correspond to an increase irn excited species and therefore a brighter “light.” 5.6 Conclusions The LIF system has been tested by applying it to determine relative ion densities. These relative ion density LIF measurements indicate changes in ion density can be produced by changes in experimental reac- tor parameters. These measurements show an increased ion density with increased microwave power (150 W to 300 W) indicating more ionizations are taking place. This data corresponds well to the total ion density mea- surements taken with a Langmuir probe. The relative metastable ion density decreases with increased pres- sure amd constant gas flow indicating fewer excited species are present in the center of the source at the higher pressures tested (particularly over the 1 .5 to 3 mTorr region). The measurements Show little density change Intensity (arbs) 1 I I I I I I I I Power = 240 W CI: Pressure = 0.75 mTorr u u D D GasFlow=208ccm Dunn ti: 2: 1cm up a a r = 3 cm D + r = 1.5 cm U r = 0 cm a D n + D u $96906“: + D d? + o u o a" n + °" 0 d] 10 9+ in tn +3 9 on + 3 ”a? 3‘” I I I I I I I u I -8 -6 -4 -2 0 2 4 6 8 Velocity (km/s) Figure 5.4: LIF Spectra at Various Radial Positions 45 Radial ECR Excitation Laser Beam Volumes (bright regions) Figure 5.5: MPDR Baseplate - Top View Showing ECR Volumes 46 with increased gas flow (10 seem to 40 seem/0.43 mTorr to 1.31 mTorr) indicating the small affect that increased neutral density has on the num- ber of ions in this gas flow/ pressure range. With the applicability of the LIF system to ion density measurements verified, future experiments will characterize this reactor with respect to gas flow, pressure changes and microwave power to determine the operation of the MPDR reactor over a wider range of experimental conditions. Chapter 6 Summary of Results 6.1 Implementation of an LIF System A laser induced fluorescence system has been implemented for the characterization of an MPDR plasma. A pulsed Nd :YAG pumped dye laser is used to excite the plasma species. The subsequent fluoresced light is collected using a five centimeter focal length lens. The light, focused onto a one millimeter fiber cable, is filtered through a monochromator and measured with a photomultiplier. A gated integrator averages the fluo- resced light intensity of each pulse. The resulting integ'ator sigial is col- lected and averaged using a computer. 6.2 Testing of the LIF System The application of the LIF system has been demonstrated for a number of measurements including ion emerges, ion velocities and rela- tive ion densities in an Argon plasma. 6.2.1 Ion Emery Measurements Ion enery measurements are crucial to understanding ion bom- bardment during plasma processing. The LIF measurements quantify the enery of ions created in the ECR volumes of a microwave plasma disk re- actor. Upper limit estimates of 0.5 eV ions give an understanding of these source energies for the MPDR plasma operating at a pressure of 0.7 5 mTorr with 240 Watts of microwave power and an argon gas flow of 47 48 20 sccm. The upper limit is necessary at this poirnt since residual power and laser broadening may artificially affect the enery estimate. 