ION IMPLANTATION - Chapter 8 Basic Concepts

ION IMPLANTATION - Chapter 8 Basic Concepts Ion implantation is the dominant method of doping used today. In spite of creating enormous lattice damage it is favored because: Large range of doses - 1 11 to 1 16 /cm 2 Extremely accurate dose control Essential for MOS V T control Buried (retrograde) profiles are possible Low temperature process Wide choice of masking materials Analyzing magnet Ion source Pump -3keV V Resolving aperature V -2keV Accelerator Focus Neutral beam gate Neutral trap Q X & Y scan plates Faraday cup Wafer There are also some significant disadvantages: Damage to crystal. Anomalous transiently enhanced diffusion (TED). upon annealing this damage. Charging of insulating layers. 1

A. Implant Profiles Range R Projected range R P At its heart ion implantation is a random process. High energy ions (1-1keV) bombard the substrate and lose energy through nuclear collisions and electronic drag forces. Concentration (cm-3) 1 21 1 2 1 19 1 18 117 Vacuum Sb As Silicon R P.2.4.6.8 1 Depth (µm) P B ΔR P.66 C P Q = 2 Profiles can often be described by a Gaussian distribution, with a projected range and standard deviation. (2keV implants shown.) ( C(x) = C P exp x R P ) 2 (1) 2 2ΔR P C x dx or Q = 2π ΔR P C P ( ) (2) where Q is the dose in ions cm -2 and is measured by the integrated beam current.

Depth (µm).4.35.3.25.2.15.1.5 4 8 12 16 2 Energy (kev) Energy (kev) B Sb P As Standard Deviation (µm).12.1.8.6.4 Phos As Sb Boron Range (µm) Std Dev (µm) Range (µm) Std Dev (µm).2 As 4 8 12 16 2 Energy (kev) Range (µm) 3 Std Dev (µm) Range (µm) B Sb P Std Dev (µm) 1.199.64.84.43.121.58.473.249 2.342.125.156.75.219.1.826.384 3.473.179.226.12.36.133.114.483 4.598.229.294.128.385.162.143.562 5.717.275.362.152.459.187.171.628 6.833.317.429.176.528.29.198.685 7.947.356.495.198.594.229.223.736 8.15.393.561.22.656.248.248.78 9.116.428.626.241.716.265.272.821 1.127.461.692.261.773.28.296.857 12.148.522.821.31.883.39.341.922 14.169.579.95.339.985.334.385.978 16.189.63.17.375.18.357.428.12 18.21.678.12.411.117.378.469.17 2.229.723.133.446.126.397.59.11 Ranges and standard deviation Rp of dopants in randomly oriented silicon.

y Implant Beam 8 4-4 5 z -5-1 4 8 x 12 Monte Carlo simulations of the random trajectories of a group of ions implanted at a spot on the wafer show the 3-D spatial distribution of the ions. (1 phosphorus ions at 35 kev.) Side view (below) shows Rp and Rp while the beam direction view shows the lateral straggle. 8-4 y 4 y 4-4 8 1 5-5 -1 z Beam direction 4 8 12 x Side view 4

The two-dimensional distribution is often assumed to be composed of just the product of the vertical and lateral distributions. y2 (3) C( x, y) = C vert ( x) exp 2 2ΔR Now consider what happens at a mask edge - if the mask is thick enough to block the implant, the lateral profile under the mask is determined by the lateral straggle. (35keV and 12keV As implants at the edge of a poly gate from Alvis et al.) (Reprinted with permission of J. Vac. Science and Technology.) The description of the profile at the mask edge is given by a sum of point response Gaussian functions, which leads to an error function distribution under the mask. (See page 7 of class notes on diffusion for a similar analysis.) 5

B. Masking Implants x m How thick does a mask have to be? C P * For masking, Concentration C B C * (x m ) ( ) = C * P exp C * x m ( * x m R P ) 2 2ΔR P * 2 C B (4) R P * Depth Calculating the required mask thickness, x m = R P * + ΔR P * 2ln C * P C = R * * P + mδr P B The dose that penetrates the mask is given by Q P = Q 2πΔR P * exp x R * P 2dx = Q * 2ΔR P 2 erfc x m R * P * 2 ΔR P x m 6 (5) (6)

