Pré-visualização parcial do texto
Baixe Glow discharge processes e outras Notas de estudo em PDF para Física, somente na Docsity!
Glow Discharge Processes Glow Discharge Processes SPUTTERING AND PLASMA ETCHING Brian Chapman É A WILEY-INTERSCIENCE PUBLICATION JOHAN WILEY & SONS, New York + Chichester - Brisbane * Toronto Copyright (& 1980 by John Wiley & Sons, Inc. AR rishis reserved. Published simultaneousty im Canada Reproduction or transiatiom of amy part of this work beyond that permitted by Sections 107 or 108 vÊ the 1976 United States Copyright Act without the permission of the copyright owner is unlawful. Requests for permission or further information should be addressed to the Permissions Department, John Wiley & Sons, Inc. Library nf Congress Catuloging in Publi Chapman, Briao N Glow discharge processes. “A Wiiey-Interscience publicatios Includes bibliographical references and index 1. Sputtering (Physics) 2, Glow discharges. 1 Title. M. Title: Plasma etching QC7O2.7.P6CAS 397.52 Bot ISBN 0-471-07828.X Printed in the United States of America 0987654321 Preface This book is based on a series of seminars held in 1978 and 1979. The seminais were intended 10 give some more insight into several practical glow discharge processes that are being increasingly used, particularly in the semiconductor industry. Thope that the text will serve as a useful general introduction to some of the scientific principles involved in these processes. Glow discharges, like so many topics in science, are incompletely understood. Results are often misinterpreted, contradictory, or irrelevant. Glow discharge science has its own subdanguage of special terms, with names that arc often misleading, and with meanings which cannot be assumed to be constant from author to author! One can easily understand the need (or the precision of scien- tific writing and sympathise with the conditions, provisos and double negatives of the author who is taking care not to make any definite statement which might be wrong. This is probably as scientific literature must be when one is close to the bordess of knowledge and ignorance, but it is rather daunting to a nesvcomer to that particular branch of science. Many of you will have had the experience of wanting to learn something about a particular area of science you're not familiar with, and so you go along for advice to the chap in your company or university who is considered the local expert. More often than not you come away with a list of references, in just about all of which it is assumed that you know the subject pretty well! And this is a particular problem in muiti-disciplinary subjects such as sputtering where you are as likely to meet some electrical engineering phase angles as you are some organic chemistry. This book is trying Lo be an introductory book. ft attempts to thread a path through all the basic material you need before you can read the much more erudite reviews on the subject, In an effort to spare readers from attacks of mental indigestion, I have selected those aspects which appear to be more useful for first-Lime acquaintance, and even these are dealt with too briefly. This text was written for readers with a wide rango of backgrounds, and the emphasis is on concepts rather than on rigorous detail. 1 have usually restricted discussíons to the general application of an idea, and have often ignored excep- v pre esto En Reproduction and Copyright Acknowledgments My thanks are due to the many authors and publishers who have permitted me to reproduce the figures and tables herein. Copyright of the authors or publishers, as relevant, is acknowledged. A tist of publishers and related publications follows: Academic Press (including Adv. Electronics & Electron Phys.) Airço Temescal American Chemical Society American Elsevier Publishing Co. American Institute of Mechanical Engineers (Trans. Met. Soc. AIME) American Institute of Physics (F. Chem. Phys., Physies of Fluids, J. Appl. Phys., Appl. Phys. Lett., Rev. Sci. Instrum., J. Vac. Sci. Tech.) American Physical Society (Phys. Rev.) Dover Publications Electrochemical Society (J. Electrochem. Soc.) Heinemann Educational Books IBM Corp. (IBM J. Res, Develop.) Institute of Electrical & Electronics Engineers (Proc. IEEE) Iapanese 3. Appl. Phys. Joint Institute for Laboratory Astrophysics (JILA) Litton Industries (General Mills Report) McGraw Hit Book Company M.LT. Press North Hoeltand Publishing Co. (Nuci, Instrum. & Methods) Optical Society of America (Optics & Spectroscopy) Oxford University Press Pienum Publishing Corp. RCA Corp. (RCA Review) Royal Society (Proc. Roy. Soc.) Societé Francaise du Vide (Le Vide, Les Couches Minces) Society for Applied Spectroscopy (Appl. Spectroscopy) Solid State Technology Standard Telecommunication Laboratories Taylor & Francis (Wykcham Publications) Tokuda Seisakusho John Wiley & Sons Contents Chapter 1: Gases ........ Masses and Numbers of Atom: Kinctio Energy and Temperature Mean Speed T 0. Moswell-Bolteman Distribution Pressure. Partial Pressures Pressure Units... .. Avogadro!s Laws . cer Number Density of Gases .... impingement Plux no Monolaver Formation Time... BRÉS ecc sm DO th né ma Mean Free Path . Probability of Collison . u Collision Frequency . 4 Energy transfer in Binary Collisions . 1 Gas Flow .. 1 Types of Gas Flow iz Pumping Speed and Throughput .. 14 Measurement of Gus Flow Rate 16 Residence Time. . 16 Fiow Velocity 16 Conductance . 17 General References 19 Chapter 2: Gas Phase Collision Processes n Collision Cross Section ne z2 Elastic and Inelastic Collisions .... 23 The Main Collision Provesses 25 Electron Volts 25 Elastic Collisions ... 26 lonizalion . 2 Fxeitation . 31 Relaxation . cus 4 Recombination . 35 3 Body Collision 35 ina Two Stage Process. . 37 Radiative Recombination 3 má xiv CONTENTS CONTENTS xr The Mechanisms ot Sputtering ..... DINDA 179 Surface Amilyois ........ 282 Spurtering Target Kinetics 180 Surface Cleaning ,, 283 Summary of the Overall Process 184 Implications for Bias Sputtorine 255 Applications of Sputtering 185 Some Other Sputtering Contigurations 256 Sputter Etching. 185 Enhancement of Jonizution . 256 Sputter Deposition ,.. 185 Hot Filament Discharges . 256 Limitations of Sputtering ... RR na o co 185 Mognetically Enhanced Sputtering Systems 260 A Conventional DC Sputterine System ooo 2. co 186 Axial Magnetic Fields 261 Choosing the Sputtering Gas 188 Magnetrons ......... 242 Choosing Lhe Pressure Range... 188 Cylindrical Magnetrons . 265 Choosing Flectrical Conditions for the Glow Discharge ... Ro 190 Circular Mupnetrons 267 Summary Lo siice cics re teria ce TB Planar Magnetrons 268 Sputter Btcihing and Deposition of Insulators . 194 General Comments... . 268 RF Sputtering . 195 Analytical and Monitoring Yechniques 20 Reaotive Sputtering ........ 195 jon Beam Systems .......0000. 212 Practical Aspects of Sputterina Systems 196 Jon Beam Sources 272 Creund Shidlds ,. ii... 196 lon Beam Sputtering ..... 2 Shutters ooo 197 Ton Beam Deposition. ... 275 Target Cooling 198 ion Plating 2% Substrate Temperature Control. . 199 Activated Reactive 28 Electrode Voltage Measurement 290 Thin Film Adhesion . 279 Sputtering as a Depasition Process... 201 Methods of Influencing Adhesion 280 Thin Film Formation 201 Conclusion 283 Life on the Substrate 203 References 284 Sputtered Atoms and Contaminants 205 Spurrering Gas Atoms - Fast and Slow 206 Chapter 7: Plasma Etching 297 Excited Neutras 208 Plasma Ashing 297 Positive lons . 208 Plasma Etching . 298 Negative [ons 209 Isotropic or “Amisotropie” Etching? 299 Electrons 309 Reactor Systems . 306 Photons 3 Etehing Mechanisms . 307 Radiation Damage: Creation and Removal . 283 Selective Etching and Plasma Polymeriza 32 Bius Techniques... cor, ns Glow Discharge Aspects of the Reactors 223 Voltage Distribution in Bias 215 Two Mare Reactors . 325 Bias - DC or RE? .. 219 Chemical Dry Ltching . 326 Control of Film Properties . 27 Trivde Plasma Etching . 326 Control of Gas Incorporetion .. 219 Gas Flow Rate Effects... 339 Application of RF Bias . 21 Low Flow Rate Region 330 Bias Sputtering Mechanism . 225 High Flow Rate Region , na 335 Analysis of Charged Particle Bombardment at the Substrate 228 Overali Flow Rate Dependenco .... 336 Bias Fraporation ..... a 3 Etehing of Aluminium and Aluminium Álloys 338 Bias Sputtering for Conformal Coverage .. 233 Silicon Etchine in Chlorine Discharpes . . 340 Backscatterins in Bias Sputterins ....... .. Corrc BA Memitoring of the Etching Process ., ,. 340 Deposition of Multicomponent Films ..... .. neo 0. 27 Plasma Deposition 341 Alloys . 