CA1291443C - Method and target for sputter depositing thin film - Google Patents

Abstract

METHOD AND TARGET FOR SPUTTER DEPOSITING THIN FILMS
Abstract A sputtering target has a sputtering surface with first and second regions of respective first and second materials.
The first region comprises a surface of a first member of the first material, such as of a circular cobalt plate. The se-cond region comprises a surface of a second member of the se-cond material, such as a platinum ring. A cobalt cover ring clamps the platinum ring to the cobalt plate. By varying the relative sizes of the first and second regions, as by changing the size of the cover ring to expose more or less of the platinum ring, the concentration of the two materuals in a layer deposited from the target onto substrates is varied.
In addition, by imparting planetary motion to substrates during deposition and sizing and positioning the exposed por-tion of the platinum ring, a radial coercivity gradient is established in the layer deposited on the substrates.

Description

METHOD AND TARGET FOR SPUTTER DEPOSITING THIN FILMS

Technical Field This invention relates to sputter depositing of thin films, such as for thin film magnetic recording discs, and to methods and apparatus fo:r manufacturing such discs. More particularly, the invention relates to a composite source target and method for fabricating by RF diode sputtering a thin film magnetic disc having a magnetic layer comprising at least two materials, such as platinum and cobalt and having a radial coercivity gradient~

As disclosed in U.S. Patent 4,610,911 of James E.
Opfer and Bangalore Natarajan entitled "Thin Film Magnetic Recording Medial' issued September 9, 1986 and assigned to the instant assignee, the magnetic properties of a thin film magnetic recording disc having a cobalt-platinum magnetic layer with selectively desirable properties may be established in accordance with the thickness of an underlying chromium film, the thickness of the cobalt-platinum film, and the concentration of platinum in the cobalt-platinum film.

It is known that the chemical composition of a sputtered film i5 usually the same as that of the cathode (target) from which it is sputtered. See Handbook of Thin Film Technoloay edited by L.I. Maissel and R. Glang and published by McGraw-Hill Book Co., New York, New York (1970) at pages 3-28 and 4-39. Thus, to sputter, for example, a cobalt-platinum film containing about three percent platinum, the target would be a homogeneous composition of cobalt and platinum with the concentration of platinum in the target being about three percent. Alternatively, on page 3-29 of this ~,," ~k la text, it is suggested that sputtered compositional or alloy films can also be obtained by the use of multiple targets, each of a single material, and that a wide range of compositions can be obtained by independently varying the sputtering rates of the targets.

There are, thus, two suggested known prior techniques for providing sputtered compositional or alloy films: utiliz-,",~ ~ .

4~

zation of a homogeneous target o~ the material~, or utiliza-tion of two or more independent target3, there being a sepa-rate target for each material.
The first o~ the~s known techniques is limited and non-variable in that the percentage composition of the various material~ in the target i~ ~ixed and determine~ the composition o~ the sputtered film. Furthermore, in th~ case of cobalt and platinum, the cost of such a homogeneous target is prohibitive. The second known technique, while permittin~
some versatility in the composition of the sputered film, by varying tha targets, is also costly. Moreovar, problems are encountered in this second approach ~hcn it is desired to ~imultaneously deposit films on opposite ~ides o~ a disc or o~her substrate. For example, to ~putter a compo~ition of two material~ on the two ~ide~ of ~uch a disc, at least four targsts are required. ~hat i~, on~ 3et of two targets are required at each ~ide of tho di~c ~or sputtering thereon.
one target o~ ~ach set being of a fir~t material and the other target o~ each 8et b~ing of a ~econd material. In addition, each of the four target~ would have to be provided with a separate power supply ~ystem i~ each were to be independently controllable. Independent control i~ nece~sary in order to vary th~ ~putkering rate ~rom each target, and thus the percentage compo~ition of each ~aterial.
Furthermor~, an arranse~nt of multiple, discrete targets does not lend itsel~ to providing discs with magnetic films o~ radially-varying cvercivity. As explain~d below, this latter characteristic i8 particularly de irable for high density recording on magnetia r~cording dlscs.
When di~cs are u~ed in a typical magnetlc recordiny disc drive appliaction, the disc i~ annular, i5 rotated, and a read-writ~ head i9 positioned to fly over the disc 90 as to read or write on concQntric tracks on the disc. Tha speed o~
travel of the head, relativ~ to th~ di3c, i8 greater and the head flie~ higher over the di3c when the head i~ reading or writing onto outer track~ at outer diamet2rs of the disc in comparison to inne~ tracks at inner diameters. I~ tha write frequency is held constant, the recording density i~ much tg higher on tracks closer to th~ inner diameter of the disc in comparison to the density on tracks toward the outer dia-meter.
A~sum$ng a disc has a ~agnetic layer with a constant radial coercivity, such as understood to be pro~ided by the above de3cribed known techniques, writing on tracks near the outer diameter of the disc i~ impoasible or unreliable unless the writing current i~ increa~ed at ~uch outer diameter tracks. Increased writing current i~ required because, as explained above, the head ~lie~ abovQ tho disc surface as the head move~ outwardly ~rom inner to outer tracks of the disc.
In order to write with a con~tant current, which in many ap-plication~ i~ highly desirable, th~ radial coercivity o~ the magn~tic layer on the diac must bs ad~u~ted 30 a~ to derrease a~ the flying height of the head increa~e~. In other words, th~ coercivity of th2 di~a should decrease with incr~a~ing radial di~tances fro~ tha center o~ the disc.
The above described prior technique~ 8im~1y do not have ox sugge~t the provi~ion of a magnetic layer with a radial coercivity gradi~nt.

isclosure o~ Inven~io~
In accordance with ~he ~puttQring target and method of the present in~Qntion, a ~puttering targst has a ~puttering surface with ~ixat and ~cond r~gions of re~pectiv~ fir~t and second material3. In th2 illu~tr~t~d preferr2d embod:LmQnt of tha invantion, the targ~t co~prise~ a plat~ of a fir~t mate-rial onto whlch is mountsd, in contact therewith, a member o~
a ~econd matrial o~ a pr~determin~d geometric shap~. By con-trolling thQ area o~ th~ member o~ tha second material ex-posed to a ~ub3trate ~uring sputtering, r~lative to the area o~ the ~irst material expo~ed during ~putt~ring, the composi-tion o~ the depo~ited lay~r ~putt~red ~rom the target may b~
d~termin~d.
More specifically, the ~irst ~aterial may aomprise a disc Or cobalt, ~or axampl~, on ths 3ur~ace o~ which is mount~d a ring o~ ~latinu~. By controlling the area o~ the platinum ring which i3 expo~ed to a ~ub~trate during ~put-tering, relative to the exposed area of cobalt, the percen-tage composition o~ platinum in the ~puttered layer may be controlled and determined. Thu~, to ~orm a cobalt-platinum ~ilm having a platinum content averaging about three percent, the exposed area of the platlnum ring should constitute about three percenk of the total sputtering target surface area, the remaining ninety-~even percent o~ this area being cobalt.
Thi3 concentration o~ platinum i~ best determinad ~rom the ratio of the width of the exposad platinum ring to the width of the exposed cobalt areas.
As one mean~ of controlling th~ extent o~ the exposed area of the platinum ring, the in~r perim~ter of tha platinum ring is cla~ped in place on the cobalt disc by a cobalt cover ring af~ixed to the cobalt disc. The cobalt di~c, platinum ring, and cobalt co~er ring ~rQ concQntric with one another.
In addit1on, in on~ sp~cific for~ ~hown, the outside dia~eter of the cobalt co~er ring i~ 1~8~ than the insid~ diameter of the platinum ring- It will thu~ ba understood that the ex-po~ed area of the pl atinu~ ring i~ directly deter~ined and controlled by controllin~ the out~id~ deameter of the co~alt coYer ring. That i8, the out~idQ d~ameter o~ tha cobalt cover ring may be altered to achiov~ sxposure o~ any desired ar0a of the platinum ring. Thu~, by decrea~ing the outside dia~eter of the cobalt r~ng, tho ar~a o~ exposed platinu~ is increased. Conver~ely, by increa~ing thQ outside diametsr o~
th~ cobalt covQr ring, the expo~ad area o~ the platinu~ ring i~ d~crea~ed. Altornately, the cobalt cover ring may be sized larger than the platinu~ ring. In thi~ case, th2 outer perimeter or margin of the platinum ring i~ clamped to the cobalt disc by the cover ring. Thu~, by controlling the inner dlamater o~ ths cobalt cover ring, thQ exposed area of the platinum i~ controlled.
Thus, tha sputt~ring ~urfa~e o~ the t~rg~t in this caqe ls compri~od o~ only cobalt and pl~tinum. Mor~over, the ex-tent o~ tha exposure o~ platinu~ is readily controlled and pr~determined as desirQd. In addition, with ~uch a two mate-rial compo~ita tar~t configuration, it is pos~ibla to depo-sit a layer on a ~ubstrate with a percentage concentration of ~he Recond ~ub~tanca which varie~ in a controlled manner at di~ferent location~ on thQ ~ub~trata. Thi~ variation is achieved by varying thQ location o~ th~ axpo~d portions of the member o~ the s~cond substance rQlativa to th~ ~xposed portions of tha plat~ o~ the firat ~ub~tance. A~ a ~pecific example, a~sume the abov~ described ring configuration and that annular disc substrate~ are ~upported for planetary mo-tion with thQ cent~r o~ ~otlon o~ th~ sub~trate~ nearly cen-tered on tho c~nt~r o~ the ring. In thi~ ca~e, thQ platinum concQntration in the sputtered layer variea in the ra~ial di-rection. This variation in platinum concentration provides a radial coercivity gr~di~nt ~ro~ inner to out~r diamet2rR of the dis~ ~ub~trate~. Moreov~r, a8 thQ ~ize 0~ tho axposed platinum ring i~ changed to shl~t the centQr o~ th~ exposed ring away ~ro2 tha center o~ th~ sub~trat~, th~ radial coer-civity gradi~nt i~ changsd~ Furthermore, the coorcivity gra-di~nt which re~ult~ ~ro~ ~ pArticular platinu~ cobalt target configuration ~ay bQ ~xparim~ntally determinad or b~ predic-ted with 80~Q acouracy by ~ath2~atlcal modeling technigues.
Itis therefore an object of an aspect of the present invention to provid~ an improvad spuktering targ~t and method for ~put-tering at le~st two mat~ri~ls onto a ~ubatr~t~.
It is an object of an aspect of the present invention to provido a sputter~ng ~ethod ~nd t~rgot ~or ~puttoring a ~hin ~il~ magnetic r~cording di~c with a layer o~ magnet~c materi-al havinq a coercivity which Yarie~ in th~ radial direction on th~ sub~trate.
It is an object of an aspect of the present invention to provid~ an improve~ ~puttering ~thod and targst ~or sput-tering a co~ponit~ layer on ~ ~ubstra~o in which tha porcen-taga concentration o~ th~ co~ponente o~ ~h~ lay~r ara accu-rately and o~iciently controll~d.
It is an object of an aspect of the present :invention to provide a relativoly low coet sputtering targ~t and ~ethod for sputtering a ~ilm o~ at lea~t ~wo conntituent materials on a subetrate.

5a Various aspects of the invention are as follows:
A target for sputter-depositing a magnetic layer having a radial coercivity gradient on a substrate which is moved relative to the target, said magnetic layer having at least two constituent materials, said target comprising:
a circular disc member of a first diameter which is formed of a first constituent material;
a ring member having an inside diameter and an outside diameter smaller than the first diameter, the ring member being formed of a second constituent material and being disposed on the base member concentrically therewith; and a clamping ring member having an outside diameter which i6 less than the outside diameter of the ring member and greater than the inside diameter of the ring member, the clamping ring member also having an inside diameter which is less than the inside diameter of the ring member, the clamping ring member being formed of ~he first constituent material and being mounted to the base member concentrically therewith and overlying a portion of the inner perimeter margin of the ring member so as to cIamp the ring member to the base member whereby the sputtering surface of the target comprises only the first and second constituent materials in a predetermined ratio of exposed areas thereof.

A method of depositing a magnetic layer having a radial coercivity gradient on a planar substrate having two sides comprising the steps of:
mounting a plurality of circular substrates for planetary motion on a substrate carrier, said substrate carrier being rotatable ~ t an axis normal to an~ through its center, of said substrates being mounted for rotary motion about ,~

~L29~L~43 5b individual, circumferentially spaced axes which are parallel to and radially displaced from the substrate carrier axis of rotation, each of said substrates being mounted in front of a circular aperture in said substrate carrier, the diameter of said aperture being greater than the diameter of said substrates;
preparing a pair of sputtering targets, each of said sputtering targets having a sputtering surface comprises of a circular disc of a first material and a concentric ring of a second material mounted thereon, said pair of sputtering targets being oriented in the vertical plane disposed in spaced apart relationship having the sputtering surfaces opposing;
disposing said substrate carrier in the vertical plane between said opposing sputtering surfaces such that each side of said substrates is exposed to a sputtering surface; and rotating said substrate carrier during sputtering thereby imparting planetary motion to said substrates about the axis of the substrate carrier, the centers of said substrates being substantially centered over said ring during such planetary motion.

A method of sputter depositing a magnetic layer having a radial coercivity gradient on a substrate comprising the following steps:
exposing said substrate to a sputter target having a sputtering surface comprised of a circular disc of a first constituent material and a concentric ring of a second constituent material mounted thereon: and simultaneously imparting planetary motion to said substrate during sputter deposition of said magnetic layer with the center of said substrate ~9~44~
5c being substantially centered over said ring during such planetary motion thereby varying the relative concentration of said constituent materials in said layer in radial directions from a predetermined location on said substrate.

,... . .

L4~3 These and other feature~, ob~ 8Ct9 and advantage~ of the present invention will become apparent with re~erence to the following description and drawing~.

Bri~f Description o~ the Drawinas Fig. 1 is a front elevational view o~ one e~bodiment of a system for making thin film ~agn~tic discs and other prod-ucts in accordance with the pre~ent invention;
Fig. 2 i~ a front i~ometric view o~ a load cha~ber of Fig. 1;
Fig. 3 is a side elevational view of a load chamber of Fig. 2, taken in the direction o~ llnes 3-3 of Fig. ~ to ~how a ~ubstrate pass through opening through which sub3trates are trancf~rred to an ad~oi~ing chamb~r of th~ ~y~tem;
Fig. 4 i~ a front iso~etric view o~ a dQposition cham-ber hou~ing o~ ths ~yst~m;
FigO 5 i~ an i~ometric Yi~W 0~ a valve a~embly used to interconnect the chambers o~ the 3y~tem o~ Fig. l;
Fig. 6 is a vertical sQctional view of a portion of th~
valv~ a88embly 0~ Fig. 5, tak0n along line~ 6-6 of Fig. 5;
Fig. 7 1~ a cros~ seational view 9~ the valvQ assembly of Fig. 5, taken along lin~ 7-7 o~ Fig. 5:
Fig. 8 i~ a varti~al ~ctional view of a portion of the valve a~e~bly o~ Fig. $, taken along lin~ 8-8 o~ Flg. 7 to show a gate portion o~ the va}ve a~sembly;
Fig. 9 i~ a rear elevational view of a radio fre~uency sputter deposition cha~ber o~ the sy~t~m of Fig. 1;
Fig. 10 i~ a vertical ~ectional view of the radio fre~
guency depo~ition chamber o~ ~ig. 9, taken along lines lO-lO
o~ Fig. 9;
Fig. 11 i~ a ~ront el~vatlonal vi~w of a wat~r cooling ~acket portion o~ on~ Pvrm o~ a radio ~re~uency sputt~rlng target a~embly utilized in the radio rrequency deposition ahamb~r o~ Fig. 97 F~ g. 12 i3 a ~ront elavational viow o~ kh~ sur~ac~ of the target oX on~ ~orm of a radio ~reguency depo~ition target assembly utilized ~p the radio ~xeguency depo~ition chamber of Fig. 9:

L4~3 Fig. 12a ttenth sheet of drawings) is a diagram showing variables in a mathematical model for calculating the percentage concentration of two substances, sputtered by the target of Fig. 12, at a point on the surface of a substrate;

Fig. 13 is a sectional view of a portion of the target of Fig. 12, taken along lines 13-13 of Fig. 12;

Fig. 14 i8 a rear elevational view of a direct current sputter deposition chamber of the sy~tem of Fig.
1;

Fig. 15 is a vertical sectional view of the direct current deposition chamber of Fig. 14, taken along lines 15-15 of Fig. 14;

Fig. 16 is a vertical sectional view through the load chamber of Fig. 1, taken along lines 16-16 of Fig.
l, and showing the load chamber loaded with a racX or tray of substrate carriers;

Fig. 17 is a vertical sectional view of the chamber of Fig. 16, taken along lines 17-17 of Fig. 16;

Fig. 18 is a cros~ sectional view o~ the chamber of Fig. 17, taken along lines 18-18 of Fig. 17, and with all but one of the substrate carriers removed;

Fig. 19 is a partially exploded isometric view of one ~orm of substrate carrier utilized in the system o~
Flg. I to support substrate~ as they are processed by the system;

~29~443 7a Fig. l9a (twentieth sheet of drawings~ is an isometric view of an altPrnate form of substrate carrier utilized in the system of Fig. 1 to support substrates as they are processed by the system;

Fig. l9b (twentieth sheet of drawings) is a vertical sectional view of the substrate supporting portion of the carrier of Fig. l9a, taken along lines l9b-19b of Fig, l9a;

Fig. 20 i5 an exploded view of a carrier loader for transferring substrate carriers from a tray to a transporter which then transfers the carriers from the load chamber to the deposition chambers of the system of Fig. l;

Fig. 21 is an exploded view of a lift-lower bellows mechanism of the loader of Fig. 20, which is utilized for lifting substrate carriers from, and for lowering substrate carriers to, the rack;

4~

Fig. 22 is an exploded view of a ~eed through utilized to deliver operating ~luid to the li~t~lower bellowc mecha-nism of Fig. 21;
Fig. 23 is a sida ~levatlonal view o~ one ~orm of a transportQr, transportar track, and transporter drive mecha-nism which tran~fers tha substrate carrier3, and thereby the substrates, betwsen the chambers of the sy~tem of Fig. 1:
Fig. 24 i~ a vertical ~ectional vi~w of the trans-porter, track, and transportQr drive mechanism taken along lines 24-24 o~ Fig. 23;
Fig. 24a 1~ a side elevational vi.ew of a portion of the transportor, track, and transport2r drive mechani~m, taken along lines 24a-24a of Fig. 24, but with the transporter shifted to a po~ition in which the tran3porter i8 ready to cros2 ~rom the chamb~r in Fig~ 23 to a chamber to the right of this chamber:
Fig. 25 is an ~xploded view o~ the transporter o~ Fig.
23;
Fig. 26 is an i~ometric vi~w o~ an end portion of a plunger which li~t8 thQ ubstrate carrier~ from the tr~ns-portar and rotate3 the sub~trat~ carri~rs during depo~ition, the plunger being ~hown in Fig. 26 in po~ition for in~ertion into a hub o~ a ~ub tr~te carrier:
Fig. 27 is an i~ometric ViQW 0~ an end portion of the plunger o~ Fig. ~6, tha plunger being ~hown in engagement with the hub Or the sub~trate carrier;
Fig. 28 i~ an ~xploded view o~ the plunger o~ Fig. 26 and o~ a plunger drive mschanism which operates tha plunger:
Fig. 29 is a schematic diagram o~ a water cooling sys-tem utiliz~d in tha sy~tom oP Flg. l;
Pig. 30 is a schematia illu~tration o~ a portion o~ the water coolinq ~y~tam ~or radlo ~r~quency sputtQring targets o~ the type ~hown in ~lg~. 9-13;
Fig. 31 i~ a schematic lllu3tration o~ a portion o~ the water cooling system ~or direct current cathode ~puttering targets o~ the type shown in Fig~. 14 and 15:
Fig 32 i~ a ~he~atic diagram o~ a vacuum system uti-lized in the ~ystem o~ Fig. l;

Fig. 33 is a block diagram of a s~eond embodiment of a system for making thin film magnetic di~c~ and other products in accordance with tha inventlon; and Fig. 34 18 a block diagram o~ a third embodiment o~ a system for making thin ~ilm magnatic discs and other products in accordance with th~ present invention.