C. Profile Evolution During Annealing 2Dt R P Implanted After Diffusion Comparing Eqn. (1) with the Gaussian profile from the last set of notes, we see that Rp is equivalent to 2Dt. Thus Q ( x R P ) 2 2π ΔR 2 P + 2Dt 2( ΔR 2 P + 2Dt) C( x,t) = ( ) exp (7) Imaginary Delta Function Dose Q Virtual Diffusion Delta Function Dose Q (Initial Profile) Diffused Gaussian The only other profile we can calculate analytically is when the implanted Gaussian is shallow enough that it can be treated as a delta function and the subsequent anneal can be treated as a one-sided Gaussian. (Recall example in Chapter 7 notes.) x C( x,t) = Q πdt x2 exp 4Dt (8) 7

1 2 Boron 38 kev Concentration (cm -3 ) 1 19 Antimony 36 kev 1 18 1 17.5.1.15.2.25.3 Depth (µm) -.4 Real implanted profiles are more complex. Light ions backscatter to skew the profile up. Heavy ions scatter deeper. 4 moment descriptions of these profiles are often used (with tabulated values for these moments). Range: ( ) R P = 1 Q xc x dx (9) Distance (µm) -.2.2 3 tilt 3 Degree Tilt.2.4.6.8 1. Distance (µm) Real structures may be even more complicated because mask edges or implants are not vertical. Std. Dev: Skewness: Kurtosis: 8 ΔR P = γ = β = 1 Q 2 ( x R P ) C( x)dx ( x R P ) 3 C( x)dx Q ΔR P 3 ( x R P ) 4 C( x)dx Q ΔR P 4 (1) (11) (12)

D. Implants in Real Silicon - Channeling At least until it is damaged by the implant, Si is a crystalline material. Channeling can produce unexpectedly deep profiles. Screen oxides and tilting/rotating the wafer can minimize but not eliminate these effects. (7 tilt is common.) 1 21 Sometimes a dual Pearson profile description is useful. Note that the channeling decreases in the high dose implant (green curve) because damage blocks the channels. Concentration (cm-3) 1 2 119 118 117 116 115 9 2 x 1 15 cm -2 2 x 113 cm-2 2 x 113 cm-2 scaled.1.2.3.4.5.6.7.8 Depth (µm)

Modeling of Range Statistics The total energy loss during an ion trajectory is given by the sum of nuclear and electronic losses (these can be treated independently). A. Nuclear Stopping R R = dx = 1 N de dx = N ( S n + S e ) E de S n (E) + S e (E) (13) (14) An incident ion scatters off the core charge on an atomic nucleus, modeled to first order by a screened Coulomb scattering potential. V(r) = q2 Z 1 Z 2 4πεr exp r a (15) Incident ion Target Recoil Scattered Ion This potential is integrated along the path of the ion to calculate the scattering angle. (Look-up tables are often used in practice.) S n (E) in Eqn. (14) can be approximated as shown below where Z 1, m 1 = ion and Z 2, m 2 = substrate. S n (E) = 2.8x1 15 Z 1 Z 2 ( 2 / Z 3 2 / 3 1 + Z ) 1 / 2 2 1 m 1 m 1 + m 2 ev -cm 2 (16)

B. Non-Local and Local Electronic Stopping Dielectric Medium Target e- V ion V ion Energy loss (kev µm -1 ) Retarding E-field Drag force caused by charged ion in "sea" of electrons (non-local electronic stopping). To first order, 1 1 1 1 1 1 Energy (kev) 11 Collisions with electrons around atoms transfers momentum and results in local electronic stopping. S e (E) = cv ion = ke 1 / 2 where k.2x1 15 ev 1/2 cm 2 (17) As N C. Total Stopping Power S E P N and electronic stopping are equal is The critical energy E c when the nuclear B: 17keV P: 15keV B N As, Sb : > 5keV Thus at high energies, electronic stopping dominates; at low energy, nuclear stopping dominates.