6.2.2 Directed Ion Velocity Measurements Plasma potential variation is one method that ions gain directed en- ery as they drift within the plasma. The velocity distribution shifts show an increase in directed enery (velocity) from the source to the processing region. These distributions are measured away from any magnetic fields and far from any surfaces: therefore, the measured shifts indicate a change in plasma potential of about 25 volts / meter over four centimeters from one centimeter to five centimeters below the magnets. This mea- surement was taken with 240 Watts of microwave power at a pressure of 0.7 5 mTorr with an argon gas flow of 20 sccm. 6.2.3 Relative Density Variation Measurements The study of metastable species in this thesis gives insight into the density of ions within various regions of the plasma. An increase in mi- crowave input power (150 W to 300 W) produces a corresponding increase in metastable ion density. These results match well with the total ion density measurements made using a double Langmuir probe. A change in gas flow rate (10 seem to 40 seem/0.43 mTorr to 1.31 mTorr) produces no significant change in ion density over the range test- ed. The increase in neutral density associated with an increase in gas flow does not necessarily result in increased ionization. An increase in chamber pressure (0.5 mTorr to 3 mTorr) with all other parameters held constant results in a decrease in metastable ion 49 density above 1.5 mTorr. This measurement, taken at the center of the source, indicates a drop in excited species at the center. 6.3 Future Research This thesis lays a foundation for future detailed studies of the MPDR ECR source using laser induced fluorescence. Using the tech- niques described here a complete study can be undertaken. Increased knowledge of the broadening mechanisms associated with enery mea- surements will refine the enery estimates. The ability to include a pro- cessirng surface with bias to simulate the resulting plasma sheaths, the addition of downstream magietic confinement and the use of other pro- cess gases will better simulate various processing conditions. Finally, ex- tending the experimental parameter space through imcreased microwave power and an increased gas flow range will facilitate a complete charac- terization of the plasma. APPENDD( APPENDIX QUICK BASIC PROGRAM FOR LIF SYSTEM CONTROL The following progam is used to control the gated integator, the translation stages, the Nd:YAG laser and the gating angle within the dye laser. The code is written in Microsoft's QuickBasic version 4.0. The gat- ed integator is controlled through an HPIB interface. The Nd:YAG laser is controlled with a D /A convertor board (Metrabyte DAS-8). The gating angle is controlled through a stepper motor (SIC-Syn Model M061-LF- 408E) and stepper motor controller (Slo-Sym Type 3 180-P1125) connected through the serial port of the computer. The stages are controlled through a translation stage controller (Newport 860-C2) connected to the parallel port. 100 5 1 REM $INCLUDE: ’QB45ETUP' DIM INlT(15). POSII3. 