Modes for Carryinq out the Invention General De~criptlo~ o~ First ~mbodiment By way o~ a speei~le examplo, th~ ~ysta~ and method of the present lnvention will be deseribed with re~pect to sevs-ral pr~erred embodiment~ in an applieation in whieh plural layers Or materials are deposited by vacuum depo~ition in a low pre~ure gas environment upon ~ ~ub~trate tQ ~orm a thin film magnetie reeording dise. How~ver, it i8 to be under-stood that thQ ~y~tem and ~ethod i~ not li~ited to thi exem-plary appllcation. That i~, the method and y~te~ i~ us2~ul generally wh~n i8 i~ desired to vaeuu~ depo~it succe~sive layers o~ material~ upon a sub~tra~e. By way o~ additional ex~mple~, ~ueh applieations ineludQ the manufaeture of thin film optieal reeording di~e~, lntegr ted eireuit ~anu~acture, and the manu~aetur~ o~ othsr produet~.
In genaral, vaeuum depo~ition, within the meaning o~
thi~ app}i~ation, amploy~ a mechanls~ ~or e~ecting atoms of coating mat~rial ~rom a ~ource Or target in a low pre sure ga~ environmsnt. Th~ coating material ato~ are e;ected with su~ici~nt energy to travel to the ~ur~ace o~ a substrate for depo~ition th~r~on. Vacuum depo~ition thereby includes tech-nique~ such ~ sputtering (including DC ~putterlng, RF sput-tering, r~active ~puttering, et~.), evaporative deposition, ion plating, and neutralizod ion bea~ coating. ~t does not ordinarily lnclude chomical vapor dopo~ition, alectroplating, or rapid ~olidi~iaation coating t~chniquo~. Ion plating i8 a variation o~ both ~putt~rlng and avaporativo depo~ition which involve~ th~ ioniz~tion o~ atom~ in th~ v por ~ollowed by at-tra~tion o~ ~om~ ~brtion o~ th~ ionized atom~ to the sub-strate with an olec~ric rield. Sinc~ ~put~ering is ~he most ~L~9~ 3 important vacuum deposltion m~thod used in the present inven-tion, and i5 repre~entativ~ o~ the oth~r method~, the re-mainder of this de~cription will concen rate on sputter depo-sition. However, the principle~ di~cu~sed hereinafter are to be conRidered a~ equally applicable to all vacuum deposition technique~.
With ref~rence to Fig. 1, a first ~mbodiment o~ the system 10 includes plural vacuum chamb~rs, which in this form include~ 8iX such chambers 12 through 22. These cha~bers are supported by a ~rame 24 ln a side-by-~ida relationship.
Ad~acent chambers are connected together by, and communicate with one another through, a tran~r p,aa~ag2way auch as valve containing hou~inge 26. Each o~ the~e valve hou~ings 26 in-clude~ a valve 2a, OnQ being ahown in da~hed lin~s in Fig. 1.
When a valve 28 between two ad~acent cha~ber~ i~ open, the adjacent chamber3 com~unicat~. with one another through the valve hou~ing 26. This per~it~ tha tran~r o~ sub~trates through th~ valve hou~ing and betw2en th~ chamber~. Con-ver ely, when the Va1Y~ 28 i~ closed, th~ adjacsnt chambers are i~olated and ~ealed by thQ valv~ fro~ on~ another. Valve 2 8 i8 operated b~twean its op~n and clo~ed po~ition~ by a solenoid controlled pn~umatic cy}inder 30, on0 o~ which is alco ~hown in da~hed lines $n Fig. 1.
Each o~ the chambQra 12 through 20 i~ provided with an independently controllable separata simil~r high vacuum pu~p-ing ~tack 34 ~or drawing a vacuu~ in the a~60cla~ed chamber.
An ind~pendently controllabl~ vacuum pumping ~tack 36 18 also provld~d ~or ~stabli~hing a ~acuum in th~ cha~ber 220 There-~ore, whenevar the valvo~ 28 a~30ciated with a particu~ar chambsr are clo~ed, tho vacuum pumping ~tack associated with that chamber i~ capable o~ ad~usting the pr~sure within such chamber to a de~ired magnitude. Furthermore, this ad~ustment may be made ind~pendently o~ the pre~urQ which exi3t8 in other ohamber~ o~ khe systQm. 0~ cour30, a ~ingl~ pumping stacX may alternately b~ u~ed ~or drawlng ~ vacuum in more than on~ chamber.
In tho ~yste~ o~ Fig. 1, cha~bsr 12 compri~es a sub-strate loa~ chamber m~an~ into which ~ubstrate~ are loaded for processing by the ~ystem. Al~o, chamber 22 comprises a sub~trate unload cha~ber mean~ from which processed sub-strates are removed from the ~y~tem. In addtion, the cham-bers 14 through 20 comprise proce ~ing or deposition chamber means within which layers of mat~rial are deposited onto sub-strate~ while po~ition~d therein. More specific~lly, each of the chambers 14 through 20 comprises a ~puttering chamber within which material from sputtarin¢J targets is sputtered onto the substrates. Further morQ, :in the speci~ic illu5-trated embodiment, becausa of the type of material being sputtered therein, chambers 14, 18 and 20 comprisQ DC
sputtering chambers whll~ chamber 16 compri~ea an RF
aputtering chamber. A pair o~ DC sputtexing cathode assemblie~ 40 are ~ounted by a circular support pl,~te 38 to the front o~ each o~ the chambera 14, 18 and 20. In addition, an RF sputt~ring cathod~ a~sambly 42 ia mounted by a circular aupport plate 39 to th~ front o~ tha chamber 16.
Similar a~semblies are mountQd to the rsar of thes~ chambers.
These as~e~bliea may bh re~dily r~placed by ~imply removing the support plata~ 38, 39 and replacing thQ aa~embliea with other asse~blie~ mounted to similar plates 38, 39.
During proce~aing, aub~trates pas~ along a proces3ing pathway through the chamb~r~ an~ are positioned between the front and rear aputtering a~ssmbliQ~ in the depo~ition cham-bers 14-20. Whan in ~uch chambers, both ~ides of th0 ub-~trate~ are si~ultaneou~ly depo~ited. That i~, the front cathode sputtering a6~e~blie~ depo~it a layer on a ~ront sur-face of each ~ubetrate and the rear cathode sputtering as~em-blies depo~it a layer on a rear surface of each substrate.
A~ explained in greater det~il below, in general, cham-bers 14 through 22 are ~vacuated with the valve 28 isolating chamber 12 ~rom ahamber 14. Substrates to be proce~ed are loaded in chamber 12 an~ then thi~ cha~ber i8 evacuated.
Thereafter, th~ sub~trates are transporte~ ~orm chamber to chamber ~or proce6sing. B~cau~ the chambers are isolatable from one another by the valves 28, tha desired operating parameters may be ~tablished within each chamber for the de-position to be per~ormed therQin. At the ~ame time, other ~L~9~3 parameter~ may be establi~hed in other chambers to optimize the deposition being performed in such other chambers. Fur-thermore, because of the isolation capabilities of the system, two ad~acent evacuated chambers may be i~olated from the other chamber~. In this ca~e, the ~ubstrates may be transported through an o~en valve 28 betwsen these chambers without losing the vacuum in either of the two chambers.
The isolation capabilitie~ of the cha~bers facilitates maintenance of the sy~tem. During the rapair or replacement of cathode assemblie~ in one or more cha~ber~, such chamber or chambers may be isolated from the other cha~bers by the valve 28 and then exposed to the ambient environment durlng thQ maintenanca proceduras. A3 a re~u:lt, thQ cathodQ assem-blies in ths othQr cha~bers are isolat~d ~rom the a~bient en-vironment and are ther~ore not expos~d to contaminants such as water vapor and oxygen. In addltion, bacaus~ o~ th~ iso-lation, a high vacuum can be mainainQd in all o~ the cha~bers except tho e being repaired. Following repair, 1es ~yst~m down tima is re~uired becau~e one does not hav~ to re establish a high vacu~m in all cha~ber~ o~ the sy~tem, but only in those chambers af~acted by t~e maintenance.
After a batch of sub~trate~ have been proces~ed, they are removed fro~ the unload chamber 22. During such removal, the unload chamber i~ isola.t~d from the ad~acent processing chamber 20 ~o that proc~ing may continue during the un-loading operation.
The per~or~anc~ of the depo~ition proces~ is monltored and controll~d utilizing a control subsystem including a pro-grammed digital computer 46 in con~unction with one or more terminals 48. Line 50 schsmatically repr0s~nt~ data lines along which signal~ are transmitted ~rom sy~tem sensors and other system compon~nt~ to the control 3ubsystem~ In addi-tion, llne 52 ~chematlcally rapr~s~nt3 control lines along which control signal~ are tran~mittad to the sy~tem for con-trolling the operation o~ valves ~nd other components o~ the system duxing ~yst2m op~ration. The programming o~ computer 46 i9 explained be~bw.

~29~3 Load. Unload, and De~o~ition Chamber~
With rQferenc2 to Fig~. 1 through 3, the hou ing ~or load chamber 12 i~ generally of a rectangular box-lik~ con-struction having ~irst and ~coned vertical side walls 56, 58, horizontal top and bottom wall~ 60, 62, and a rear wall 64. In addition, a perimeter flange 66 i~ attached to the front edge~ o~ the top, bottom and side wall and surrounds an opening leadlng to the interior of the chamber. A door 68 i~ mounted at one ~ide by hinges 70 to the flange 66. The door includes a perimet~r ~lange 72 which abuts the ~lange 66 wh~n the door i3 closed. A seal 67 ~Fig~. 16, 18) is pro-vided b~tw~n th~ ~langes 66 and 72 to tightly ~eal the door 68 again~t ths chamber flange 66 when thQ door i~ closed.
pair of latchQs 74 are pivotally mounted to th~ free edge of the door. When plvoted to a latch~d pos~tion, as ~hown in Fig. 2, latch roll~r~ 76 o~ the~ latche~ abut the rear sur-face of the ~langa 66 and aecuxe the door clo~ed. The lower edge of the door i9 suided to it~ closed position by a roller 78 supported by a bracket 30 a~ to project ~orwardly form the lower edge o~ the p~rimeter flange 65. Therefore, the door is guided to its clssed position and tightly held in place when latched.
A~ bo~t shown in Fig. 3, the wall 58 is provided with a v~rtically elongated ~ub~trat~ pasa through opening 82.
opening 82 co~municate~ with the interior o~ the valve housing 26 when the ~yetem iB asssmbled as ehown in Fig. 1.
A similar pa99 through opaning i~ provided through the ad;a-cent ~ide wall o~ th0 ad~oining depo~ition cha~b~:r, a~ ex-plained below. Th~re~ore, when the valve 28 i~ open, the two chamber~ communicate with one another through the~e pas~
throuyh opening~ and the valve hou~ing. Aa a re~ult, when the ~alva 28 i3 open, the trans~er o~ ~ubatrates between ad~acent chamberY i~ par~itt~d.
The bottom wall 62 o~ chamber 12 is providod with an opening 83 ~Fig. 3) through which a vacuum i~ e~tablished by the pumping ~tacX ~ (Fig. 1). A cylindrical pumping stack attachment ~lange 84 ~urround~ opaning 83. Flange 84 pro-, , . .
, .

jects downwardly ~rom the bottom wall 62 and, as shownin Fig. 1, the pumping stack 34 is attached to flange 84.

Sealed view ports 86 are provided through the top wall 60 and side wall 56 to enable an operator of the system to visually inspect the interior of chamber 12.
Ports, one being indicated at 88, through the rear wall 64 o~ chamber 12, are provided for passage o~ system components such as transporter drive mechanisms and loader drive mechanisms into the chamber. In addition, other openings, not shown, are provided for pressure gauges, air supplies and the like.

When the chamber 12 is mounted to the frame 24, a pair of support bars 90, connected to the undersicle of chamber bottom wall 62, rest on a horizontal plate portion of the frame 24. This provides a stable support for the chamber. The frame itsel~ is leveled so that the chambers are aligned vertically and the~openings 82 are in a straight line.

The unload chamber 22 is a mirror image of the chamber 12 and for this raason will not be described in detail .

With reference to Figs. 1 and 4, all the deposition chamber housings are of similar construction. For this reason, the deposition chamber housing will be described with re~erence to the housing ~or chamber 14 shown in Fig. 4. Further~ore, the deposition chambers are similar to the load and unload chambers 12, 22.
Therefore, components o~ chamber 14 which correspond to similar components o~ the unload and load chambers are correspondingly numbered.

,~ A 7 ., ' ~

~;~9~4~3 14a Deposition chamber 14 differs from the load chamber 12 in that it lacks a hinged door and a perimeter flange 66. Instead, a front plate 92 i5 provided at the front of the deposition chamber. The front and rear walls 64, 92 of chamber 14 are provided with circular openings 94, 96. The sputtering assembly support plates 3~ and 39 are secured to walls 64, 92 to close these openings and mount the sputtering assemblies 40, 42 in position for deposition within the chambers. Also, because the chambers 14 throuqh 20 are each intermediate to chambers adjacent to each side wall thereof, openings 82 through which the substrates pass are provided throuyh each of the side walls of the~e chamber~. Cen-~equently, substrates may be passed from one chamber to the next during operation o~ the system 10. The top wall 60 of the deposition chamber~ i~ detachably mounted to a flange 61 provided at the upp~r edges of the chamber front, rear and side walls. A ~eal i8 po~itioned between these co~ponents 60, 61. Access to the intQrior o~ the deposition chambers is thereby provided fro~ above.
Each of the cha~ber~ 12 through 22 are o~ rigid durable constru¢tion and are form~d o~ a 3trong mat~rial ~uch as, for example, stainles~ steel or aluminum.

Isolatio~ Valve9 The valve as~embli~ for salectively i801ating the re spective chambar~ 12 through 22 ~ro~ each other are illus-trated in Fig~. 5 through 8. A3 pr~viou~ly mentioned, each valv~ a~sembly include~ a valve housing ~6 within whieh a valve 28 i~ poaitionad and operated by a pneumakic cylinder 30 to selectiv~ly open and close th~ val~ hou~ing. When the valve i~ open, a pathway 1~ provided through tha valve housing and batwe~n ad~ac~nt chamber~0 Con~er~ely, when the valve i~ closed, th~ ad~acent chambers are isolated, that is sealed, fro~ ono another.
~ ore ~pecifically, tho valv~ housing 26 includes a first hollow box ~ction 100 which de~in~s an internal first valve passag~way 102 and a ~econd hollow box section 104 which defines an internal second valve pas~ageway 106. The valve hou~ing al30 include~ a hollow bonnet 108 intermediat~
the sections 100 and 104. The valv~ passageways 102 and lOÇ
communicate with on~ another through the valve bonn2t ~xcept when a valve 28 compri~ing a gate valve 110 is ~hi:Eted to a closed po~ition, a~ ~hown in Flg. 7. When cloae~, the valve 110 seal~ rir3t valve pas~ageway 102 ~rom the ~econd valve pa~agaway 106.
~ he ~irs~ and second valvo pa~ag~way~ 102 and lOfi are o~ the sams cross 3ectional size and shap~ as thQ chamber ~ide wall pa53 ~hr~gh openings 82. For that matter, in the illustrated embodiment, t~e opening~ 82 are Rized to permit 9LX9~443 the passage of components which are three inches (7.62 centimeters (c~)) wide and twenty-two inches (55.88 cm) high.
The valv~ section 100 is provided with an attachment flange 112 which is ~Qcured to a wall 58 of one of the chambers with the chamber pass through op~ning 82 aligned with ~he first valve pa-q~ageway 102. ~180, the valvs section 104 i~ pro-vided with an attachment flange 114. Flange 114 is secured to a wall 56 of an ad~acent chamb2r with the chamber pass through opening 82 aligned with tha valvs pa~sageway 106.
Seal~ 113 and 115 seal the connection betwesn the respective flanges 112, 114 and wall8 58, 56.
Therefor~, with th~ valve 110 ~oved to its op~n posi-t$on 3hown in da~hed line~ in FigO 7, 3ubstrate~ may ba transferred through the valve housing 26 between ad~acent chambers such as in the direction indicated by the arrows 118. Convar~ely, whon the valve i~ in ths clo~ed pocition shown in Fig. 7, th0 ad~acent cha~ber~ are 6ealed from one another by the valve. When ~ealed, ~ubstrate transfer b~tween the chambers i~ blocked and dif~erent ga~ pressure environment~ ~ay be maintained in the chambers. The valve 110 provides s~ective ~ealing betwe~n the c~ambers. The illu~tratsd valve ha~ a maximu~ leak rate of 1 ~ 10-9 atmosph~res p~r cubic centi~ter per ~econd when sealed against a one atmo~ph~re di~erential in either direction across th valve.
The bonnet section 108 i~ o~ rectangular box-like con-struction with parallel spaced apart ver~ical ~ide wa~ls 122, 124 and an end wall 126. The other end of the bonnet section is clo~ed by a cover 133 mounted to a ~langs 132. ~ top wall 128 and bottom wall 130 co~plete the bonnet. One ~ur~aca o~
the valve 110 engage~ the interior ~urracs o~ wall 124 as the valve is moved b~tween open and clo~ed positions. A valve ~eal 134 carried by valvo 110 i8 po~ltioned between the valve and wall 124. Seal 134 surrounds the valve pa~sageway 102 for ~ealing purposes when the valv~ i~ clo~ed. Rollers, ~or exampl~, 138 in Figs. 7 and 8, bQar against interior surfaces o~ the bonnet wal~ 122 and urga the valvs 110 against the wall 124. More ~p~ci~ically, the roller~ 138 are pivoted to ~:9~L~413 valve 110 by links 139 (Fig. 7). As th~ valve approaches a closed position, the roller~ 138 which lead the motion abut the end wall 126. Continued motion of the valve 110 causes the link3 139 coupled to such roller~ to pivot so that rol-lers 138 bear again~t wall 122 and urge valve 110 against wall 124.
A~ previou~ly mentioned, a cylimder 30 is l~tilized to shift the valve between its open and closed po~itions.
Cylinder 30 is pneu~atically operated and, a~ shown in Fig.
6, ha~ a piston 140 positionad within a cylinder housing 142.
A pi~ton rod 144 extends ~rom pi~ton 140, thro~lgh a seal, and into ths bonnet 103 wherein the end o~ the pi~ton rod en~age~
the valve 110. An air~low valve 146, controlled by a 301enoid 152, directs air either through a conduit 148 or a conduit 150. With air directQd through conduit 148, the piston 140 is shlfted to the right a~ ~hown in Fig. 6 and the valve 110 is open. Conver~ely, with air directed through conduit 150, the piston 140 i~ shi~ted to the le~t in Fig. 6 amd the valvQ 110 i~ clo~d. Solenoid 152 control the position of the valve 110 in re~ponsQ to control signals generated by th~ computer 46 (Fig. 1). Conductor~ 154 deliver power to the sol~noid.
First and ~econd valve po~ition sen~ing limit switche~
156, 158 ar~ provided ~or dekecting the respective open and closed position~ o~ the valve and tran~itting a signal indi-cating the valve po~ition to the computex. With r~erence to Fig. 6, when the valve i8 in an open po~ition as shown in this ~igure, a spring biased ~tem 160 o~ ~en~or 156 is posi-tioned in an annular groove 162 ~ormed in ~he pis~on rod 144.
When th~ ~tem 160 is in this position, a valve open indica-ting Aignal i9 transmitted by th~ eensor 156 to the computer.
At the same time, the etem 164 o~ tha sen~or 158 is held in a retracted po~ltion by the pi~ton rod 144. In contrast, when the valve iB in a closed po~ition, the ~te~ 164 i~ positloned in an annular groove 168 ~or~ed in the piston rod. When ste~
164 i~ in groove 168, a valv~ clo~ed indicating ~ignal is ~ent from the ~ens~F 15~ to ~he computer. At the same time, the piston rod 1~4 holde tha stem 150 in a retracted posi-~9i44~

tion. In this manner, the position of each valve ismonitor~d and controlled by the computer.

Thus, valvs housings 26 provide one form of a transfer passageway through which chambers of the system 10 may communicate with one another. Furthermore, the illustrated valv~ structure provides one form of effective means for selectively isolating the respective chambers from one another.

Deposition Processinq Chambers The processing chambers 14, 16, 18 and 20 in which sputtering takes place are shown in various ones of Figs. 9 through 15. During substrate processing, as explained below, substrates are first transported from the load chamber 12 to the deposition chamber 14.
Sputtering is performed in chamber 14 to simultaneously deposit an underlayer, e.g., chrome, on each side of substrates positioned in chamber 14. Thereafter, the substrates are transported to chamber 16 wherein a second layer is simultaneou~ly sputtered onto each side of the substrate. The second layer may comprise a magnetic material, such as a cobalt platinum layer.
From chamber 16, the substrates are transported to chamber 18 wherein a third layer is sputtered simultaneously onto both sides of the substrates. This third layer may be of chrome and comprises an oxidation barrier which minimiæes diffusion of potentially corrosive oxygen through the third layer to the magnetic layer. The partially processed substrates are then transferred to processing chamber 20. In chamber 20, a wear layer, such as of carbon, is simultaneously sputtered onto both ~ides of the substrates to complete "
~.
:-~291a~3 18a the processing. From chamber 20, the substrates are transferred to the unload chamber 22 for subsequent removal from the system.

Radio Frequency S~utterin~ Chamber In the illustrated embodiment, chamber 16 comprises a radio frequency deposition chamber and is described with reference to Figs. g through 12. First and second vertically oriented radio frequency cathode assemblies 42 are supported within the chamber 16 along the front and rear walls of the chamber. Inasmuch as these assemblies are simi-lar, only the front a~sembly will be described in detail. Aspreviously mentioned, a~sembly 42 is mounted to support plate 39 which ls in turn mounted to the front wall g2 of the depo-sition chamber. An optional central cylindrical view port may be used to provid~ visual acce~s to the i~terior of the chamber through plate 39. An annular target inaulator 172 is secured to the ~upport plate 39. Th~ in~ulator upport~ a water cooling jacket to which a ~puttering target 174 i mounted. The sputtering surfac~ 176 of target 174 is paral-lel to the front wall o~ thQ cha~ber and al~o to the front ~urface of substrat~s positioned in ths deposition chamber.
The watar cooling ~acket include~ a ~acket rrOnt 17a to which a ~acket back plat~ 180 is ~ecured. The ~acket front 178 and tho ~ac~st back plata 180 are ~ormed o~ an electrlcally conductivs mat~rial ~uch as copp0r. A~ ~hown in Fig. 11, tha ~ack~t front 178 i~ ~nnular and include~ an outer cixcular rib or wall portion 1~2 and an inne.r annular hub 184. The ~acket back plato 180 1~ an~ular and when mounted to jacket front 1781 as ~hown in Fig. 10, has it8 outer ~urface ~lush with th~ out~r ~ur~aces of wall 182 and hub 184. Channel~ 190 aro ~orm~d in the ~urface of ~acket front 17~. The~ channel~ ar~ ~eparated by chann~l de~ining walls which abut tha innQr sur~ace o~ back plate 180 to closa the channel~ wh~n th~ ~ackot back plate and ~ack~t front are as~embled. ~hus, togQth~r with the back plate 180, these chann~l~ provido a circuitou~ cooling water ~l~w path ~hrough the cocling ~ack~t. Thu~, cooling wat~r enter~ an inlet 192 and flow~ in the channel~ in the direction o~ arrow~ 194 to an outlet 196. Thie cooling water maintains the operating temperature~ o~ the ~putterin~ targ~t~ 174 at de~ir~d levels.
Water ~upply and return lines 198, 200 (Fig. 9) are re-~pectively conn~cted to inlet 192 and outlet 194 to circulate cooling water to and ~rom the coollng ~ack~. The conduits 198, ~00 may be electrlcally conductlvo and used to supply RF
power to tha target etructur~ a~ wall a~ the coolant fluid.
Typlcally, however, RF power i~ supplied along water supply line 198 wh$1e wa~er return lina 200 15 o~ an insulating mat~rial, ~uch as pla~tic. A wat2r line shield 202 is mounted to the support plate 39 and protects the water supply and return lines at the location where they enter chambar 16.
Seals, some being numbered at 204, seal the chamber 16 so that a high vacuum may be drawn by the vacu~m pumping stack 34.
In the Fig. 1 sy~tem, chamber 16 is th~ rhamber within which deposition o~ the working magnatic layer of a thin film magnetic disc is accompli~hed. In the illustrated embodi-ment, this magnetic layer i~ formed by sputtering a target composed o~ cobalt and platinum.
To understand the sputtering process, basic in~ormation concerning the materials tran~port ~y~kem ~escribed in detail below i8 needed. In general, sub~trates 260 to be processed are supported by a carrier 220 (Fig. 19) with the carrier and substrates b~ing tran~ported ~rom chamber to chamber by ro-bot~ or transporters 222 (Fig. 10). The transporter 222 i~
supported on a trac~ 224 and driven by a tranRporter drive mechanism 226. During ~puttering, tho carriers 220 are sup-ported in a vertical plan~ with th~ ~ubstrates 260 centered betwsen the two targ~t assembli~ 42 o~ the dQposition cham-ber. More peci~ically, the transporter 222 positions the carrier in the c~ntar of the d~position chamber 16. When so positioned, a plunger 228 i~ operat~d by a plunger driv~ me-chanism 230 to ~lrst shi~t th~ plung~r axially to insert a CarriQr gripping tip portion 232 o~ the plunger into a hub 278 (Fig. 19) o~ th~ carrier. The plunger tip then grip~ the carrier and li~t3 it upwardly from th~ tran~porter 222. The tran~porter 222 is then driven to a parked position within tha chamber, but out o~ the way og the cathode as~emblies 42 and tha depo~ition proces~. Additionally, the plungar 228 i9 rotated to therQby rotate the carrier. The di~cs 2~0 are supported (i.e., by ~heave~ 288 (Fig. 19) or in groove~ 283 (Fig. l9a)) such that rotation o~ the plung~r cause~ the discs to move in a pl~natary manner past th~ ~puttering tar-get~ 174. An opening 238 (Fig. 12) is provided thrcugh the target 174 to permit pas~ag~ of the plunger 228 through the target and into th~depo~ition chamber.

~ -~
., ~

Referring again to th~ target 174 used in depositing the magnetic working layer, the target may be a homogeneous cast mixture of platinum and cobalt with the percentage of the platinum being controlled to establish the magnetic properties of the resulting sputtered layer. As one example, a ninety~six percent cobalt to four percent platinum target is suitable.
However, because of the expense and difficulties of casting a homogeneous target, in the illustrated embodiment, the target 174 is formed by mounting a platinum ring 206 concentrically to the surface of an annular cobalt plate 208. A concentric cobalt ring 210, with an outside diameter which is less than the diameter of the platinum ring, holds the platinum in place. The ring 210 has an annular recess 212 for receiving the inner margin of the platinum ring. Threaded fasteners 214, recessed into cobalt ring 210, secure the cobalt ring 210 to the plate 208 and thereby clamp the platinum ring 206 in place. Cobalt plugs 216 overlie the 2n fasteners 214. Plugs 216 are press fit into the fastener receiving recesses of ring 210. Thus, the sputtering surface 176 of the target 174 is entirely of cobalt, except for the exposed portion of the platinum ring.

