MIMS Workshop 23 - F. Hillion MIMS Workshop 1/ Practical aspects of N5 Tuning Primary column : small probe, high current, influence of Z. Dynamic Transfer and scanning. Cy and P2/P3. LF4, Q and chromatic aberration. Second order aperture aberration. PHD and aging effect. QSA. Mass Fractionation at ES. 2/ New developments Motorized slits/diaphragm and hexapole. 24 by the author. This information may not be used without referencing the author.
Primary column High Current In order to get a high probe current at the sample, L alone or L and L1 must be used. L alone Probe current vs L (Cs+) 1 I probe (pa) 1 1 1 1 Iprobe Iprobe-Th Experimental conditions : Differential pumping tube Diam. 2.5 mm Length 17 mm 1 FCp = 33.45 na 1 3 35 4 45 5 L (Volts) D1-1 (3 microns) Simulations made without beam stops.
Primary column High Current Use of L and L1 Probe current vs L1 for different L Probe current vs L1 for L=17 35 1 3 1 25 Iprobe (pa) 2 15 1 L=18 L=17 L=16 Iprobe (pa) 1 1 1 L=17 Iprobe-Th-17 5 1 5 1 15 2 25 3 35 4 L1 (Volts) 1 5 1 15 2 25 3 35 L1 (Volts) Experimental conditions : FCp = 33.45 na D1-1 (3 microns) Simulations made without beam stops. Maximum beam current : 29 na FCp = 5 na D1-1 (75 microns)
Primary column Small probe Main Aberrations : (Probe size) 2 = (gaussian size) 2 + Σ (aberrations) 2 Aperture aberration : ½ Cs α 3 Chromatic aberration : Cc α E/E Alpha being the half aperture at the sample. Aberration coefficients Cs and Cc are linked to the optical properties of the immersion lens. Respective values are 66 mm and 16 mm. Alpha is controlled by D1. For a given probe size one can determine an optimum value for D1 which maximize the probe current. 1 nm D1 = 23 microns Simulations made at 8keV with Cs + : Source size 4 microns, E = 1 ev
Primary column Small probe Without L1 1 nm probe size can be reached. In order to reach smaller probe size, L1 must be used. Probe current vs L1 Demagnification vs L1 Irelatif (%) 12 1 8 6 4 2 Idem L1= Iexp Ith Demagnifaication 1,2 1,8,6,4,2 Idem L1= 5 55 6 65 7 75 5 6 7 8 L1 Volts L1 (Volts)
Primary column Small probe Aberration coefficients vs focal length Dependence of lateral resolution upon Z. Cc is proportional to f.83 Cs is proportional to f 2.93 E focal length : 6 mm For Z = 1 microns As the demagnification G <<1 f = Ζ G = 1.3 % As alpha is proportional to 1/f, aberr. = No significant effect on lateral resolution Log (Cs/1), Log(Cc) Cc (mm), Cs/1 (mm) 5 45 4 35 3 Cs 25 Cc 2 15 1 5 5 1 15 2 25 Focal length (mm) y =,833x + 1,729 7 6 5 4 3 2 1 Cs Cc -1-2 3 3,5 4 4,5 5 5,5 Focal length y = 2,932x - 1,548
Primary column Practical rules for D1: Small probe : L1 = use D1-2 or D1-3 Probe size : 1 12 nm 6 < L1 < 7 use D1-3 or D1-4 Probe size : 7 1 nm L1 > 7 use D1-4 Probe size < 7 nm Large current : Always use D1-1 D1 size 3 2 15 1 microns With the new D1 (5 holes) : D1 size 75-3 2 15 1 microns
Dynamic Transfer and Scanning How does it works?: B1, B2 and B3 rotate the primary ion beam around the center of D1. B3 in addition to its action on the primary ion beam is in charge of the Dynamic transfer. B3 is powered so as to cancel the secondary ion beam motion
Dynamic Transfer and Scanning Tuning of the scanning system: B3 and B1 are set at their theoretical values respectively : 496 and 37 bits B2 is the free parameter and can be tuned independently in X and Y. Procedure : Implant a large area 7 microns without D1, Reduce the scanning field to 1 microns and set up D1-2 and ES5, Increase the scanning field to 6 microns and tune B2X and B2Y independently to get an homogeneous image yield on the whole area. Check the raster relationship (microns bits) and the large field coefficient. Standard values are respectively 5 microns 19 bits and 1. Theoretical values for B2 : B2X = 317, B2Y = 348
Dynamic Transfer and Scanning Why this tuning can change? Beam position at ES (microns) 4 3 2 1-1 -2-3 -4-3 -2-1 1 2 3 Simulation conditions : B3 = Volts Position at ES Angle at C.O. Scanning area (microns) Sec. Ions emitted at different position on the sample Conclusion : ES must be kept constant 2,5 2 1,5 1,5 -,5-1 -1,5-2 -2,5-3 Rotation angle at C.O.(mrad) Simulation conditions : B3 = 1 Volts Sec. Ions emitted at 3 microns ES reference : 7 Volts B3 (Volts) C.O. position (mm) B3 and C.O. position vs ES 1,8 Position,6,4,2 -,2 -,4 -,6 B3 -,8-1 685 69 695 7 75 71 715 ES (Volts)
Cy, P2 and P3 How does it works?: P1, P2 and P3 allow to center the beam in LF2 and ES in the vertical plane. These 3 small deviations (-6, +12, -6 ) are globally achromatic. Cy allows to center the secondary ion beam in LF3 and ES in the horizontal plane. C2 is then tuned to center the beam in SS1.
