US20020056813A1 - Target locking system for electron beam lithography - Google Patents
Target locking system for electron beam lithography Download PDFInfo
- Publication number
- US20020056813A1 US20020056813A1 US09/290,785 US29078599A US2002056813A1 US 20020056813 A1 US20020056813 A1 US 20020056813A1 US 29078599 A US29078599 A US 29078599A US 2002056813 A1 US2002056813 A1 US 2002056813A1
- Authority
- US
- United States
- Prior art keywords
- field
- target
- aperture
- stage
- writing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000609 electron-beam lithography Methods 0.000 title description 16
- 239000000758 substrate Substances 0.000 claims abstract description 42
- 238000012937 correction Methods 0.000 claims abstract description 28
- 238000000034 method Methods 0.000 claims description 23
- 238000010894 electron beam technology Methods 0.000 claims description 14
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical group [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims description 5
- 229910052799 carbon Inorganic materials 0.000 claims description 5
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 claims description 5
- 239000010931 gold Substances 0.000 claims description 5
- 229910052737 gold Inorganic materials 0.000 claims description 5
- 238000007654 immersion Methods 0.000 claims description 4
- 238000002039 particle-beam lithography Methods 0.000 claims description 3
- 239000002245 particle Substances 0.000 claims 12
- 230000000977 initiatory effect Effects 0.000 claims 3
- 238000010408 sweeping Methods 0.000 claims 3
- 238000011065 in-situ storage Methods 0.000 abstract description 8
- 238000010586 diagram Methods 0.000 description 16
- 239000000463 material Substances 0.000 description 8
- 238000005259 measurement Methods 0.000 description 5
- 239000012528 membrane Substances 0.000 description 5
- 239000004065 semiconductor Substances 0.000 description 5
- 230000001351 cycling effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000005415 magnetization Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003094 perturbing effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- AHTSCRNWVSEMPI-UHFFFAOYSA-N gold;methane Chemical compound C.[Au] AHTSCRNWVSEMPI-UHFFFAOYSA-N 0.000 description 1
- 230000003116 impacting effect Effects 0.000 description 1
- 238000005305 interferometry Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000001459 lithography Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
- H01J37/00—Discharge tubes with provision for introducing objects or material to be exposed to the discharge, e.g. for the purpose of examination or processing thereof
- H01J37/30—Electron-beam or ion-beam tubes for localised treatment of objects
- H01J37/317—Electron-beam or ion-beam tubes for localised treatment of objects for changing properties of the objects or for applying thin layers thereon, e.g. for ion implantation
- H01J37/3174—Particle-beam lithography, e.g. electron beam lithography
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B82—NANOTECHNOLOGY
- B82Y—SPECIFIC USES OR APPLICATIONS OF NANOSTRUCTURES; MEASUREMENT OR ANALYSIS OF NANOSTRUCTURES; MANUFACTURE OR TREATMENT OF NANOSTRUCTURES
- B82Y10/00—Nanotechnology for information processing, storage or transmission, e.g. quantum computing or single electron logic
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B82—NANOTECHNOLOGY
- B82Y—SPECIFIC USES OR APPLICATIONS OF NANOSTRUCTURES; MEASUREMENT OR ANALYSIS OF NANOSTRUCTURES; MANUFACTURE OR TREATMENT OF NANOSTRUCTURES
- B82Y40/00—Manufacture or treatment of nanostructures
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
- H01J37/00—Discharge tubes with provision for introducing objects or material to be exposed to the discharge, e.g. for the purpose of examination or processing thereof
- H01J37/30—Electron-beam or ion-beam tubes for localised treatment of objects
- H01J37/304—Controlling tubes by information coming from the objects or from the beam, e.g. correction signals
- H01J37/3045—Object or beam position registration
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
- H01J2237/00—Discharge tubes exposing object to beam, e.g. for analysis treatment, etching, imaging
- H01J2237/30—Electron or ion beam tubes for processing objects
- H01J2237/317—Processing objects on a microscale
- H01J2237/3175—Lithography
- H01J2237/31752—Lithography using particular beams or near-field effects, e.g. STM-like techniques
- H01J2237/31754—Lithography using particular beams or near-field effects, e.g. STM-like techniques using electron beams
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
- H01J2237/00—Discharge tubes exposing object to beam, e.g. for analysis treatment, etching, imaging
- H01J2237/30—Electron or ion beam tubes for processing objects
- H01J2237/317—Processing objects on a microscale
- H01J2237/3175—Lithography
- H01J2237/31761—Patterning strategy
- H01J2237/31766—Continuous moving of wafer
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S977/00—Nanotechnology
- Y10S977/84—Manufacture, treatment, or detection of nanostructure
- Y10S977/887—Nanoimprint lithography, i.e. nanostamp
Definitions
- the present invention generally relates to particle beam lithography systems and, more particularly, to a particle beam lithography system with in situ calibration and calibration methods therefor.
- Electron beam (e-beam) lithography tools are commonly used in semiconductor manufacturing to form sub-micron shapes on a semiconductor wafer. Shapes are formed by directing a beam of electrons from a source at one end of a column onto a photoresistive layer on a substrate at an opposite end of the column. A typical substrate may be 200-300 mm in diameter or larger. These submicron shapes may be formed either by writing the shape directly onto a photoresistive layer on the substrate, wherein the substrate is a semiconductor wafer; or, by writing the shape onto a photoresistive layer on a substrate which is used as a mask, subsequently, to print the shape onto the semiconductor wafer.
- the first type is referred to as “blind mode” or a “dead reckoning mode” and is commonly used in mask making.
- the substrate is a featureless blank coated with resist and all of the patterns are placed by dead reckoning.
- the second mode which may be referred to as the “registered write mode” or a “direct write mode,” is commonly used in direct write applications, i.e. writing directly onto a semiconductor wafer, in what are referred to as device fabrication runs.
- the patterns In the registered write mode case, the patterns must be precisely located relative to previous levels which requires registration targets built into each level and the substrate as well. Regardless of the mode employed, accurately placing or repeating sub-micron shapes at precise locations across a distance of 200-300 mm demands precise beam registration.
- pattern writing may be interrupted periodically, depending on the particular tool's inherent e-beam drift, to check tool registration and, whenever registration error exceeds an acceptable tolerance, to adjust the beam.
- the substrate is held on a stage opposite (beneath) the beam source and this registration measurement involves diverting the stage to position a registration target under the beam. Then, the beam is scanned over the registration target, the target's location is measured and the target's measured location is compared against an expected result. Any measured errors are corrected by adjusting the beam or adjusting stage positional controls. Then, the stage is returned to its former position to resume writing the mask pattern. This measurement and reregistration can be time consuming.
- the registration measurement takes place at a stage location other than where the pattern is actually written. Consequently, even after measuring and correcting errors, moving the stage back into position from the measurement area may actually introduce errors, such as from the stage slipping or from other move related stresses. Also, to assure complete accuracy, the beam should be reregistered, frequently, preferably at each field. However, when throughput is a consideration, as it nearly always is, it is impractical to correct the beam registration prior to printing each field.
- a SPLEBL type system also would require a sophisticated method of extracting positioning information during exposure. While such a sophisticated method may be feasible with a Gaussian or fixed beam shape; it is highly improbable that such a method could be developed for a variable shape beam such as a described in U.S. Pat. No. 4,243,866 entitled “Method and Apparatus for Forming a Variable Size Electron Beam” to Pfeiffer et al.
- U.S. Pat. No. 4,119,854 entitled “Electron Beam Exposure System” to Tanaka et al. teaches an e-beam exposure system that may be compensated for e-beam drift and workpiece drift.
- the system of Tanaka et al. uses a pair of x and y lines as a reference target. The relative position of the stage with respect to the x and y lines is determined using differential interferometry.
- a coil is included for refocusing the beam onto the x,y reference target lines.
- refocusing the beam introduces hysteresis into a system such as that taught in Tanaka et al. The hysteresis, itself reduces beam accuracy.
- the present invention is an e-beam lithographic system capable of in situ registration.
- the preferred system is a Variable Axis Immersion Lens (VAIL) e-beam system and is a double hierarchy deflection system.
- VAIL Variable Axis Immersion Lens
- a controllable stage moves a substrate with respect to the beam axis placing the intended substrate writing field within an aperture on a field locking target.