The area or width of the platinum ring which is exposed determines the platinum to cobalt ratio which is sputtered onto a substrate. Moreover, over a limited range (i.e., from approximately a zero to a twenty percent platinum concentration), the higher the platinum concentration, the hiyher the coercivity of the resulting magnatic layer. Therefore, by adjusting the magnitude of the exposed area of the platinum ring, a degree of control of the coercivity of the resulting disc is achieved.

~,'9' ' i 'i 9~D~43 21a In general, to obtain a film of a desired platinum concentration percentage, the ratio of the exposed area of the platinum ring to the total target area should equal this desired percentage. Thus, to form a magnetic layer having a platinum concentration of three percent, the exposed area of the platinum ring should constitute about three percent of the total target sputtering surface area, the remaining ninety-seven percent being cobalt. The area of the platinum ~29~l~43 ring 206 which i~ e~po~ed, and thereby tha platinum concen-tration, is readily controlled by controlling the outside diameter of the cobalt cover ring 210. The diameter o~ cover ring 210 may be varied a~ de~irad to Qxpo~e the desired area of the platinu~ ring. Therefore, the percentage content of platinum in the sputt~ring magnetic layer i3 readily adjustable, controllable and predetermined as desirad.
As a more specific example, as~ume plural ninety-five millimeter discs are supported (a~ shown in Fig. 17) on sheaves 288 mounted on a circular carrier 220 and spaced at a radius o~ 7.28 inches (18.49 c~) from the center of the car-rier to the c~nter o~ the 3hsæve~. In thi~ example, also as-sume that in deposition chamber 16 a two inch (5.08 cm) horizontal spacing exist between the ~ront and rear ~puttering ~urfaces 176 ~nd the ad~acent surface~ of substrates 260. In addition, a~sume thQ target 174 hac a cobalt plate 208 which i8 about 0.25 inche~ (0.64 cm) thick and i~ about twen~y-~our inche l60.96 cm) in ou~side diametar. Al o, as~ume the platinum ring 206 iR about 0.35 inches (0.76 cm) thick, 12.6 inche~ (32 cm) in outside diameter and 11.6 inch~a (29.46 c~) in inside diameter. In addition, a~sume th~ cobalt covering 210 ha~ an in~ide diameter of about 10.5 inche~ (26.67 cm) and an outside diameter o~ 12.28 inche~ (31.19 c~). Also, a~sume the thicknes~ o~ tha cover ring 210, where it contacts the cobalt plate 208, is about O.Q96 inche~ (1.24 cm~. Thus, the inner diamet~r of the expo~ed portion o~ the platinum ring is 12.28 inche~ (31.19 cm). When planetary motion i~ impart~d to the ~ubstrates as explained in connection with a description of carrier 220 balow, and 5putt9xing i9 per~ormed as explained below, the re~ulting magnetic layer has approximatoly a three to four percent platinu~ concentration. Al~o, when this speci~ic platinum ring i~ ~ubstantially totally exposed, the resulting platinu~ concentration is about t~n percent, although this varios with di~erant substrat~ ~izes. Also, a zero p~rcent platinum concentration results when th~ platinum ring is totally 2Pvered by cobalt. Other results are obtained for other di3c ~ize~ and geome~rie~.

. ,~
....... .

Each of the cathod~ ~puttering a~semblie~ ~2 is powered by a commercially available 30urc~, such as a three kilowatt radio ~requency diode source produced by Plasma Products, Inc. and designated model number ~FS-3000D. In addition, commercially available radio frequency automatic matching networks 674 (Fig. 30), such as nstwork model number AMN-300E
available fro~ Pla~a-Therm, Inc., are employed in a conven-tional manner.
During sputtering in chamber 16, ~ub~trates 260 are placed in the previously evacuated cha~ber. The chamber is then pressuriæed with approxlmataly ~eYen microns of argon sputtering ga~. The ~puttering ga~ iB ignitad in a conven-tional manner to provid~ a pla~ma ln thQ chamber. Also, power i delivered to the targ~t 174 to cause sputtering.
The aarrier and sub~trat~ are grounded through the plunger 228. As the plunger rotate~, planetary motion i8 i~parted to the sub~rat~s and the targ2ts depo~i~ co~alt and pl.atinum on the ~ubstrate~ supported by the carrier. With 1800 watt~ of power deliYered to each targ~t 174, in approxi~ately two and one half minutQs, a ~our hundred ang~trom magn2tic layer is produced. Although tha thicknas~ may be varied and still re-sult in a ~ati~actory ~agn~tic thin ~ilm r~cording ~i5C, a four hundred angstrom layer ic highly sati~factory.
~ 130, wh~n supported ~or plane~ary motion, the 8Ub-tra~es move relative to th~ ~puttQring surfacQ 176 duringsputtering. ~oreover, any givQn point on th~ substrate is continuou~ly shl~t6d to points on the target ~putter~ng sur-fac~ 176 which ar~ inter~ected or mappQd by a horizontal line pro~ecting from the given point to the ~puttering surface.
HorQ sp2cl~ically, any given point on the ~ubstrate maps in-wardly and outwardly ~piraling path~ on the sputtering sur-~ace 176. Thu~, the given point and other point~ on the sub-~trate sur~ac~ are not con~tantly ~putt~red by the same re-gion or rQgions of the sputtering surfac~ 176 during deposi-tion. AB a re~ult, any non-uniformities in ~puttering from particular region~ o~ tha ~arget 174 t~nd to be averaged so that a layer of cop~istent thickne~ i3 sputter2d onto the sub3trates. That i8, substrate ~otion relative to the target 1~9~3 is such that non-uniformities in sputtering from particular regions of the target are uni~ormly integrated or averaged over the ~puttered surface of the substrate.
~ urthermore, the deposition rate is uniform to within ~ive percent at the subatrats plan6~ at locations from approximately three and onehalf incheæ ~. 89 cm) to ten inches (25.4 cm) from th~ center o~ the plunger 228. Thus, the system i~ u~able in producing various sized thin ~ilm magnstic discs by supportlng such disc~ at location~ on the carrier where uni~orm d2po ition ocauxs. common disc sizes proces~ed by the system include ninety~lvls millimeter (three and one-half inch) diam~tQr diacs, one hundred thirty millim~ter (~ and one-~ourth inch) dia~atsr disc~, and two hundred ten millimeter (eight inch) diamater diac~. ~agnetic coercivity is affect~d by the thickn~s~ of th~ sputtered magnetic chromium layer. Th~re~ore, by controlling these thickness2~ from disc to di~c, the re~ulting disc~ ha~e a con~ist~nt coercivity. For exa~ple, the coercivity ~ay be controlled to wi~hin ~w~nty oersted~ ~ro~ di~c ~o disc.
Furthermore, the u e oP a target 174 with a platinum ring 206, enables th~ establish~ent of a radial coercivlty gradient in th~ rasultant di~c. When di~cs are u~d in typi-cal magnetic recording disc drive application~, the disc ia annular, is rotated, and a r~ad-write head i8 po~itioned to fly over and read or writQ on concentric ~rack~ on the disc.
The sp~ed of travel o~ th~ head, rslative to the disc, is greater and th~ head ~lies higher over the disc when the head is reading or writing onto outsr tr~cks at outer diameters of the disc in comparison to inner tracks at inner diameters.
Also, in magnetic recording discs, the recordiny density is much high~r on tracks approachlng the inner diamet~r of the di~c in comparison to the density on tracks toward the outer diam~ter.
A~uming a di~c haa a magnotic layer ~ith a constant radial co~rcivity, writing in track~ near the outar diam~ter o~ the disc i~ lmpo~ible or unreliable unle~ the writing current i9 incre~ed a~ ~uch outsr diameter tracks.
Increa3ed wrlting current i3 re~uired becaus~ the head flies ~29~4~3 higher above the disc surface as the head moves outwardly from inner to outer tracks of the disc. In order to write with a constant current, which in many applications is highly desirable, the radial coercivity of the magnetic layer must be adjusted so as to decrease as the flying height of the head increases. In other words, the coercivity of the disc should decrease with increasing radial distance from the center of the disc.

Therefore, discs with a radial coercivity gradient are desirable, with the radial coercivity decreasing in a radial outward direction from inner to outer dia~eters on the disc. In the present system such a gradient established by progressively decreasing the concentration of the platinum in the cobalt of the magnetic layer from inner to outer diameters of the disc. As the platinum concentration decreases, the coercivity decreases. The gradient is also enhanced by varying the thickness of the first sputtered chromium under layer as explained below.

In the illustrated embodiment, by sizing the platinum ring 206 such that the center of the exposed portion of the ring is nearly centered on the center of the sheaves 288 (Fig. 19) of the carrier 220 (Fig. 1~), a radial coercivity gradient is produced which is about fifty oersteds from inner to outer diameters of the discs. As the platinum ring size is changed to shift the center of the ring away from the center of the sheaves 288, the radial coercivity gradient approaches zero and then reverses.

The percentage platinum concentration at locations on a substrate, and thereby the radial coercivity gradient, which results ~rom a particular platinum , cobalt target configuration may be experimentally measured. In addition, the percentage platinum ~,`' .
's~

~L2~ 43 25a concentration resulting from sputtering with a target 174 comprises of a platinum ring 206 concentrically mounted on an annular or circular cobalt plate 208 may be predicted with some accuracy by the following mathematical model, which is descri~ed with reference to Fig. 12a.

~,, dt3 In this model, the following daPinitions are u~ed:
Target Plane: The plane 176 de~ined by the surface of the cobalt plat~ 208.
Substrate Plane: The plane which i~ parallel to the target plane and which contains the surfaces of the disc sub-strate~ 260 being sputtered from the karget as the substrates rotate on the substrate carrier 220.
An equation (Equation A) de~cribing sputtering ~rom a single infinitely narrow ring of a homog~neou~ target to any arbitrary point ln the eubstrate plane i~ given in a prior art publication, entitled Handbook oX Thin Film Technolo~y, edited by Mal~sQl and Glang, published 1970, at page 1-58, as ~ollows:

= Cs ~1 + ~ e /h)2 ~ (5/h~23 ds (A) h2~tl - (e /h)2 + ~h)2]2 + 4 (e/h)2]3/~

Where:

N - th~ deposition rate (atoms per unit ti~e) ~t a point Pl at a radius e in the ~ub~trate plan~.
C = a con~tant proportional to the 3puttQr rate or yield of the target material.
s = a variable repr~nting the radius of the target ring, fro~ th~ origin Cl of the target, in the target planQ.
e = a variable repre~enting the radiu~ from the origin C2 of the sub~trate plan~ to the point Pl. The origin C2 of the aub~trate plane being on a line nor~al to the target plane and pa9~ing through the origin Cl o~
the targ~t plane.
h ~ a v?riable repre~entlng the distance separating the targe~ plan~ and the ~ub~trate plan~ ~i.e., the dl~tanca ~rom Cl to C2).

For a ring 2~6 o~ platinum QxposQd on a cobalt plate 208, the ring 206 havlng an in~ide radiu~ o~ S1 and an out-~ide radiu~ of S2 ( each radius being measured ~rom center .
Cl), the equation (Equation B) can be integrat~d as follows:
s N (e) Cpt / 5~1 + ( /h)2 -~ ~s/h)2] dg h2 Sl [[1 - (e/h)2 + (SIh2]2 ~, 4(e~h~2~3/2 Simllarly, ror a target sur~ace 176 with an out~ide ra-diu~ oî S3 extending to an in~ide radills o~ SOI and which is entirely o~ cobalt except ~or the above de~crib~d platinum ring 206, the ~ollowing equation ~guation C) can b~ written:
S

Co( Q ) D CCo ~ [ ~ /h)2 + (s/h ] d~ _ 2 3/2 h J [[1 - (e /h)2 ~ (~/h) ] + 4 (e/h) ]
J ~3 El + (e/h)2 + ~s/h)2L ds ._ c [[1 - (e/h)2 + (s/h)2]~ + 4 (e/h)2l3/2 ( ) ;

In tll~ above equation~, th~ criptY Pt and Co refer re~pectively to pl~tlnu~ and cob~ltO ~or a point Plon a disc 260 located OTl a carri0r 220:
e. eO+~
where e o ~ the radius ~rom tho center C3 o~ the disc s~strate 260 to th~ center C2 ~ the ~ trate plane; and r,~", coordin~es oi~ a point ~1 on the ~ trate disc 260 r~l~tive to its gaoDIetric cent~r C3.
Not~: a8 an approxi~nation, the cent3r C3 o~ tha sub-strate iY a~awll~d to b~ at ths ~anter Or the ~upporting sheave 288. Thi~ i~ valid ~hen th~ 3h~zw~ diæmater i9 ~imi-lar to ~h~ dlam~t~r og th~ c~ntQr hol~ in th~ ~i3c 260.
Equation~ (B) and ~C) becom~:
N~t ( r ,~) ~ Npt ( O ~ r co~) l~Co ~r,~) ~ Nco ( O + r cooe) where Npt ( eO ~ r co~) and Nco ( ~ + r ao~e) imply the sa~o function~l depQndenco de~crl~ed ln eguations (~) and (C) with e O ~ r co~ ~ubstitlatfid ~or e .

2~
The motion ef a point Pl on ths di~c ~ub~trate 260 as it undergoe~ planetary motion during rotation of the sub-strat~ carrler 220 i~ accountQd for by integrating over the angle ~: 2 Npttr) ~ 1 ¦ Pt ( eO + r c~s~

NC (r) - 1 ~ Co ( eO + r COse~) de-ThQn, the alloy compo~ition (percentage platinum%Pt(r)) for a point Pl at radiu3 r on the disc ~ub~rat~ i8 giv~n a~:

Pt (r) - 100 ~
Co ( ) Pt ( Al~o, th~ thickne~ o~ th~ d~po~itlon ~t a radius rl relative to the thickn~3s at anothor radiu~ rO i9 approximat~d a~:
.
~1ckness (rl) ~ co(rl) ~Npt(rl) Thickne~ (rO) NcO(rO) ~ Npt(r ) Th~ abov~ int~grals ar~ b~t ~v~luated u~ing standard nu~eri-cal techniquse. Fro~ the~o int~grals, th~ p~rcentage concen-tration o~ platinu~ at ~p~ci~iQd radial di~tanc2s from the center o~ th~ ~uh trata m~y bo calcul~d. In add$tion, the radial ~oncentratlon gradient May al~o b~ calculated and used in predlcting the p~r~orm~nc~s o~ diw~ produc~d ~rom ~ given target conf~guratlon.
A~ a 8p~Ci~iC ex~mpl~, th~ atomic p~rc~ntage conc~ntra-tion of platlnum at polnt Pl, was calculat~d to be 5.0% when th~ following para~et~r valu~ were u3ed:

~0 ~ 7.28~inch~D (18.49 c~) r - 1 inch (2.54 c~) ~9~ 4~

SO = 0 in~he~ (o cm) Sl = 6 . 076 inche~ (15 . 43 cm) S2 - 6 . 3 00 inche~ ( 16 . 0 cm) S3 - 1~ . O inche~ (30 . 48 cm) h 2 2 inclla~ ( 5 . 08 cm) Pt = 1. 14 c~o The relativ~ putt~r rat~ for cobalt and platinum can }: e estlmat~d~ ~ro~ pu~ ah~d tabl ~a o~ s~tes y~ds . ~o~
exampl~, at table 2, page 4-40 o:~ tha abov~-mentioned Handbook o~ T2~ F~ C~}QlO~~ th~ 3putter yi~lds i~r cobalt and platinum sputt~red ln ~gon with an ivn bombarcaing ~ne~gy Or 600 volts are gir~n a~ 1.4 and 1.~ resp~cti~ely.
The ratio o~ Cpt ~o Cco i~ then 1.14, a~ ~e~ i~orth abo~e~
Th~ abovQ calculat~d p~are~ntagQ conclantr~tion compares well with ~n average m~a ured platlnu~n concentration o~ 4 . 89~
as measur~d by Ruth2rford sackscatter Spectro~copy, ~or a sa~pIe which wa~ 3putter~3d ~ing the geometry d~scribed by the param2ter valu~3~3 list~d abo~o.
: ~ Sputtering 3hi~1d3 240 ar~ also provided within the de-position chambers to focu~ the d~po3ition on the ~ubstrate and to shield other area~ o~ th~ cha~ er from undesired depo-sition~.
The illustrated depoaition cha~er 20 i~ li}c~ chamber 14 and 18. How~svor, it m~y be a radio fr~sluQncy sputtering chamb~r lik~ cha~bor 16. In thi~ oase, unlike chamber 16, cha~er 20 d~po~it~ a wear re~stant matarial on substrate~
po~itionad thereln.
A~ an example, radlo frequ~ncy r~activo sputter~ng o~ a cobalt-oxide wear layermay b~ e~ployed. In thi~ exampls, a cobalt targ~t i~ u~ed and the cha~ber 20 i~ pres~urized to approxi~ately 30von micron~ with a sputtering ga~ com~rised o~ twenty p~rcent oxygon and eighty p0rcent argon. A typical ~puttering ti~e i~ 5.6 minute~ at two kilowa~ts power to the sputtering target~. Thi~ r~ul~s in a wear lay~r o~ approxi-mately fiv~ hundr~a an~troms. Suoh a layer ha3 provided satls~actory wear r~8i~tanca when sub~ectad to ten thousand :
. ,......................... ~ .

computer disc drive head start/stop cycleq. Alternately, as another example, DC ~puttering may ba employed in chamb~r 20 to deposit a carbon wear layer a~ explained below.
Such wear layars provide protection to the underlying layer~ depo it~d on the 8ub~trate8. In connection with understanding thi~ wear protection, a~ume tha ~ub~trates comprise magnetic r~cording di~cs used in computer disc drives. Whenever th~ pow~r i8 shut off to an operating disc drive, the rotating disc slows down and thQ head of the disc drive ceases to fly and beginq to drag on the disc. The wear layer increases thQ ll~a of tha di~c by minimizing wear from the head dragging on the diac when powlsr is shut ofr.
B~causQ th~ chamber~ are isolatable ~rom one another as explained above, the parameters affecting ~puttsring, ~uch as sputtering ga~ pres~ure, sputt~ring ga~, ~puttering time and power, in the individual aha~ber~ may ba optimized for the particular ~puttering deposition b~in~ pQr~or~Qd.

Direct Curr~n~ Sputterin~ Chambers Th~ deposition chambsr~ 14, 18 and 20 are best under-stood with reference to Fig~. 14 and 15. Ele~ent~ in these chambers which hav2 countarparts in thQ prev~ou31y de~cribed spu~tering chamber 16 are nu~b~red with corrQsponding numbers and there~ore will not be de~cribed in detail.
In the illu~tr~tQd Flg. 1 ~ystem, chambers 14 and la are Qach de~ign~d to deposit chromiu~ layers, and cha~ber 20 is design~d to deposit a carbon lay~r, on ~ubstrate~ posi-tioned ~ithin th~ chambers. Thi~ depo~ition i~ acco~plished by direc~ curren~ sputt~ring. C~mmerc~ally availabla cathode ~puttering a~8emblias 40 may be utilized ~or this purpose.
For example, one suitabl~ asee~bly comprise~ a direat current planar magn~tron ~putt~ring cathod~ availabla rrom Vac-~ec Sy~tems and ~old under the trademark Fl~xi~ag. The~e cath-ode~ have ~ive inch (12~7 cm) by ten inch (25.~ cm) rectangu-lar water cooled, ~iv~ kilowatt rated targat~. Such cathodes may be powered by commsrciall~ availabl~ ~ive kilowatt ources, such as fPQ~ Advanc~d Energy System~.

~L~9~L443 AB shown in Figs. 14 and 15, two ~uch cathode assem-blies 40 may be provided at the front and two at the rear of the chambers. Al~o, the front and rear cathode aRsemblies are at equal distances fro~ the plane containing substrates 260 in the ehambers. Referring to the right hand portion of Fig. 15, the two front catho~e asaemblie~ 40 ar~ ~ecured to th~ circular support plat~ 38 which in turn ls fastened to the ~ro~t wall 92 of the depo~ition cha~er. Th~ cathode as-sem~lies 40 are aool~d in a conventional manner via water in-let and outlet line~ 198, 200 (Fig. 14). In addition, power i~ deliv~red to the cathode a~semblie~ vla power cabl~s 248.
For purpo~e3 o~ ¢larity, tho water lines, and the uppermost power cabl~s, ha~e bo~n eliminated ~rom the Fig. 15 view of the~e cathod~ ~s~mblia~. Each o~ ths czthode a~e~blies 40 include~ a cathodQ housing 250 in~erted within a corr~pond-ingly shaped opening through the ~upport plate 3B. A DC
sputtaring target a3sembly 252, including a target 254 mou~ted to a water cool~d iacket 256, i~ ~upportsd within th2 cathode h~u~ing 250. An in~ulator 25~ separate~ the cathode hou~ing from the tar~et assembly. Clampa 2Sg hold asse~bly 252 in plac~. During sputtQring, ~at~rial i~ ~puttered from the surfaco o~ target 254 to the ~ub~trates 260 as the subs~rates ar~ carriod pa~t tha target by a carrier 220 (~ig.
19). Th~ t~rget~ 254 in chamb~rs 14 and 18 are of chromium while the target~ 254 in chamber 20 are o~ aarbon. A cover plate 251 ~nclo~s the cathode housing 250 wherQ it a~erge~
fro~ the support plata 38. Suitable ~eal~, som~ being number~d a~ 242, ~eal the chamber~ 14, 18 and 20.
During a typical 3puttering process in the Fig. 14 chamb~r, the substratea 260 on carrier 220 are moved in a planetary motion pa~t the targets 254. ThQ cklamber is pre~uriz~d to approximately 7 microns with argon and a plasma i~ ignited. When th~ targ~t~ ar~ aputtered at, ~or example, an applied pow~r o~ approxi~ataly thrae hundred volta and two amp~ ~or approximatoly ~iV4 minutes, a ~irst chromtu~ underlayer of approxi~ately 3000 angstr~ms is depoaited on the ~u~trat~.