Cy, P2 and P3 In order to maintain the mass spectrometer settings unchanged it is recommended to re-center the secondary ion beam in ES with CY and P2/P3. (*) Caution: While using the Sec. Ion Beam centering software which determines the optimum values for Cy and P2/P, the relative ratio P2/P3 has to be properly set. This ratio allow to maintain the sec ion beam parallel to the horizontal axis while changing P2and P3. This ratio is very sensitive to the setting of LF2. P3/P2 relative ratio vs LF2 As LF2 is generally set to 125 bits, the relative ratio P3/P2 must be set at.36.,6,5 Standard value.36 This coefficient can be introduced in the Setup (Tuning section) Ratio P3/P2,4,3,2,1 11 115 12 125 13 LF2 (bits)
Mass spectrometer aberrations EnS, LF4, ES, AS M/M = G W ES + K θ θ 2 + K E θ E/E + (K β β 2 + higher order terms..) H, AS, ES H, LF2, LF5 W ES : Entrance slit width (or beam waist) G : magnification of the spectrometer K θ θ 2 : second order aperture aberration term, K E θ E/E : chromatic aberration term, with θ being the half aperture angle in the radial plane and E/E the relative energy spread of the secondary beam.
LF4, Q and Chromatic aberration Q and LF4 act both on beam focalization and on chromatic aberration compensation. But as LF4 has been set very near from the energy slit, it acts mainly on chromatic compensation.
LF4, Q and Chromatic aberration Tuning of LF4: Set up ES3, Record 3 HMR spectra for each LF4, corresponding to 3 different values of EW offset, Select the optimum value for LF4 which corresponds to a motionless mass line Center line vs EW offset Slope vs LF4 CL (microns) -5-1 -15-2 -25-3 -35-4 -45-5 LF4 = 177 LF4 = 171 LF4 = 165 LF4 = 159 LF4 = 153 slope,6,4,2 -,2 -,4 -,6 Optimum Value -1-5 5 1 15 155 16 165 17 175 18 EW offset (bits) LF4 (bits) 7.5 Volts Experimental conditions : Cs + on Silicon Wafer, ES3, Q = 37
LF4, Q and Chromatic aberration As Q acts on Chromatic compensation, LF4 cannot remain unchanged as Q varies. Thus, For each value of the Quadrupole Q one can determine an optimum value for LF4. Experimental conditions : Cs + on Silicon Wafer, ES3 LF4 vs Q LF4(bits) 158 156 154 152 15 148 LF4 = 9921 22.5 * Q This empiric relationship can be introduced in the setup (Keyboard section) 146 37 371 372 373 374 375 376 Q (bits)
Second order aperture aberration The mass spectrometer is corrected for second order aperture aberration in the radial plane by the Hexapole H. In the vertical plane, the beam shape has been transformed from a circular one to a slit one, leading to a dramatic reduction of aberration effects. Experimental conditions : Cs + on Silicon Carbide, ES5 and AS5 Second order aberration Second order aberration 12 12 CL (microns), I (a.u.) 1 8 6 4 2,2,4,6,8 1 AS5y (mm) Center Line C- intensity CL (microns), I (a.u.) 1 8 6 4 2-2 7,8 7,9 8 8,1 AS5x (mm) Centerr Line C- intensity Radial Plane (θ 2 ) Vertical plane (β 2 )
PHD Aging effects After aging Before aging Count per second (cps) per milivolt 7 6 5 4 3 2 1 1 2 3 4 5 6 Discriminator threshold Pulse height (milivolts) Aging leads to a detection efficiency decrease with time (ion dose)
PHD Aging effects Evolution of PHD parameters with 32 S - ion flux of 1.41 6 cps over 3 hours Max D evolution D/G evolution 5,5 3,5 Ln (Max D ) 5,4 5,3 5,2 5,1 5 D/G 3 2,5 2 4,9 5 1 15 2 1,5 5 1 15 2 Time (mn) Time (mn) The evolution of Max D with time can be expressed as : Max D exp ( t τ ) with t: exposure time, and τ D : fitting parameter D