Damage Production 3 kev As ion Damage cylinder 3 ev Si Knock-on Consider a 3keV arsenic ion, which has a range of 25 nm, traversing roughly 1 atomic planes. n = E n 2E d = 3, 2 15 = 1 ions (18) 1 atomic planes t =.1 ps t =.5 ps Molecular dynamics simulation of a 5keV Boron ion implanted into silicon [de la Rubia, LLNL]. Note that some of the damage anneals out between.5 and 6 psec (point defects recombining). t = 6. ps 12

Amorphization For high enough doses, the crystal becomes amorphous and loses all long range order. At this point, the arrangement of lattice atoms is random and the damage accumulation has saturated. Surface.2µm 1E15 1.5E15 2E15 4E15 5E15 1E16 Crystalline Amorphous Cross sectional TEM images of amorphous layer formation with increasing implant dose (3keV Si -> Si) [Rozgonyi] Note that a buried amorphous layer forms first and a substantially higher dose is needed before the amorphous layer extends all the way to the surface. These ideas suggest preamorphizing the substrate with a Si (or Ge) implant to prevent channeling when dopants are later implanted. 13

Damage Annealing - Solid Phase Epitaxy Amorphous 5 min 1 min 15 min 2 min SPE regrowth Max damage below amorphization threshold 1 nm EOR damage If the substrate is amorphous, it can regrow by SPE. In the SPE region, all damage is repaired and dopants are activated onto substitutional sites. Cross sectional TEM images of amorphous layer regrowth at 525 C, from a 2keV, 6e15 cm -2 Sb implant. Concentration Implanted Profile EOR loops In the tail region, the material is not amorphized. Damage beyond the a/c interface can nucleate stable, secondary defects and cause transient enhanced diffusion (TED). Residual damage Surface a/c interface Depth 14

Goals: Annhilated I & V per implanted ion 1 1 <311> 1.1 I-dimer 15 Damage Annealing - +1 Model Remove primary damage created by the implant and activate the dopants. Restore silicon lattice to its perfect crystalline state. Restore the electron and hole mobility. Do this without appreciable dopant redistribution. Bulk I&V recombination Surface V recombination Initial excess I & V Surface I recombination 1-8 1-6 1-4 1-2 1 1 Seconds Ribbon-like defect <11> 311 Capture radius In regions where SPE does not take place (not amorphized), damage is removed by point defect recombination. Bulk and surface recombination take place on a short time scale. "+1" I excess remains. These I coalesce into {311} defects which are stable for longer periods. {311} defects anneal out in sec to min at moderate temperatures (8-1 C) but eject I TED.

Dopant Activation When the substrate is amorphous, SPE provides an ideal way of repairing the damage and activating dopants (except that EOR damage may remain). At lower implant doses, activation is much more complex because stable defects form. 1. Fraction Active 1. 1 x 1 15 cm -2.8.6.4.2 4 5 6 7 8 9 T A ( C) 3 x 1 12 cm -2 1 x 1 14 cm -2 Sub Amorphous Amorphous Fraction Active.1 8 x 1 12 cm -2 2.5 x 1 14 Reverse Annealing 2 x 1 15.1 4 5 6 7 8 9 1 T A ( C) Plot (above left) of fractional activation versus anneal temperature for boron. Reverse annealing (above right) is thought to occur because of a competition between the native interstitial point defects and the boron atoms for lattice sites. 16

Concentration (cm -3 ) Concentration (cm-3) 1 19 1 18 1 17 116 1 21 1 2 1 19 1 18 117 1 16 1 sec 1 C 2 min 8 C.8.16.24.32.4 Depth (µm) As profile Boron profiles Initial Transient Enhanced Diffusion Final.5.1.15.2.25.3.35.4 Depth (microns) TED is the result of interstitial damage from the implant enhancing the dopant diffusion for a brief transient period. It is the dominant effect today that determines junction depths in shallow profiles. It is anomalous diffusion, because profiles can diffuse more at low temperatures than at high temperatures for the same Dt. The basic model for TED assumes that all the implant damage recombines rapidly, leaving only 1 interstitial generated per dopant atom when the dopant atom occupies a substitutional site (the +1 model) [Giles]. TED effects may be very non-local. After 9 C, 1 sec anneal, the amorphous As surface profile recrystalizes by SPE without much TED. The buried boron layer is drastically affected by the +1 interstitials in the As tail region. 17

Si Self-Interstitial Density (cm -3 ) 1 19 1 17 1 15 1 13 1 11 Si Self-Interstitial Density (cm -2 ) 1-6 sec 1-5 1-4 Atomic Level Understanding Of TED +1 Initial Damage 1-3 1-2 1-1.2.4.6.8 1 Depth (µm) 1 15 1 14 1 13 1 12 815 C 738 C 75 C 67 C 1 11 1 1 2 1 4 1 6 Time (seconds) 18 {311} clusters form rapidly and then are stable for extended periods (sec - min), driving TED by emitting I while they shrink. By.1 sec (75 C), the {311} defects have formed and C I is down to 1 13 cm -3 (SUPREM). * But C I 1 8 cm -3 at 75 C, so the enhancement is > 1 5! On a much larger time scale, the {311} clusters decay. These act as an ongoing source of excess interstitials which drives TED. TED lasts hours at very low T, minutes at intermediate T and msec at very high T.