~2~4~3 It has been found that the thickne~s of the chromium underlayer hae an effact on the eoercivity of an overlying cobalt platinum magnetic layer. That ie, with increasing thicknes~es of the chromium underlayer, the coercivity of the magnetic layer is increased. ~his coercivity increase~ at a rate of about se~en o~r~tod~ per one-hundred ang~trom~ o~
chro~ium underlayer thickne~s. Thl~ increasing coercivity is pro~a~ly due to an epitaxial s~fect be we~n the underlayer and the cobalt platinum lay~r. ~y controlling the consisten-cy o~ the thicknes~ o~ the und~rlayer from disc to disc, additional control of the consistency o~ the coercivity of the thin ~ilm magn2tie reeording dises i~ maintained.
Furthermore, by v~rying the thicknes~ of the underlayer in the radial direction, a radial coQrciv~ty gradient may be established in th0 re~ulting di~c. With the eputter~ng cathode~ 40 positioned in the eonfigura~ion illustrated in Fig8. 14 and 15, and with th~ sub~trate~ moved ln a planetary manner during ~puttering, the re~ulting ehromium underlayer $g somewhat thicker at in~er than ~uter radial positione of the substratee. Th~refor~, thi~ ehro~ underlayer deposition aleo eontrihutes to th~ pr~viou~ly described desired higher to lower radial eoereivity grad~ent moving fro~ inner to outer positio~s on the di~e~. ~k ha~ al~o been found that the coereivity of the r~eulting thin ~ilm magnetic recording discs i~ more pr2dictable and mor~ consi~tent ~rom disc to disc, i~ the time betwe~n spultaring o~ th~ chromium underlay~r and cobalt platinum }ayer i6 limited to no more than about five minute~. With the sy~tem of the present invention, this i~ ea~ily accompli~had because the ~ub~trates ara readily tran~ar~ed ~rom cha~ber to chamber.
The chamber 18 in the pre~arr~d ~mbodim~nt is al~o uti-lized to ~putt~r a chrome outer layer onto tha ~ubstrate~.
Thls chroma outer layer sQrve~ th~ ~unction o~ providiny an oxy~sn di~usion b~rrier to protQct the cobalt platinum layer from oxidation or corroslon. A chrome outer layer o~
approximat~ly 250 angstrom~ i9 ~uitabls ~or thi~ purpose.
Consequ~ntly, in c~a~ber 20, although ~hown wlth ~our cathode asaemblies 40, only on~ ~ront and one rear assembly 40 are ~L~9~44~
~3 typically u~ed. With a two target chamber, thi~ outer layer i~ deposited by sputtering the targets at, for example, an applied power of approxi~ately 0.7 a~p and ~hree hundred volts for two and onQ~half minutes. A seven micron arqon sputtering ga~ environment i~ suitable.
In sputtering a carbon wear layer in cha~r 20, ~our carbon cathode assemblies 40 are u~ed, two at the front and two at the rear o~ th~ chamber. To produc~ a 400 angstrom wear lay2r, the targ~ts are sputterea at, for example, an applied power of approxi~ately thre~ amp~ and threa hundred volt~ ~or three and one-hal~ minutQs. A seven micron argon sputtering ga~ environment is al~o ~uitable for thi~ wear layer depo~ition.
Although de~cribed above with speci~ic ~puttering operation~ in the specific proces~ing chambers, ona can easily replacs the previously de~cribed ~puttering a~semblies with other vacuum deposition a~e~blis~ a~ de~ired. Thi~ is readily acco~pli~hed by simply remo~ing the plate~ 38, 39 and replacing them with plate~ conta~ning dif~rently con~i~ured targets. Al~o, fewer or more deposition chamber~ may be em-ployed depending upon tha number of layers to be deposited onto a ~ubstrat~.
Mat~rials ~andlina Sv~tem The materlal~ handling systQm ~or transferring and handling thQ sub~trates during proc~ing i8 shown in Figs.
16 throuyh 28. Thi~ system includ~s the planetary 6ubstrate carriers 220, one being ~hown in Fig. 19, for carrying sub-strate~ 260 during proc~ssing. Another component o~ the ma-terials handling system comprise~ rack~ or trays 270, one positionQd in load cham~er 12 and on~ in unload chamber 22.
The tray 270 in the load cha~bQr 12 ~upport~ carrier~ 220 prior to proce~sing while the tray in the unload chamber 22 ~upport~ carriers ~ollowlng proce~ing. In addition,a load mechanlsm 272, and a aimilar unloading mechanlsm, are provided in the re~pective load and unload cha~bers. The~e latter mechani~m~ trans~r carri~r~ 220 to and from transporter 222. ~ The transportQrs 222, tracks 224 and tran porter drive assembl~es 226 compri3e ~urther components ~9~L~L43 of the materials handling system. In addition, the plungers 228 and plunger drive 230, are also included in the materials handling system.

Planetary Carriers and carrier SupPort Trav In the system of the present invention, a carrier means, such as carriers 220 (Figs. 19, 19a) are provided for supporting the substrates for movement during deposition in the high vacuum, high temperature environment typically found in sputter deposition chambers. In addition, such carriers impart a planetary motion to substrates supported thereon while minimizing particle generating from frictional engagement of metal parts. This planetary motion enhances the uniformity of deposition on the substrates because the substrates are not continuously sputtered from the same region of the target. As a result, this motion compensates for and averages the effects of non-uniform sputtering from particular regions of the target. Moreover, these carriers permit simultaneous deposition of both sides of the substrates 260 without requiring complex mechanisms for turning the substrates over during deposition.
Furthermore, the carriers 220 are readily adapted to support substrates of varying sizes.

With reference to Fig. 19, one form of planetary carrier 220 comprises a circular planar pallet or carrier chassis plate formed of aluminum or other electrically conductive material. A central opening 276 is provided through the carrier plate. A hub 278 is inserted through opening 276 and secured in place by a hub clamp ring 280. The hub is engages by the load and unload mechanisms 272 (Fig. 20), as set forth below, to transport carriers 220 to and from the trays 270 (Fig.
17). In addition, the hub is engaged both by the .!,, 1~3~4~3 34a plunger 228 (Fig. 15) and by the transporters 222 during various steps of the process, as explained below.
Portions of the carrier plate 220 are removed to provide plural, generally circular, ~puttering openings 282 through the carrier plate. A substrate supporting structure is provided for supporting the substrates 260 in the openings 282 so that one surface of the substrate is exposed :,. ;
. ., . ~

~L2~

to sputtering target~ through th~a openings. A~ shown, the substrat~ ~upport may be an integral part OI the carrier plate and comprise plural thin ~poke~ 284 extending from the perimeter of th~se opening~ to a central hub re~ioTI 286. As shown, three such 8pOl~,Q3 may be employed and ar~ ~paced one-hundred and twenty degraes apart about the hub regionL Sub-Rtrates ~upportillg ~h~ave~ 288 are rigidly ~ecured by a fas-tenar 289 to the hub r2glons 286 and ~uppor~ tha substrates 260 a~ shown in Figs. 17 and 19. Th~ 3heave~ are po~itioned at equal radial di~tanc~s ~rom the cent:er of op~3ning 276.
The size of the openings 28~ is varied depending upon the ~iz~ o~ the di~c3 being procee~ed. Thu~, lar~r and few-er openings 282 are pro~,rid~d when larg~r disc~ ar~ handled by the ~y~tQm. For examplQ, opQnings may b~ provided to handle nine nin~ty-fiv~ millime~t2r di~cs, ~ix one-hundred and thirty milll:neter discs, or three ~wo-hundred and ten mlllimeter discs. The Fiq. 19 carrier 220 can accommodate thin, planar trates of variou~ ~iz~ and ~hape~. All 1:hat is required i~ that the substrate have~ a circular holQ concentric with the center of gravity of th~ trate and sized to fit onto a sheave 288. Thu~, while round sub~trate~ with concentric hole~ are illustrated and pr~erred ~or the ambodlment de-scribed, 3ub~tratea o~ virtu~lly any shape may be supported in this manner.
The sheave~ 288 ara groov~d around their circum~erence much like the groove provid~d in pulley wheQl3. The grooves ~re ~orm0d to acco~odate the thicXnes~ o~ the sub~trate to be proce~3ad. With the plan~ of the carr~er plate in a ver-tical orientation a~ ~hown, the grooves o~ the sheave~ are also in a common vertical plan~. tn addition, substrate~
260, with interior hol~3 o~ a di~mater D2, hang ~ro~ the groove Or the qheave~ and aonta¢t tha circular sur~aoe at the base Or th~ shQavQ groove. Slnc~ ~ub~tr~te~ 260 merely rest in the shQa~e groove~, loadlng and unlo~ding o~ ~ub~tratQs 260 onto the carrier 220 13 gre~tly ~impllried. This circular ~heave sur~ace 1~ o~ ~ diameter Dl and is les~ than D~. ~otation o~ the planetary carrler 220 at a preselected speed about it~ center by the plung~r 228, as ~xplained 3L~9~

below, causes a corre~ponding rolling of the substrates on the sheaves. For each revolution of th~ carrier, each substrate 260 somplete~ a ~raction of a revolution on i~s sheave given by th~ ratio Dl divided by D2. There~ore, the orientation of the substrat~ z60 relative to a fixed sputtering target i~ gen~rally dlfferent after each revolution of the plan~tary. Similarly, the orientation of the ~ubatrata~ 260 relativs to the ~pokea 284 continuously varia~. A~ a rQsult, sputtering o~ the back ~ide o~ the ~ubstrates ~ay be per~ormed through the op~ning~ 282 without the ~poke~ 284 leaving shadow~ on tha aub3trate3 and inter-~erlng with the deposltion. Con~equ~ntly, ~imultaneous depo-sition of ~aterial~ onto both sid~ o~ th~ disc ~ub~trate i5 po~ ible and the r~sulting disc ~urfaces have ~ub~tantially uniform propertiaR.
Furth~rmore, circumferential uni~ormity o~ the depos-it~d fil~ on th~ aub~trat~ i8 enha~c~d by this planetary mo-tion. That i~, variatlons in sputt~ring by di~erent por-tion~ of the sputtsring target~ tend to be averaged because of the planetary tra~el o~ the sub~trate during ~puttering.
In addition, as previously expla~ ned ~n conn~ction with the depo~ition o~ th~ cobalt platinum layer, layers with radial film concentratlon gradiQnt~ may be ~puttered onto the sub-strate~ to vary the radial coercivity in a de~ired manner.
Further~ore, thQ rolling Or th~ sub~trates on the sheaves re-sult~ in ~ubstantially no contaminating particle generation a~ each cubetrate ~imply roll~ in a ~h~ave groove a~ the ghaave i9 rotated. In addition, ~uch a sub~trate carrier re-qyire~ no lubrication. There~ore, contamination ~xom that ~ource i~ aliminated.
In addition, such a carrier 220 i~ relatively inexpen-stve, i~ compatibla with simple load and unload tooling me-chanis~, and i8 una~ectQd by hlgh temperature~ and high vacuums encountered in typ~cal ~puttaring operations. As ~ention~d, the carrier plat~ i~ typically o~ alu~inum while the sheav2~ 2~8, hub components 278, 280, and fasteners 289 are typically of s~inle3s st~l. The carrier pla~e i~ also typically of ~tainle~ ~t~el or oth~r high temperature ~2~4~3 resistant material if the temperature of the deposition process exceeds about one-hundred and eighty degrees Celsius. The carrier 220 provides a ground plane for grounding the substrates 260 and electrically isolating the deposition environment, such as the sputtering plasma in a two-~ided deposition process.

The carrier 220 shown in Fig. 19a also imparts planetary motion to the substrates by supporting the substrates for rolling within an annular groove as the carrier is rotated. In this form of carrier, the spokes 284, central hubs 286 and sheaves 288 are eliminated.
Instead, a circular groove 283 of a cliameter Dl is provided at the circumference of each of the circular openings 282. As shown in Fig. 19b, each subs~rate 260, of a diameter D2 which is less than Dl, contacts the groove 283 and thereby rolls in the groove as the carrier is rotated.

The Fig. l9a form of carrier is also suitable ~or substrates of various sizes. In addition, the substrates need not have a central opening. Howeverr the outer perimeter of the substrate must be substantially circular for smooth rolling action.

For each revolution of the carrier 220, each substrate 260 completes a fraction of a revolution on its groove given by the ratio of Dl divided by D2.
However, to provide stable support of a substrate supported in this manner in a groove 283, the ratio of ~1 to D2 must be only slightly greater than one. This requirement does not exist for the Fig. 19 form of carrier because, in the Fig. 19 form with the center o the disc 260 supported on a sheave 288, Dl and D2 need not be close to unity for stable support. In general, the greater the difference between Dl and ~2, the greater the randomness of exposure of the substrate . ._ ~2~ 3 37a surface to different regions of the target surface as the carrier is rotated. Furthermore, the greater the randomness, the better the compensation for non-uniform deposition from different regions of a target and the better the uniformity of the deposition. Thus, the Fig.
19 form of carrier has some advantages over the Fig. l9a form of carrierO Also, somewhat higher partial generation may re-~;, ., ~, .æl .

a3 sult from the Fig. l9a carrier than the Fig. 19 carrier.Otherwise, th~ Fig. l9a carrier posses~es the advantag~s and ~eatures previouqly explained in conn~ction with the de~crip-tion of th~ Fig. 19 carrier.
Referrlng to Figs. 16, 17 and 18, the rack or tray 270 has a frame which include~ front and r~ar ~upport plates 296, 298. Three horizontal planetary ~upporting rod~ 300, 302 and 304 are support~d by th~ plat~ 296,298. Thq rod~ 300, 302 and 304 are eaah provid~d with plural axially spaced apart annular groove~ 306. Each groove o~ each rod i~ aligned in a vertical plan~ pa~sing through a corre~ponding groove o~ each o~ the othQr rods. Further~ore, the plate3 296, 29~ ~upport the rods so that corrQsponding groove~ o~ tha rods are po~i-tion~d in an arc o~ a radiu~ which Qquals the radills of the carri~rs 220. Consoquently, a~ ~hown i~ Fig. 17, the car-riers ne3t within thQ corr~sponding grooves and are ~upported at thr~e location~ by the rod Qcau~e th~ rod~ are po3i-tioned beneath ~u~trata~ 260 ~upported on the carriers 220, th~ po~ibility of aontamination of thQ sub~trate~ by par-ticle~ ~rom tha rods i8 ~inimi~ed.
As ~hown in Fig. 17, a pair o~ parallel, horizontal, 3paced apart rails 308, 310 ar~ ~upp~rtod ~ro~ the floor 62 of chamber 12 and extsnd sub~tantially ~rom front to rear of the chamber. Th~e rails ar3 parall~l to chamber walls 56, 58 and hava n upper tr~y engaging portion which i~ o~ circu-lar cros~ section. Grooved rollors 312 are pivotaliy mounted to the tray 270 and each engaga the upper por~ion of rail ~08 at two location~. Flat rollQrs 314 arQ also pivotall~
mounted to tho tray. Each rollar 314 Qngage~ the upper por-tion of rail 310 ~t ons point. Thexe~or~, as the tray is ~lld on rail~ 308, 310 into and out o~ the chamber, the roll~r3 312, 314 and rails 308, 310 coop~rate to establi#h a plan~ which ~upport# ~h~ tray. Furth~rmora, rod 308 in co-op~ration wlth roller~ 31Z d~in~ a lin~ ~long which tha tray ~lide~ into ~nd out o~ th~ chambar 12. Furthar~ore, a stop 316 (Fig. 16) li~it~ the d~pth o~ in~ertion o~ tha tray i~to the chamber to a pa~ticular point. Con~eguantly, the tray is easily and pr~ci~ely po~ltioned at th~ ~ame location each lX~ 3 tim~ it i8 placed into th~ chamber. In addition, a stop 318 (Fig. 16) i~ mounted to the rail 310 following the posi-tioning of th~ tray within the chamber 12. Stop 31~ prevents th~ tray fro~ rolling toward door 68 after it is in position.
Note, for purpose~ of clarity, th~ sub~trates and ~heaves have been omitted from the carrier~ 220 ~hown inFigs. 16 and 18. An identical tray supporting structura i9 also proYided ln unload cha~ber 22.

L~--9b~L~ Ig~ sbanis~s The loader 272 ~or loading carriers 220 from the tray 270 and onto th~ tran~port2r 222 1~ shown in Fig~. 16-22.
The unload chamber 22 i~ provided with an unloadar which is a mir~or ~mage o~ the loader ln chamber 12. Con~e~u~ntly, the unload~r will not be d~cribed in d~tail.
In g~neral, thQ loadQr 272 h g an upwardly extending load arm 320 with a carrier handling finger 322 projecting outwardly ~rom th~ fxe~ ~nd o~ ar~ 320 ~n th~ direction of door 68. The arm 320 i~ ~upport~d at it~ lo~er end by a bel-low~ block 324 which is capabl~ o~ v0rtical upward and down-ward mov~ent. A bellowo 3~8~bly ind~cate~ generally at 326 (Fig. 21), and do~crib~d ln dotail b~low, i~ ~upplied with air to shi~t th~ block 324, and th~reby th~ ar~ 320 and finger 322, upwardly and downwardly~ Th~ bellows block 324 is mounted to a trav~ling body 330 which i~ 31ida~1y mounted to a pair o~ spaGod apart upp~r and lower horizontal guide rails 332, 334~ Rails 332, 334 are parallal to wa:Ll 56 and extend ~ro~ the gront to the r~ar o~ the cha~ber. A hex drivo ~crow 336 i~ coupled to.the travel~ng block 330, as ex-plained below, and driven by a r~varsible step motor 338.
When driven, the driv~ ~cr~w ~hi~t~ th~ traveling block 330, and thus the arm 320, eithar ~orwardly toward the ~ront o~
chamber 12 or r~rwardly.
Eloctrical drive pu13~ 9 d~ r~d to the ~tep motor under tha control o~ tho computor 46. By ~onitoring ~he num-ber o~ pulsQ~, th~ po~ition o~ tha travali~g block 330 and arm 320 along the~guid~ rail~ ls known. ~n optional sha~t encoder i5 utilized to ~onitor the rotation o~ the motor and thus of drive ~crew 336. ~he shaft encoder provides feedback to the computer of th~ movem~nt of kh~ drive ~crew in re-sponse to the step motor pulse~. In addition, as e~p~ained below, tha comput~r control~ t~ air which i~ suppli~d to a pair o~ bellow~ 392, 394 (Fig. 21) which operat~ as axplained below to rai~Q and low~r bQllow~ block 326. ~herefore, the upward and downward movement o~ the ~r~ 320 i~ controlled by the computer.
In operation, the loader i8 aapable of automatically moving along a tray 270 o~ carriers 220 in oha~ber 12, re-trisving a ~ingle carrier from th~ tray, and then loading the retrieved carrier onto a transport~r 2a2. Thi~ operation ls sequencad as ~ollow~. At th~ start o~ the ~equence, a trana-port2r 222 is po~itionad out3id~ o~ tha ohamber 12. Also, tha traveling block 330 i~ po~itionad a~ a ho~s po~it$on ad;acent th~ x~ar wall 64 o~ th~ ch~mb~r 12, ~uch a~ ~hown inFig. 16. Th~ traveling blocX 330 i~ th~n driv~n ~orwardly by motor 338 until the ~ingar 322 i~ ert~d fully into the hub 278 o~ th~ r~armos~ carriQr on th~ tray. The b~llows block 326 i~ then raised to raisQ th~ arm 320. This causes the finger 322 to contact thG hub 278 and li~t th~ carrier out o~ the tr~y ~ The trav~l ing body 330, and thus the arm 3~0 and Yupport~d carrier 220, i8 then driv~n rearuardly to a pc~ition which center~ th~ carrl~r 220 over the cent~r of the track 224. Th~ tran~port~r 222 i~ then drlv~n into the cham-ber 12 until upwardly axtending arm~ 340, 342 o~ the trans-port~r 222 are positioned beneath th~ carrier hub 278. The arm 320 is thQn lowsr~d by b~llow~ 392, 394 a3 explained below, to cau~a the carrisr 2~0 to res~ on the ar~s 340, 342 o~ the transporter. The travQling bedy 330 iB then driv~n to it~ homa po~ition ad~ac~nt to tha rear wall 64. Whan the traveling body 330 1~ thu~ out o~ th~ way, tran~port~r 222 i~
mov~d to th~ next chamb~r and carria~ th~ load~d carrier 220 with it. A~ter the carrier 2Z2 ha8 exlted rro~ chamber 12, the sequence i8 again r~pe~ted ~o ~hat, upon re~urn Or the transport~r, the n~xt carrior i8' ln poaitlon ~r loading.
Thi~ ~equence i~ ~4peatad until ths l~t carr~er is loaded on~o th2 transport~r and the tray 270 is e~pty. Then, the ~racuum 18 removed ~rom chamb~r 12 while chamber 14 is isolated, the door 68 i~ openedr another tray o:e carriers is insertad into chambar 12, and tha door is closed. Following this, the vacuum i3 reeatablished in chamber 12 and loading of carrier~3 from the tray ~nd onto th~ transporter is continu~d .
Th~ detall~ OI th~ portion o~ the loader snechani~m 272 utilized ~or shifting the traveling body 330 along the guide rods 3 3 2, 3 3 4 are shown in Fig . 2 0 .
~ sre speciIically, a cha~ber wall mounting bracket 343 is ~nounted to the chamber ~ide wall 56 as shown in Flgl 18.
The ~orward end o~ each o~ the guide rod~ 332, 334 i~
Iastened to the bracket 343 ~9 indicat~d ln Fig. 20 while the raarward end o~ th~e rods i~ Ia t~n~ad to the r~ar cha~nber wall 64. Upper and lower op~nlng~ 345, 344 ar~ provided through the tra~teling body 330. }~all ~ushing~ ~not ~hown) within the~o op~ning~ slidably r~ce~Y~ th~ ro~p~ctiva upper and lower rods 332, 334. Tho hex drive ~cre~r 336 iE~ threaded through an elongated nut 348 ~nd ha~ it~ forward end 350 sup-ported by a bearing 352 in ~ bs~rin~ block 354 mounte~ to the bracket 343. The nut 348 i9 ~ecured ts: a ~ount 356 and held in plac~ by a cover 358. Mount 3S6 in turn i8 rigidly mounted to the trav~ling body 330. Consequently, when drive screw 336 ia rotatQd in a ~ir~t direction, th~ traveling block shi~t~ in a ~orward dlr~cti~n along guide rail~ 332, 334. Conv~rsely, wh~n tha scrQw 326 is ro~ated in th~ oppo-si~ dlr~ction, th~ ~raveling block ~hi~t~ r~arwardly.
Th~ driv~ ~crew 336 i8 couplQd to ~ha step motor in the following ~nner. Tha rear end 360 Or ~crew 336 i~ connected to a tor~ionally r~gid Plexibl~ coupling 362. Coupling 362 i~ conneated to and drivan by A ~ha~t ~nd 3G4 pro~ecting from on~ ~nd o~ ~ commercl~lly ~v~ilabl~ s~led rerrorluidic rotary faed through couplar 366~ Such ~ are co~mercially available ~rom ~arro~luidic3 Corporatio~ undor thQ trad~mark Ferro~luidic TM seal~. A 3ha~t end 36~ pro~e~ting ~rom the other end o~ coupl~r 366 ~upport8 a hub in~ert 370 which is conn~ct~d to a hu~371 o~ large diam~ter timing pulley 372~
A timing belt 374 couple~ timlng pulley 372 to a smallar ~9~443 tlming pulley 376. Pull~y 376 i8 driven by the step motor 338.
The coupler 366 is po~itioned within a s2aled hou~ing 378 (Fig. 16j ~cured to cha~ber wall 6d by a connector 380 (Fig~ 20). A c~ to hou~ing 378 i~ provided thr~ugh a plug 379 ~or tha purpos~ oP parmitting tightening o~ coupler 366.
The drive ~crew 336 pa~a~ through tha oha~ber wall 64 and engage~ the couplar 366 within housing 378. Because the coupler 366 is ~aal2d, rotaton i~ transmitted through the coupler while a high vacuum i8 maintained within chamber 12.
Motor 338 i~ ~upported by a bracket 383 (Fig. 16) which is mountad to houslng 378 by a motor mount 382.
Thu3, ~tQp ~otor 338 i8 oparatlvely couplad to the drive screw 336 for rotat~on o~ th~ 3crew in either dir~c-tion. In addition, the po~it~on of the tra~eling block 330 along the guide~ 332, 334 relativ~ to a referenc~ location may be deter~ined ~ro~ the nu~ber o~ driv~ ~tep~ through which the ~crew 336 ha~ be~n dri~n by the step ~otor. Fur-thermore, th~ ~tep~ ar~ ~lectrically controlled and monitored by the comput~r 46 80 that tha po~ition o~ th~ traveling block is known.
. Th~ bellow~ block 324 i~ raised and lo~red by alter-nately ~ressurizing bellows 392, 394 (Flg. 21) to thereby rais~ and low~r th~ arm 3~0. The vertical ~otion of b~llows block 324 i9 guide~ by a pair o~ verti~al pin~ 384 (Fig~. 16, 20), mount~d with~n the traveling body 330. The~e pin8 ex-tend through v~rtical op~ning~ 386 through the b~llows ~lock 324. Pin~ 384 are slidably coupled to the bellow~ block by bushing~ 388~ one being ~hown in Fig. 20.
A~ ~hown ln Fig. 21, tho bellow~ a~embly 326 includes an upper ~tainl~s~ ~t~el b~llow~ 392 mount~d by bellows holding clamps 3g3 to an upp~r sur~ace o~ the bellows block 324 with a ~ealiny ga~k2t poeltioned b~twaen th~ bellow~ and block. A ~imil~r lower bellow~ 394 i~ mount~d in the sama ~anner to ~h~ undar~ld~ o~ th~ bellow~ block. Thes~ b~llows are suitable ~or operatlon in a high vacuum environment wi~h-.
out 1 eaking ga~ ~r~ th~ bellow~ into the ~nvironment. When the bellows block 324 and travsling block 330 are a~se~bled, ~X9~443 the upper bellows contacts an upper surface of the traveling bl~ck while the lower bellows contacts a lower surface of the traveling ~lock. Tharefore, when the upper bellows is pressurized, the bellows block 324 and attached arm 3~0 are shifted downwardly. Conversely~
when'lower bellows 394 is pressurized, the arm 320 is raised.