60). WAVARRIS. 60) DIM AVEOFF( 15) DIM INARY%(7) DIM OUTARY%(7) BYTE% = 0 : MSTEP = 32 GRATORD = 4: ISC& = 7: ADDRBr = 728: MAXLENGTH‘K) = 20 TIMEOUTNALI = 1! 'Initializc serial port for control of stepper motor OPEN ‘COM1:9600.N.8.2.CS.DS.CD' FOR RANDOM AS #1 ‘Set limit on forward and reverse operation of stepper motor PRINT #1, CW1?) PRINT #1. “<1 L18 +25000" PRINT #1, “<1 L19 ~25000' PRINT #1, “<1 L41 0" 'Initialize das8 data acquisition board MD% = 0 BASADR% = &H300 FLAG% = 0 CALL DASSiMD%. BASADR%, FLAG%) MD% = 14 FLAG% = 0 OP% = 9 CALL DASB(MD%. OP%, FLAG‘XI) LOCATE . . 0 PRINT “SET-UP" PRINT ' 1) OPl to Reed Relay" PRINT " 2) Reed Relay red and black to Laser Remote“ PRINT ‘ 3) Q-SW ADV SYNC to GI trigger” PRINT " 4) 0P4 to RESET AVG on back of GI“ PRINT PRINT "Enter filenamc to store data (.dat assumed) --" INPUT FILE l8 FILE$ = ‘\boxcar\data\" + FILE1$ 4» “.dat' FILE2$ = ‘\bourcar\data\' + FILE1$ + 'z.dat" FILE3$ = "\boxcar\data\" + FILEI$ + ‘y.dat" CLS GOSUB TIMEIT CLS OPEN FILES FOR OUTPUT AS #2 OPEN FILE” FOR OUTPUT AS #3 OPEN FILE3$ FOR OUTPUTAS #4 'imitializc HPIB interface by: set-up address for GI CALL iorcmoteIADDR&) 52 IF PCIB.ERR <> NOERR THEN ERROR PCIB.BASERR “Set value for length of time before timing out CALL iotimeoutIisc&. timeoutval!) IF PCIB.BRR 0 N OERR THEN ERROR PCIB.BASERR 'Clear channel CALL iocleariiscaz) IF PCIB.BRR <> NOERR THEN ERROR PCIB.BASERR R3 = "‘ S& = 0 WHILE (88: AND 1) = 0’ Poll HPIB line for problems GOSUB SPOLL WEND PRINT 'Set up menu PRINT 'ENI'ER CONDITION --" PRINT " 1 -- Laser on” PRINT " 2 -- Laser ofi“ PRINT " 3 -- Automatic data collection“ PRINT " 4 -- Manual wavelength adjustment (release motor torque)" PRINT " 5 -- User controlled motorized wavelength adj ustment" PRINT “ q -- Quit data collection” RS = INPUT$( 1) PRINT N = 0 A3 = ‘RDA" + CI-IR$(13) 'set to read from channel A IF RS = “'1" THEN OP1% = 8: OP2% = 0'set data board to ’turn on laser IF R$ = ”2" THEN OP1% = 9: OP2% = For turn ofi'laser IFR$='1” ORR$= ‘2'THEN GOSUB LASER PRINT PRINTTIMEI. 'MINUTE(S) '10 GO" GOSUB RESETIT GOSUBWRITEA GOSUB READA mm = B$ FOR Z = 1 TO TIME NEXTZ GOSUBWRITEA GOSUB READA IF OP1% = 8 THEN OP1% = 9: OP2% = 1: GOSUB LASER DIFF& = VALIB$I - VALIINI'NI CLS PRINT DIFFGI END IF IF RS = '3' THEN CLS 53 PRINT “ENTER CHOICEz' PRINT “ 1 -- Increasing wavelength” PRINT " 2 -- Decreasing wavelength“ PRINT " 3 -- Return to lower limit” PRINT " 4 -- Return to upper limit” PRINT “ 5 -- Position scan (increasing only)“ PRINT “ 6 -- Single Averaged Data Point“ PRINT “ e -- Back to Main Menu” PRINT S$ = INPUTSII) 'If 1 is chosen the gating is changed after each data poirnt to increase the wavelength 'of light. The length of the wavelength step is chosen by choosing the number of points ’takem per quarter angstrom of wavelength scan IFS$= “1" ORS$= ‘5'THEN PRINT “Enter 15. 