Given this picture, we can model the {311} behavior as follows: (where Cl n is a cluster with n interstitials) C I t = Cl t = 1 2 5 6 7 8 9 1 11 12 Temperature C k f C I Cl k r Cl = growth shrinkage 19 I + Cl n Cl n +1 (19) The most important part of the transient is while the {311} clusters are evaporating I, maintaining a constant supersaturation of I. During this period, dopant diffusivity enhancements are constant and given by (see text): max 1 C I 1 = exp E b E 6 F * C I 4πa 3 (21) C I kt Interstitial Supersaturation CI max /CI * 15 14 13 (2) Note that the diffusivity enhancement is as large as 1, at low T and falls off to 1-1 at RTA temperatures. These calculated values agree with experimental measurements.

Concentration (cm -3 ) 1 21 1 2 1 19 1 18 1 17 1 16 As-implanted 1 min, 75 C.1.2.3.4.5 Depth (µm) Estimating the Duration of TED 1 min, 75 C 1 5 1 4 1 1 1 1 Interstitial Supersaturation Ratio Over time the interstitial supersaturation decays to zero and TED ends. Example - Boron TED (TSUPREM IV). Note that C I / * C has dropped from 1 4 I to 1 2 in 1 min at 75 C. The excess I diffuse into the bulk and recombine at the surface. Note the relatively flat interstitial profiles (dashed) except at the surface where recombination is occurring. C I max Surface R P Surface Recombination Flux Dose Q Diffusion flux The flux towards the surface is d I C max I / R P where R P is the range of the implant. The time to dissolve the clusters is given by the dose divided by the flux (see text): τ enh = 4πa3 R P Q d I exp E b kt (22) 2

1 6 Time (seconds) C I max /CI * Enhancement 14 1 2 1 1-2 5 6 7 8 9 1 11 12 Temperature C 1, 1 1 1 1 τ enh Steady-state {311} s decay Time Rapid exponential decay Plot of Eqn. 22. This matches the experimental data on page 18 of these notes (Fig. 8-33 in text). Note that TED persists for hours at low T and disappears in msec at very high T. Thus the general picture of TED that emerges is as shown on the lower left. Because the {311} clusters exist for much longer times at low T, there can actually be greater dopant motion during low T anneals (below). Q P Concentration (cm -3 ) 121 1 2 1 19 1 18 117 1 16 21 9C 8C 7C 7C 8C 9C.1.2.3.4.5.6 Depth (µm) 15 1 4 1 1 1 1 Interstitial Supersaturation Ratio

1 18.18 micron.2 Microns -.2 Concentration (cm-3).25 micron 1. micron -.4 -.2.2.4 Microns 2D SUPREM simulation of small MOS transistor. Ion implantation in the S/D regions generates excess I. These diffuse into the channel region pushing boron (channel dopant) up towards the surface. Effect is more pronounced in smaller devices. 22 1 17.5.1.15.2.25 Depth (microns) Result is that V TH depends on channel length (the "reverse short channel effect" only recently understood). (See text - Chpt. 7 - for more details on this example.)

Summary of Key Ideas Ion implantation provides great flexibility and excellent control of implanted dopants. Since implanted ion energies are >> Si-Si binding energy ( 15 ev), many Si lattice atoms are displaced from lattice positions by incoming ions. This damage accumulates with implanted dose and can completely amorphize the substrate at high doses. The open structure of the silicon lattice leads to ion channeling and complex as-implanted profiles. TED is the biggest single problem with ion implantation because it leads to huge enhancements in dopant diffusivity. Understanding of TED has led to methods to control it (RTA annealing). Nevertheless, achieving the shallow junctions required by the NTRS will be a challenge in the future since ion implantation appears to be the technology choice. 23