TA ! UNCLASSIFIED 2 1 15 . 12 ..! A . ORNL 6. 499 MIT Oran-p-499 : É CT 30 1964 PAR METER MSHHURATATS IN THIN DIELECTRIC FILMS J. V. Cathcart Mestils roul Coroico Division, Oak Ridyo Hition1 Laboratory Onk Ridge, Tenne 8800 37831 This paper attempts to suomarize the advantı:808 and linitutions or ellipcape try nod x-ray diffractions as methols for charootorizing thin films. The results of # study of oxide films on copper crystals are used to illustrate the kinds of information that can be obtained. In the x-ray studiesd a stendurd xorey diffractometer was employed equipped with a doubly bont Lif monochromtor and a thoriw-rictivated No I scintilation counter as a detecta. This wait was used to e:cominc approprinte Brass reflections from the cueo film. Ibe key requirement or the wtbod is that tube film be a single crystal or at least highly orientod. Rocking curves for such fllus yield a nensure of the moseic spread of tbe l'ils 418 well is thicknes3 lata (0.6., the thickness at #film formed at room temperature on copper was showa to be 20 A). Analysis of the Iloe contours obtained during a 20 scan of the specimens also gives information regarding the thickness, average strain, and strain gredient 10 the films. In contrast to an a-ray diffractometer, an ellipsaneteral 18 a relatively little known device. It is an optical Ipstrument, which operates on the ! . A priociple that when plade polarized light 18 rorlacted from a film-covered surface, 10 general, the components of the electric vector parallel and perpendicular to the place of incidence wdergo different phase changes Research sponsored by the v.s. Atomic Energy Commission under contract with the Union Carbide Corporation. TEGAL NOTICE - The report o n Ouwer worth Mother Outud montes, ne tietotu, meremangat Owl A. Mehm et oprocento them, a nd a lapte, wo maput more my, w u, Newstarted the mar, art and the - 1. A Honda m ot en stor rotting out we are, ut, men det in the report As with the per mette a whole a r Iekta Whega o mercato Owo, w what we courte, that o amplegenw wtrmotrat Online, a person med utrotar memores, me, ar non m odo per Meyer wat w the Center, er we played with controller cou suffer dirrerent (egrees are our lituie reluction un'interlection. Conscuently the reflected light is in genertl elliptically poli ricci, The thickne:33 tabu the refrictive lpiex of the pilin manly be coiucul i'rou the degree of clliyticity of the reflected bean. An ollipsometer 10 cupable of detecting changes 10 livoroxe 'i lze thic.nous of approx 1 do Itu measurements are bonuostructive and nocy be made fra distance. Only 'i relatively smu11 specimen surf, ce orea is required, 600 edge artucts are eliminuted altogether. For these ro: 9003 the ellipscue ter 180 londs itself to rote studies to which the growth of Illu is involved. The diondvegtegen of ud ellipsaeter are relateứ in lürge nca sure to the complexity of the optical equations which must be use1 to describe the reflection process.* Iven in dealing with the cnse of an 1sotropic film on an isotropic gubatrute, a computer 13 a virtual necessity. The equations applicable to optically is alsotropic systems have been worked out but have yet to be tested exporizontally. Reading3 taken with the ellipoameter are rolated to the relative amplitudo reduction, tan y, epi the relative pause retardntica, 4, surfered au nflactioq by the parallel and perpendicular cataponouto of the electric vector or the incident light; ten and A are lo turn related to tbe Fresnel coefficients for reflection for tibe syatan: ten y exp(14) RP/RS. be auperscripts p and s refer, respectively, to the reflection coefficiente at the parallel and perpendicular components of the E vector. po And Rº are rativer complicated functions of a number of variables, all of which are directly measureble except Dp and d, tube index of refraction and thickness, respectively, of the film, Dous (1) tm0 .p(10) - f'(0.g., is). Blace Ey. (1) contolna two unknowns, it cunnot ke solved directly, u graphical solution 1:y be obtained to follows: Ithoush a) Opo obtoins a 300 of tony Hou a vo lues by woasuriac tbe je juaptities for soveral films of differeot but wa'k Dowo thickpeas. d) a vulve is no sumed for og eind sets of ten and Aare colculated for severni film thickne 0908. c) The calculated epid.perimentul tuo v vs A curves are then cappared and the calculuted cw.vo adjusted (by selecting new value: for na) until the two curve 3 coincide. The matching of the curves 13 tüken as an indication that the correct volue of ng has been chosen. Edition (1) By then be used to 8830clate thickness values with each pair of ton and a values. specific probleus (13socisted with the use of the ellipsometer ere best illustrated hy the repults of u study of thin crizo films on copper,4,5 Both copper and cuprous oxide bove cubic crystal structures and in the bulk are optically isotropic, while attempting to mensure the thickness of such films an tiba (zu), (110), (111), upá (100) of a corper single crystal, we observed, bowever, thet for the (32) and (110), a rotation of the specimen about its surface normal produced periodic changes in the upparuit film thickness ladicated by the ellipsometer. Maxima and platua la tibo ppannt nilu tlakms occurred at 90° lutervals as shown in Fig. 1. anleg -. .a Kamination of the (301) and (110) under polarizing