37 Conctusion 341 Compounds . o 241 References and Bibliography 343 Rostoration af Stoiohiomen 243 i 4 Reuctive Sputtering - Agaia q o 243 Appendices ....lccclccccs 351 Sputter Etching... 244 Patiora Production 5a Index lol css sis s scr e eee 2 AD] Etch Topograpby... cc... . o corro 249 Glow Discharge Processes 2 GASES molecules and hs wall, and these collisions cause a continual interchange of energy. This energy has some steady state distribution, which can be derived by statistical techniques. One wouid expect that the overall Kinetic energy of the - and the average kinetic energy of crch constimen: molecule, wouid depend on the absolute temperature W (Figure 1-1) where mt: is lhe mass of'a molecule, cê is the mego square speed, and k is the ity, known as Bolizmann's constant and having the value of 1.38 10775 ergsidos K relevant constant of proportion: o Figure 1-1, Íme MEAN SPEED E Ti foltows that the 7 an square speed is given by 3 KE/m. However, a more useful parameter is e mean speed € which is not equal to the square root of e? (since imean? and “root mean square" are differently defined) but can be shown to have a similar value Vá = (o ) Ta [SK am has a valve 0f 3.94 10º emjsec, which is a litile more han 880 mph. So whilst the atoms aré moving very rapidiv, they're not treveliing so much fasicr than à jumbo jet, and are considerablw slower than Concorde at 1450 mph. (Hopefully (he planes have 4 larger mean free path!) Superman - “faster than a specding bullet” — is neck and neck with Concorde £Figure 1.2). Fes argon atoms at 20 MAXWELL-BOLTZMANN DISTRIBUTION Jt'is implicit in our usage of a men speed € that atoms travel both slower and faster than €, which is as one would expect in any multiple collision process. By considering the random nature of these collision processes, Maxwell and een MAXWELL-BOLTZMANN DISTRIBUTION 3 1.480 Mites Per Hour Figure 1-2. Relative spocds of argon atoms and other familiar bodies Bolumamn (independentiy) were able to show that the number dn of molecules out of à total of n that have speeds between c and «+ de, is given by Se ( me? do dn fim exp (o! do mê (x) “Pla y N This Maxwell-Boltzmam distribution function is shown in Figure 1-3. Although the detail of the distribution is not important for our present purposes, we should point out that while some atoms travel very much faster and some very E de 0 al+ ——. c [o] Figure 1-3, Maxweli-Boltzmana distribution 4 GASES such slower than the mean, nearly 90% of them at any time have speeds between half and twice T. Again, without wanting to stress the detailed form of the Maxwell. Bolizmann function, it is worihwhile to note that the mean speed (6) which we discussed carlier (and indeed the mean of can he obtained from the distribution function since: fe y other parameter) PRESSURE Hwe place a molecule in an enclosure, then because it has some energy, it will continually be in motion, bouncing of! the walls of the enclosure and off other molecules. Each time 4 molecule hits a wall, it will exert a force and the total force per unit wall arca due to all the atoms is known as pressure (Figure 1-4). Consider a section A of the surface of a wall (Figure 1-5). Imagine an x-axis perpendicular to the wall, and let A have an area of 1 em?. Consider the elastio impact (an assumption of kinetic theory) of an atom of velocity % on the wall. lts momentum is mã, and since it bounces back at ihe same speed, the change of mx. Hf the volume density of atoms wiih velocities between X and * + dX is n(X)dX, then the total number that strike the wall per unit time are contained in a cylinder of base A and length numerically equal o X, and this is just n(X)xdX. The rate of change of momentum due to these is 2 mn(X)x? di. But force is equal to the rate of change of momentum (force = mi = mdvidt), and therefore the total force per unit area on the wall (otherwise known as pressure) is the integral of this over all velocíties x: rmomentum is * p= f múgitai The function n(%) is just the one-dimensional version of the Maxwell-Boltzmann distribution, and subslituting this and integrating yields - p= nmã? But x 497422 Since X, Y. and 2 ure symmetrical, then given by: PRESSURE 5 This expression tells us directly how the pressure in a plasma processing chamber depends directly on the density of gas atoms n, their mass m, and their mear square specd. If is also clear why à temperature increase, for example due 10 one of fhe plasma processes, leads to a pressure increase: higher temperature means higher kinetic energy which means higher c? X, ad A Figure 1-4, Wall collisions cause pressure Figure 5. p- 8% 8 GASES IMPINGEMENT FLUX We saw earlier how lhe flux (i.e. number per unit area per unit time) of atoms bombarding the walls of a chamber exerts a pressure. 