30, 45 or 60 ONLY" REDO: INPUT “Enter number of data points PER quarter amgstro --': ND4% IF (ND4% 0 15) AND (ND4% 0 30) AND (ND4% 0 60) AND (ND4% <> 45) THEN GOTO REDO INPUT “Enter lower laser n-number --": N1 INPUT “Enter upper laser n-number --"; N2 L1 = N1 / gatord: L2 = N2 / gatord ND%=ND4%‘(N2-N1):NPOS=0 (111$ = “H07”: WAVE = L1 IF ND4% = 60 THEN MSTEP = 8: DISP = .0042 IF ND4% = 45 THEN MSTEP = 12: DISP = .0063 IF ND4% = 30THEN MSTEP = 16: DISP = .0083 IF ND4% = 15 THEN MSTEP = 32: DISP = .0167 END IF ' Same as 1 except decreasing wavelength. Step should have already been set with 1. IFS$= ENDIF ‘2' THEN INPUT 'Enter lower laser n-mumber --’: N1 INPUT ‘Emter upper laser n-number --"; N2 L1 = N1 / gatord: L2 = N2 / gatord ND%=60‘(N2-N1):NPOS=0 (”1'3 = “H06": DISP = -.0042: WAVE = L2 '3 and 4 return the gating angle to its original position after a scan 1FS$= ‘3'THEN 1FNPOS<>0THEN FORY: ITONPOS+ 1 PR1NT#1,"<1" PRINT#1.'H06" FORZ: 11025: NExrz NEXTY ELSE PRINT “Already atn=": N1 ENDIF GOTODONE ENDIF 54 IFS$="4"THEN IFNPOS <> OTHEN FORY = 1 TO NPOS + 1 PRINT #1, “<1" PRINT #1. “H07“ . FORZ= 1'1‘025: NEXTZ NEXTY ELSE PRINT “Already at n =": N2 END IF GOTO DONE ENDIF 'If 6 is chosen from the main menu the data gathering is stopped to allow input of a ’particular pressure or power setting BEGIN: IFS$='6"THEN INPUT “Enter Power or Pressure --": PARAM INI'I‘OT‘ = 0 OP1% = 9: OP2% = 1 GOSUB LASER PRINT #3. PARAM: FORY = 1 TO 13 GOSUB RESETIT ’reset gi GOSUB WRITEA “send the command to read channel A (RDA) GOSUB READA 'read data from channel A mm = B8 'save the initial reading FORZ= l T‘OTIME: NEXTZ GOSUB WRITEA GOSUB READA 'write to g and read in data mm = VAL(B$) - VAL(IN1T$) 'data point is final-initial 'iins odd thenlaserwasofi'andwillbe turned on 'else iinseventhenlaseron andwillbe off PRINT INl'i‘tY): PRINT #3. INITIY): 1F(Y/2i)<>1NT(Y/2i)TI-IEN 1FY< 13TI-IEN 'iflastthisislast laserofi' OP 1% = 8 0P2% = 0 GOSUB LASER END IF 'set DASB for Laser On ELSE OP1% = 9 'set DASB for Laser Off OP2% = 1 GOSUB LASER END IF NEXT Y PRINT AVEOFFIZ) = (1Nl'1‘(1) + INIT(3)) / 2 'average the laser ofi' AVEOFF(4) = (INl'I‘(3) + 1NIT(5)) / 2 “values surrounding the AVEOFF(6) = (1NT115) + INI'I‘(7)) / 2 AVEOFFIB) =(1Nn‘(7) + INIT(9)) / 2 AVEOFFIIOI = (INNS) + 1Nl'I(11)) / 2 55 AVEOFF(12) = (INITII 1) + INI'Ir13)) / 2 'laser on one PRINT #3. ' ' ' each of the 3 intermediate data points is laser on minus average laser ofi' FORY=2TO l2 STEP2 INITOT= 1N1TOT‘+ (INlTiY) - AVEOFF(Y)) NEXT Y ’ the ultimate data point is the average of the three intermediate ones DIFF& = INITOT / 6 PRINT PARAM. DIFF& PRINT #2. PARAM. DIFF& IF mm 0 'E' THEN GOTO BEGIN ELSE S$ = ‘e' END IF END IF IFS$="C' ORS$= ‘E'THEN GOTODONE TDIR%=0 'Choosing 5 allows control of the z-axis of the translation stage. Similar control of 'of the other two axes can be set up. IF S$ = '5' THEN PRINT ‘SET z-axis TO TOPMOST POSITION" PRINT INPUT “Enter number of positions --": PNUM INPUT ”Enter distance between positions (cm) --"; DIS END IF PRINT PRINT ‘NUMBER OF DATA POINTS:". ND% 'Choosimg 1. 