- the field locking target is located between the optics section and the substrate and the aperture is sized to permit the beam to write the field.
- the field locking target includes alignment marks around the aperture.
- a differential interferometric system measures the relative positions of the field locking target and the stage. As the stage is moving into position for writing a field, the beam is swept to hit the alignment marks, checking system alignment.
- the beam control data (coil currents and electrostatic deflection plate voltages) required to hit the marks are stored, and drift correction values calculated and the field beam control data is compensated. Writing resumes on the newly positioned field with the beam control data corrected by the calculated drift correction values.
- the field locking target may include a mechanical adjustment for fine tuning aperture location. After an initial calibration of the system, the field locking target is adjusted to position the alignment marks at their expected location.
- FIGS. 1 A-D show the pattern relationship between spots, rectangles, subfields, fields and a skeleton
- FIG. 2 is a cross-sectional diagram of the preferred embodiment e-beam lithography system
- FIG. 3 shows a sectional view of one corner of the field locking target
- FIG. 4 shows a diagram of operation of the preferred embodiment system
- FIG. 5 is a flow diagram of target mark acquisition and centering
- FIG. 6 is a flow diagram for target locking and calibration of the preferred embodiment system
- FIG. 7 is a flow diagram of a default A1 cycle
- FIG. 8 is a flow diagram of an A1 cycle in which the locking target is scanned
- FIG. 9 is a flow diagram of the A1 cycle wherein target locking is performed in conjunction with pattern writing.
- FIGS. 1 A-D represent the hierarchical relationship between the individual components printed to form a pattern.
- FIG. 1A represents a typical substrate pattern that is divided into an array of fields 40 to accommodate a pattern which is much larger than the preferred e-beam lithography system's writing field.
- the collection of fields 40 is referred to as a skeleton 42 .
- Each field 40 is identified by the x,y address of its center point.
- each field 40 is divided into multiple sub-fields 44 , each identified by the x,y address of its center point.
- each subfield 44 may include one or more rectangles 46 , 48 , 50 and 52 . Rectangles are logical groupings and are not specifically addressed.
- each rectangle 46 , 48 , 50 and 52 is formed from groups of individual spots 54 .
- the preferred e-beam lithography system is a Variable Axis Immersion Lens (VAIL) e-beam system and is a double hierarchy deflection system.
- VAIL Variable Axis Immersion Lens
- VAIL lens system Although the preferred embodiment system is a VAIL lens system, the present invention is not intended to be limited to VAIL lens e-beam systems and other deflection systems may employ the present invention as well with appropriate modification.
- substrate is not intended as a limitation, but merely refers to material bearing the surface upon which the pattern is being printed.
- FIG. 2 shows a cross-sectional diagram of the preferred embodiment e-beam lithography system 100 .
- the preferred embodiment system includes an optics section 102 with a registration focus coil 104 a , an autofocus coil 104 b , beam deflection coils 106 , 108 , a projection lens axis shifting yoke 110 and beam deflection plates 111 .
- registration focus coil 104 a and autofocus coil 104 b are represented as two concentric coils in FIG. 2, these two coils could be combined into a single shared common coil with a summing circuit combining their individual currents into a single driving current.
- An e-beam source 90 emits a beam represented by arrow 112 , which, during writing, travels to a target field on a substrate held on carrier 114 .
- Autofocus coil 104 b adjusts beam focus for target height variations resulting from substrate imperfections, thickness variations, etc.
- double deflection yokes 106 , 108 magnetically deflect the beam 112 ; and axis shifting coil 110 shifts the variable axis of the projection lens to follow the deflected beam 112 .
- the relatively slow magnetic deflection from coils 106 , 108 determines the subfield location, while within the subfield, the beam 112 is deflected by the high speed electrostatic deflection plates 111 .
- a passive field locking target 116 permits the beam 112 to write the pattern in the substrate's target field through an aperture 118 .
- the preferred aperture is rectangular and is large enough to permit writing an entire field.
- substrate subfields are placed within the field locking target aperture 118 and electrostatic deflection is used to write spots which form the pattern shapes.
- the subfield is defined as being over marks on the field locking target 116 adjacent to the aperture 118 ; and, the beam is deflected accordingly, as represented by arrows 112 ′.
- the marks on the field locking target 116 are scanned, in situ, with the electrostatic deflection, to provide near real time positional feedback information.
- the marks on the field locking target 116 are located 4 mm above the normal writing plane of the substrate.
- magnetic deflection is used to raster the beam to an x,y address for a selected subfield 44 .
- the sub-field 44 is patterned using vectored electrostatic deflection. Rectangles are not specifically addressed by any element of deflection. Instead, the shaped beam is electrostatically positioned at each spot 54 for sufficient time to allow deflection settling and resist exposure time. So, groups of spots 54 are written to form a rectangle 46 , 48 , 50 or 52 . Rectangles 46 , 48 , 50 and 52 are written to form a subfield 44 . Subfields 44 are written to form a field 40 . Fields 40 are written to form a skeleton 42 which completes the pattern.
- the beam is magnetically deflected to the next selected subfield 44 .
- the stage moved to position the next selected field 40 in the aperture 118 , and so on, until the skeleton 42 is complete.
- e-beam systems may include a coil for refocusing the beam onto a reference target, as with any electromagnet, hysteresis is introduced into the system each time the electromagnetic lens refocuses the beam.
- This focusing hysteresis effect of the electromagnetic lens occurs as a result of superposition of the field from the current flowing through the lens windings with the fields induced in the surrounding magnetic medium.
- the first time a magnetic lens is energized a field is created which is the vector sum of these two superimposed fields.
- the coil current is turned off, a residual magnetic field remains from magnetization of the surrounding medium. If, prior to the cycling of this lens, the electron beam had been in focus (as happens whenever the beam is moved onto the field locking target), the beam will be out of focus when it is returned to the substrate.
- the registration focus coil 104 a is continuously cycled between these two locations. This continuous cycling occurs even if the reference target is not being written or scanned and insures that the off state magnetization stabilizes. With the off state magnetization stabilized, an correction offset may be applied to the projection lens. Thus, the focusing hysteresis effect is compensated for in a preferred embodiment system.
- the preferred embodiment e-beam system 100 includes a differential interferometric system 120 .
- the interferometric system 120 directs a laser, represented by arrows 122 , to laser targets 124 and 124 ′ to measure the relative position of the field locking target 116 to the stage mirror assembly 126 .
- Laser target 124 is mechanically coupled to field locking target 116 and laser target 124 ′ is attached to a stage mirror assembly 126 .
- the carrier 114 is kinematically clamped to the stage mirror assembly 126 at points 128 .
- the stage mirror assembly 126 is flexure mounted to a stage base 130 at points 132 .
- An x or y drive 134 is attached to an appropriate side of the stage base 130 to drive the stage, typically under computer control, in the x or y direction; and, once in place, to lock the stage in place.
- a mechanical centering adjustment 136 provides a fine adjustment for the field locking target 116 to precisely place it with respect to the beam.
- the beam could be centered by increasing the magnetic deflection of the beam, but at the expense of increasing system noise.
- a pattern is formed by first positioning the stage, under laser control 120 , 122 , at a corresponding x,y address of a selected field 40 .
- the stage pauses at the selected field 40 and beam exposure begins and continues until the field pattern is complete.
- the mechanical centering adjustment 136 allows the field locking target 116 to be centered both on an off center field. Mechanically centering the field locking target avoids the noise that would be otherwise induced by providing the deflection field with a sufficient range to electrically compensate for misalignment.
- FIG. 3 shows a perspective view of a corner of a field locking target 116 .
- the field locking target 116 is a carbon substrate with an aperture 118 tailored to be, preferably, 0.2 mm larger than the field size in each direction, e.g., for an 0.9 ⁇ 0.9 mm field size the aperture is 1.1 ⁇ 1.1 mm.
- Aperture sidewalls 160 and the bottom of the field locking target 116 are coated with gold to provide a radiation barrier.
- a perpendicular grid (not drawn to scale) or mesh is formed on the upper surface by crisscrossing lines 162 , 164 spaced at the subfield periodicity.