Pressurized air for operating the bellows 392, 394 is delivered by a pair of air lines (not shown) which pass through an upper ~eed through housing 396 attached to the chamber wall 56 (Figs, 17 and 223. As gasket seals housing 396 to the wall 56. A flexible stainless steel bellows conduit ~98 is connected from the upper housing 396 to a lower bellows feed through housing 400 mounted to the bellows block 324 lFigs. 17 and 21). A
gasket 426 seals housing 400 to the bellows block 324.
The air delivery lines pass through conduit 398 and enter housing 400.

To connect the conduit 398 to the feed through housing 396, a cylindrical insert 402 (Fig. 22) is inserted within the end of conduit 398 and a compression ring 404 is then forced over the outside of the conduit.
A retainer plate 406 holds the compression ring, and thus the attached conduit, to the underside of the feed through housing 396 with a gasket seated between the housing and compresæion ring. The lower end of the conduit 398 ~Fig. 21) is connected to the lower bellows ~eed through housing 400 in the same manner by a respective insert 410, compression ring 412, gasket, and retainer plate 414.

43a A first of the air lines entering housing 400 is connected to a flow controller 416 which extends into an opening ~17 in an air flow block 418 and communicates through the block and an aperture 420 with the interior of upper bellows 392. The second of the air lines entering housing 400 is connected to a flow controller 422 which extends into an opening 424 in the air flow block and communicates through an aperture (not shown) leading to tha interior of bellows 394. Flow controllers 416, 422 permit unrestricted flow into the bellows and restricted flow out of the bellows to smooth the lifting and lowering movement of the arm 320.

~LZ9~L443 To lower the arm 320, a solenoid operated computer con-trolled air valvo is op~ned to permit the flow of air through the first air line and into th~ upper bellows. To lower the arm 320, another comput~r controlled Rolenoid operated air valve i~ opened to permit the flow o~ air into th~ second air line and into the lower b2110w3.

Plun~er and Plun~er Drive ~echanism The detail~ of the plunger 228 and plunger drive mecha-ni3m 230 can be und~r~tood with rQ~erenc~ to Figs. 10 and 26-28. The plunger 228 i~ de~igned to accompli~h three func-tions. First, it is movabl~ axially to po~ition the carrier gripping tip 232 o~ th~ plunger lnto tha hub~ 278 o~ the planatary carrier~ 220 (Fig. 19) when each carri~r i8 pogi-tioned by a transportar 222 in alig~m~nt with th~ tip of the plunger. Following in ertion, the plunger grlpping ~ip 232 is operated to grip the hub o~ kh~ planetary carrier and lift the carrier upwardly from th~ tran~portsr. Lifting and clamping action i~ acco~plishad util~zing rolling contact bçtwee~ ~urfac~s of khe plunger tip an~ interior o~ the hub.
That i~, t~ plung~r tip ha~ a minor ~h~ft with ~ccentrically mounted baaring~ which ar~ rotat~d rQlativ~ to a ma~or sha~t with a fixed protru~ion. A~ thi0 rotation occurs, th~ di~-tance betwesn the b~aring and protrualon increa~es until these elements grip the int~rior of the hub and li~t th~ hub from the tran~port~r in one contlnuou~ motion. Then, the plunger i~ rotated by th~ plung~r driv~ machanism 230 during sputtering to ~hor~by rota~ th~ carriQr ~20 and move the 3ubstrate 260 a~ pr~viou~ly explained. A~t~r sputter~ng, ro-tation is stopped. Th~ carri~r 220 i~ then lowered onto a tran~porter 22~, and the carrier 220 1~ rel~a~ed ~rom the hub 278 in one ~otion and the plunger is withdr~wn ~ro~ the hub.
~herea~ter, the tran~portor tr~n~ar~ th~ carrler to the next chamber ~or ~urth~r proc~ing.
Th0 clamping and li~ting action o~ the plunger tip 232 is illu~trated in Fig~. 26 and 27. Speci~ically, the plunger 228 includes a ~a ~ r outer ~ha~t or ~pindle 436. A ~ixed protru~ion 438 pro~ects outw~rdly ~rom a portion o~ the peri-'a3 phery o~ the front face of the end of ma~or shafte 436.~hu~, the protrusion 438 is offset from the central longitu-dinal axis of the shaft 436. More than one ~uch protrusion may be utilized if de~ired. A rotatable minor shaft 440 (Fig. 28) Qxtends within the ~ha~t 436 and has its longitudinal axis parallel to, but off-center from, the longitudinal axi~ o~ th~ maior ~ha~t 436. Ths outer end of shaft 440 term$natee in a head 442 ~rom which an eccentric pin 444 projects. Bearings 446 and ~n outer bearing shield 448 are secured to thi~ pin and thereby hav~ centers which are eacentric to thQ longitudinal axis o~ the minor shaft.
An air actuated cylind~r a0~embly 470 (Fig. 28) is operatively coupled, as explainad balow, to th~ minor shaft ~40 ~or rotat~ng this ~ha~t. As the ~inor ~haft 1~ rotated relative to ~a~or sha~t 436, th~ di~tancQ or sep~ration between the center of thQ pin 4~4 and ~ha out~ide surface of fixed pro~ection 438 increase~ a~ shown movin~ fro~ Figs. 26 ~o 27. ~ha splndl~ or plunger ~ip 232 has its longitudinal axi~ disposed in a horizon~al line nonmal to the plane of carrier 220. Prior to in~ertion o~ the plunger tip into the hub 278, tha minor ~h~t i~ ~irst rotat~d relativa to the ma;or sha~t to an orientation which ~ini~ize~ the distance between th~ c~nter o~ pin 444 and th~ outer ~ur~ace of protrusion 438, as ~hown in Fig. 26. The tip 232 is also rotated, by a motor 510 a~ explained below, to position protru~ion 438 in a down position, beneath pin 444. This provides maximu~ clearance ~or ~asy insertion o~ the tip 232 into the hub. Thu~ oriented, the tip 232, includ mg pro~ection 43~ and pin 444, i~ insertod into tha hub ~78 of a carrier 220. After insertion, th~ minor ~ha~t 440 is rotated relativ~ to the maJor ~ha~t 436 to bring th~ baaring 446 into rolling contact with thQ lnner sur~aca o~ the hub 278 and li~t thQ carrier 220 ~rom it~ ~upporting transportar 222.
Additional rotation og th~ minor sha~t 440 cau~es further li~ting o~ the hub until eventually the carrier hub 278 i8 cla~ped and ~ripped by the bearings 446 and ~ixed pro~ection 448. The eccantri~ b~arings 446 are prevented from rotating over the center of the m~or ~har~ 436. That is, the ~9~443 interior aurface of the hub 278 i9 ~ized to be gripped by projections 438 and bearings 446 before the b~aring~ ~ove to an over c~nter position. When engaged in thi~ manner, the hub pre~ent~ further rotation of the minor shaft 440.
Aft~r the cla~ping and lifting action i5 complete, and the tran~porter 222 is moved away ~ro~ th~ ~puttering targets in a chamber, the plung~r drive ~echanism 230 rotate~ the major shaft 436 and ~h~ supportQd carrier during ~he d~posi-tion process. Upon completion o~ proce~sing, plunger tip 232 i8 stopped, with the protru~ion 438 in its down positlon, in the ~ams orientation a3 when the tip was in~erted into the hub 278. In addition, tran~porter 222 i~ po~itioned under the hub 27a of the plung~r supported carrier 220. Th~ minor shaft 440 i8 th~n rotat~d in ths oppo~ite dir~ction fro~ that pre~iously described to lower the hub onto the txansporter and r~lea e the hub. The plungsr tip a32 i~ th~n withdrawn fro~ th~ hub 80 that th~ tran~port~r 222 may transfer t~e carrier to another chamber.
Ther~ are a numb~r o~ advantag~ to thi~ type of plunger. First, there ar~ tring~nt r~guirements ~or po~i~ioning o~ a tran porter 222 and it3 suppOrted plane~ary carrier 220 in a chambsr. That i8~ the hub 278 need n~t be perfectly align~d with the cent~r o~ tha plung~r tip 232 in order ~or th~ plunger tip to bo in~arted into the hub. More-over, becau3a o~ thQ po~ive cla~ping action by the plunger tip, good ~l~ctrical aontact i~ mad~ between th~ plunger 228 and the hub 278. During ~puttering, a~ previously ~en~ioned, grounding of the ~ub~trates i~ acco~pli~hed through the carrier and plunger. Also becau~e o~ the po~itive clamping action, the rotating carrier will be m~intained in a vartical plane, perpendicular to th~ longitudinal plung3r axis. Thi~
~inimizee di~c suhstrate wobbling, motlon out o~ a vertical plane, in the ~heave groove~ ~nd thu~ Minimizes thi~ poten-tial ~ource o~ unde~irabl~ particles. Al~oj ~uch wobbling could modulato th~ ~putt~rin~ by p~riodically moving certain areas o~ the substrate3 clo~ar to the sputtering targets and thereby cau~ing a ~riation in tha thicXne~ of the deposi-tion on euch ~ub~trate areas. Finally, ~his clamping action ~91~43 eliminate~ relative rotation between the hub 278 and plunger tip 232 during sputtering to thereby eliminate partial gene-ration that could otherwise result ~rom such relative rotation.
Referxing to Fig. 28, the minor ~haft 440 is rotata~ly ~upported within the ma~or sha~t 436 by a pair of bearings 450 separated by a spacer 452. ~a~or shaft 436 extends through the wall o~ ths depo~ition cha~ber. A coupler 454 connects the inner end o~ th~ minor ~ha~t to a BtUb ~haft end 456 of a commercially available s~aled rotary motion ferro-~luidic ~ee~ through coupl~r 458. An 0-ring ~eal, not ~hown, is provided to seal ~ead through 45B at its connection to ~ha~t 436. The other ~tub shart end 460 o~ the reed through is coupled by a ~ushing 461 to an elongated dri~Q screw re-ceiving h~lix nu~ 462. As a re~ult, rotation of ~he helix nut 462 cau~Qs the stub ~ha~t ~n~ 460, 456 and t~ minor s~aft 440 to rotat~ and thsr~by producas th~ preYiou~ly de-scribed cla~ping action. Th~ feed through 458 and helix nut 462 are po~itioned within ~ hollow Qxternal ma~or ~haft ex-tension 464 (seQ al~o Fig. 10) to ~hich a plunger rotating drive pulley 466 is ~ixedly mounted. A pneumakic actuator mounting collar 468 i~ ~ixQdly connected to pulley 466. The shaft extension 464 i8 threadedly connected to ma~or ~ha~t 4 3 6 . A ga~k~t i8 provid~d between the~e two 3ha f t component3 where they ~oin together.
A computsr controll~d ~olenoid actuat~d pneumatic cylinder as~embly 470 i~ coupled by collar 468 to the drive pulley. ~he pneumatic cylinder a~sembly 470, a~ explained below, i~ designad to ~electively rotate the helix nut 462 to cause a corre~ponding rotatlon o~ the minor ~ha~t and, there-by, the plunger li~ting and clamping actoin. More specifi-cally, the pn~umatic cylinder a~embly 470 includes an actuator or pi~ton cylinder body 472 clamped in place by col-lar 468. A piston a~aembly i~ poaitioned wlthln body 472 and include~ a pi~ton head 474 to which a pi~ton rod, having a ~irst exten~ion ~ectisn 476 and a second drive screw section 478, i8 mounted. ~ ~lat sidQd ~lot in the end o~ extension ~ection 476 ~it~ over the ~n~ o~ drive ecr~w section 478 such ~g~4~3 that linear movement of piston head 474 results in linear movement of drive screw section 478. Drive ~crew section 478 comprises a non-rot~ta~le high helix drive screw ~hich is in-serted into the rotatable helix nut 48 whQn the apparatus i~
assembled. As tha piston head 47~ slid~ within the body ~72 toward the collar 468, the drive screw section 478 rotates helix nut 462 relative to ma~or ~hsf~ 436 and also rotates the minor ~haft 440 relative to th~ msjor ~ha~t. Thi~ con-~erts linear motion o~ the pi~ton into pivoting motion of the minor ~ha~t. A pi~ton retuxn ~pring 480 ~iases pi~ton head 474 in the oppo~ite direction away ~rom collar 468. Guide pins 482, in~erted through internal bore3 o~ th~ piston head, guide the sliding mov~ment of the pi~ton head. Thes~ guide pin~ 482 al30 pre~ent rotation of the pi5ton head relative to the shaft p~rtion~ 436, 464. The ~nd o~ body 472 i~ closed by a valve body 484 to which a ~urc~ o~ air i~ coupled by a rotary air union 486. A pair o~ ~low controls 488, like con~
trol~ 416, 422, control the ~low o~ air through val~e body 484 to the interior o~ the body 472.
~ computer actuated solenoid controlled air valve is operated to deliver air to a~ bly 470 as required to shift drive crQw section 478 ~orwardly toward th~ spindle tip.
Thi~ rotat~ thQ minor sha~t 440 ~o a~ ~o li~ and cla~p the carrier 220. ~h~ co~puter al~o conkrol~ thl~ air valve to relieve air pr~sur~ fro~ th~ p~ton head 474 a~ required to lower and r~ e th~ carrior 220. Wh~n air pressurQ is re-li~v~d, pring 480 ~hi~t~ drivo ~crew s~ction 478 rearwardly and cau~s the lowering and relea~ing o~ th~ carrier.
The plungQr drive a~sQ~bly 230 includs~ a chamber wall attachm~nt plate 490 which i8 mounted to tho r~ar wall o~ the deposition chamb~r a~ ~hown in Fig. 10. Three horizontal guide ~ha~t3 492 pro~ect outwardly ~rom plat~ 490 and away fro~ the depo~ition ch~mbor. A ~otor carriage plate 494 is slidably mount~d to the ~nds o~ tha guide ~ha~ts 492 a~ter th2 carriags plata 494 i~ po~i~ion~d on th~ guid~ ~ha~ts.
The carriage platQ 494 1~ ~aalQd ko the rear chamber wall by a flexible~bellows 500. In addition, a rotary Rha~t vacuum seal 502 i~ po~ition~d within an annular projection ~9~443 4g 504 of the carriage plate 494. Shaft extension 464 extends through the rotary seal 502. A pair of 0-ring gaskets ~ur-round shaft extension 464 to seal th~ ~pace between this shaft extension and the interior surface o~ saQl 502. A pair of external o-ring gask~t~ ~shown in Fig. 28, but unnumbered) surround seal 502 to sQal the space between 5~al 502 and the carriage plate projection 504. Rotary ~eal 502 permits rota-tion of shart 464 and thsreby the rotation of the major shaft 436. This results $n a corresponding rotation of a supported carrier 220 during sputtering. Because of the sealing accom-plished by seal 502 and bellows 500, the deposition chamber is ~ealad against leakaga through the plunqer drive as~2mbly.
Axial ~hifting of the plunger 228 to in~ert and with-draw the plunger tip 2~2 i~ acc~pli~h~d by a pne~matic cylinder 506. Cylinder 506 has ita hou~ing connected to the carriagQ plat~ 494 and it~ pi~ton rod connacted to the plate 490. A computer controlled ~olenoid actuated valv~ delivers air thro~gh a flow controller S08, like c~ntrollers 416, 422, to cylindar 506 to extend and rotract the piston rod as re-quired. When th~ piston rod i~ retracted, the carriags 494 i~ shifted axially toward the depo~ition chambex and the plung~r tip 232 is in~erted into th~ hub of a carrier. In contra~t, when tha pi~ton rod i8 ext~nded, the carriage is shifted in the opposite directlon and the plungsr tip is withdrawn ~ro~ the hub. ~ellow~ 500 provide~ a vacuu~ ~eal while permitting the axial motion of th~ plung~r tip, A plungor rotatlon 3tep ~otor 510 is mounted by a mounting block 512 ~o th~ carriagQ pla~e 494. When he mo~or 510 i9 snQrgizQd by ~lectrical pulse~, a drive pu~ley 514, mount~d to tha motor ~ha~t 516, rotat~ in steps. Drive pul-loy 514 i3 coupled by a ti~ing b~lt 518 (~e~ Fig. 10) to the pullay 466 mounted to the sha~t 464. Consequently, wh~n the motor 510 i8 opQrated, th~ oxten~ion sha~t 46~ and its con-nected ma~or sha~t 436 rotats. Consequcntly, whan a carrier 220 i~ gripped by the plungr tip 232, motor 510 i9 operated to rotate the carrier and mov~ ~ubstrate~ 260 on the carrier in a planetaxy ~as~ on pa~t ~puttering target~ in the cham-ber.

The computer 46 controls the electrical drive pulses transmitted to motor 510. These pulses are monitored and counted to determine the degree and rate of rotation. Also, feedback to the computer is provided by signals from an op-tional conventional shaft encoder. This shaft encoder in-cludes a reflector 522 coupled by coupler 520 to the motor shaft. A conventional optical through beam sender 524 senses the position of reflector 522 and thus of the motor shaft.
Signals from the sensor 524 are transmitted to the computer and used to track the shaft position and thus the rotational position of the plunger. Therefore, for example, the plunger may be rotated to always position protrusion 438 in its down position following processing so that the plunger tip 232 is in position for easy withdrawal from a carrier 220 and inser-tion into the nest carrier.
Transporter, Track and Track Drive Mechanisms The transporter or robot 222, track 224, and track drives 226 are shown in Figs. 15, 23, 24, 24a and 25. These mechanisms are designed to transfer carriers 220 from one chamber to the next chamber when the valve housing 26 between the chambers is open.
In general, a transporter 222 includes an elongated body 530 (Figs. 23, 24, 25) supported at its front and rear ends by respective wheel supported trolleys 532, 534. These trolleys travel along the track 224 from chamber to chamber.
The transporter arms 340, 342 are vertically extending, parallel, spaced apart, and are mounted at their bases to the respective sides of the body 530. Each of the arms is provided with a respective arcuate cradle or saddle 540, 542 at its upper end. The hub 278 rests in these cradles (Figs.
23, 24) with the carrier 220 positioned between the arms, when the carrier is loaded onto the transporter 222. The hub ring 280 and a section 544 of hub 278 act as spacers to maintain the separation between the carrier 220 and the arms 340, 342.
As shown in Figs. 23 and 25, the arms 340, 342 are dis-placed from the center of the trolley body 530 toward one end 1~9~43 of the body. With this construction, following the loading of ~ carrier onto a plunger in a processiny chamber, the transporter 222 is moved to a parked position adjacent wall 58. Thi moves the arms 340 and 3~2 out of the way of the sputtering targets so that there is no need to remove the transporter from the chamber prior to sputtering, if desired.

Each of the trolleys 532, 534 is pivotally mounted to the underside of the body 530 as shown in Fig. 25.
That is, a shoulder screw 5S0 mounted thereon is inserted into a recess at the underside of the trolley body~ A cover plate 556 holds this assembly within the recess. The lower end of screw 550 is threaded into an opening 558 formed in the upper surface of a trolley ~5 b~dy 560 of the trolley 532~ An annular spacer 561 maintains a separation be~ween the elements 530 and 560.
Trolley wheels 562, which comprisa bearings, are each press fit onto a dowel S64 which is then pressed into an opening 566 of the trolley body to secure thP wheel to the trolley. This construction also allows for compliance along the plane of the track since each trolley pivots on its own center. The trolley also is provided with non-metallic bumpers 568, 569.

The ends of a track in a chamber are spaced from the respective ends of the tracks 224 in adjacent chamber~. Thus, a gap exi9t5 :Ln the tracks batwaen the chambers. These gap~ are located within the isolation valve housings 26 and the valves 110 (Fig. 7) slide in these gaps to close and .isolate the chambers without interference by the tracks. This arrangement of two trolleys per transporter 222 enhances the smooth transfer of the transporter across these gaps between tracks 224 in adjacent chambers. Also, the distance ~, ~
P , 1 129~4~3 51a between the front and rear sets of wheels of each trolley is greater than the distance across the gap.
This facilitates travel of the trolleys across the gaps without skipping.

The track assembly 224 comprises an elongated straight rigid trolley supporting track 580 supported at walls 56 and 58 by track mounts 582, 584. A trolley receiving recess 586 is formed in the upper surface of the track 580 and is bounded by ~irst and second track side walls 588, 590. The , . . .

~ ' ~a.Z9ï~

trolleys 532, 534 fit within thi3 rece~ and are guided in a linear direction along the longitudinal axi~ o~ th~ track by the side walls 588, 590. The bumper~ 568, 569 guide the trolley~ along the trac~ and prevent unde~irabl~ particulate generating ~tal-to-metal contact b~tween the trolley 560 and tracX wall~ 588, 590~ The track ~ positioned in the cham-bers to guide the transporter 222 and ~upported carrier ~20, with the support~d sub~trate3 positioned in a plane centered between tha front and rear ~putt~ring target assemblies 40 or 42. This enhancas the unl~orm sputtering of the s~strates 260 during th~ previou~ly describ~d deposition processes.
Fir~t and second elong~t~d cover strip~ 592, 594 ara moun~ed to the upper surface~ o~ th~ r~p~ctive walls 5a8 and 590.
Covers 592, 594 pr~vent th~ trolley~ from li~ting upwardly out o~ tha track. An elongated chain guiding slot 596 is providad in th~ ~loor o~ th~ rece~s 586. Another such chain guiding ~lot 598 i~ providsd at khe undar~id~ of the track 580 ~or purpose~ expl~i~ad below~
The transport~r 222 i~ driY~n along the track by a chain driva mechan~sm 226 a3 follows. Spsci~ical~y, as ~hown in Fig. 23, a contlnuou~ loop o~ chain 600 spans the ~amber and i~ supported at its re~p~ctiv~ ~nd~ by toothle~ pulleys 602, 604. From pulley 602, thQ low~r sectoin o~ the chain pa~C~s over and i~ drivanly ~ngagad by a drive sprockQt 608.
An idler wheal S10, in coop~ration with th~ pull~y 602, main-tain~ the chain 600 in contact with thQ driva ~procket. The 810t 596 provide~ clearanc~ ~or th~ chain 600 where it passe~
over the drivQ sprocket 608. The pullsy 604 i~ mounted to a ~ensioning block 612. ~lock S12 i~ ~hirtabl~ toward and away ~rom wall 58 by a te~sion ad~ust~ent ~crew fil4 to ~hereby ad~ust tha ten~ion in the chain 600. Other optional chain ten~ian ~d~ustment ~echani~m~ are egually suitable. For example, pulley 604 may b~ ~t~tionary ~nd idler wheel 610 may be ~ovable to ad~ust the chain ten~ion. A ch~in guard 616 mo~nted to ths und~r~ide o~ thQ track 580 guid~ the travel of th2 lower ~ection o~ th0 chain. In addition, the upper section o~ the ch~4n 600 pa~e~ through the chain guiding slot 596 and und~rneath tho rQ~p~ctive trolleys 532, 534.