2 or 5 brings the program here for the actual data collection FOR MOTOR = 1 TO ND% (1 = 1 P081: M$ = "' INITOT = 0 OP1% = 9: OP296 = 1 - GOSUB LASER PRINT #3. WAVE: FORY = 1 TO 13 Reset the gated integator to 0 GOSUB RESETTI‘ ’send the command to read channel A (RDA) GOSUB WRITEA GOSUB READA 'read data from channel A INIT$ = B$ 'save the initial reading FOR 2 = 1 TO TIME: NEXT 2 GOSUB WRITEA GOSUB READA ‘write to gi and read in data INITIY) = VALIBSI - VALIINITSI ’data point is final-initial ’iins odd then laserwas offandwill be turned on 56 'else iinseventhenlaseron andwillbe off PRINT #3. INI'IIY): IF(Y/2i)<>INI(Y/ 2i)TI-IEN IFY< 13THEN 'iflast thisislast laserofl' OP 1% = 8 ‘then don’t set up for OP2% = 0 GOSUB LASER END IF 'set DASB for Laser On ELSE OP1% = 9 'set DASS for Laser Ofl' OP2% = 1 GOSUB LASER END IF NEXT Y PRINT AVEOFF(2) = (INIT(1) + INITI31) / 2 'average the laser “off“ AVEOFF(4) = (INlT(3) + INI'115)) / 2 'values surrounding the AVEOFF(6) = (INIT(5) + INI'I(7)) / 2 ' laser ‘on" values AVEOFF(B) = (INlTl7) + INlT(9)) / 2 AVEOFF(lO) = (INl’T(9) + INIT(11)) / 2 AVEOFF(12) = (INTI‘il 1) + INlT(l3)) / 2 Taser on one PRINT #3, " " ’ each of the 3 intermediate data points is laser on minus average laser ofi’ FORY: 210 12 STEP2 INI’I‘OT = INITOT + (INthY) - AVEOFFM) NEXT Y ' the ultimate data point is the average of the three intermediate ones DIFF& = INTTOT / 6 PRINT WAVE. DIFF&, “Data point #": MOTOR. NPOS POSIIq. MOTOR) = DIFF& WAVARRIq. MOTOR) = WAVE PRINT #2. WAVE. Drrra IFS$='5'ANDq NOERR THEN ERROR PCIB.BASERR RETURN This subroutine reads the HPIB line to get the data B8 from the Gate Integrator READA: B$ = SPACE$(20) S& = 0 WHILE (S& AND 1) = 0 GOSUB SPOLL IF IS& AND 128) THEN CALL ioentersIADDR&, B3, MAXLENGTH‘XI, ACTUAL.LENGTT-I%) IF PCIB.ERR <> NOERR THEN ERROR PCIB.BASERR END IF WEND RETURN This subroutine polls the HPIB line for errors or busy signals SPOLL: CALL iospoHiADDR&, S&) IF PCIB.ERR <> NOERR THEN ERROR PCIB.BASERR RETURN This subroutine sends the command to release motor torque to the stepper motor MANUAL: CLS INPUT “Are you sure you want to IDSE motor positiom": MS IFM$=‘y'ORM$="Y’Ti-IEN PRINT #1, '<1 N000 068’ FORZ= 1 T025: NEXTZ PRINT #1, “<1 H01” FOR Z = 1 TO 2: NEXT 2 CLS 58 PRINT ”Press any key to lock motor and continue...” DOWI-IILE mm = ": LOOP PRINT #1. “<1 N000 G69" FORZ= 1T025: NEXTZ PRINT#1. ”<1 H01” FORZ= 1TO25: NEXTZ PRINT #1. '<1 H09" FORZ= 1TO25: NEXTZ END IF RETURN This subroutine allows step by step movement of the grating angle. The length of 'each step depends on the number of data points chosen for the interval measured MO’ICON: CLS INPUT ‘Are you sure you want to LOSE motor position”: M$ CONT: IFM$=YORM$=TTHEN PRINT “Increase. Decrease or End (1. d or e)" M$ = INPUT$(1) IFM$=‘1'ORM$= "I'THEN FOR Y = 1 TO MSTEP PRINT #1. “<1 H07” FOR Z = 1 TO 25: NEXT Z NEXT Y END IF IF M$ = 'd' OR M3 = 'D" THEN FORY: l TOMSTEP PRINT #1. “<1 H06” FORZ= l T025: NEXTZ NEXTY ENDIF IF M$ = 'e" OR M$ = "E" THEN RETURN ELSE M$ = 'Y' GOTO CONT END