- the lines 162 , 164 are of a material selected as having a much higher atomic number (high Z) and, therefore, scatters electrons much more than the carbon background material. So, for the e-beam, selecting contrasting materials with a difference in atomic number of at least one order of magnitude provides a strong material contrast to the scanning electron beam.
- the carbon background material has a relatively low atomic number (low Z) providing a high contrast to the high Z gold lines of the preferred embodiment, which results in a high contrast back scatter signal. Then, back scattered electron detection is used to determine the position of the electron beam relative to marks.
- the preferred field locking target 116 is gold on carbon, other material combinations may be substituted, provided they do not degrade the resulting back scatter effects by reducing the signal to noise ratio or by adversely affecting the e-beam environment. Further, the preferred field locking target 116 individual material thicknesses are not critical, but are selected to be thick enough to block stray electrons or x-rays from reaching the photosensitive pattern layer on the substrate, which would inadvertently expose resist. Thus, the carbon thickness, the bottom gold layer thickness, and the sidewall thickness must be considered together in determining an appropriate thickness for each.
- Subfield 166 includes four vertical scan rectangles 170 that identify the y position of the locking target's horizontal bars 164 .
- Subfield 168 includes four horizontal scan rectangles that identify the x position of the locking target's vertical bars 162 .
- the average of the scan rectangles 170 positions in A1 cycle subfields 166 and 168 serve to identify the location of the field corners relative to the field locking target bars 162 and 164 . This positional relationship serves as the basis of computing the field distortion. Individually, each of the four subfield scans 170 establish the x and y components of subfield distortion.
- multiple alignment bars 162 and 164 there are several advantages to providing multiple alignment bars 162 and 164 .
- Second, in the preferred embodiment e-beam lithography system 100 only drift errors are measured at the four corners of the aperture 118 .
- This measurement provides sufficient information to correct X and Y translation, X and Y skew (rotation), X and Y magnification and X and Y trapezoidal distortion.
- more precise corrections for even higher order terms, such as barrel-pincushion, symmetric and anti-symmetric quadratic distortion etc. may be generated by including additional target locations for measuring additional error information.
- four vertical 170 and horizontal 172 scans may be made at each calibration location, with each scan located at the nominal subfield limits. These scans provide data to monitor and correct subfield level distortions.
- system 100 operation can be understood with reference to the simple flow diagram FIG. 4.
- the preferred embodiment system 100 continuously cycles through 4 distinct phases, an A1 phase 180 , an A2 phase 182 , a B phase 184 and a C phase 186 .
- the beam is scanned across reference targets 162 , 164 at the four corners of aperture 118 .
- the registration marks on the substrate are scanned during the A 2 phase 182 .
- the pattern is printed onto the main field of the substrate during the B phase 184 .
- the C phase 186 is preparation for the next field, i.e., the table is moved to shift between fields and, if necessary, beam servos may be adjusted, autofocus height data may be collected, autofocus and field locking coil digital to analog converters (DACs) may be loaded and the pattern may be setup.
- DACs digital to analog converters
- the preferred stage 128 includes a permanently mounted system calibration grid. Prior to beginning normal operation, the aperture 118 must be centered, mechanically, with respect to the exposure field. To that end, the beam is scanned over it's full deflection, in focus on the target grid. Back scattered electrons from the calibration grid and the field locking target 116 are detected and displayed. The field locking target's x/y position relative to the deflection field is delineated by the change in focus between the back scattered image of the calibration target and the field locking target. The image from the field locking target 116 is centered in the deflection field using mechanical adjustment 136 .
- the beam is scanned to find bars 162 and 164 in each of the four corner A1 subfields 166 and 168 .
- the field is centered and in step 190 , the four expected locking target 166 , 168 locations, i.e., at the corners of the aperture 118 , are scanned to determine if beam to target alignment is sufficiently accurate. If the locking targets are found at their expected location, no further adjustment is necessary, and so, in step 192 , target acquisition is complete.
- step 194 a check is made to determine whether a mark 170 was detected in each scan. If not, then in step 196 , for each locking target 166 , 168 where a mark 170 was not detected, the mark deflection location is adjusted; and, using an appropriate search delta, such as half the length/width of a subfield, the area in the proximity of the expected mark location is searched. If a mark is initially detected in each scan in step 194 or, after finding the marks in step 196 , the individual centering of each mark 170 is checked in step 198 .
- step 200 for each off center mark, the mark deflection location is adjusted by an appropriate amount to center the mark. After either finding that all of the marks are centered in step 198 or, after centering off center marks in step 200 , the A1 deflection is loaded in step 202 and, the A1 cycle is begun.
- step 204 the system calibration grid, which is at the writing plane, is scanned and subfield center and corner coordinates are measured and adjusted to bring the deflection field and subfields into alignment with the grid. Then, the deflection control parameters are perturbed by a preselected amount and, the A1, A2 and B phases are re-executed. using the writing plane calibration grid located on the stage, the A2 and B deflections are calibrated to generate correction factors. Then, in step 206 , a relationship is established between the affect of the correction factors in the writing plane and at the field locking target.
- the system calibration grid mounted on the stage at the writing plane height is scanned. Then, the subfield center and corner coordinates are adjusted to bring the deflection field and subfields into alignment with the grid. Finally, the deflection control parameters are varied, intentionally, an identical amount during each of an A1, A2 and B cycle. Any relative displacement between the A1 and A2, B phases is measured and translated into writing plane corrections. After this initial calibration, substrate patterns are formed normally, cycling through all four phases A1, A2, B and C.
- field locking is done regardless of the system's operating mode, whether in INACTIVE, STANDBY or ACTIVE mode. Further, whether in STANDBY or ACTIVE mode, the beam is scanned over the reference targets 166 , 168 during every A1 cycle 180 to maintain the targets 166 , 168 in thermal equilibrium. Without continually scanning the reference targets 166 , 168 , target temperature will vary unpredictably due to the irregular power input. This temperature variation will cause the reference targets 166 , 168 to exhibit unacceptable thermal drift. Although the reference targets 166 , 168 are scanned every cycle, processing scan results is optional in STANDBY mode, but processing is mandatory in ACTIVE mode to recalibrate the system.
- the A2 correction normally overrides any corrections derived from an A1 cycle, making the A1 cycle is superfluous.
- both cycles may be used to perform a field locking diagnostic in a field locking diagnostic mode.
- a registration target that is fixed onto the XY stage 114 , is moved into the deflection field and the A2 target scan is performed repeatedly on a the registration target.
- the A1 scans are not processed.
- the A2 scans are not for generating corrections; but instead, for monitoring field translation drift and distortion.
- drifts are induced, deliberately, by perturbing the upper column, followed by A1 field locking scans. Corrections are calculated from the A1 field locking scans and applied to the tool. If the system is functioning properly, the A2 scans should not measure any drift because, the drift will have been detected in A1 cycle and applied as a correction for the A2 and B cycle.
- FIG. 7 is a flow diagram of an initial setup A1 cycle 180 , wherein the locking target is not scanned. Prior to centering the deflection on the field locking target 116 , stray electrons may reach the substrate. Therefore, prior to centering the beam should not be active whenever there is a substrate on the stage. However, once the locking target has been acquired as described above with reference to FIG. 5, the beam is active in every A1 cycle.
- step 210 upon entering the A1 cycle, status bits are set to indicate that the A1 cycle is in process (A1IP). Then, for the first time, in step 212 the status bits are checked to determine if the A1 cycle is in process as indicated by the A1IP bits being set.
- the autofocus coil 104 b is set to the calibration grid focus value grid location and the registration focus coil 104 a is set to its active value.
- the beam feedback status (BF_STATUS) is placed in the system data path and the beam feedback state (BF_STATE) is set to zero to disable mechanical stage error corrections.
- beam feedback information is collected to provide stage positional information in the nanometer range. The collected information is used as a fine tune adjustment to compensate beam placement for stage vibrations and mechanical inaccuracies from the coarser mechanical placement resolution. Typically, stage placement mechanical accuracy is to within what is known as the beam feedback capture range.
- the actual stage position is irrelevant and, so, beam feedback is disabled.
- step 216 the magnetic deflection is driven to actually deflect the beam to each of the corner subfields 166 and 168 to scan the reference target.
- step 218 the autofocus coil is set to the target value (i.e., the field address) and the focus locking coil is turned off.