~Z9~3 Each of the trolley bodi~ 560, a~ shown in Fig. 24a, has a row of downwaxdly projecting chain engaging ta~th 620. These teeth travel in th~ ~lot 596 (Fig. 24) and are engaged by the upper ~ection o~ the driva chain. Con~eguently, when the chain is driven in either direct~on, th~ transporter 222 is correspondingly driven.
The samQ link (i.~., 622 in Fig. 24a) alway~ engages the same tooth of a transporter in thl~ cha~ber. ~herefore, by monitoring the po~ition o~ thi~ link, the poBition of the transporter in the chamber i~ known. The transporters have ~our positions within a chamber, corr~ponding to four posi-tions of the link. The~e po~ition~ include a load po~ition in which the hub 278 i8 centered on the plunger 228, a parked po~ition in which the tran~porter i~ mov~d ad~acent ~o a wall 58 to shift tha arm~ 340, 342 out Or the way of the ~put-tering targets, a rear crossing po~ition in which the trans-porter i~ positioned for a tran~fQr to the left, and a ~or-ward cro~sing po~ition in which th~ trans~orter i8 po~itioned for a trans~er ~o tha right. In addltlon, tw~ positions of the link ar2 u~d when a tran~porter i~ not engaged on the chain containlng th~ link. ~he~e additlonal link positions include a rear cro~sing offs~t, in which tha chaln i~ posi-tion~d for entry o~ a tran~porter from a cha~ber to the le~t (l.e. in Fig. 23), and-a forward cros~ing of~aet, in which the chain is po~itionQd ~or ~ntry o2 a tran~porter from a chamber to thc right.
In ~ l~ft to right tran~ar (a~ shown in Fig. 24a), the track chain in the right chamber is shifted to its rear cro~sing o~et. Then, the track chain in the left cha~ber i~ shi~ted to it~ forward crossing position which po~itions the rorward ~-ooth 620 to tha point Or contact with the chain link which is b~yond the link 622. As chown in ~ig. 24a, the receiving ahain i9 in a prop~r poBition wh~n thQ top Or the roller link 622 i9 spac~d ~u~t b310w ~axaggerated in Fig.
~4a) the lower edg0~ o~ ~irst kooth 620. ~hi~ alignment re-duce~ the wear and the potential binding of the chain. After th~ transportQr is~riv~n to the right, to the point of con-tact as pr2viously dQ~cribed, the chain driva in the left ~291~43 cha~ber is haltsd. The chain drive~ in the left and right chambers are then driven simultaneously, in synchronization, in a direation whioh mvv~ the transporter to the right and into the right chamber. In a right to left transfer, the track chain in the left chamber i~ po~itioned at it~ forward crossing offset. Then, the track in th~ right chamber is positioned at it~ r~ar cros3ing po~ition. Ths chain dri~es are again ~imul~aneou~ly dri~en i~ ~ynGhronizat~on to drive the transporter into the left cha~ber.
The 3am~ link (i.e., link 622 in Fig. 24a) alway~ en-gages tho ~ame tooth o~ a transporter.
With referenca to Figs. 15 and 24, th~ drivo ~ochanl~m 226 includes a st~p moto~ 626 driY~nly conn~cted by a b~lt to a drive pulley 628. ~h~ pull~y i~ coupl~d through a rotary seal 630 to a flexible coupling 632 located within the cham-ber. A~ shown in F~g. 24, the ~l~xibl~ coupling 632 ls con-nectad ~o the drive ~prockot 6~8. There~ore, when the ~otor 626 is operat~d ~o driv~ ~h~ drive sprockst ei~her in the clocXwisa or co~nterolockwi~e dirQctlona, the chain i9 driven in the corresponding direction. The computer 46 control~ the transmission o~ driv~ pul~es to th~ motor 626. By counting these pulse~, the computer track~ the po~ition o~ the chain link~ 622 and thus the po~ition of transporters 222 in the system. ~ shaft encoder (not shown, but integral with the motor) i~ utilized to monitor the move~nt of the mokor drive sha~t in a con~entional manner. Signals from the shaft en-coder are tran~mitted to ths computor to proYid~ feedback o~
the position o~ the chain within th~ chamber. Of course, limit ~witche~ ~r optical detectQr~ may al80 be used to moni-tor the posltion~ o~ thQ tran~portQr.
A singlo tran3porter 222 may b~ utilizad to trans~er a carrler 220 from tha lo~d chamber 12 through the daposition chambers 14-20, and to the unload cha~ber 22. In this case, a~er the carrier i8 unloaded in cha~ber 22, this transporter i9 raturned to chamb~r 12 to rec0ive the n~xt carrier. How-ever, ln th~ illu~trated prsferred embodiment, to sp~ed the processing, three ~ch transporter~ ar~ e~ployed. The ~irst transport~r ~rav~ls between cham~er~ ~2, 14 and 16. The ~9~443 second transport~r travel3 between chambers 14, 16 and 18.
Finally, tha third of these tran~porter~ travels between chambers 20 and 22. Therefore, under the control of computer 46, certain of the transporter~ ar~ tran~porting carriers 220 in certain part~ o~ the sy~t~m whila other tran~porkers are transporting other carrier~ elsewhere in the sy~tem.

Nater Coolinq Syst~3m The water cooling syatem ~or the cathode a~semblies 40, 42 in the processing chamberR 14 through 20 i9 ~hown in Figs.
29-31. The Fig. 29 cooling ~y~tem i3 a clo~ed loop sytem.
Alternately, water ~rom ~ municipal water supply or other ~ource may be util~zed and r~tur~ed to the at~rm drain~ or a system sew~r after U5~.
With referenc~ to Fig~. 29-31, cool water from a refri-g2ration apparatus 636 i8 directQd through a ~ain shut-o~f valve 638, a ~ilter 640, te~peratur~ and pr~sure switches 642, 644 and to branch lin~ 648 and 650. The temperature 8witch 642 i~ interlocke~ with a ~ain huk-of~ valve for turning off water flow in the ~ent ths cooling water temper ature exc~eds a predeter~lned level, 3uch a3 70 degrees fahrsnheit (21.11 degrees Cel~iu~). Thl~ shut-o~ valve is al o closed and an alarm i~ triggored i~ the pres~ure sen~ed by pres~ure switah 544 exceed~ a predetermined level, ~or ex-ample, Rixty p~ig ~413,685.4 pa~cal). Water ent~ring line 6~B is directed through th~ wator ~ackets o~ the cathode as-semblie~ 40, 42 of tha chambers 14 and 16 and then returned via a lin~ 652 to a ~ain return line 654 and then to the cooling apparatu~ 636.
Similarly, coollng water i~ ~ed through the water ~acket~ o~ the cathoda as~emblie3 Or chamber# 18 and 20 and returned via a branch line 656 to the main return line 654.
~anually operatod ~hut-o~ valv~ 658 are provided ~or ~hu~-ting o~ the water rlow a~ ~sired. The coollng water aupply sy~tem utillzad for chambers 18 and 20 i~ ldentical to that utilized ~or chamber 14. There~ore, the cooling 9y8tem for chambers 18 and 20 ~ill not be de~cribed in detail.

~lZ9~4~3 Cooling water ~lowing along lines 648 iB directed as shown by the arrow~ to bran~h linea 660, 662 leading to the respective chambers 14, 16. From line 660, the cooling wat~r is fed through lines 198 at the re~pectiYe ~ront and back sides of chamber 14. At each ~ide of the chamb~r, water flows through one cathode a~sembly 40, through a coupling line 667, through another cathode assembly 40, and is re-turned via return li~a 200 to a branch line 664. From line 664, the water flows via lines 652 and lin~ 654 to the cooling apparatus 636. Isolation valves 666 are positioned in the water supply lines between linQs 660 and the respec-tive lines 198. Similar isolation valve~ 668 are interposed b~tw~en the llnes 200 and tho return line 664. When a set of valves 666, 668 a330ciatQd with a ~low path through a ~et of cathodes 40 at the ~ront or rsar o~ th~ cha~ber 14 are clo~ed, the rQspQctive ~et o~ cathode~ is i~olated ~rom the water ~upply sy~te~ for rapair or oth~r purpose~. Also, com~
puter monitored water ~low ~w~tches 670 are positioned be-tween the line~ 200 and 664. Ths~ switche~ enable the com-puter to detect water ~lowing thxough tha ~ront and rear sets of cathode as~emblies 40 and to block energi~ation o~ the cathode assemblies in the event cooling wat~r i~ not being delivQred to the a85emblie9.
In cha~b~r 16, cooling water ~rom th~ lin~ 662 is directed through isolation valve~ 672 and thorugh RF matching networks 674 to the ra~peGtivo cathode as~e~blie~ 42. The outlet lines 200 ~rom these cathoda asse~blies pa~s through the RF networks 674, i~olation valve 676 and water flow switchRs 678 to a water return branch line 680. From line 680, water i~ returned to line 652 and via line 6454 to the cooling apparatu~ G36. The pairs o~ valva~ 672, 676 operate like the valves 666 and 668 to selectively i~olata the cathode as~e~blies 42 ~rom thQ w~ter cooling system. In addition, the water ~low 3witche~ 678 operate lika the previously d~cribed ~witch~ 670.

~29~43 yacuum Pumping and Sputterinq Ga Su~ly Systems The sputtering gas supply and vacuum pumping ~ystem utilized in the embodiment of Fig. 1 are shown in Fig. 32.
Sputtering gas is ~upplied ~rom one or more sputtering gas sy~tem~ ~84 to the chamber~ 14 I:hrough 20 for the sput-tering processes. On~ ~uch sy~t~m is typically employed for each type o~ sputtaring gas which i8 used. Argon or other sputtering gas from a source 686 i8 ~ed through a regulator 688, pa~3t a manually controlled shu1:-off valve 690, and through a two micron filtex 692. From filter 692, the sput-tering gas i~ deliv~rcd via a conduit 694 through a f`low in-dicator 696, a computer actuated sol~noid controlled valve 698, and through a needla valvQ 7û0 to the deposition chamber 14. Needle valve 700 is ad~u~ted to pro~ridel the~ appropriate gas flow rate to the chamber. Th~a solenoid controlled valve 698 is op~ned and closed in ra~pon~e to commands from the computer 46 to deliver ~puttaxlng gas to chamber 14 as reguired. A conduit 702 deliver~ the sputtering ga~ from fileter 692 to other chambers utilizing ~he ~a:~e type o~ gas.
Each such cha~ber i~ provided w~th it~ own ~low indicator, sol~noid controllad valve and na~dl~ v~lv~.
Each o~ ths vacuum pumping ~tack~3 34 are con~tructed from commQrcially available colaponents. Furthermore, the pumping stack0 34 ar~ identi~al and will be described in con-nection with ths pumping ~tack u~ed for chamber~ 12 and 14.
Each vacuum pumping stack 34 include~ a cryo co:mpressor 706 which is coupl~d to a cryo pu~p 708. The p~Llnp 708 is coupled to a variabl~ sp~3ed orifice throttle valve 710 in coD~Qunic:ation with a cryo trap 712. l'hs trap 712 i~
selectively coupled to the chamber 12 by a high vacuum valve 714. The ~ryo trap 712 i~ provld~d with a van~ 716.
Suitabla solenoid a¢tuatad v~lva~ 720, 722, 724 and 726 are included in lines leading ~o tho ~y~tem ~or purposes explained below. Howl3ver, in gen~ral, valve 720 compri~as a cryo syatem regener~tion valve, valv~ 722 comprisss a rough vacuum valve, val~ 724 co~prises a vacuum system purging valve, and valve 72~; compri~e3 a chamber venting valve. In ~29~3 addition, a liquid nitrogen ~ill control valve 728 is also included. Furthermore, gauges numbered as 730, 732 and 734 are provided for monitoring the vacuum sy~tem.
Th~ fir~t three chamberc 12, 14 and 16 are coupled by a rough vacuum line 738 to a mechanical rough vacuum pumping system 736. A ~imilar rough pumping ~ystem i~ provided for the chamber~ 18, 20 and 22 and iE~ c:oupled to these chambers by a rough vacuum line 739. A rough ~acu~m crossover valve 790 permits selective coupling o~ mechanical pumping systam 736, via line 739, to chambQrs 18-22 and coupling of the other mechanical pumplng systQm, via line 738, to chambers 12-16 a~ de~ired. ~he mechanical pumping 3y8tem 736 includes a mechanical pump 740, a comput~r controlled solenoid opera-ted shut-o~ valv~ 742, and a bellows 744. Also, a molecular sieve 750 i3 po~itioned in rough line 738 between the pump 740 and valves 720, 722 Or ~ach of the chambers 12, 14 and 16. A sieve heater (not ~hown) i~ provided within sieve 7~0.
A similar ~ieve and h~ater i3 provided i~ rough line 739 ~or cha~ber~ lB, 20 and 22. In addition, pr@ssure gaugs~ 752 and 754 are provided ~or monitoring the ~tatu~ of the mechanical rough pumping 8y8tam.
~ i~uid nitrog~n i~ suppli~d to each o~ the cryo traps 712 via a lin~ 785 ~rom a liquid nitrog~n ~upply syst~m 760.
The liquid nitrogen supply sy~tem include3 first and second liquid nitrog~n tanks 762, 7G4, pr~sur~ relief valve~ 766, 76~, and 770, and oomputer controlled 301enoid actuated flow valv~s 782, 784 and 786.
The gauge~ 730 monitor th~ chamber pxes6ura and include a rough Yacuum gauge ~or monitoring the establi~hment o~ the rough pre~ure in chamber 12. This rough vacuum gauge i~ of the commercially avallabla type which tran~mits an eloctrical slgnal corre3ponding to the gauge pre~sur~. Thi~ ~lectrical signal i9 tran~mittQd to tha computer ~6 ~or monitoring of the chamb~r pre~ure. Gauge~ 730 al~o include a high vacuum ion gau~a ~or monitoring th3 vacuum in chamber 12 when a high vacuum i8 being e~tabliah~d a~ explained below. In addition, gauges 730 include ~ ther~istor gauge.

. ~

~LX91D~3 A ~uitable rough vacuum gauge i~ a series 275 Convec-tron gauge manufactured by the Granville-Phillips ~ompany.
Suitable ion and thermistor gauges are Perkin-Elmer DGC~III
gauges. In addition, gaug~s 732 co~prise cryo temperature gauges and gauges 734 may compri~e Convectron gauges. The gauge~ 730 which monitor the chamber pressure are the same for chambers 12, 16 and 20 except that, in chamber 16, a ca-pacitancs monometer gauge i~ used as the rough pressure gauge. The gauges 730 for the chambers 14 and 18 comprise capacitance monometer gauges such a~ model 227 gauges produced by MXS In~trument~. In addition, the gauge3 730 for chamber 22 compri~e~ a rough vacuum Convectron gauge. Also, although not shown in Fig. 32, the radio ~requency sputtering chamber 16 includes a conventional hot filament for heating the sputtering gases as required.
Each of the above gauges, like the above described rough vacuum gauge, may be of the type which generate~ elec-trical ~ignals corre~ponding to the para~eter being measured.
Such signals are tran~itted to and monitored by the computer 46.
The vacuum pumping system 36 u6ed in chamber 22 i like the pumping ~yskems utilized in chambers 14, except that a throttle valve 710, cryo trap 712, liquid nitrogen fill valve 728, and sourcQ o~ liqu~d nitrogen is not used. Because ful-ly proce3~ed Bubstrates are received in cha~ber 22 and then unloaded, it i8 not a~ important to establish a~ high a vac-uum in thi~ latter chamber a~ in the other cha~bers. For that matter, by placing the components o~ the throttle valve in contact with the cryo pump, the cryo trap 712, liquid ni-trogen supply, and nitrogen ~ill valve~ 728 may be eliminated from the other chambers a~ w~ll.
The operation o~ the vacuum pumping sy~tem can be understood with referenc~ to Fig. 32~ As~ume that a tray 270 o~ sub~trate contain~ng carrier~ 220 have ju~t been loaded in chamber 12 and the door to this cha~ber ha~ been closed to ~eal the chamber. Also a~sume that a rough vacuum has been established in the~ryo portion of the pumping system by the mechanical pump 740 via valves 720 and 724. In ~his case, 1~931 ~

valve3 724 and 726 are closed. Also, the high vacuum valve 714 c108~ the path bekw~en the chamber 12 and the cryo trap 712. However, th~ rough valves 722 and roughing pump valve 742 are open~ ~ump 740 draw3 a rough vacuum ~r~m the chamber 12 via a path through valve~ 714, 722 and 742. A~ter a rough vacuum has been established in th~ sy~tem, for example, one millitorr (0.13 pa~cal), valve 722 i~ closed. The high vacu-um valve 714 ia then openQd a~d th~ cryo pump 708 i~ operated to continue the e~tablish~Qnt o~ the desir~d vacuum in cham-ber 12.
~ iquid nitrogen ~rom ~ub-sy3tem 760 i~ delivered via liquid nitrogen ~ill control valve 728 to the ~ryo trap 712 to assi~t in the sstabll~hm~nt o~ the high vacuum. After the high vacuum is establish~d, it i~ maintained within chamber 12 dus to the tightly ealed natur~ o~ thi3 chamber. In addition, becau~e the cha~ber 12 is selectiYely isolatable by th~ ieolation valva~ ~ro~ the ad~oining chamber, a v~cuu~ may be e~tabli~hed in thi~ ~hamber without int~r~ring with a previously ~stablish~d vacu~ inside t~ ad~oining chamber.
It is important to e8tabli8h an extrem~ly high vacuum in the chamber 12 prior to opening this chamber to the adjoining chamber. For oxamplo, a vacuum on the order of 1 x 10 7 torr (1.33 x 10 5 pa~cal) may be established in chamber 12. Otharwise, it h~s b~sn ~ound that som~ conta~ination, for example wat~r vapor ~rom ~re h ~ub~trate~ loaded ~nto chamber 12, remain3 when the carrier0 220 are transported into chamber 14. This water vapor can intar~ere with the uni~ormity o~ di~c~ produced by teh proc~ss. Purging gas ~uch a~ ~iltered nitrogen, 1~ de}ivered along a llne 788 through valv~ 724 and 726 at desired timea to purge the vacuum pumping ~ystem and al~o to eliminate the vacuum within chamber la prior to opaning the door and loadlng o~
additional sub~tr~te~ to bc proce~ed.
In the proc~s~ing chamber~ 14 through ~0, ~ollowing the initial e3tabli~hment o~ a hlgh vacuum in the~e chamber~, the chambers are pres~urized ~o the de~ired pres~ure with sput-tering ga~ ~rom ga~sy~tem 684.

,,;

A further understanding of the vacuum system will be apparent from the computer logic descriptions and algorithms set forth belowO

Computer Control System As previous~y mentioned in connection with Fig. 1, a programmed digital computer 46 in conjunction with terminals 48 are used to monitor and control the system.
The computer ~6 may comprise, for example, a Hewlett Packard Model 1000 Programmable Digital Computer. The control software used in the computer is designed to control the various sub-systems of the processing system of Fig. 1, including the vacuum pumping sub-system, the materials handling sub-system, and the sputtering sub-system.

In the control instrumentation, the positions of the drive shafts of the six track drive motors 626 are monitored via the track drive encoders. The track drive motors 626 are stepped by motor drive pulses under the control of computer 46. The computer monitors these drive pulses and therefrom, together with the drive pulse count and feedback from the encoders, the position of the transporters in the system is established.
Similar step motors 3~8 and 510 control plunger rotation and the position of the traveling block 330 of the loader/unloader mechanisms. Encoder feedback is also provided for the loader and unloader drive mechanisms.

rrhus, there are twelve step motor axes which are monitored by the computer. Each of the plunger axes, when the optional encoder feedback system is not used, employs a commercially available step motor controller having an indexer and driver card. one suitable step motor controller is available from Superior ~lectric ~b'`
~'. ~' ~'~9~3 61a Company. The indexer produces a timed series of pulses as required to sequence the windings of the step motor 510 and produce rotation of the motor shaft and plunger 228. ~he computer communicates with the step motor controller, as explained below, programs the indexer card with the desired velocity, acceleration and other parameters, and control the operation of the indexer to produce the ... ..

~L~9~4~3 desired outputs ~or step motor operation. ~he driver card amplifies the indexer pulses to a level required to produce a useful amount of torque at the shaft of each of the step motors 510.

The instrumentation for controlling the chain drive motors 626 for the transporter drive mechanisms 226, and the loader/unloader motors 338, include an indexer card, a driver count card, and a count compare card. Such a count compare card is also available as a part of the step motor controller available from Superior ~lectric.
The function of the indexer and driver cards is the same as described above for the plunger motors 510. The count compare card is used to count pulses transmitted from the shaft encoders associated with each of the chain drive motors and the loader/unloader motors. This allows closed loop monitoring of the operation of the step motors so that the actual motor shaft rotation, and thus the distance traveled by the driven component, may be verified after a move is complete.

Communication be~ween the step motor controllers and computer 46 is through commercially available interface cards in the step motor controllers. These interface cards link the computer to the indexer and the count compare cards. While a step motor is vperating, the indexer cards generate motion busy signals which are sensed by a da~ mLL,L ~ cont~l unit and fed through a conventional data ac~uisition control unit interface to the computer. The data acquisition control unit may comprise a Hewlett-Packard Model 3497A main frame computer which employs commercially available function cards. Two or more such units are typically used in the system.

, . ~
., , ~, ~29~3 62a The data acquisition control units interface the electrical hardware of the system of Fig. 1 to the computer. Four different types of functions cards are used in the data acquisition control unit. The first type of function card is a sixteen channel digital input card used to sense motion busy signals from the indexer cards in the step motor controllers, and also the state of all of the limit switches of the system o~ Fig. 1.
These digital cards produce signals corresponding to the state of the sensed components. These ~L29~443 signals are then read by the computer through the data acqui~
sition control unit.
A second type of function card i8 an eight channel high voltage actuator card used to control solenoid~ employed in the system of Fig. 1. E~ch card co~tain~ eight programmable relay~. In respon~e to signals from the~e cards, ~olenoids are operated to supply air pre~ure to control the valve cylinder~ 30, cylinder~ 470, 506 of the plungers 228, and other air controlled components of the system. In addition, other components are also controlled by such actuators, in-clud$ng the ~putteirng power supplies.
A third type o~ ~unction card utilized in the data ac-quisition control unit i~ a twenty channel analog multiplexer card u~ed to gate a Yalect2d analog voltagQ ~ignal into an internal volt meter of the data acgui$itlon contrsl unit.
Each card contains twenty relays ~uitable ~or gating of low level analog signals. The digital co~puter selects the analog ~ignal o~ intere3t, by progra~ming a corresponding multiplex~r relay, from the multiplexer cards. The computer then read~ the internal voltmet~r through tha data acquisition control unit ~nterfac~. The voltage is then converted to a repre3entation of tho physical parameter, such as pre3~urs, by tha computer. ~ a re~ul~, tha compu~er is interfaced between vacuum and other physical parameter monitoring in~truments in the ~ystem.
A flnal typ~ o~ function card i8 a dual voltage digital to an~log outp~t card. These card~ are utilized to convert a nu~b~r generated by the computer 46 to a correspond$rlg con-trol volt~ge lsvel. ~ach card contain~ ~wo channels o~ digi-tal to analog capability for gen~r~ting voltage sigrlals of from plU3 to minus t~n vol~s, at ~i~teen milliamps. The com-putor, vla thes~ digital to analog cards, directly controls devicQ~ which r~quire voltage re~erence~. For example, the output power ~rom the r~dlo ~requency generator~, used in the radio freguency sputtering chaDber 16 in the system o~ Fig.
1, i8 controlled in this way.
;,;

~L29~3 Each of tha above data acquisition control units in-cludes an internal clock which may b2 read by the computer when the system i~ energized to establish a system time.
The vacuum gauges descri~d in Fig. 32, provide vacuum measurements u~ed in conkrolling the operation of the vacuum pumping sy~tem~ 34, 36. The vacuu~ gauges have analog out-puts which intar~ace to the computer 46 indirectly through the multiplexer cards de~cribed above. The computer monitors the signal~, a~ ~xplained abov~, to obtain a digital repre-sentation o~ th~ pressurQ sensed by the in~tru~ent. The in-struments in thi~ group include the oonvectron gauges, tem-peratur~ sensor~ and capacitance mono~eter~.
All computer control function~ ar~ implemented with so~tware in the computer 46. Tho internal hard di~c o~ the computer i~ u3ed for gQneral m~s ~torage of data and pro~
grams. ~n internal micro-Ploppy di~c i~ u~ed for input and output o~ progxam~ and d~t~. In additlon, th~ te:r~inals 48 in Fig. 1 may compris~ a 8y8te~ conBole and log ~or moni-toring the performance of thi~ ~y~tem, an operator co~mand entry terminal, and a color graphic~ terminal for dlsplaying the ~tatu o~ the sy~tem. A printer may be employed to pro-duc~ hard copy outputs ~rom th~ control ~ystem.
The operation o~ th~ co~putsr 46 to control the system o~ Fig. 1 will b2 readily appar~nt to one skilled in the art whan con~idarad ln con~unction with th~ above tn~or~ation and the ~ollowlng log~c d~3criptions. Ths de~cription which ~ol-low~ explain3 the control algorlthms lmplemented to accom-plish the ba~ic control proce~ in the ~ystem of Fig. 1.
The control proc~s~e~ ara cat~gorlz¢d according to u5e.