PHD Aging effects 1/t (mn-1) 1/τ Comparison between two EMs version,6 Si,5 Si Small,4 C,3 C,2 Si O Large S,1 Si O Si Si 3 6 9 12 Intensity (cps x 1E3) 1/ τ proportional to the ion dose Large 1 MaxD Small 1 MaxD Small 2 R3/28 Large 2 MaxD Large 3 MaxD Large 4 MaxD EM dimension (w x l x h): Small: 7 x 3 x 35mm Large: 7 x 6 x 65mm The Aging effect has been reduced by a factor ranging from 5 to 22 thanks to the Large version
PHD Aging effects An abundant isotope and a weak one are recorded simultaneously with two different EMs. The EM detecting the abundant isotope exhibits a change of its detection efficiency due to aging effect while the other remain unchanged. The relative variation of isotopic ratio R can be expressed as : R/R exp 1 τ 2 1 ( τ ) The fitting parameter τ R is related to τ D by an approximate relationship given by: ( ) D τ R Thus a 1% variation of MaxPHD leads to 5 per mil variation of any isotopic ratio measurement. t R
QSA Effects on Isotopic ratio measurements Assuming a Poisson statistics, the correction factor is given in a first order approximation by : Ncor = Nexp (1 + K/2) (1) Where Ncor is the real number of ions reaching the first dynode and Nexp the number of pulses counted with a given threshold and K is the ratio secondary over primary. Kcor = Kexp / (1 - Kexp/2) 12, 1, δ34exp = <δ34cor> +.69 Kcor * 1 Delta 34/32 (per mil) 8, 6, 4, 2,, -2, -4, Delta exp. Delta cor. Kcor/2 Experimental conditions : Primary ion : Cs +, 1 pa -6,,,5,1,15,2,25 Sample : pyrite Kcor
QSA Effects on PHD Semi-empirical model : 45 4 35 Np : electron-ion yield on 1 st dynode Ne : electron-electron yield on other dynodes For K =.195, Np=11, 22, 33 and 44 are added properly weighted by the probability of detecting simultaneously 1, 2, 3 or 4 ions. PHD (cps/mv) 3 25 2 15 1 5 1 2 3 4 5 Threshold (mv) 34S 32S 45 K=.1 4 K=.1 4 35 PHD (cps/mv) 35 3 25 2 15 1 5 PHD (cps/mv) 3 25 2 15 1 5 K=.195 1 2 3 4 5 1 2 3 4 5 Treshold (mv) Threshold (mv) 34S, Np =1.7, Ne = 3 32S, Np =11, Ne = 3
Mass Fractionation at the entrance slit Due to the presence of leaking Bfields along the secondary ion trajectories, the secondary ion beam at the entrance slit is mass fractionated. These Bfields are mainly produced by the two ion pumps in charge of pumping the analysis and the central chamber. Two coils have been added to cancel this effect at the entrance slit. Intensity (cps) Cy (Volts) 28Si 12C
Mass Fractionation at the entrance slit Tuning Set up ES3 or ES4, select two mass lines (12 and 28 or 28 and 56, ) Record SIBC spectra corresponding to different values of intensity in the coils, then plot C.L. vs Bhor or Bvert. Select the optimum values which correspond to exact coincidence between the two mass lines Horizontal Bfield compensation Vertical Bfield compensation 6,5 1,5 P3-35 V (Volts) 6 5,5 5 4,5 4 3,5 3 2,5 2 12C 28Si Optimum value Cy (Volts) 1,5 -,5-1 -1,5 12C 28Si Optimum value 2 4 6 8 1 5 1 15 2 Bhor (bits) Bvert (bits) Experimental conditions : Cs + on Silicon Wafer, ES3
Motorized slits/diaphragm and hexapole In order to increase the reproducibility and throughput of the N5 4 main elements will be motorized. H Fully under computer control including : Preseted positions. Different positions for presputtering and analysis. Automated software to center each element. Compatible with existing instrument. D1 ES AS
Motorized slits/diaphragm and hexapole Diaphragm D1 Horizontal Movement D1 arm with 5 holes Flange Vertical Movement