- the BF_STATUS is placed in the system data path, the BF_STATE is set to one to enable mechanical stage error corrections and, simultaneously, the A1IP bits are turned off.
- step 212 the A1IP bits are checked again. However, the recheck in step 212 indicates that the A1IP bits are off. So, in step 220 , an A2 HOLDOFF signal is checked to determine whether to continue to the A2 cycle. If, in step 220 , the A2 HOLDOFF signal is active, the system enters a wait cycle in step 222 . Periodically, the A2 HOLDOFF signal is rechecked in step 220 until it is found to be inactive, at which point the system proceeds to an A2 cycle 182 .
- FIG. 8 is a flow diagram of an A1 cycle in which the locking target is scanned, but which otherwise is substantially the same as the flow diagram in FIG. 7. Accordingly, identical steps with identical function are identified with identical numbers.
- the locking target is scanned, measured and the scan results are fed back.
- the preferred e-beam lithography system enters the A2 cycle from the A1 cycle of FIG. 8, the system resumes default cycling and begins processing the A1 scan feedback signal in step 224 .
- the mesh on the upper surface of the field locking target 116 provides an initial guide for the preferred system, aiding in initially locating suitable reference targets.
- the beam is sequentially deflected around the periphery of the field aperture 118 , during an A1 cycle 180 , at expected mark locations.
- a searching algorithm is used to locate center subfield marks at each of the respective marks and, the corresponding mark command addresses are modified accordingly.
- FIG. 9 is a flow diagram of the A1 cycle wherein target locking is done as a pattern is being written but, which otherwise is substantially the same as the flow diagram in FIG. 8. Accordingly, identical steps with identical function are identified with identical numbers. So, in step 216 ′′, as the pattern is being scanned, the registration focus coil 104 a is checked in step 260 to determine whether correction is needed. This check is done by perturbing the field locking coil from a current value and returning for a second A1 iteration in step 216 ′′. After the second A1 iteration, in step 262 , correction terms are computed and passed forward for application to pattern position controls during the A2 cycle 182 and B cycle 184 . Also, during pattern writing, in step 264 , after entering the A2 cycle 182 , the patterns are adjusted using update corrections generated during the A1 cycle.
Abstract
Description
- 1. Field of the Invention
- The present invention generally relates to particle beam lithography systems and, more particularly, to a particle beam lithography system with in situ calibration and calibration methods therefor.
- 2. Background Description
- Electron beam (e-beam) lithography tools are commonly used in semiconductor manufacturing to form sub-micron shapes on a semiconductor wafer. Shapes are formed by directing a beam of electrons from a source at one end of a column onto a photoresistive layer on a substrate at an opposite end of the column. A typical substrate may be 200-300 mm in diameter or larger. These submicron shapes may be formed either by writing the shape directly onto a photoresistive layer on the substrate, wherein the substrate is a semiconductor wafer; or, by writing the shape onto a photoresistive layer on a substrate which is used as a mask, subsequently, to print the shape onto the semiconductor wafer.
- Further, there are two broad types of writing modes used in electron beam lithography. The first type is referred to as “blind mode” or a “dead reckoning mode” and is commonly used in mask making. In the blind mode, the substrate is a featureless blank coated with resist and all of the patterns are placed by dead reckoning. The second mode, which may be referred to as the “registered write mode” or a “direct write mode,” is commonly used in direct write applications, i.e. writing directly onto a semiconductor wafer, in what are referred to as device fabrication runs. In the registered write mode case, the patterns must be precisely located relative to previous levels which requires registration targets built into each level and the substrate as well. Regardless of the mode employed, accurately placing or repeating sub-micron shapes at precise locations across a distance of 200-300 mm demands precise beam registration.
- However, even if the beam is registered adequately when pattern printing begins, during the course of writing the pattern, the e-beam may exhibit what is referred to as drift, i.e., exhibiting increasing inaccuracy in one direction as time passes. So, in order to maintain adequate precision, pattern writing may be interrupted periodically, depending on the particular tool's inherent e-beam drift, to check tool registration and, whenever registration error exceeds an acceptable tolerance, to adjust the beam.
- Normally, the substrate is held on a stage opposite (beneath) the beam source and this registration measurement involves diverting the stage to position a registration target under the beam. Then, the beam is scanned over the registration target, the target's location is measured and the target's measured location is compared against an expected result. Any measured errors are corrected by adjusting the beam or adjusting stage positional controls. Then, the stage is returned to its former position to resume writing the mask pattern. This measurement and reregistration can be time consuming.
- Furthermore, for this e-beam registration approach, the registration measurement takes place at a stage location other than where the pattern is actually written. Consequently, even after measuring and correcting errors, moving the stage back into position from the measurement area may actually introduce errors, such as from the stage slipping or from other move related stresses. Also, to assure complete accuracy, the beam should be reregistered, frequently, preferably at each field. However, when throughput is a consideration, as it nearly always is, it is impractical to correct the beam registration prior to printing each field.
- Consequently, attempts have been made to perform registration in place while writing the pattern, i.e., in situ. One in situ approach, suggested by MIT as reported in “Spatial-phase-locked Electron-beam Lithography:Inital Test Results”, pp2342-5J. Vac. Sci. Technol. B. Vol. 11, No 6 November December 1993, is referred to as the Spatial-Phase-Locked E-Beam Lithography (SPLEBL) system. Implementing a SPLEBL type system would require including special registration patterns on every substrate and a blanket exposure of the substrate for the registration cycle. However, such a blanket exposure is unusable for high sensitivity resists. Further, a SPLEBL type system also would require a sophisticated method of extracting positioning information during exposure. While such a sophisticated method may be feasible with a Gaussian or fixed beam shape; it is highly improbable that such a method could be developed for a variable shape beam such as a described in U.S. Pat. No. 4,243,866 entitled “Method and Apparatus for Forming a Variable Size Electron Beam” to Pfeiffer et al.
- Another in situ registration technique has been proposed for x-ray membrane exposure by Scientists at Naval Research Laboratories (NRL) as described, for example, by Perkins et al. in “Improving Pattern Placement Using Through the Membrane Signal Monitoring,”J. Vac. Sci. Technol. B, Microelectron. Nanometer Struct. (USA) Vol. 16, No. 6 November December 1998 pp3567-71. This technique requires fixing a registration grid to the carrier, directly under the x-ray membrane. The registration grid forms a Schottky diode junction with the carrier, allowing it to function, simultaneously, as a high gain detector of incident electrons. However, this registration technique is limited to high transmission membranes and its resolution capability has not yet been evaluated. Further, below the membrane, the beam diverges and scatters rapidly, which limits the usefulness of this technique.
- U.S. Pat. No. 4,119,854 entitled “Electron Beam Exposure System” to Tanaka et al. teaches an e-beam exposure system that may be compensated for e-beam drift and workpiece drift. The system of Tanaka et al. uses a pair of x and y lines as a reference target. The relative position of the stage with respect to the x and y lines is determined using differential interferometry. A coil is included for refocusing the beam onto the x,y reference target lines. However, refocusing the beam introduces hysteresis into a system such as that taught in Tanaka et al. The hysteresis, itself reduces beam accuracy.
- Thus, there is a need for in situ registration methods for e-beam lithography system and more particularly for VAIL e-beam lithography systems.
- It is therefore a purpose of the present invention to improve e-beam lithography system registration;
- It is another purpose of the present invention to reduce the time required for reregistering e-beam lithography systems;
- It is yet another purpose of the present invention to improve VAIL e-beam lithography system accuracy;
- It is yet another purpose of the present invention to improve VAIL e-beam lithography accuracy without impacting e-beam tool throughput.
- The present invention is an e-beam lithographic system capable of in situ registration. The preferred system is a Variable Axis Immersion Lens (VAIL) e-beam system and is a double hierarchy deflection system. A controllable stage moves a substrate with respect to the beam axis placing the intended substrate writing field within an aperture on a field locking target. The field locking target is located between the optics section and the substrate and the aperture is sized to permit the beam to write the field. The field locking target includes alignment marks around the aperture. A differential interferometric system measures the relative positions of the field locking target and the stage. As the stage is moving into position for writing a field, the beam is swept to hit the alignment marks, checking system alignment. The beam control data (coil currents and electrostatic deflection plate voltages) required to hit the marks are stored, and drift correction values calculated and the field beam control data is compensated. Writing resumes on the newly positioned field with the beam control data corrected by the calculated drift correction values.