Materi~l~ Hand,Ling ~ te~ ontrol ProçQs~e~
o~d-Car~le~ Pxoce~s This proce~s is u~ed to tran~er the next available carrier 220 ~ro~ ~he planstary tray 270 to the ~inger 322 of th~ loadar arm 320 in pr~parztion rOr tran3~rring the carrier to the transporter ln the load chamber 12.
1) I~ a carrier a~o ~ 3 on the laoder ~ing~r 322, then skip to st~p 8.

1~9~ 3 2) ~f the planetary tray 270 i~ empty, then terminate with error.
3) I~ th~ first tran~porter 222 is in chamber 12, then open the chamber 12 to 14 gate valve 28 and move the first transportex ~ro~ cha~ber 1~ to aha~ber 14.
4) Put loader ar~ 320 ln a down po3i~ion.
~) Move the loader arm 320 to tray position of next avail~
able carrier 220 and insert ~inger 322 in hub 278.
6) Put loader ar~ 320 in up position.
7) Move loader ar~ 320 to load po~ition with carrier 220 alignQd with track.
8) Finished.
Ad~anaed~ al P~cess Thi~ procQss i~ u~d to tran~fer a carrier fro~ a source ch~ber to a de3tinatlon chamber, the dsstination chamber b~ing th~ next ch~mber in th~ proc~s~. Three ~ub-proce3~es are used $n this proc~ss and will be d~scrib~d here be~ore the main A~anc~-Material Process.

dvanced-Material~ateri~l-to-carrier sub-~roce~3 Thi~ gUb-prOCQ~8 1~ used to move a carrisr 220 from a plunger 2~8 to a transport~r 222 in a chamber.
Algorithm:
1) I~ trAnsporter 222 not in ch~mber, then terminate with error.
2) If cha~ber i~ 12, or cha~ber is 22, carrier 220 is on th~ transporter in tha cha~bsr, then skip to st~p 16.
3) I~ carrier 220 is not on the plunger 228, then terminate with ~rror.
4) I~ plungex 1~ inserted (in rorward position in chamber), then eklp to ~tep 9.

5) Chsck ~or op~rator rsque6ted pau~Q.

6) ~ove transportsr 2~2 to park po~ition.

7) Check ~or opQrator requ~t~d pau~.

8) Insert the plunger (move ~orwardly~.

9) Check rOr operator roquestad pau~s.

10) I~ tran~porter~not at load po~ition then ~ove it to load position.

11) Check for operator reque~ted pause.

12) Release the plung~r grip.

13) Check for op~rator rQquested pau~e.

14) Withdraw the plunger 228 (mo~ rearwardly).

15) Check for operator reque~ted pause.

16) Finished.

Advance-Material/Material-to_Plunger ~ub-process Thi~ sub-process i8 used to move a carrier 220 from a transporter 222 to a plunger 228 in a chamber.
Algorithm:
1) I~ chamber i~ 12 or chamber 18 22 or carrier 220 ia on the plunger, then ~kip to ~tep 14.
2) If tran~porter 222 is not in the cha~ber, then terminate with error.
3) I~ carri~r 220 is not on th~ tran~port~r 222, then ter-minate with error.
4) Check for operator r~qussted pau~e.
5) I~ plunger 228 i~ in~art~d into th~ chamber, ~hen with-draw th~ plunger (movo r~arwardly).
6) I~ plung~r tip 232 i~ in clamping position, r~lease it.
7) Chack ~or op~rator requ~t~d pause.
8) I~ transport~r not at load poaition, then moYe transpor-ter to load po~ition.
9) Check ~or op~rator r~que~t~d pau~e.
10) Insert plunger (mo~e ~orwardly) into hub 278.
11) Check ~or operator requ2sted pause.
12) Clamp tha plunger tip 232 onto hub 278.
13) Check ~or operator regue~ted pau~e.
14) Finished.

Ad~anc~d ~at~rl~1 Loader-materlal-to-c~rier sub~~roce9~
This sub-proce3~ i~ used to mov~ a carri~r 220 ~rom the loader 272 to a ~ran~porter in load chamber 12.
Algorithm: ~
1) If carrier 2~ not on loader arm 320, then termina~e with error.

~L~9~43 2) If loader arm 320 i9 in down posltion, then put it in up positio~.
3) If loader ar~ 320 i~ at load posikion, then skip to step 6.
4) If transport~r 222 i~ in load chamber 12, then open chamber 12 to cha~ber 14 valv~ and move transporter to chamber 14.
5) Move load~r to load position.
6 ) MOVQ tran~porter to load position in cha~ber 12.
7) Put loader arm 320 ln down po~ition.
8) Move loadQr arm 320 to park position adjacent rear wall o~ chamb~r 12.
~in Advance-Ma~rial Proce~s Algorithm: (for main procoss) 1) I~ carrier 220 not ln ~ouxc~ ch mber, th~n t~rminat2 with ~rror.
2) I~ carri~r 220 ln de~tination cha~b~r, th~n p~us~ with error.
3~ Repeat step 2 until caxrior 220 not in destination cham-ber or untll proce~q is a~ort~d by op~rator.
4) Determine which tranaportQr will be u~ed to transport the carri~r a~ follow~:
Source Cha~ber Carri~r U~ed 5) Record current ~tate o~ valve between ~ource chamber and de~tinatio~ chamb~r ~opened or clo~ed).
6) I~ valve clo~d then open it.
7) I~ source ch~mb~r ls cha~ber 12, then exQcut~ Loader-mat~rial-to-carri~r ~ub-proc~ or el~ ex~cute the following two ~ub-~tep3:
7a) Move tran~por~er ~electod in ~t~p 4 to load posi-tion in~ource chamb~r.

9~L443 7b) Exacut~ Material-to-carriQr sub-process in source chamber.
8) Move transporter ~elected in step 4 to load position in destination chamber.
9) Xf sourca chamber is a transporter home chamber (chamber 14 ~or fir~t transporter, cha~ber 18 for seoond transporter, or chamber 22 ~or third transport~r) then execute the following two ~Ub-Bt~p~:
9a) EXQCUt~ Material-to-plunger s~b-proce~s in desti-natlon chamb~r.
9b~ Move tran portsr ~elect~d in step 4 ~o source chamber.
10) Close the gate valvQ betw~n ~ourc~ and desitlnation cha~b~r~.
Unload-Carrier P~ocess This proces~ i~ u~ed to tran~er a carrisr 220 ~rom a tran3porter to the next availabla po~it~on in the planetary tray 270 in the unload cha~ber 22.
Algorith~:
1) If tran~porter i~ not at load poaitlon then terminate with error.
2) If loadQr arm 320 i~ not at park po~ition in unload cha~ber 22, then termina~ with ~rror.
3) If carri~r 220 i~ on th~ load~r arm, ~hen ~erminate with ~rror.
4) I~ carri2r 220 i3 not on the transporter, then skip to step 13.
5) Put load~r arm 320 in down po~ition.
6) ~ove loadQr arm 320 to load posltion.
7) Put load~r arm 320 in up po~ition.
8) Move transporter to park position.
9) Ig planetary tray 270 i~ ~ull, then ~kip to step ~2.
10) MOVQ load~r arm 320 to naxt available tray position.
11) Put loader arm 320 ln down po~ition.
12) ~ov~ load~r arm 320 to park po~ition.
13) Finished.

~L~9~1L4~3 ~9 SE~Ltkerinq Proces~
Proce~s-Material This proces~ i~ used to per~orm a deposition in a selected chamber. Several ~ub-processe^~ are used in this process and ara dQ~cribed first be~or~ deccribing main sput-tering proce~s. Some sub-proa~s~ mentioned below are de-scribed abov2 in ~atQrial~ ~andling System Control Processe~.

SE ~terinc~Froaess~
Ba~ Chambex Sub-l~rocess Thi~ sub-proco3s i3 used to bring th~ pre~sure o~ 3put-tering gas in the chambar to the required l~v21 before igni-ting a plasma.
Algorithm:
1) Check for operator requ~tad pau~.
2) Clo~ throt~le valve 710.
3) open proce~s gas valv~ 698.
4) Wait for proces~ ga~ stabilization tim~.
5) Walt one s~cond.
6) Read chamber pres~ura.
7~ R~paat~ ~tep~ 5 and 6 until chamber prQs~ur~ is within process g~3 pre~surs tolerance, or until a time out occur~.
8) I~ timed out, th~n pau~ with ~rro~ (an oporator re-quested ratry will caus~ execu~ion o~ steps 5 through 8 again with ~nother time out interval).
9) Repeat ~t8p8 5, through 8 until the operator ha~ con-tinued or aborted th~ proc~3 when a tim~ out occurs or until a tlme out doe~ not occur.

~ut~q~inq P~o~e~
S~ar~S~utt~ Qnltor 5u~-~Eoaes~
Thi~ ~ub-procas~ ~tart~ a ~puttsr monitoring process, set ~orth below, which run~ ~oncurrently with ~he ~ain sput-tering proce~. The ~unction o~ tha ~puttar monitoring pro-ce~s i~ to monitor ~ e depo~ition operation.
Algorithm:

~9~L4D~

1) Set handshake flag to false.
2) Start the sputter monitoring process.
3) Wait ~or handshake ~lag to be set to true ~this insures that the sputtQr monitoring proce 8 iS running before continuing).
SputtQr Proces~
Ram-~owe~ Sub-P:roce~
This sub-process i~ used to ra~p up the power level of the RF power generator~ in ~ cham~er to the desired sput-tering power.
Algorithm:
1) CalculatQ tha power increment ~or ~ront and rear power ~upplies a~ ~ollows:
front increment: ~ ~ront-powsr level/~tep~-in-ra~p Rear-incr~ment: = rear-power-l~vol~tep~-in-r~mp.
2) Set front-accumulation and rQar-accu~ulation to O.
3) Increment front and re3r accumulation~ a~ rollow~:
front accumula~ion: o ~ront-accu~ula~ion ~ front-incre~ent r~ar-accumulation: ~ r~ar-accumulation + rear-increment 4) Set front and rear power level to the ~ront and rear accumulation~.
5) Wait for ti~a-per-step.
6) Rep~at ~tsp~ 3 through 5 ~or 1 to (3tep~-in-ramp-l).
7) Set ~ront and rear pow0r level~ to the ~ront-power-level and r~ar-powar-level (thi~ i~ tha actual regu~red power lav~l).

Sutt~r ~rocess $~art-powor Su~-~ro~a~
Thls sub-proces~ i~ use~ to ignite a plasma in a cham-ber.
Algorithm:
1) I~ watsr Slow not pres~nt in target~ then pause wltherxor.
2) Close ~Ipo~er of~ relay contact~ ~or all ~pu~tering power supplie~in the cha~ber.

~29~443 3) Close "power on" relay contacts for all sputtering power upplies in the chamb~r.
4) Turn on tesla coil.
5) I~ power ~upplies are RF thQn execute the ~ollowing two sub-staps:
5a~ Wait ~or te31a pra-ignite ti~
5b) Execute Ra~p-power Bub-proce~s .
6) S~t the pla~ma-on flag to true to indicate to the sput-ter monitoring proca~s that depo~ition has started.

Sutter Proces~ea stoP-Pow~r Sub-E~ocQsE~
Thl~ sub-proce~s 1~ uaed to terminata BpUttarin~ in a chamber .
Algorith~:
1) II pow~r ~upplie~ ar~ RF then executa th~ rollowirlg ~ub-step~:
la) Set p~wer output level o~ all supplie~ im the chaDlber to O
lb) Wait for te~la prQ ignite time.
2) Open "powar o~f" r~lay contact~ ~or all ~puttering power 5uppl ies in ~he chamber.
3) Set the plas~a-on ~lag to ~alse to indicate to the 5pUt-ter monitoring proc~s that the depo~ition i8 complete.

SPut~er, P~o~es~e~.
Spu~ y~ ub-~roce~s Thi~ sub-proc~ is ueed to ~tart plung~r 228 rotating, to iynite a plasma, and to time th~ depoaition proces3 in a chamber.
Algorithm:
1) Execute Start-aputter-monitor 3ub-proce~.
2) Check ~or operator reque~ted pause.
3) Lock the proce~ in it~ memory partition (~or aacurate timing o~ the deposition proces~).
4) Calculate plunyer 2~8 rotation time a~ follows:
rotation-tim~4~ proc~ time + 5 second~

lX9~L44~3 If the sputtering power supply i~ an RF ~upply, then add an addltional time to the plunger rot~tion for plasma ignition, power supply ramp up, and power 8Upply ramp down.
rotation-time: = rotation-time + te~la-pre-ignite-time +
(steps-in-ramp * time-per-step) + tesla-post ignite-time.
5~ Start plunger rotation.
6) Execute Start-power sub-process.
7) Walt for proce~s time.
8) Execute Stoprpow~r 3ub-process.
9) Unlock the proce~ ln it3 memory partition ~so other proce~seR can ~hare th~ memory).
10) Clo~ the proce~ ga~ v~lve 69B.
11) Open thQ throttle valve 710.
12) Wait ~or plunger 228 to ~op rotating.

S~ut~r Proces~ea Process-Materlal Algorithm:
1) Check ~or op~rator reque~ted pause.
2) ~xecuts Mat~rial-to-plunger ~ub-proces~ in selected chamb~r.
3) If a tran~port~r i~ in the chamber, move it to park poaition.
4 ) C10~Q th~ chamber gate valve3.
5) Wait for th~ pre-depo~ition delay.
6) Check ~or op~rator requested paus.
7) Execute Back~ill-chamber sub-process.
8) Check ~or operator rQ~uested pause.
9) Execute Sputter-ln-cha~ber ~ub-proces3.
10) Ch~ck for operator reguest~d paus~
11) Wait ~or the post-depo~ition delay.

~2~4~

Sputter Proces~es Sputter Monitorinq ~roce~s Thi~ proces~ is used to ~onitor th~ deposition in the selected chamber. It is started autv~atically by the start-sputter-monitor BUb-prOCe~B described above.
Algorithm:
1) Set handshake flag to true to i~dicate to the deposition process that the sputter monitoring process i~ running.
2) Wait for the plas~a-on ~lag to be set to true or until a time out occurs. A time out will occur a~ter waiting for the process tim~ for the depoE~ition proces~.
3) I~ time out ha~ occurred then ter~inata.
4) Wait rOr t~ala poAt-ignite time.
5) Op~n "power-on" r~l~y contacts for all ~puttering power ~upplie~ in the chamb~r.
6) Turn o~ the te la coil.
7~ Read chamber prQ~sur~.
8) Calculate 5um and ~um of sguare~ of pre~surs.
9) Wait for 1 second.
10) If no water flow in targets th~n turn off power supplies and terminate with 2rror.
11) R~pQat step~ 9 and 10 until deposition process is over or until ti~ to sampl~ thQ chamb~r pressure again.
12~ Repeat ~tep~ 7 through 11 until d~position process is over~
13) Calculat~ mean and standard deviation of pressure sample~.

Automatic De~o~ition Proc~s~as ~his proces~ i~ used to cycle a ~ingle carrier 220 ~rom the load chamber 12, through all proc~a~ chambers 14-20, and to th~ unload chamber 22, in an automatic mode run. ~he re-quired number o~ r~petitions o~ this proaa~ ara invoked when an auto~atic run i~ skarted. Many o~ ~he previously defined sputtering proce~a~ are u~ed by this process.
Algorithm:
1) Execute Load-CRrrier process.

~9~4413 2) Execute Advance-Material proce~ for chamber 12, set current chamber to cha~ber 14.
3) If proce~sing required in current chamber, then execute the Proces~-~aterial process for the cuxrent cha~ber.
4) Execute Advance-Material proces~ for the current cham-ber, and set the ourrent cha~ber to the next chamber.
5) Repeat steps 3 and 4 ~or current chambers of 14 through 20.
6) Execute Unload-Carri~r procss~.

Vacuu~ Pum~in~ SY~tem ~roces~es Tha~e proces~e~ are us~d to draw a vacuum in a selected chamber. Several ~ub-proce~se3 ar~ us8d and ~ra de~cribed below.

Vacuum Pum;Qinq__y~e~ Proces~es Relea~e rou~h hine~ Sub-process ~his sub-procR~Y is us~d by the maln vacuum processes ln conjunction wlth a Get-rough lines ub-process ~o manage and shar~ the u~e of th~ two rough pumping lines. It re-leases ownership of any rough lin~s owned by the ~ain process making them availabl~ ~or u~ when not usad by the main pro-cess. It al~o control~ a rough cro~ valv~ 790 to couple the two rough pumpo (742 and ona not ~hown ~n Fig. 32) to aither chambers 12, 14 and 16 or cha~bsr~ 18, 20 and 22.
Algorithm:
1) I~ both rough pump3 ar0 on lins and at least one rough line i8 owned by th~ ~ain proae~s, then close the rough cros~ valve and skip to ~tep 3.
2) I~ a-t le~st one rough pu~p i5 on lin~ and both rough linc~ are owned by the ~aln proce~s, then open the rough cro~3 valvs and 3kip to ~tep 3.
3) I~ the right rough line i~ owned by ths main procos~, then rsle~e owner~hip.
4) I~ the le~t rough line i8 ownod by th2 main process, then rolea~e ownership.

~LX914q~3 Vac:uum Pum~n~ SYstem Proc:e~e~3 Put rouqh PumPc-on-l ine Sub-~roces~
Thi~ sub-prQce33 i~ used by the main vacuu~ processes to put the rough pumps on line. Thi~ i3 normally re~Iuired after a power failure. lgorithm:
oth rough pump!3 are on 1 ine then skip to 5tep 6 .
2) Attempt to obtain owner~hip of both rough lines, try for no longsr than 1 ~econd each.
3 ) I~ ft rough pump i~ not on line AND the pressure at the pump i~3 les~ that that required to put it on line AND tha le~t rough line i~ owned aND (the right rough line i3 owned OR the rough cro~s valve 1~ clol3ed) then open the le!fk rough pump cuto~ valve.
4) ïi~ right rough pll~p if~ not on line AND the pres~ure at th~ pump i9 le~ than that rQ~ulr~d to put it on lin~
AND l:h~3 right rough linQ i owned AND (the 1~ ft rough line i~ own~d OR the r~ugh cro~s v21ve i~; closed) then open the right rough pump cuto~f valv~.
5 ) E3cecutis R~leas~--rough 1~ n~ su}:\-proces~ .
6) Fini~hed.

Vacuu~ PuEp~n~ SY~te~ Proce~e~
Pu~p-rou~h ~ina-~o~n Sub-~roc~s~
Thi~ ~ub-proG~s~ i~ used to veri~y that the spe~ified rough lina can be pumped to th~ prQs~ure required by the main v~cuum processes for u~e o~ the rough line. It assumes that th2 rough pump to be u~ed 1~ on line.
Algorith~:
1) Walt ~or the rough line pre~ure to reach th~ proper prQ~sure. ~ima out a~ter the rough lin~ time ou~ value.
2) I~ step 1 timed out, then exeauta the ~ollowing sub-!3te~pB:
2a) Clo~e the rough pum~ valv~
2b) Pau~e with error ~9~4~3 2c) If a retry i~ requasted by the op~rator, then open the rough pump valve and ~xecute steps 1 and 2 again with a new time out int~rval.
3) Repeat step~ 1 and 2 until step 1 doe~ not time out or until the proce~s is continued or abort~d by the operator.

vacuum ~ ~in~ Syst~m Proce~se~
Ge~ouah Line~ Sub-roc~s~
This sub proce~s is used by the ]~ain processes in con-junction with the RelQa~e-rough line~ ~ub-proces~ to manage and sh~re tho use o~ th~ two rough lin~s. It ~btains owner-ship o~ th~ ~peci~iod rough lina and al~o the other rough linQ i~ it i8 reguirod by tha main vacuu~ process. It veri-fies that th2 rough line3 can b~ p~p~d down. It handle~ the case when on}y on~ rough pump i~ o~ }in~ by opening the rough cross~valva.
Algorithm-1) Ex~cu~e Put-rough pu~p~-on-llne ~ub proce~s.
2) If both rough pu~p~ ~r~ on lin~ th~n execut~ steps 3 through g el~ ~kip to 8t~p 10.
3~ Obtain owner~hip o~ th~ rough line.
4) Close all cha~ber rou~h and cryo reg~nerats valv~s ~or ths rough lin~.
5) Clo~e ths rough cros~ valve.
6) Executo Pump-rough linQ-down sub-proces~ ~or the rough line.
7) I~ th~ other rough line 1~ raquired by the main vacuum proces~, then attempt to obtain ownership of the other rough line, but only try ~or 10 ~acond~.
~) Ir owner~hip i~ obtained in ~t~p 7, then ex~3cute the rollowing sub-~tep~:
8a) Close all chAmber roug~ an~ cryo regen~rate valves ~or the other rough lln~
~b) Ex~aute Pump-rough llna-down ~ub-pro~2s~ ~or th~
other rough lin~
8c) Open th~ ~ough cros~ valve.
9) Skip to ~tep 15~

9~

10 ) I f no rough pUmp8 are on 1 ine then termina~e with error.
11) Obtain ownership of both rough lines.
12) Close all chamber rough and cryo regenerate valves ~or both rough line~.
13) Open the rough cross valve.
14 ) ~xecute pump-rough line-down ~ub-proc~ss for the rough line. This ha~ the effect of pumping both rough lines with one rough pump since only o~e pump is on line but the rough cro~s valve i~ open).
15) Finished.

Vacuum P~pin~ Svste~ P~oce~es Hi~h VacUum Sub-P~ocess This sub~proce6~ i~ u~ed to put ~ chamber o~ the ~ystem into a high vacuum modQ. A rough-cha~bar sub-proce~s i5 uged in thi~ main High Vacuum Sub-proc~s. It i documented here before the ~ain process.

Vacuu~ PumpLng_~yg~_m Proce6se~
Hiah Vacuu~ Sub-Process Rou~h-Cha~ber Sub-Process This ~ub-pr3cas~ is u~ed to rough a chamber to a lower cros~over pressure. Th~ cros~ov~r pre~ure i~ at a pre~eter-mined level (i.e. ona hundred microns) where the rough vacuum has been e~tablishad to a low enough level ~or the high vacu-um to be drawn by the high vacuum portions o~ the vacuu~ 5ys-tems.
Algorithm-1) I~ chamb~r pre~sure l~s~ than or equal to lower cross-over pres~ure, than ~ip to step 10.
2) Execute Get-rough llne~ ~ub-proc~s~.
3) C10~Q tha as~ociaked cha~ber gate valve~ 28, proces~ gas valve 698, vent valve 726, rough valve 722, and high vacuum valve 710.
4 ) Check ~or oparator requested paus2 .
5) open chamber rough valve 722.
6~ Wait for chamb4r pre~sure to reach lower cro~30ver pres-aure, ~ait no long~r than th~3 chamber rough tim~ out.