wicroscope cont immed the fact tibat tabe alde os tibien planes was optically anisotropic. A variation in the degree of anisotropy with film thickness was also noted with a maxim 1n and stropy occurring at about 300 A. No such bebevior wo oborved for the axide on the (10) and (100). It was possible to 12c1on1130 the inetö .11t: tively thiough : consilieri tion oj the - epitexiled stre390's between the onico in the wctrl. The 30 results illustiste the 319:3:31 bilities of using the elitesche tor to stuciy str: 1n raisotropy na chunges in tho legitu of uti in eni 30trolu in this fils. They :,190 under score, huurovo!, the ongor or' consi Jaring the oxide fis uniforra, homogeneous fllu in: onih stie the need for the developwont of a wore reulistic olitical woel for the l'1113 11' the full potential of the cllipuaxeter 1a to be roullzeit. Fortunetely, the uploution of the i-ray tochnique mentioned xbove provideu consideruble inaight into the structure:1 details of the film. The physicul jictue of the oxide as indicated by the i-r1.y deti rafty be suwrized og follows: In generul the oxide film 1s subject to cpitaxial forces in the 36 plane of the surflice, cousing tihe oxide to be compressed 10 a interad direction; normal to the surfuce there is a corresponding expansion of the lattico. Å strain grodient exists in the film, which implies the existence of a corresponding gradient in the index of rorruction of the oxide. Likewise, the average lattice circiteter of the film varies with thickness, implying that the average index of refraction of the oxide aldo 18 a function of file thickness, Ad attdrept was made to utilize these findings in the formulation of new optical model for the oxide. The film was considered to consist of u loyers each of which was treated as a bomogeneous, isotropic medium. The Index of refraction was allowed to vary from layer to layer, and the avere se Ladex of refraction of the film was assumed to change with film thickness. Applying the rocursion formulas developed by Wolters for reflection frow multi- Layered filma, ellipsometer data for oxide films on the (w) of copper were used to test the new model. Compared with the results obtained using the classical model, a better it was achieved between the calculated and experimental ton vs A curves, and the film thickness values were la mich better agreement with those obtained from the x-ray data. RUPERENCES NE 1. B. . Borie, acte Cryst. 13, 512 (19u); B. 3. Borie und C. J. Biarks, Ibid. 14. SOG (1961). 2, A, B. wiaterbottan, J. Out. soc. . 38, luminoth? (1913); K. Danske videnskabernes so luket skrifter, 1955 MP , ko . 3. P. L. McCrackin, . Paudeigldri, R. R. Stromwork, and H. L. 3telaberg, I. Res, Nat. Bux. 3ti... A. 1 313-317 (1963). 1. B. 8. Borta Ce J. Beras, and J. V. Cuthicart, cte Mat. 1, 6:4;1457 (16). 5. J. V. Othcart, J. E. my person, and G. F. Peterson, ctu Mit. lv, 6.49–7:3 (1963). 6. 1. Wolter, "Optix Jlnner Babichten," Ila nubuch der Physik, Vol. XXIV, pp. 461-34), Springer-Verlag, Berlin, Götingen-fleidelberg, 1956. . A CO T . L . . . M S . FIGURE C:\PTION: Mb. 1. Vri:tion of the Freint Olde Mlu mici.Ni 3.3 on How Crystallog'sible Flines of Courie!" 6.3 is junction of the ingle or Rotation r.bout the Film Nord, sifter Critheart, Ovier300, ani Petersen (reference 5). M8. 2. Vorlation of Degree of inisotropy of Oide on the (110) of Copper as e Punctiou of Film Thickne 33. ifter Cthcart, Epperson, and Petersen (reference 5). BY . . r . sinonim .. .. .. ... UNCLASSIFIED ORNL-LR-DWG 65311 - 100) APPARENT FILM THICKNESS (A) 60 120 480 240 300 360 ANGLE OF ROTATION ABOUT SPECIMEN NORMAL (deg).. Variation of Apparenf Oxide Film Thickness on Four Crystallographic Planes of Copper as a . Function of the Angle of Rotation About the Film Normal. „EIGT www.elektroninio pat meningkatan come se Mar ibor in the per te n e rti i ne....*. UNCLASSIFIED ORNL-LR-DWG 55313 OXIDATION TEMPERATURE DEGREE OF ANISOTROPY = .:[(max-TMIN)/TMAx] X 100 A 150°C o 200°C o 250°C o 300°C . 90 200 400 600 800 1000 · AVERAGE APPARENT FILM THICKNESS (A) 1200 Variation of Degree of Anisotropy of Oxide on the (110) of Copper as a Function of Film Thickness. Fig. 2 12 .mt! WETTER DATE FILMED 12/ 21 /64 : ' E . marrin . XLR RA E LEGAL NOTICE – 4 This report was prepared as an account of Government sponsored work. Neither the United States, nor the Commission, nor any person acting on behalf of the Commission: A. Makes any warranty or representation, expressed or implied, with respect to the accu- racy, completeness, or usefulness of the information contained in this report, or that the use of any information, apparatus, method, or process disclosed in this report may not infringe privately owned rights; or B. Assumes any liabilities with respect to the use of, or for damages resulting from the use of any information, apparatus, method, or process disclosed in this report. As usod in the above, "person acting on behalf of the Commission” includes any em. ployee or contractor of the Commission, or employee of such contractor, to the extent that such employee or contractor of the Commission, or employee of such contractor prepares, disseminates, or provides access to, any information pursuant to his employment or contract with the Commission, or his employment with such contractor. . 1. 30k L SYV 941 1 5 KU 4 UT . * NET 29 M 1.1 - - . . END 2.