- The field locking target may include a mechanical adjustment for fine tuning aperture location. After an initial calibration of the system, the field locking target is adjusted to position the alignment marks at their expected location.
- The foregoing and other objects, aspects and advantages will be better understood from the following detailed preferred embodiment description with reference to the drawings, in which:
- FIGS.1A-D show the pattern relationship between spots, rectangles, subfields, fields and a skeleton;
- FIG. 2 is a cross-sectional diagram of the preferred embodiment e-beam lithography system;
- FIG. 3 shows a sectional view of one corner of the field locking target;
- FIG. 4 shows a diagram of operation of the preferred embodiment system;
- FIG. 5 is a flow diagram of target mark acquisition and centering;
- FIG. 6 is a flow diagram for target locking and calibration of the preferred embodiment system;
- FIG. 7 is a flow diagram of a default A1 cycle;
- FIG. 8 is a flow diagram of an A1 cycle in which the locking target is scanned;
- FIG. 9 is a flow diagram of the A1 cycle wherein target locking is performed in conjunction with pattern writing.
- Referring now to the drawings, and more particularly, FIGS.1A-D represent the hierarchical relationship between the individual components printed to form a pattern. FIG. 1A represents a typical substrate pattern that is divided into an array of
fields 40 to accommodate a pattern which is much larger than the preferred e-beam lithography system's writing field. The collection offields 40 is referred to as a skeleton 42. Eachfield 40 is identified by the x,y address of its center point. In FIG. 1B, eachfield 40 is divided intomultiple sub-fields 44, each identified by the x,y address of its center point. In FIG. 1C, eachsubfield 44 may include one ormore rectangles rectangle individual spots 54. - The preferred e-beam lithography system is a Variable Axis Immersion Lens (VAIL) e-beam system and is a double hierarchy deflection system. U.S. Pat. No. 4,376,249 entitled “Variable Axis Electron Beam Projection System” to Pfeiffer et al. and U.S. Pat. No. 4,544,846 entitled “Variable Axis Immersion Lens Electron Beam Projection System” to Langer et al., both assigned to the assignee of the present invention and incorporated herein by reference, are examples of VAIL e-beam systems. Although the preferred embodiment system is a VAIL lens system, the present invention is not intended to be limited to VAIL lens e-beam systems and other deflection systems may employ the present invention as well with appropriate modification. Furthermore, as used herein, substrate is not intended as a limitation, but merely refers to material bearing the surface upon which the pattern is being printed.
- FIG. 2 shows a cross-sectional diagram of the preferred embodiment
e-beam lithography system 100. The preferred embodiment system includes anoptics section 102 with a registration focus coil 104 a, an autofocus coil 104 b, beam deflection coils 106, 108, a projection lens axis shifting yoke 110 and beam deflection plates 111. Although registration focus coil 104 a and autofocus coil 104 b are represented as two concentric coils in FIG. 2, these two coils could be combined into a single shared common coil with a summing circuit combining their individual currents into a single driving current. An e-beam source 90 emits a beam represented byarrow 112, which, during writing, travels to a target field on a substrate held oncarrier 114. Autofocus coil 104 b adjusts beam focus for target height variations resulting from substrate imperfections, thickness variations, etc. In the preferred Vail lens system, double deflection yokes 106, 108 magnetically deflect thebeam 112; and axis shifting coil 110 shifts the variable axis of the projection lens to follow the deflectedbeam 112. The relatively slow magnetic deflection fromcoils 106, 108 determines the subfield location, while within the subfield, thebeam 112 is deflected by the high speed electrostatic deflection plates 111. - A passive
field locking target 116 permits thebeam 112 to write the pattern in the substrate's target field through anaperture 118. The preferred aperture is rectangular and is large enough to permit writing an entire field. During normal pattern writing, substrate subfields are placed within the field lockingtarget aperture 118 and electrostatic deflection is used to write spots which form the pattern shapes. During registration, the subfield is defined as being over marks on thefield locking target 116 adjacent to theaperture 118; and, the beam is deflected accordingly, as represented byarrows 112′. Then, the marks on thefield locking target 116 are scanned, in situ, with the electrostatic deflection, to provide near real time positional feedback information. Typically, the marks on thefield locking target 116 are located 4 mm above the normal writing plane of the substrate. - So, within the
main field 40, magnetic deflection is used to raster the beam to an x,y address for a selectedsubfield 44. Then, thesub-field 44 is patterned using vectored electrostatic deflection. Rectangles are not specifically addressed by any element of deflection. Instead, the shaped beam is electrostatically positioned at eachspot 54 for sufficient time to allow deflection settling and resist exposure time. So, groups ofspots 54 are written to form arectangle Rectangles subfield 44.Subfields 44 are written to form afield 40.Fields 40 are written to form a skeleton 42 which completes the pattern. - After forming all of the
rectangles subfield 44, the beam is magnetically deflected to the next selectedsubfield 44. After forming all of thesubfields 44 in the selectedfield 40, the stage moved to position the next selectedfield 40 in theaperture 118, and so on, until the skeleton 42 is complete. - As indicated hereinabove, while some e-beam systems, e.g., Tanaka et al., may include a coil for refocusing the beam onto a reference target, as with any electromagnet, hysteresis is introduced into the system each time the electromagnetic lens refocuses the beam. This focusing hysteresis effect of the electromagnetic lens occurs as a result of superposition of the field from the current flowing through the lens windings with the fields induced in the surrounding magnetic medium. The first time a magnetic lens is energized, a field is created which is the vector sum of these two superimposed fields. When the coil current is turned off, a residual magnetic field remains from magnetization of the surrounding medium. If, prior to the cycling of this lens, the electron beam had been in focus (as happens whenever the beam is moved onto the field locking target), the beam will be out of focus when it is returned to the substrate.
- However, in the e-beam system of the present invention, the registration focus coil104 a is continuously cycled between these two locations. This continuous cycling occurs even if the reference target is not being written or scanned and insures that the off state magnetization stabilizes. With the off state magnetization stabilized, an correction offset may be applied to the projection lens. Thus, the focusing hysteresis effect is compensated for in a preferred embodiment system.