~29~43~

7) Clo~e chamber rough valve 722.
8) If timed out then pause with error (an operator reques-ted retry will cau3e steps 5 thro-lgh 8 to be executed again wi~h a new ~lme out value).
9) Repeat ~tepB 5 throuyh 8 until no time out or until operator has continued or abortsd the process.
10) Finished.

Vacuum Pumplna Sy~tem Proce~ses Hiqh-Vacuum Sub-Pxoc:e 9 Algorithm:
1) I~ high-vacuum valve 710 is opened then skip to ~tep 10.
2) Check for operator re~ue~ted pau~a.
3) Cl03e the chambar g~te valva~.
4) Chec~ for operator reque~ted pause.
5) Execut~ Rough-chamber ~ub-proc~Q.
6) Wait for cro~sov~r d~lay.
7) Read chambar pre~ur~.
8) Repeat 8tQp~ 4 through 7 until cha~ber pre~sure is less than tha upp~r crossover pra~ur~, or until the number of allowed itera~ion~ ha~ b~en ~xhaustad.
9) ExecutQ Relaase-rough lin~3 ~ub-proces~.
10~ I~ chambQr pre3~ure not le~e than uppQr crossover pres-sure, then pau30 with orror ~an operator reguasted retry will cau~6 ~tep~ 4 through 10 to b~ executed again with a ~et o~ iteration~).
11) Rep~at step~ 4 through 10 unt$1 chamber pres~ure is less than uppQr cro~ov~r pres~ure, or until process has been continued or aborted by the operator.
12 ) ChQck ~or operator r~quasted ~au3e.
13) Close th~ as~ociated chamber isolation valve~, process ga~, vent, rough and high-vacuum valve~.
14) open the chamber hlgh-va~uum valve 710.

Vent Proc~
Thi~ proce~s ~ used to vent a chamber o~ thQ 3ystem to atmosph~ric pra~ ure.

~X~ 3 Algorithm.
1) Close the chamber isolation valves, process gas, vent, rough and high-vacuum valves.
2) Check for operator requested pause 3) Open the chamber vent valve 726 4) Wait for vent time.
5) Close chamber vent valve.
Vacuum Pumpin~ System Processes Rou~h and Chill Cry~o Pump Process This process is used to rough and chill a cryo pump. It is used by the Regeneration and Recover pxocesses. Several sub-processes are used only in this process. There are documented before the main process.
Algorithm:
1) If cryo pressure is less than or equal to lower cryo crossover pressure, then skip to step 12.
2) Isolate the cryo pump frnm the rest of the vacuum pumping system.
3) Execute Get-rough lines sub-process.
4) Check for OperatQr requested pause.
5) Turn cryo compressor off.
6) Set bakeout complete flag(s) to false indicating that the molecular sieve(s) for the rough line(s) is being used and will reguire(s) baking out.
7) Open the cryo regenerate valve 720.
8) Wait for cryo pressure to reach the lower cryo crossover pressure, but wait no longer than the rough cryo time out valve.
9) Close the cryo regenerate valve 720.
10) If step 8 timed out, then pause with error (an operator re~uested retry will cause steps 7 through 10 to be executed again with a new time out interval).
11) Repeat steps 7 through 10 until the cryo pressure is less than the lower cryo crossover pressure, or until the operator continues or aborts the process.
12) Finished.

. . .

~2~3~4~3 Vacuum Pumpi~ Svstem Proce~es Rouqh and Chill Cryo_~Pum~ rocess Chill-cryQ_Sub-proces~
This ~ub-proc~s~ is ~s~d to chill a cryo pump to a de-sired chill ~ndpoint temp~ratur~.
Algorithm:
1) If cryo te~peratur~ i~ lesY than or equal to the chillcryo endpoint temperature, then turn cryo compressor on and ~kip to stQp 7.
2) Check ~or operator requ~ted pau~e.
3) Turn cryo co~pres~or on.
4) Wait ~or cryo temperature to reach chill cryo endpoint temperature, but wait no longer ~han the chill cryo tima out.
5) If ~top 4 tim~ out, th~n exQGut~ the ~ollowl~g sub-step~:
5a) Turn cryo compr~3~0r of~
5b~ Pau~e wlth error (an op~rator retry w~ll cause ~ep~ 3 throuqh 5 to be Qx~cuted again with a ne.w time out int2rval).
6) Repeat stsps 3 throu~h 5 until cryo temp~rature i3 less than or sgual to the ch~ll cryo endpoint t~mpsrature, or until oparator continua~ or ~borts the prOCe~8.
7) Finished.

~L~cuu~ nplnq SY~ 2roc2s3e~
Rouqh_an~ ChiLl Crvo Pum~ Process (Main Process) Algorithm: :
1) Check ~or operator r~qu~ted pau~.
2) Execut~ the rough-cryo sub-proce~3.
3) Wait ~or cryo cro~ov~r delay.
4~ Read cryo pra~ur~.
5) Repeat ~tep~ 2 through 4 un~il cryo preesure i~ le~s than uppar cryo ¢ro~80ver pres~ura, or until the maximum allowed it~ra~ions have been ~xecut~d.
6) ExecutQ the R~a~e-rough line~ ~ub-procQ~e.

~29~L443 7) I~ cryo pressure i3 greater than the uppar cryo cross-over pra3sure, then pau~e with error (an operator re-quested retry will cause steps 2 through 7 to execute again with a naw set of iterations).
8) Repeat ~tep~ 2 through 7 until cryo pressure is less than the upper crcss-ov~r pr~s~ure, or until operator has continued or aborted the proces~.
9) Ex~cute Chill-cryo sub-proce~.

Vacuum Pumpin~ Syst~m ~^oceasQs Si~ve-Bakeout Proae~s Thi~ proce~ us~d to bak~ out one or both the molecular ~ieve traps 750 o~ th~ roughing pump ~y~tem.
Algorithm:
1) Execute Get-rough lines ~ub-proces~.
2~ If the bakeout-complete ~lag ~or the rough lin~ i~ true, then skip to ~tep 11.
3) Determine if both ~i~va~ are to be bakad out. Both ~ieve~ are to b~ b~ked out if both rough line3 are owned by the main process.
4) Turn on the siev~ heat~r for th~ rough linQ.
5) If both ~ia~e~ ar~ to be baked out, then turn on the sievQ h~ater for the rough l$ne.
6) Wait for the bakeout delay tim~.
7) Execut~ Pump-rough lina-down ~ub-proces~.
8) Turn of~ th~ sieve he~ter ~or the rough line.
9) Set th~ bak~out-complete ~lag ~sr the rough line to true, indicating that thQ ~i~v~ for the rough line has been baked out.
10) I~ both ~iev~ wara baked out, then ~xecute the ~ollowing 9ub-step~:
lOa) Turn of~ the ~i~ve ha~t~r for the other rough llne lOb) Set the bakeout-complete fla~ ~or the other rough line to true. 1) Execute Releaae-rough llne~ ~ub-process.

~g~ 3 ~ .

Vacuum Pum~in ~ ~tem Proces~es Re~eneration Proce~s Thi~ proces~ i~ used to regenerate a cryo pump of the system. Several ~ub-proce~sas are u~ed in this regeneration prcces~ and are described first.

Vacuum~ umDlnq~ Svs.tem ProceYs Reganer~tion Proces~
Start-bak~out Sub-proces~
This ~ub-proce~ initiate~ the ~i.ave bakeout process.
Algorithm:
1) Start the ~i~ve-bakeout proce~3.

Vacuum Pumnina Sv~t~m ~roce~s R ~enera~ion ~rocess PuroLe-and-warm-cryo Sub-~oces~
Thi~ purge-and warm ~ub-procs~ i3 u~ed to bring the cryo pump to room temperature. ~h~ cryo pu~p i5 purged for an additional time after thi~ t~p~rature i8 achieve.
Algorit~m:
1) ExecutQ Start-sieve -bakaout sub-proc~
2) Clo~ the high-vacuum, purge and regenerate valve~ for the cryo pumping apparatu~.
3) Check ~or op~ra~or equaated pau~e.
4) Turn oryo compr~s~or of~.
5) Open cryo purge valv~.
6) Wait ~or tha cryo to roaoh roo~ temp~rature, or for purge time out whichever ia ~ir~t.
7) I~ ti~2d out, then pau~e with ~rror (an operator re-quested ret~y will cau~e 8t9pB 6 and 7 to be executed again ~ith a new timQ out valu~).
8) Repea~ ~tep~ ~ and 7 until room te~psrature i~ reached or until the operator ha~ continued or aborte~ ~he pro-C~
9) Ch~ck ~or op~rator r~guestsd p~U~Q.
10) Wait ~or the additional purge ti~e.Check ~or op~htor reque~t~d pau-Q.
.. .

-~

~''3~4~3 12) Wait for bakeout-complete flag to be set to true, but wait no longer than the bakeout delay plus the rough line time out (this flag indicates that the sieve bake-out started in step 1 is complete).
13) Check ~or operator rsqu~ted pause.
14) Close cryo purg~ valve.
15) If timed out in Bt~p 12 then ter~inate.

vacuum Pumpln System Process Reqeneration Proce~s fMa~n Process) Algorithm:
1) Generate operator mas~age to turn the cold trap ~ill controller 728 o~.
2) Record ~tate o~ high-v~cuum valve 710 (opened or clo~ed).
3) C10B~ the high va~uum valv~ 710, purge valve 724, and regeneration v~lves 720 ~or the cry.
4) Check ~or opeartor requested pau~e.
5) ~xecute Purge-and warm cryo 3ub-proce3~.
6) Ch~ck for op~rator regue tad pau~e.
7) Execute rough and chill cryo pump proce~.
8) Check ~or operator r~que~ted pau~s.
9) I~ recorded state o~ high-vacuu~ valve i~ opened, then execute tho high vacuum, sub-prooe~.

Vaouum ~ ein~ s~ste~ Proce~
Recov~ Proces~
Thi~ proce~s i~ used to recover the cryo pump3 a~ter a power failure. Thi~ process is auto~atically started afker a power ~ailur~.
Algorithm:
1) Determin~ 1~ regenoration i~ allowed.
2) Executa Put rough pu~ps-on-line ~ub-process.
3) Read cryo te~peratur~ ~nd pre~sure.
4) I~ cryo temperature i~ les~ than or equal to the chill cryo endpoin~ temperature OR cryo pressure is less than or aqual to ~e low~r cryo cros~over pressur2, then start the rough and chill cryo pump proces~ for the cur-rent chamber and ~kip to step 6.
5) If regneration is allowed, then start th~ regeneration proce ~ for the current cha~berJ or else log a message lndicating th~ cryo ne~d~ to b~ rege~erated.
6) Repeat ~tep~ 3 through 5 for cha~bers 12 through 220 operation o~ thQ Fir3t Embodiment Substrates, ~uch a~ el~ctrole~s nickel plated aluminum substrate~, are fir~t suitably cleaned prior to processing.
For example, the 8ub3trat~ may be sprayed with 1,1,1 tri-chloroothane which ha~ been dis~illed and ~lltered through a 10 micron filter. This remov~s the majority of the polishing abrasiv~ and r~si~ua ~rom the substrata~. Substrates are then ultrasonically clsan~d in a dzgr~a~er bath containing 1,1,1 trlchloroe~han~ haated ~o 159~162~Cal~iua (C~ and fil-tered to 10 micron~. ~he ~ubstrate~ are rai ~d through the v~por zon ab~va thi~ ~ath. Th~n, th~ aub~trates are lowered into a second bath containing 1,1,1 trichloro~than~ heated to 159-162C and filtared to 10 mlcron~ and again ultrasonical-ly c~eanad. The sub~tr~t~ ar~ ral~ad through the vapor zone abo~e thi~ bath and low~r0d into 3tlll ano~her bath contain-inq 1,1,1 trichloro~thane, which has b~en distilled ~rom a boiling ~ump at 159-162C. Fro~ thi~ last bath the sub-strates are rai~ed into th~ vapor zone, allowod to drain, and then 810wly rai~ed at ~ 310w rata to clear tha vapor zone.
Thes~ lattar 3t~ps ~liminata ~vaporation marks ~ro~n the sub-strate ~ur~acas. ~h~ cleaned ~ubstrates are ~tored in an en-closed box. Additional cleaning ~ay b~ p~rformsd a~ needed.
A~t~r loading of ~ tray o~ Aubstrata~ lnto chamber 12 ~in a cloan roo~), a vacuum i~ e~tablish~d in chamber 12.
Also, a vacuum i~ establi3h~d in cha~ber~ 14 through 22 as well. In addition, cathode a89e~bli~ 40, 42 may be pre-sputt~red to bring tha~o cathode a~0embl i9~ to a ~teady state operation.
The load~r 272 in chamber 12 ~hen picks up the flrst of the carrl~r~ 220 ~r~m the tray and po~ition~ it in th~ center of the track in cha~ber 12. A transporter 222 then enters ~9~43 chamber 12 and is loaded with the carrier. The loaded transporter then travels to chamber 14, wherein the plunger 228 in this chamber is inserted into the hub 278 of the planetary carrier and grips and lifts the planetary from the transporter. The transporter is then shifted out of the way of the sputtering cathode assemblies 40 in chamber 14. The plunger 278 then rotates the planetary carrier 220 and supported substrates and sputtering of a first layer (for example, of chrome) is performed. Following sputtering, the transporter 222 in chamber 14 transfers planetary carrier 220 to chamber 16 and then returns to chamber 12 to fetch another carrier for chamber 14. A second transporter delivers substrates processed in chamber 16 to chamber 18 and also delivers substrates processed in chamber 18 to chamber 20. In this way, successive chrome, cobalt-platinum, chrome and carbon layers are sputtered onto the substrates. Finally, a third transporter transfers the processed substrates from chamber 20 to chamber 22. The planetary carrier 220 is unloaded from this third transporter by an unloader mechanism 272 and placed onto a tray 270. In this manner, the processing continues.

After the last planetary carrier 220 is lifted from the tray in chamber 12 and delivered to chamber 14, the vacuum in chamber 12 is relieved and the door 68 to this chamber is opened. The next tray of substrates is then loaded. Chamber 14 is isolated form chamber 12 during this loading operation. After the desired vacuum is reestablished in chamber 12, a first transporter from chamber 14 is returned to chamber 12 to obtain the first planetary carrier 220 from this new tray.

In a similar manner, after the tray in chamber 22 is filled, chambers 20 and 22 are isolated and chamber 22 is opened to permit replacement of the filled tray . .
. ~ , . ...

~lZg~43 85a with an empty tray. The vacuum is then reestablished in chamber 22. Thereafter, carriers are again transferred between chambers 20 and 22.

The isolation valves in housing 26 permit the isolation of the chambers from one another so that the parameters affecting sputtering may be optimized in each of these chambers. In addition, the processing speed is enhanced because , '.J., ,, ~<

~91~4~

processing may continue while additional trays o~ substrate containing planetarie~ are load~d and unloaded from the re-~pective chamber~ 12 and 22.

Pre~errQd Embodimant of ~i~. 33 Another embodim~nt of tha inventio~ is sho~n in Fig.
33. This embodiment lncludes cha~ber~ 12, 14, 16, 20 and 22 like those hown in the embodim~nt o* Fig. 1. In addition, isolation valve~ are al~o provlded for ~lectiv~ly i~olating these chamber~ from one another.
As can ba se~n from Fig. 33, this sQcond embodiment of the invention eli~inate~ tha proce~sing chamber 1~. There ~ore, during processing, the carri~r~ 220 are transrerred in the ~ollowing ~e~uQnce through the cha~ber~ o~ thi~ e~bodi-ment. From load chamber 12, a carrier 220 i~ delivered to chamber 14 for sputtQrlng o~ th~ und~rlayer onto the 9ub-~trate~. From cha~bQr 14, thQ carrier 220 i3 trans~erred to chamber 16 Por depo~ition of ~he ~econd la~er. From chamber 16, rather than trav~lling to a chamb~r 18, the carrier 220 is return~d to chamber 14 for depo ition of the third layer.
In thi~ ca~e, both the ~ir~t and third d~po~ited lay~rs are of the same material, such a~ chrome. The sputtering, po~er and other param2t~r~ are ad~u~ted in chambex 14 to ad~ust the deposition o~ thi~ th~rd layer. ~or ~xample, to make the third layer thinner than the ~irst layer. From chamber 14, the ~ub~trate~, now cont~ining thrae d~pos$ted layero, are trans~erred to chamb~r 20 ~or d~po~ition o~ the ~ourth layer.
Finally, fro~ cha~ber 20, the carrier~ are delivex~d to the unload chamber 220 A single transport~r 222 may be utilized to per~orm thi~ sequenc~. How~ver, three tran~portera are pre~erred, as in the ca~e o~ th~ Fig. 1 embodiment. ~h3 ~ir~t o~ these trans~oxters travel~ ~rom chamber 12 to chambar 14. The ~econd Or the~e tran~porter~ travel~ b~twen cha~b~r~ 14, 16 and 20. Finally, the third Or ~hese ~ran~porter~ travels be-tween chambar3 20 and 22.
The ~mbodimefl~ o~ Fig. 33 is ~lightly ~lower t~lan the embodiment o~ Fig. 1, becau~e o~ the ~act that the cha~ber 14 is utilizQd for two d~poRitions. Nevertheless, this embodi-ment illustrate~ the principle that the system does not re-quire transportation o~ the carriers 220 in on~ dîrection from ona end of the ~ystem to another.

Pre~erred Embodiment o~ Fi~. 34 Still another embodimant o~ the invention ic shown in Fig. 34. This embodim~nt i~ like the embodiment o~ Flg. 1, except that th~ load cha~ber 12 ha~ b~en modified to include a pair of load chamber~ 12a, 12b like the previously de-~cribed chamb~r 12. In addition, an inter~ace chamber 12c is also employ~d. Th~ inter~ace chamb~r 12c i8 position2d be-tween the load chamber~ 12a, 12b ancl th~ fir~t processing ahamber 14.
Also, in tha e~bodiment four transport~r~ 222 are utilized. A ~ir~t tr~n3port~r trav~l~ between the cha~bers 12a a~d 12c on a track 224~ loacted at the rear of chamber 12c. A ~cond transportar trav01s betwaen chambers 12b, 12c and 14. The third transport~r travQls betwe~n chambers 14, 16 and 18. Finally, th~ ~ourth tran~porter travels between cha~ber~ 18, 2 0, and 22.
In th~3 Fig. 1 ~ystem~ ~ub~tantial tim i~ required to e3tablish a vacuu~ in cha~ber 12 to a desirsd high vacuum l~vel be~or~ tr~n~ar~ betwQen chambsrs 12 and 14 ar~ psrmit-t2d. In ~om~ cas~, a delay in proc~inq occur~ becau~e ch~mber 14 i~ empty ~or a p~rlod o~ time until the de~ired vacuum ia ostabliahad in cha~b~r 12, and carrier~ 220 ar~
again transferr~d fro~ chamb~r 12 to chamber 14. The embodi-men~ of Fig. 34 eli~ina~s any ~u~h d~}ay.
: Specl~ically, a tray 270 o~ carrier~ 220 i po itioned in chamber 12a and al~o in chamber 12b. A vacuum is esta-blished ln theaa chamb2rs. Th~ carriers 220 are loaded ~rom one o~ th~ chamb~ra, ~or exa~ple chamber 12b, onto a tran~-porter which carriaa th~ carrier~ through the lnterrace cham-ber 12c and to tha chamb~r 14 ~or procassing as proviously explainad. When ahamber 12b is empti~d of carriers, proces~ing con~inu~ by uaing carriar~ ~rom chamber 12a.
That i~, the rsar tran~por~er 2~2 obtains a carrier from ' ~9~443 chamber 12a and carries it to interface chamber 12c. The plunger 228, which may be like those previou~ly described, is then utilized to pick the planetary toward the front cha~ber 12c and load~ it onto a transporter. ~hi~ latter transporter carries the carrier to hamb~r 14 to continue the substrate processing. While carriar~ are being tran~ferred from chambar 12a, a n~u tray iA place~ in chamber 12b and a vacuum is reestablished in this chamber. Thereafter, when chamber 12a i emptied of it~ carriers, the ~y~tem then utilizes carriers fro~ the replenishQd chamber 12b. Also, while the carrier~ are u~ed ~rom chamber 12b, a new batch o~ carriers is loaded into chamber 12a. Thu~, in tha embodiment of Fig.
34, alternately operating load chamber~ are pro~ided for delivering a continuous supply o~ carriers to downstr~am chamber~ ~or procassing.
Also, th~ unload chamber may comprise two unload cham-bers, like the load chamb~rA 12a and 12b, togethQr wikh an unload inter~ace chamber like 12c. However, this is not typically used. That i8, unifor~ity in the ~rocessed discs is particularly af~ected by contaminants, such as water vapor, carried by unproces~ed sub~trates from chamber 12 to cha~ber 14. ~y pumping chamber 12 to a high vacuum, such effectR ar~ ~inlmizad. How~v~r, the disc~ are far less sensitive after they h~vo be~n completoly processed.
Consequently, di~a6 may bo trans~errQ~ ~rom chamber 20 to chamber 22 wi~hout waiting ~or th~ establi~hment of a vacuum which i~ a~ high a~ the vacu~m in chamb~r 12.
In addition, ~putt~r ~tching ~or ~leaning purpose~ may be perfor~ed in chamb4r~ 12 and inter~ac~ chamber 12c a~ de-sired~ To accompli~h sputter Qtching, ~puttering a semblies and a plungar i8 utilized in the3~ ¢ha~bers. Sputter ~tching i~ accompli~hed by n~tive}y biasing the plunger 228, and thus the carrier which i~ mounted to thQ plunger and th~ ~up-port~d ~ub~trat~. q'hi~ cau9~ po~ltive ions in the plasma to bombard the ~u~trate~ and remov~ a ~mall quanti~y of ma-terial ~rom the sub~trats ~ur~ace~. Substantially uniform etching occur~ bea~s~ the plan~t~ry ~otion ~mparted to the substrat~s during ~tching expossA both substrate sur~ace3 to 1~9144~3 8~
the pla~ma~ Al~o, additional procea~ing apparatus 790 may be placed in chamber~ 12 or 12c. Such apparakus ~ay comprise commercially availabls ion gun~ which bo~bard tho ubstrates with ion~ ~or cl~aning purposes. Altarnately, or in addi-tion, such apparatu~ 790 may compris~ ~ub~trate heaters for warming the ubstrates prior to delivary to chamber 14.
Having illu~trated and dQscribQd thQ principla~ of our invention with refQrence to ~averal pr~ferred embodiment3, it should be apparQnt to tho~ persons skillad in the art that such invention may b~ modi~i~d in arrangement and d~tail without departing from such principl~. We claim as our in-vention all such modi~ication~ as co~ within the trua spirit and ~cope o~ the ~ollowing cl~ims.

Claims (2)

1. A target for sputter-depositing a magnetic layer having a radial coercivity gradient on a substrate which is moved relative to the target, said magnetic layer having at least two constituent materials, said target comprising:
a circular disc member of a first diameter which is formed of a first constituent material;
a ring member having an inside diameter and an outside diameter smaller than the first diameter, the ring member being formed of a second constituent material and being disposed on the base member concentrically therewith; and a clamping ring member having an outside diameter which is less than the outside diameter of the ring member and greater than the inside diameter of the ring member, the clamping ring member also having an inside diameter which is less than the inside diameter of the ring member, the clamping ring member being formed of the first constituent material and being mounted to the base member concentrically therewith and overlying a portion of the inner perimeter margin of the ring member so as to clamp the ring member to the base member whereby the sputtering surface of the target comprises only the first and second constituent materials in a predetermined ratio of exposed areas thereof.

2. A target according to Claim 1 in which the first constituent material is cobalt and in which the second constituent material is platinum.