1f we consider Figure 1-5 and the argument used earlier to derive the pressure relationship. then we can show that: Fluxfunit area = f n(iidi and on integration, (his yields an impingement flux of nc/4 per unit area (Figure 1-7). Figure 1-7. Flux = E per unit area For argon at 5C0 millitorr and 20º€, the impingement itux is 1.8 102º atoms per em? per second. Even the base of the um size hole we considered earlier will be subjected to 1.8 10!? argon atoms per second at 500 mtorr, dropping onty to 3.6 10º at 1 miort. Generally several gas species are present in the vacuum charaber during a plasma process, causing every surface in the chamber, including the substrate, to be constantiy bombarded by a large flux of many gas species; Uhis may be particularly significant during thin film deposition. MONOLAYER FORMATION TIME The preceding surface impingement rate leads us to the concept of monolayer formution time. The substrate is being bombarded by many gas species other tan the depositing thin film or the reactive ctching species. The lime to form a monolayer of any species on the surface is based on the assumption that every atom striking the surface remains there, which is not really Iruc except at low MEAN FREE PATH 9 temperatures, but does give us an accurate idea of the fluxes involved. The formation lime is obviously inverselv proportional to the impingemént lux. À typical sputter deposition rate is one monolayer per second (- 50'8 atoms per em? per second for a typical atom of 3 À diameter). This ra gas pressure of not much more ihan 10º torr. Benee, if a contaminant gas has a partial pressure of only 10"$ torr, every welcome sputicr depositing atom artiving at lhe substrale is accompanied by a contaminant atom! Additionally, the argon used as the spuitering gas may have a partial pressure typically fous orders of magnitude higher, so that each single condensing atom of the growing thin film may be accompanied by 10 000 argon atos. Tt is therefore hardly surprising that some of this argon is trapped in the growing film, ihough as we'l see later, we can control the amoumi of this trapped argon. By comparison. industrial vacuum evaporation process wheroby aluminium is deposited at aboul Uamjfmin in a pressure of 102 tor — which would conventionally be regarded as poor vacuum -- the fluxes of depositing aluminium and background gases at the substrate are about equal. it's all a question of relative rates : corresponds inan MEAN FREE PATH The mean fiee path is (he average distance travelled by a gas atom bevweea collisions with other gas atoms (Eigure 1-8) and obvionsly decreases at higher pressures (greater densities of gas molecules). At ) miorr and room temperature, Mean fres petha Colligion frequeney Ta Tigure 1-8. Random motion and coliisions of aa atom w GASES te mfp of argon is about & em and most other gases are within a factor of three of this. (These figures are relevant only for atoms of thermal energy). Kinetic theory assumes that there are no interactions between gas utoms other than during collisions. The probability of collision, and hence mean free path, depends or atom size, and so will vary from gas to gas. It's easy to derive an approximate value for 4 mean free path. If an ideal gas atom has a diameter of 3 À, it will collide 5f the centre comes within 3 À of the tre of another atom The cullision area therefore has a radius of 3 À or an 8 107!