- For tracking and selecting stage location, the preferred
embodiment e-beam system 100 includes adifferential interferometric system 120. Theinterferometric system 120 directs a laser, represented byarrows 122, to laser targets 124 and 124′ to measure the relative position of thefield locking target 116 to thestage mirror assembly 126. Laser target 124 is mechanically coupled tofield locking target 116 and laser target 124′ is attached to astage mirror assembly 126. Thecarrier 114 is kinematically clamped to thestage mirror assembly 126 atpoints 128. Thestage mirror assembly 126, in turn, is flexure mounted to a stage base 130 atpoints 132. An x ory drive 134 is attached to an appropriate side of the stage base 130 to drive the stage, typically under computer control, in the x or y direction; and, once in place, to lock the stage in place. A mechanical centering adjustment 136 provides a fine adjustment for thefield locking target 116 to precisely place it with respect to the beam. Alternatively, instead of including mechanical centering adjustment 136, the beam could be centered by increasing the magnetic deflection of the beam, but at the expense of increasing system noise. - A pattern is formed by first positioning the stage, under
laser control field 40. The stage pauses at the selectedfield 40 and beam exposure begins and continues until the field pattern is complete. It should be noted that the mechanical centering adjustment 136 allows thefield locking target 116 to be centered both on an off center field. Mechanically centering the field locking target avoids the noise that would be otherwise induced by providing the deflection field with a sufficient range to electrically compensate for misalignment. - FIG. 3 shows a perspective view of a corner of a
field locking target 116. Preferably, thefield locking target 116 is a carbon substrate with anaperture 118 tailored to be, preferably, 0.2 mm larger than the field size in each direction, e.g., for an 0.9×0.9 mm field size the aperture is 1.1×1.1 mm. Aperture sidewalls 160 and the bottom of thefield locking target 116 are coated with gold to provide a radiation barrier. A perpendicular grid (not drawn to scale) or mesh is formed on the upper surface by crisscrossing lines 162, 164 spaced at the subfield periodicity. The lines 162, 164 are of a material selected as having a much higher atomic number (high Z) and, therefore, scatters electrons much more than the carbon background material. So, for the e-beam, selecting contrasting materials with a difference in atomic number of at least one order of magnitude provides a strong material contrast to the scanning electron beam. Thus, the carbon background material has a relatively low atomic number (low Z) providing a high contrast to the high Z gold lines of the preferred embodiment, which results in a high contrast back scatter signal. Then, back scattered electron detection is used to determine the position of the electron beam relative to marks. - Although the preferred
field locking target 116 is gold on carbon, other material combinations may be substituted, provided they do not degrade the resulting back scatter effects by reducing the signal to noise ratio or by adversely affecting the e-beam environment. Further, the preferredfield locking target 116 individual material thicknesses are not critical, but are selected to be thick enough to block stray electrons or x-rays from reaching the photosensitive pattern layer on the substrate, which would inadvertently expose resist. Thus, the carbon thickness, the bottom gold layer thickness, and the sidewall thickness must be considered together in determining an appropriate thickness for each. - At each corner of
aperture 118 are two A1 cycle subfields represented by overlapping dottedline boxes 166 and 168. Subfield 166 includes fourvertical scan rectangles 170 that identify the y position of the locking target's horizontal bars 164.Subfield 168 includes four horizontal scan rectangles that identify the x position of the locking target's vertical bars 162. Also, the average of thescan rectangles 170 positions in A1 cycle subfields 166 and 168 serve to identify the location of the field corners relative to the field locking target bars 162 and 164. This positional relationship serves as the basis of computing the field distortion. Individually, each of the foursubfield scans 170 establish the x and y components of subfield distortion. - There are several advantages to providing multiple alignment bars162 and 164. First, with a single mark as in, for example, the Tanaka et al. system, the target field must be located, precisely, with respect to the single mark. This requirement is relaxed or eliminated for a preferred embodiment system by making multiple alignment marks (bars) 162 and 164 available at each corner for registration. Including multiple bars 162 and 164 assures that a registration mark is always located to within a grid period of the edge of the
aperture 118. Second, in the preferred embodimente-beam lithography system 100, only drift errors are measured at the four corners of theaperture 118. This measurement provides sufficient information to correct X and Y translation, X and Y skew (rotation), X and Y magnification and X and Y trapezoidal distortion. Optionally, more precise corrections for even higher order terms, such as barrel-pincushion, symmetric and anti-symmetric quadratic distortion etc. may be generated by including additional target locations for measuring additional error information. Third, at each corner of the preferred embodiment, four vertical 170 and horizontal 172 scans may be made at each calibration location, with each scan located at the nominal subfield limits. These scans provide data to monitor and correct subfield level distortions. - Having described the
preferred embodiment system 100,system 100 operation can be understood with reference to the simple flow diagram FIG. 4. Unlike prior art systems, thepreferred embodiment system 100 continuously cycles through 4 distinct phases, anA1 phase 180, anA2 phase 182, a B phase 184 and a C phase 186. During theA1 phase 180, the beam is scanned across reference targets 162, 164 at the four corners ofaperture 118. In the direct write mode, the registration marks on the substrate are scanned during theA2 phase 182. The pattern is printed onto the main field of the substrate during the B phase 184. The C phase 186 is preparation for the next field, i.e., the table is moved to shift between fields and, if necessary, beam servos may be adjusted, autofocus height data may be collected, autofocus and field locking coil digital to analog converters (DACs) may be loaded and the pattern may be setup. - The preferred
stage 128 includes a permanently mounted system calibration grid. Prior to beginning normal operation, theaperture 118 must be centered, mechanically, with respect to the exposure field. To that end, the beam is scanned over it's full deflection, in focus on the target grid. Back scattered electrons from the calibration grid and thefield locking target 116 are detected and displayed. The field locking target's x/y position relative to the deflection field is delineated by the change in focus between the back scattered image of the calibration target and the field locking target. The image from thefield locking target 116 is centered in the deflection field using mechanical adjustment 136. - Then, as shown in the target mark acquisition and centering flow diagram of FIG. 5, during an
initial A1 phase 180, the beam is scanned to find bars 162 and 164 in each of the four corner A1 subfields 166 and 168. The field is centered and instep 190, the four expectedlocking target 166, 168 locations, i.e., at the corners of theaperture 118, are scanned to determine if beam to target alignment is sufficiently accurate. If the locking targets are found at their expected location, no further adjustment is necessary, and so, instep 192, target acquisition is complete. If, however, the locking targets are not found at their expected locations instep 190, then, proceeding to step 194, a check is made to determine whether amark 170 was detected in each scan. If not, then instep 196, for each lockingtarget 166, 168 where amark 170 was not detected, the mark deflection location is adjusted; and, using an appropriate search delta, such as half the length/width of a subfield, the area in the proximity of the expected mark location is searched. If a mark is initially detected in each scan in step 194 or, after finding the marks instep 196, the individual centering of eachmark 170 is checked in step 198. If the marks are not properly centered, then, instep 200, for each off center mark, the mark deflection location is adjusted by an appropriate amount to center the mark. After either finding that all of the marks are centered in step 198 or, after centering off center marks instep 200, the A1 deflection is loaded instep 202 and, the A1 cycle is begun. - This is immediately followed by an
A2 phase 182 and B cycle 184 where, as represented in the target locking and calibration flow diagram of FIG. 6. First instep 204, the system calibration grid, which is at the writing plane, is scanned and subfield center and corner coordinates are measured and adjusted to bring the deflection field and subfields into alignment with the grid. Then, the deflection control parameters are perturbed by a preselected amount and, the A1, A2 and B phases are re-executed. using the writing plane calibration grid located on the stage, the A2 and B deflections are calibrated to generate correction factors. Then, in step 206, a relationship is established between the affect of the correction factors in the writing plane and at the field locking target. - Immediately after locating the center subfield marks, during A2 and B cycle, the system calibration grid, mounted on the stage at the writing plane height is scanned. Then, the subfield center and corner coordinates are adjusted to bring the deflection field and subfields into alignment with the grid. Finally, the deflection control parameters are varied, intentionally, an identical amount during each of an A1, A2 and B cycle. Any relative displacement between the A1 and A2, B phases is measured and translated into writing plane corrections. After this initial calibration, substrate patterns are formed normally, cycling through all four phases A1, A2, B and C.