IF END IF RETURN This subroutine either turns on the laser beam or turns it off depending on OPl% LASER: MD% = 14 FLAG% = 0 OP% = OP1% CALL DASB(MD%. OP%. FLAG%) FOR Z = 1 TO 4000: NEXT Z RETURN This subroutine prints the current data file to the screen for a quick scan SHOW: CLS CLOSE #2 OPEN FILE$ FOR INPUT AS #2 59 WHILE NOT EOFIZ) INPUT #2. WAVE. DIFFGI PRINT WAVE. DIFF& WEND PRINT CLOSE #2 OPEN FILE$ FOR APPEND AS #2 RETURN This is a menu of options shown when 'q" is chosen from the main menu DATEND: CLS PRINT 'Enter 1 -- To Show Data” PRINT " 2 -- Resume Data Collection” PRINT " 3 -- Change sample time” PRINT " 4 -- Change gating order” PRINT “ e -- Exit progam” R$ = INPUT$(1) IFR$='e'ORR$='E'THENGOTOQUlT IF R3 = '1' THEN GOSUB SHOW IF R3 = '2' THEN PRINT F Rs = '3' THEN GOSUB TIMEIT IF R3 = ‘4' THEN GOSUB CRATE RETURN This computes the loop timing from the minutes entered by the user TIMEIT: INPUT ”Enter sample time in minutes--"; TIMEl CLS TIME=TIME1 ‘413000 PULSES=TIME1 ‘600 TIMEl ‘60sec‘ 10 pulses/sec RETURN This allows changing the grating order for data point calculations GRATE: CLS INPUT “Enter new gating order --": gatord RETURN This routine moves the translation stages the specified number of cm MOVE: IF ((BY'TE% AND &H2) = 0) THEN CALL INT860LD(&H17. INARY%0. OUTARY%0) BYTE% = BYTE% OR &l-160 CALL IN'T86OLD(&H17. INARY%0, OUTARY%0) IF (((BYTE%AND &H2) = 0) AND ((BYTE%AND &H8) = 0)) THEN BYTE% = BYTE% AND &I-IDF CALL INT860LD(&H17. INARY%0. OU'TARY%0) END IF END IF IF T01R96 = -1 THEN “move z-axis forward BYTE% = BYTE96 OR &H23 INARY%(O) = W INARY%(3) = 0 60 ‘velocity of z-axis motor (cm/sec) at maximum dial setting in rev 22 = 16. I CALL INT860LD(&H17. INARY%0. OUTARY%0) GOSUB TMOVE INARY%(0) = BYTE% AND &HFD CALL IN'I‘860LD(&Hl7. INARY%0. OUTARY%0) ELSEIF 'IDIR% = 0 THEN ‘move 2 in reverse BYTE% = (BYTE% AND &HFE) OR &H22 INARY%(0) = BYTE% INARY%(3) = 0 ‘velocity of z-axis motor (cm/sec) at maximum dial setting in rev ZZ = 16.3 CALL INT860LD(&H17. INARY%0. OUTARY%0) GOSUB TMOVE INARY%(O) = BYTE% AND &HFD CALL INT860LD(&H17. INARY%0. OUTARY%0) END IF RETURN This subroutine times the movement of the stage for a know stage velocity (22) TMOVE: START = TIMER S'IDPI'T=START+ZZ‘DIS WHILE (STOPIT > TIMER) WEND RETURN This routine resets the gated integator by sending a pulse to the RESET AVE input RESETIT: MD% = 14 OP% = OP2% CALL DA38(MD%. OP%. FLAG%) FOR X = 1 TO 350: NEXT X OP% = OP1% CALL DASB(MD%. OP%. FLAG%) FOR X = 1 TO 3000:NEXT X RETURN QUIT: CLOSE END LIST OF REFERENCES [ll [2] [3] [4] [5] [6] l7] [8] [9] 61 LIST OF REFERENCES J. Asmussen.” Electron Cyclotron Resonance Microwave Dis- charges for Etching and Thin-film Deposition,” J. Vac. Sci. Tech- nol., A 7, 3, (1989) J. Hopwood, M. Dahimene, D. K. Reinhard, J. Asmussen, “Plas- ma Etching with a Microwave Cavity Plasma Disk Source,” J. Vac. Sci. Technol., B 6, l, (1988). T. Roppel. D. K. Reinhard, J. Asmussen, “Low Temperature Oxi- dation of Silicon Using a Microwave Plasma Disk Source,” J. Vac. Sci.. Technol., B 4, l (1986). J. Hopwood, D. K. Reinhard. J. Asmussen. “Charged Particle Densities and Energy Distributions in a Multipolar Electron Cy- clotron Resonant Plasma Etching Source.” J. Vac. Sci. Technol., A 8, 4, (1990). L. D. Radziemski, editor, Laser Spectroscopy and its Qplica- tions, (Marcel, Dekker, New York, 1987). J. W. Daily. “Saturation Effects in Laser Induced Fluorescence Spectroscopy,” Applied Optics, Vol. 16, No. 3. (1977). D. J. Trevor. N. Sadeghi, T. Nakano. J. Derouard. R. A. Gottscho. P. D. Foo. J. M. Cook, “Spatially Resolved Ion Velocity Distribu- tions in a Diverging Field Electron Cyclotron Resonance Plasma Reactor,” Appl. Phys. Lett., 57, 12, (1990). R. B. Wright, C. E. Young. M. J. Pellin. D. M. Gruen, “High Reso- lution COntinuous Wave Laser Induced Fluorescence Spectros- copy of Sputtered Zr Atoms,” J. Vac. Sci. Technol., 20, 3, (1982). . H. R. Griem, Plasma Smctroscopy. (McGraw Hill, New York, 1964). [10] [11] [12] [13] [14] [15] [16] [17] [18) 62 A. C. Eckbreth, P. A. Bonczyk, J. F. Verdieck. “Combustion Diag- nostics by Laser Raman and Fluorescence Techniques,” Prog. Energy Combust. Sci.. Vol. 5, (1979). R. B. Wright, M. J. Pellin, D. M. Gruen. “Velocity Distribution of Sputtered Zr Atoms as Determined by Laser Induced Fluores- cence Spectroscopy.” Surface Science. 1 10. (198 l). N. Sadeghi. T. Nakano. D. Trevor. R. A. Gottscho. “Ion Transport in an Electron Cyclotron Resonance Plasma”. to be published J. Appl. Phys. T. Nakano. N. Sadeghi. R. A. Gottscho. “Ion and Neutral Temper- atures in Electron Cyclotron Resonance Plasma Reactors”, Appl. Phys. Lett.. 58. 5. (1991). E. A. Den Hartog. H. Persing. R. C. Woods, “Laser Induced Fluo- rescence Measurements of Transverse Ion Temperature in an Electron Cyclotron Resonance Plasma”, Appl. Phys. Lett. , 57 . (1990). S. D. Rosner. T. D. Gaily. R. A. Holt. “Laser Fluorescence Mea- surement of Relative Electron Impact Cross Sections for Meta- stable States of Ar“ and Xe+,'”' J. Phys. B: Atom. Molec. Phys., Vol. 9, No. 16, (1976). Y. Watanabe. M. Shiratam. S. Ogi. N. Kunihiro, “Study on Elec- tron Density Dependence of Metastable Ar+ Density in Pulsed- discharge Plasma by Using LIF Method,” Jap. J. Appl. Phys., Vol. 26, No. 1, (1987). S. G. Hansen. G. Luckman. G. C. Nieman. S. D. Colson. “Forma- tion and Decay of Metastable Fluorine Atoms in Pulsed Fluoro- carbon / Oxygen Discharges Monitored by Laser Induced Fluorescence.” Appl. Phys. Lett.. 56. 8. (1990). K. Ninomiya, K. Suzuki, S. Nishimatsu. O. Okada. “Diagnostics of Microwave Plasma by Laser Induced Fluorescence,” J. Vac. Sci. Technol., A 4. 3. (1986). [19] [20] [21] [22] [23] [24] [25] [26] [27] 63 M. Heaven, T. A. Miller. R. R. FYeeman, J. C. White, J. Bokor, “Two-Photon Absorption, Laser Induced Fluorescence Detection of Cl Atoms.” Chem. Phy. Lett.. V0186. No. 5-6, (1982). G. S. Selwyn, L. D. Baston. H. H. Sawin, “Detection of Cl and Chlorine-containing Negative Ions in RF Plasmas by Two-Photon Laser Induced Fluorescence,” Appl. Phys. Lett.. 51, 12. (1987). N. Hata, A. Matsuda, K. Tanaka, “Spectroscopic Diagnostics of Plasma-Chemical-Vapor Deposition from Silane and Germane.” J. Appl. Phys., 61. 