$ can? and this will sweep ont a volume of gas as the atom moves. At E mtorr, equivalent to 4 1012 atoms/ec, there is one atom in each 2.8 191º em?. So the average distance that the atom of 2.8 10-13 cm? collision area must move to swecp out a volume of 2.84 107! * cm? is 10 em, whichis close to the actual value for argon quoted above PROBABILITY OF COLLISION Just as there is a distribution of atom speeds, so there is a distribution of free path Jengths between collisions. We might like to know, for example, the probability P(x) of an atom travelling a distance x without colliding. But, con- sidlering events over the next smal) increment of lengtlt Ax, P6ct 2x) = P(x) x (probability of travelling Ax without colliding) = Pg (1 - 2X) hero 1/X is the probability per unit length of making a collision. By expanding the above expression and integratine, we find thar: Piso = *N (9) = expf- 5 ) since P(0, the probability of travelting zero distance withcut colfiding, is clearly unily. By « similar argument, and using (he result derived above, the number dn of atems colliding herweer x and x + dx is ) da = ngxemp(. YE n = noxexpi-ç ID o XexF AfR where no is the total number of atos. Let's now calculate the mean free path X between collisions. By the usual averaging process, Lian x=l San e eine ENERGY TRANSFER IN BINARY COLLISIONS n fe x col: Jos fio ol. Ps esl io = A (Integrating by parts) So we have the useful results that if the mean free path between collisions is À, then the probability of collision per unit length is 1/), and the probability of travelting x without colhision is exp(- x/A). Let's apply this finding to the sputter deposition of copper in a diode system. Suppose that the substrate is 5 cm away from the target electrode, and let's guess that the mean free path of copper in argor is also 5 cm at | mtorr. Under these circumstances, the probability of a sputtered atom reaching the substrate with- out colliding with ar: argon atom is u exp -. =exp- q = 037 wltich is al lcast a reasonable chance. However, we need raise the pressure to only 14 mtorr before the chance is less than the proverbial one in a million. Under these circumstances, transport is by diffusion. COLLISION FREQUENCY We've already met the idea of a collision Frequency in the discussion of mean free path. H is the average number of collisions which a gas atom makes per unit time and so is equal to T/A (Figure 1-8) in argon at 50 mtorr pressure, the collision frequency is 4 105 per second. ENERGY TRANSFER IN BINARY COLLISIONS We shall consider here the collision between two particles of masses mj and mr. Assume that my is initially stationary and that m; coltides with velocity vj at an angle É to the line joining the centres of mj and my at the moment of collision (Figure 1-9) By conservation of linear momentum mycos 8 =mju, t'mpo, (1 By conservation of energy Xmy? = em (uy? + vin? 8) + emu? (2) 16 GASES MEASUREMENT OF GAS FLOW RATE lt often occurs that we necd to measure gas flow rates in a plasma system. This might be because the system is not fitied with a low meter, or because an installed meter is either not calibrated for the gas df interest or is of dubious telability. This measure of flow rate cam usually be done without formal recati- bration, at least to a very good approximation É The prerequisites arc a knowledge of the volume of the chamber and of the pressuro within it. The volume of the chamber cam be reasonably estimated by Phtysical measurement, but 4 much more accurate technique isto expanda small known volume cf gas at atmospheric pressure into the evacusted chamber and note the resulting pressure. For example, using 10 cc of gas at STP, ihe final chamber pressure is 392. miliitorr. Since the number of gas molecules is con- served, then 0x 760 = 392 103 V. So Y = 19.4 litres To measure a given flow rate, the chamber is firsl evacuated and segled, At time t = 0, the gas is allowed to flow into the chamber and the resulting pressure Tise is noted as à function of time. This should result in à linear increase in Ssure, at least up to a few torr. Suppose, in the example above, à pressure rise “1 millitorr per second is recorded. Knowing the chamber volume, the flow rare is thus 3.1 10% x 19.4 1/s, which is 6 10 dsor48scem RESIDENCE TIME À parameter which depends directly on flow rate is the residence time 1 ofa gas molecule, i.e. the mcan Lire it remains in the process chamber before being Pumped away. TÉ the volume of the chamber is V litres and this is being pumped atihe rate of 8 litres/second, then the mean residence <íme is just V/S. Since S is not measured directly, we use instead: Vopv DV Sp Q To give a practical example, a plasma «deposition system might operate at p= Lorr, V = 100 litres, and Q = 160 