- As noted above, in the
preferred embodiment system 100, field locking is done regardless of the system's operating mode, whether in INACTIVE, STANDBY or ACTIVE mode. Further, whether in STANDBY or ACTIVE mode, the beam is scanned over the reference targets 166, 168 during everyA1 cycle 180 to maintain thetargets 166, 168 in thermal equilibrium. Without continually scanning the reference targets 166, 168, target temperature will vary unpredictably due to the irregular power input. This temperature variation will cause the reference targets 166, 168 to exhibit unacceptable thermal drift. Although the reference targets 166, 168 are scanned every cycle, processing scan results is optional in STANDBY mode, but processing is mandatory in ACTIVE mode to recalibrate the system. - Further, in a direct write operation, the A2 correction normally overrides any corrections derived from an A1 cycle, making the A1 cycle is superfluous. However, both cycles may be used to perform a field locking diagnostic in a field locking diagnostic mode. In this diagnostic mode, a registration target that is fixed onto the
XY stage 114, is moved into the deflection field and the A2 target scan is performed repeatedly on a the registration target. The A1 scans are not processed. The A2 scans are not for generating corrections; but instead, for monitoring field translation drift and distortion. Then, drifts are induced, deliberately, by perturbing the upper column, followed by A1 field locking scans. Corrections are calculated from the A1 field locking scans and applied to the tool. If the system is functioning properly, the A2 scans should not measure any drift because, the drift will have been detected in A1 cycle and applied as a correction for the A2 and B cycle. - FIG. 7 is a flow diagram of an initial
setup A1 cycle 180, wherein the locking target is not scanned. Prior to centering the deflection on thefield locking target 116, stray electrons may reach the substrate. Therefore, prior to centering the beam should not be active whenever there is a substrate on the stage. However, once the locking target has been acquired as described above with reference to FIG. 5, the beam is active in every A1 cycle. - Thus, even though the
beam 112 is inactive in the initialsetup A1 cycle 180, to maintain a stable duty cycle and to cancel out hysteresis effects, focus coils 104 a, 104 b, and deflection yokes 106, 108 and lens shifting yoke 110 are cycled. In step 210, upon entering the A1 cycle, status bits are set to indicate that the A1 cycle is in process (A1IP). Then, for the first time, instep 212 the status bits are checked to determine if the A1 cycle is in process as indicated by the A1IP bits being set. - If the A1IP bits are set then, in
step 214, the autofocus coil 104 b is set to the calibration grid focus value grid location and the registration focus coil 104 a is set to its active value. The beam feedback status (BF_STATUS) is placed in the system data path and the beam feedback state (BF_STATE) is set to zero to disable mechanical stage error corrections. During normal operation, beam feedback information is collected to provide stage positional information in the nanometer range. The collected information is used as a fine tune adjustment to compensate beam placement for stage vibrations and mechanical inaccuracies from the coarser mechanical placement resolution. Typically, stage placement mechanical accuracy is to within what is known as the beam feedback capture range. During registration, when the beam is directed to theregistration target 116, the actual stage position is irrelevant and, so, beam feedback is disabled. - Then, in
step 216 is the magnetic deflection is driven to actually deflect the beam to each of the corner subfields 166 and 168 to scan the reference target. Once scanning the reference target is complete, instep 218, the autofocus coil is set to the target value (i.e., the field address) and the focus locking coil is turned off. The BF_STATUS is placed in the system data path, the BF_STATE is set to one to enable mechanical stage error corrections and, simultaneously, the A1IP bits are turned off. - Then, in
step 212 the A1IP bits are checked again. However, the recheck instep 212 indicates that the A1IP bits are off. So, instep 220, an A2 HOLDOFF signal is checked to determine whether to continue to the A2 cycle. If, instep 220, the A2 HOLDOFF signal is active, the system enters a wait cycle in step 222. Periodically, the A2 HOLDOFF signal is rechecked instep 220 until it is found to be inactive, at which point the system proceeds to anA2 cycle 182. - FIG. 8 is a flow diagram of an A1 cycle in which the locking target is scanned, but which otherwise is substantially the same as the flow diagram in FIG. 7. Accordingly, identical steps with identical function are identified with identical numbers. In the A1 cycle of FIG. 8, when the system enters the A1 cycle, in
step 216′, the locking target is scanned, measured and the scan results are fed back. Additionally, once the preferred e-beam lithography system enters the A2 cycle from the A1 cycle of FIG. 8, the system resumes default cycling and begins processing the A1 scan feedback signal instep 224. - As described with reference to FIG. 3, the mesh on the upper surface of the
field locking target 116 provides an initial guide for the preferred system, aiding in initially locating suitable reference targets. When field locking is initially activated, the beam is sequentially deflected around the periphery of thefield aperture 118, during anA1 cycle 180, at expected mark locations. In practice, however, because of deflection field distortions and mechanical positioning errors, the marks are seldom found at these initially expected locations. So, a searching algorithm is used to locate center subfield marks at each of the respective marks and, the corresponding mark command addresses are modified accordingly. - FIG. 9 is a flow diagram of the A1 cycle wherein target locking is done as a pattern is being written but, which otherwise is substantially the same as the flow diagram in FIG. 8. Accordingly, identical steps with identical function are identified with identical numbers. So, in
step 216″, as the pattern is being scanned, the registration focus coil 104 a is checked instep 260 to determine whether correction is needed. This check is done by perturbing the field locking coil from a current value and returning for a second A1 iteration instep 216″. After the second A1 iteration, in step 262, correction terms are computed and passed forward for application to pattern position controls during theA2 cycle 182 and B cycle 184. Also, during pattern writing, instep 264, after entering theA2 cycle 182, the patterns are adjusted using update corrections generated during the A1 cycle. - While the invention has been described in terms of preferred embodiments, those skilled in the art will recognize that the invention can be practiced with modification within the spirit and scope of the appended claims.
Claims (26)
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US09/290,785 US6437347B1 (en) | 1999-04-13 | 1999-04-13 | Target locking system for electron beam lithography |
EP00105794A EP1045427B1 (en) | 1999-04-13 | 2000-03-18 | Target locking system for electron beam lithography |
AT00105794T ATE361547T1 (en) | 1999-04-13 | 2000-03-18 | ELECTRON BEAM LITHOGRAPHY SYSTEM WITH TARGET LOCKING DEVICE |
DE60034625T DE60034625T2 (en) | 1999-04-13 | 2000-03-18 | Electron beam lithography system with target locking device |
JP2000109706A JP3344713B2 (en) | 1999-04-13 | 2000-04-11 | Electron beam lithography system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US09/290,785 US6437347B1 (en) | 1999-04-13 | 1999-04-13 | Target locking system for electron beam lithography |
Publications (2)
Publication Number | Publication Date |
---|---|
US20020056813A1 true US20020056813A1 (en) | 2002-05-16 |
US6437347B1 US6437347B1 (en) | 2002-08-20 |
Family
ID=23117558
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/290,785 Expired - Fee Related US6437347B1 (en) | 1999-04-13 | 1999-04-13 | Target locking system for electron beam lithography |
Country Status (5)
Country | Link |
---|---|
US (1) | US6437347B1 (en) |
EP (1) | EP1045427B1 (en) |
JP (1) | JP3344713B2 (en) |
AT (1) | ATE361547T1 (en) |
DE (1) | DE60034625T2 (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050181584A1 (en) * | 2004-01-09 | 2005-08-18 | Applied Materials, Inc. | Ion implantation |
US20060266953A1 (en) * | 2005-05-27 | 2006-11-30 | Uwe Kramer | Method and system for determining a positioning error of an electron beam of a scanning electron microscope |
WO2012110465A3 (en) * | 2011-02-16 | 2012-11-01 | Mapper Lithography Ip B.V. | System for magnetic shielding |
US8389962B2 (en) * | 2011-05-31 | 2013-03-05 | Applied Materials Israel, Ltd. | System and method for compensating for magnetic noise |
US20140127914A1 (en) * | 2012-11-05 | 2014-05-08 | Nuflare Technology, Inc. | Charged particle beam writing apparatus and charged particle beam writing method |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE10011202A1 (en) * | 2000-03-08 | 2001-09-13 | Leica Microsys Lithography Ltd | Method for aligning an electron beam to a target position on a substrate surface |