8. (1987). B. L. Preppernau, D. A. Dolson. R. A. Gottscho. T. A. Miller, “Temporally Resolved Laser Diagnostic Measurements of Atomic Hydrogen Concentrations in RF Plasma Discharges.” Plasma Chemistry and Plasma Processing, Vol. 9, No. 2, (1989). G. S. Selwyn, “Spatially Resolved Detection of O Atoms in Etch- ing Plasmas by Two-Photon Laser Induced Fluorescence.” J. Appl. Phys., 60. 8. (1986). L. F. Dirnauro. R. A. Gottscho. T. A Miller, “Two-Photon Laser Induced Fluorescence Monitoring of O Atoms in a Plasma Etch- ing Environment,” J. Appl. Phys., 56, 7. (1984). K. E. Greenberg, P. J. Hargis. Jr.. “Laser Induced Fluorescence Detection of SO and $02 in SF6/02 Plasma Etching Discharg- es.” J. Appl. Phys., 68, 2. (1990). J. A. H0pwood, Macroscopic Promrties of a Multimlar Electron Cvclotron Resonance 1VIicrowave-cavit_v_ Plasma Source for Aniso- tropic Silicon Etching, Ph.D. Dissertation, Michigan State Uni- versity, 1990. P. Mak. G. King, J. Hopwood, T. Grotjohn, J. Asmussen. “Influ- ence of Static Magnetic Field Configuration and EM Field Pat- tern on ECR Discharge Performance”, presented: IEEE International Conference on Plasma Science, 199 l . [28] [29) I30] [31] [32] [33] [34] [35] [36] [37] [38] 64 P. Horwitz. W. Hill. The Art of Electronics. (Cambridge University Press, Cambridge. 1988). p. 587. J. Hopwood. D. K. Reinhard. J. Asmussen. “Experimental Con- ditions for Uniform Anisotropic Etching of silicon with a micro- wave Electron Cyclotron Resonance Plasma System,” J. Vac. Sci. Technol., B 6. 6, (1988). M. Dahimene. Development of a Microwave Ion and Plasma Source Immersed in a Multicusp Electron gclotron Resonant Magpetic Field. Ph. D. Dissertation. Michigan State University, (1987). T. A. Miller. “Optical Emission and Laser Induced Fluorescence Diagnostics in Reactive Plasmas.” J. Vac. Sci. Technol., A 4, 3. (1886). T. A. Miller. “Novel Techniques for Plasma Diagnostics: Electron Paramagnetic Resonance and Laser Induced Fluorescence Spec- troscopy.“ Plasma Chemistry and Plasma Processing. Vol. 1 . No. 1 . (198 1). S. M. Rossnagel. J. J. Cuomo. W. D. Westwood. eds.. Handbook of Plasma Processing Technology, (Noyes. New J ersey. 1990). Y. Arata, S. Miyake. H. Matsuoka. “Laser Induced Fluorescence and Doppler Free Polarization Spectra in a Low Density Hydro- gen Plasma,” Jap. J. Appl. Phys.. Vol. 26. No. 3. (1987). D. M. Manos. D. L. Flamm. eds.. Plasma Etching: An Introduc- tigp. (Academic Press. San Diego. 1989). M. Venugopalan. “Aim a Laser at Your Plasma and Expose Its Secrets.” Research and Development. June 1989. J. N. Ross. “Perturbation Spectroscopy of an Argon Laser Dis- charge Using a Tunable Dye Laser.” J. Phys. B. Vol. 8. No. 4. (1975). K. Ninorniya. K. Suzuki. S. Nishimatsu. O. Okada, “Diagnostics of Microwave Plasma by Laser Induced Fluorescence.” J. Vac. Sci. Technol., A 4. 3. (1986). ll \ \| (“MI Ill) “I W" IN I! III) Ill) N 4|) ll") IIII H I ll 3 1 2930088527 1 1