scom (2 1/5). Then lx 100 = 4— = 50 seconds So the residence time can be quite substantial FLOW VELOCITY Es sometimes uscful to know the velocity of a gas, and by this we mean Lhe velocity of the stream, rathe: than the much higher instantancous vel oeities of CONDUCTANCE Y the constituent atoms; the flow velocity is analogous to the drift velocity of conduction electrons in a solid or gas. Consider the high pressure radial flow reactor shown in Figure 1-LH.1f v is the Now velocity of the gas stream down the central pumping port, n is the molecular density of the gas, À is the cross- sectional area of (he port, and Q the gas throughput, then G=nvà For typical values of QQ = 160 scem = 2 tl/s, n (1 torr) = 3.5 101º jce and A = 10 cm?, then v = 204 cmjsec, much siower than typical thermal speeds of 10º em/sec. CONDUCTANCE The conductance F of an orifice or à Jengtb of tubing is a measure of its ability to transmit a gas flow. It is deiined by: Q=FAp conductance is thus where Ap is the pressure differential across the element. & rather analogous to clectriçal conductance, and the relationship above is rather like Ohnt's Law, with Q being equivalent to the current, and Ap to the applied potential. A problem in using the concept of gas conductance is that, unlike the electrical analogue, its value changes according to conditions, which is a manifestation of the nature of the low changing from molecular to transition to viscons. Never- theless it is a useful concept if we remember the limitations. The same limitations apply to the closely related parameter of pumping speed. And although the conductances of certain geometries can be calculated for molecular flow or for viscous flow, they cannot as yet be calculated for the transition region, which is where, as we have seen, our plasma systems oflen operate. The glow discharge processes we're considering take place at relatively high pressures as far as high vacuum pumps are concerned, and so limiling conduc- tances are often used to reduce the operating prossure in the process chamber to a pressure more compatible with the working range of the pump. Let's consider bow this works: initially the proce: amber is evacuated to some base pressure limited by real and virtual Icaks of magnitude g. If the base pressure is Po and the pump speed S, then (Figure 1-12 a): a=pe8 A high vacuum system for plasma applications might typically have a pumping speed of 1000 1/s and be able to initially evacuate the system to 107 torr. In 18 GASES This caso the Lhroughput of the pumping system (equal to the leak tate) would be 105º torr litres per second, or about 2 107 molecules per minute. Let's now suppose that the plasma process calls for a flow rate of Q, and that by adding this flow of gas the chamber pressure rises to py, which must still be in the cfficient working range of the pump. Let's also assume that the pumping speed remains constant at S; both of these last conditions would be satisfied up toa pressure of about ! mtorr in an oil disfusion pump. Theo (Figure [12 by: a+0=pS The hypothetical practical process detailed above may cali for a gas flow of Q= 80 scem or 1 torr litre/second. Then . ato J0t+1 s 1000 We may have satisfied the flow requirement, but perhaps a higher operating pressure pa is required (if a lower operating pressure is required — hard tuck!). This higher pressure is realized by adding a limiting conductance F (Figure 1-12 c) at the throar of the pump such that: Pr = Imtorr Pr-pi = HQ+) Limizing Congucianee arq tt to [a tal te te Atdase pressure p,,0-9,8 Now add required gas flows, q + Q= p,S Then throttle to establisa required pressure p, 0 + 0=F (pp | Figure 1-12. Pressure-flow relationships GENERAL REFERENCES 19 So, if the required p; is [OO míorr, then E needs to be abou: O torr liires/ second, reducing the effective pumping speed by three orders of magnitude. But note that the pressure in the pump and the real pumping speciul arc unchanged by the throttling operation, remaining at p, and S respeclively GENERAL REFERENCES 4. D. Cobine, Gaseous Conductors, Dever, New York (1958) S. Dushman, Scientific Foundation of Vacuum Technique, 2nd edition, ed. J.M. Lafferty, Wiley, New York and London (1962) 3.K. Roberts and À. R. Miller, Heat and Thermodynarnics, Blackis, London and Glasgow (1960) 