US6661015B2 (en) * | 2000-09-15 | 2003-12-09 | Ims-Ionen Mikrofabrikations Systeme Gmbh | Pattern lock system |
US7160475B2 (en) * | 2002-11-21 | 2007-01-09 | Fei Company | Fabrication of three dimensional structures |
US6908255B2 (en) * | 2003-06-25 | 2005-06-21 | International Business Machines Corporation | Remote clamping mechanism via vacuum feedthrough |
JP3889743B2 (en) | 2003-12-05 | 2007-03-07 | 株式会社東芝 | Charged beam drawing method and drawing apparatus |
US7477772B2 (en) * | 2005-05-31 | 2009-01-13 | Asml Netherlands B.V. | Lithographic apparatus and device manufacturing method utilizing 2D run length encoding for image data compression |
JP4520426B2 (en) | 2005-07-04 | 2010-08-04 | 株式会社ニューフレアテクノロジー | Electron beam drift correction method and electron beam writing method |
US7456414B2 (en) * | 2005-09-28 | 2008-11-25 | Applied Materials, Inc. | Beam re-registration system and method |
JP4980628B2 (en) * | 2006-03-06 | 2012-07-18 | 日本電子株式会社 | Hysteresis elimination method using automatic transition buttons |
WO2008030929A1 (en) * | 2006-09-06 | 2008-03-13 | Massachusetts Institute Of Technology | Nanometer-level mix-and-match scanning tip and electron beam lithography using global backside position reference marks |
JP2009218474A (en) * | 2008-03-12 | 2009-09-24 | Jeol Ltd | Method for suppressing beam position drift, method for suppressing beam size drift, and charged particle beam plotting device |
DE102019200696B4 (en) * | 2019-01-21 | 2022-02-10 | Carl Zeiss Smt Gmbh | Apparatus, method and computer program for determining a position of an element on a photolithographic mask |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3874916A (en) * | 1972-06-23 | 1975-04-01 | Radiant Energy Systems | Mask alignment system for electron beam pattern generator |
JPS5398781A (en) | 1976-11-25 | 1978-08-29 | Jeol Ltd | Electron ray exposure unit |
JPS5530811A (en) * | 1978-08-25 | 1980-03-04 | Chiyou Lsi Gijutsu Kenkyu Kumiai | Single field alignment method |
JPS57183034A (en) | 1981-05-07 | 1982-11-11 | Toshiba Corp | Electron bean transfer device |
JPS62287556A (en) * | 1986-06-05 | 1987-12-14 | Toshiba Corp | Axis-aligning method for charged beam |
AT393925B (en) | 1987-06-02 | 1992-01-10 | Ims Ionen Mikrofab Syst | ARRANGEMENT FOR IMPLEMENTING A METHOD FOR POSITIONING THE IMAGE OF THE STRUCTURE ON A MASK TO A SUBSTRATE, AND METHOD FOR ALIGNING MARKERS ARRANGED ON A MASK ON MARKINGS ARRANGED ON A CARRIER |
US4818885A (en) | 1987-06-30 | 1989-04-04 | International Business Machines Corporation | Electron beam writing method and system using large range deflection in combination with a continuously moving table |
JPH0831405B2 (en) * | 1988-06-01 | 1996-03-27 | イーエムエス イオーネン ミクロファブ リカチオンス ジステーメ ゲゼルシャフト ミト ベシュレンクテル ハフツング | Ion projection lithographic apparatus and method |
JPH0722010B2 (en) | 1989-09-28 | 1995-03-08 | 株式会社日立製作所 | Electron beam writer |
US5043586A (en) * | 1990-10-25 | 1991-08-27 | International Business Machines Corporation | Planarized, reusable calibration grids |
JP3148353B2 (en) | 1991-05-30 | 2001-03-19 | ケーエルエー・インストルメンツ・コーポレーション | Electron beam inspection method and system |
JP2788139B2 (en) | 1991-09-25 | 1998-08-20 | 株式会社日立製作所 | Electron beam drawing equipment |
US5424548A (en) | 1993-09-21 | 1995-06-13 | International Business Machines Corp. | Pattern specific calibration for E-beam lithography |
US5552611A (en) | 1995-06-06 | 1996-09-03 | International Business Machines | Pseudo-random registration masks for projection lithography tool |
JPH09246135A (en) * | 1996-03-04 | 1997-09-19 | Hitachi Ltd | Electric charge particle beam drawing device |
US5793048A (en) * | 1996-12-18 | 1998-08-11 | International Business Machines Corporation | Curvilinear variable axis lens correction with shifted dipoles |
US6218671B1 (en) * | 1998-08-31 | 2001-04-17 | Nikon Corporation | On-line dynamic corrections adjustment method |
-
1999
- 1999-04-13 US US09/290,785 patent/US6437347B1/en not_active Expired - Fee Related
-
2000
- 2000-03-18 DE DE60034625T patent/DE60034625T2/en not_active Expired - Lifetime
- 2000-03-18 EP EP00105794A patent/EP1045427B1/en not_active Expired - Lifetime
- 2000-03-18 AT AT00105794T patent/ATE361547T1/en not_active IP Right Cessation
- 2000-04-11 JP JP2000109706A patent/JP3344713B2/en not_active Expired - Fee Related
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050181584A1 (en) * | 2004-01-09 | 2005-08-18 | Applied Materials, Inc. | Ion implantation |
US20060266953A1 (en) * | 2005-05-27 | 2006-11-30 | Uwe Kramer | Method and system for determining a positioning error of an electron beam of a scanning electron microscope |
WO2012110465A3 (en) * | 2011-02-16 | 2012-11-01 | Mapper Lithography Ip B.V. | System for magnetic shielding |
US8884253B2 (en) | 2011-02-16 | 2014-11-11 | Mapper Lithography Ip B.V. | System for magnetic shielding |
US8389962B2 (en) * | 2011-05-31 | 2013-03-05 | Applied Materials Israel, Ltd. | System and method for compensating for magnetic noise |
US20140127914A1 (en) * | 2012-11-05 | 2014-05-08 | Nuflare Technology, Inc. | Charged particle beam writing apparatus and charged particle beam writing method |
US9117632B2 (en) * | 2012-11-05 | 2015-08-25 | Nuflare Technology, Inc. | Charged particle beam writing apparatus and charged particle beam writing method |
Also Published As
Publication number | Publication date |
---|---|
US6437347B1 (en) | 2002-08-20 |
DE60034625T2 (en) | 2008-01-03 |
ATE361547T1 (en) | 2007-05-15 |
EP1045427A2 (en) | 2000-10-18 |
DE60034625D1 (en) | 2007-06-14 |
EP1045427B1 (en) | 2007-05-02 |
JP3344713B2 (en) | 2002-11-18 |
EP1045427A3 (en) | 2000-12-27 |
JP2000349021A (en) | 2000-12-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6437347B1 (en) | Target locking system for electron beam lithography | |
US4818885A (en) | Electron beam writing method and system using large range deflection in combination with a continuously moving table | |
US6222195B1 (en) | Charged-particle-beam exposure device and charged-particle-beam exposure method | |
US6573516B2 (en) | Electron-beam lithography method and electron-beam lithography system | |
Maile et al. | Sub-10 nm linewidth and overlay performance achieved with a fine-tuned EBPG-5000 TFE electron beam lithography system | |
US4119854A (en) | Electron beam exposure system | |
JP4454706B2 (en) | Electron beam exposure method and apparatus, and device manufacturing method | |
US6392243B1 (en) | Electron beam exposure apparatus and device manufacturing method | |
US5798528A (en) | Correction of pattern dependent position errors in electron beam lithography | |
US10867774B2 (en) | Multi charged particle beam writing apparatus and multi charged particle beam writing method | |
US4338508A (en) | Inscribing apparatus and methods | |
KR100327010B1 (en) | Electron beam exposure apparatus and exposure method | |
JP2960746B2 (en) | Beam irradiation method, electron beam drawing method, beam irradiation apparatus, and electron beam drawing apparatus | |
JP2002353112A (en) | Close electron beam projection aligner, and methods for measuring and calibrating inclination of electron beam in the close electron beam projection aligner | |
EP0431864B1 (en) | Method of detecting and adjusting exposure conditions of charged particle exposure system and such a system | |
Tennant et al. | Evaluation of a 100 kV thermal field emission electron-beam nanolithography system | |
US4835392A (en) | Ion-projection apparatus | |
US4424450A (en) | Hybrid moving stage and rastered electron beam lithography system employing approximate correction circuit | |
US7053388B2 (en) | Dual-mode electron beam lithography machine | |
JPH0922859A (en) | Compensation method for sample surface height in electron beam exposure process | |
US6818906B1 (en) | Electron beam position reference system | |
US6908255B2 (en) | Remote clamping mechanism via vacuum feedthrough | |
JPH02309627A (en) | Electron beam exposure device | |
Russell et al. | Development of SEM-based dedicated IC metrology system | |
JP2002043207A (en) | System and method for electron beam proximity exposure |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW Y Free format text: INVALID ASSIGNMENT;ASSIGNORS:HARTLEY, JOHN G.;KENDALL, RODNEY A.;REEL/FRAME:009941/0345 Effective date: 19990311 Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW Y Free format text: INVALID ASSIGNMENT;ASSIGNORS:HARTLEY, JOHN G.;KENDALL, RODNEY A.;REEL/FRAME:009908/0781 Effective date: 19990311 |
|
FPAY | Fee payment |
Year of fee payment: 4 |
|
FPAY | Fee payment |
Year of fee payment: 8 |
|
REMI | Maintenance fee reminder mailed | ||
LAPS | Lapse for failure to pay maintenance fees | ||
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20140820 |
|
AS | Assignment |
Owner name: GLOBALFOUNDRIES U.S. 2 LLC, NEW YORK Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:INTERNATIONAL BUSINESS MACHINES CORPORATION;REEL/FRAME:036550/0001 Effective date: 20150629 |
|
AS | Assignment |
Owner name: GLOBALFOUNDRIES INC., CAYMAN ISLANDS Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:GLOBALFOUNDRIES U.S. 2 LLC;GLOBALFOUNDRIES U.S. INC.;REEL/FRAME:036779/0001 Effective date: 20150910 |