Semiconductor Equipment and Materials International
3081 Zanker Road
San Jose, CA95134-2127
Phone: 408.943.6900, Fax: 408.943.7943
hb khghgh1000A5331
Background Statement for SEMI Draft Document 5331A
NEW STANDARD: TEST METHOD FOR IN-LINE MEASUREMENT OF SAW MARKS ON PV SILICON WAFERS BY A LIGHT SECTIONING TECHNIQUE USING MULTIPLE LINE SEGMENTS
Notice: This background statement is not part of the balloted item. It is provided solely to assist the recipient in reaching an informed decision based on the rationale of the activity that preceded the creation of this Document.
Notice: Recipients of this Document are invited to submit, with their comments, notification of any relevant patented technology or copyrighted items of which they are aware and to provide supporting documentation. In this context, “patented technology” is defined as technology for which a patent has issued or has been applied for. In the latter case, only publicly available information on the contents of the patent application is to be provided.
Today the dominant technology for slicing bricks to produce silicon wafers for photovoltaic solar cells is multiple wire sawing. With this technology it is possible to reliably produce wafers with thicknesses between 100 – 300 µm. Wire sawing is a mechanical process where steel wires with a diameter between 100 and 160 um are used to transport slurry into the cutting zone of the silicon brick or, in the case of diamond wire saws, cut the silicon directly using only a lubricant but not slurry. The slurry typically contains SiC particles, which remove the silicon material in an abrasive process. In case of a diamond wire small diamond pieces are attached to the steel wire. In both cases the abrasive process generates grooves and steps along the wire direction in the surface of the Si wafers. These grooves and steps, called saw marks, negatively impact the quality of the wafers and subsequently the quality of the solar cell made from these wafers. The maximum depth/height of saw marks is part of the wafer specifications. Therefore these saw marks need to be reproducibly characterized regarding their depth/height by an in-line high throughput measurement method. This document proposes such a measurement method.
The corresponding SNARF was approved by the PV Committee in its meeting in Dresden on October 11, 2011. The draft document was approved for yellow letter ballot in cycle 1 of 2012 by the PV Committee in its meeting in Dresden on October 11 to be adjudicated in Berlin in March 2012where it failed. Immediate re-ballot of the document in cycle 3 2012 was approved to be adjudicated in the PV Materials Committee meetings in Munich in June 2012 in conjunction with Intersolar Europe 2012.
Check under Calendar of Events for the latest update.
Review and Adjudication Information
Task Force Review / Committee AdjudicationGroup: / PV Silicon Materials TF / Europe PV Materials Committee
Date: / Wednesday, June 13, 2012 / Wednesday, June 13, 2012
Time & Timezone: / 11:00 to 13:00 CET / 16:00 to 18:00 CET
Location: / International Congress Centre Munich (ICM) / International Congress Centre Munich (ICM)
City, State/Country: / Munich, Germany / Munich, Germany
Leader(s): / Peter Wagner / Peter Wagner
Hubert Aulich (PV Crystalox)
Standards Staff: / Kevin Nguyen (SEMI NA)
408.943.7997
/ Kevin Nguyen (SEMI NA)
408.943.7997
This meeting’s details are subject to change, and additional review sessions may be scheduled if necessary. Contact the task force leaders or Standards staff for confirmation.
SEMI Draft Document 5331A
NEW STANDARD: TEST METHOD FOR IN-LINE MEASUREMENT OF SAW MARKS ON PV SILICON WAFERS BYALIGHT SECTIONING TECHNIQUE USING MULTIPLE LINE SEGMENTS
1 Purpose
1.1 Silicon (Si) wafers for PV applications cut from a Si ingot or Si brick by multiple-wire sawing contain artifacts characteristic for this cutting process, so called saw marks.
1.2 The saw marks consist of topographic features, such as grooves, steps (see R2), on or in the Si wafer surface and that extend along the wire direction.
1.3 Saw marks may significantly impact the quality of wafers. They interfere with printing the contact fingers on solar cells. Extreme saw mark dimensions may interrupt the contact fingers or create too wide fingers.
1.4 Saw marks are frequently specified for Si wafers for solar cells with respect to their maximum peak-to-valley within a finite distance, or window.
1.5 Standardized test methods providing reproducible values for saw marks are required to specify this aspect of wafer quality.
1.6 Process and quality control during manufacturing of wafers requires continuous monitoring of saw marks with a non-contact method that supports high throughput.
2 Scope
2.1 This test method determines maximum peak-to-valley of saw marks of multi or single crystal Si wafers that typically run across the entire wafer surface and along the wire direction.
2.2 It describes an in-line, non-contacting and non-destructive method that determines the height changes of steps and grooves of clean, dry as-cut silicon wafers supported by two belts that move the test specimen through the measurement equipment.
2.3 This test method covers square and pseudo-square PV Si wafers, with a nominal edge length ≥ 125 mm and a nominal thickness ≥ 100 µm. It applies to both single-crystalline and multi-crystalline Si wafers.
2.4 The test method is intended for in-line high throughput measurements. Therefore it is mandatory to operate the measurement system under statistical process control (SPC, e.g. ISO 11462) in order to obtain reliable, repeatable and reproducible measurement data.
2.5 The test method is based ona light sectioning technique (see R1) where patterns ofline segmentsor spots of light are projected onto a wafer surface and the saw marks are oriented perpendicular to the direction of wafer transport.
2.6 Other measurement techniques may also provide similar information about saw marks of a wafer as compared to this test method, but they are not the subject of this test method.
NOTICE:SEMI Standards and Safety Guidelines do not purport to address all safety issues associated with their use. It is the responsibility of the users of the Documents to establish appropriate safety and health practices, and determine the applicability of regulatory or other limitations prior to use.
3 Limitations
3.1 Loading of wafers onto the measurement system must ensure that the saw marks are transverse, not parallel, to the direction of travel through the measurement system. Measurements made on wafers loaded with saw marks parallel to the direction of travel will not be suitable for specifying the wafers’ saw mark topography.
3.2 Wafer surface height vibration caused by transport during measurement may adversely impact measurements.
3.3 The transverse locations where the measurements are taken on the wafer surface are critical. When comparing measurement results obtained by different equipment or by subsequent measurements on the same equipment it should be verified that the measurements are taken at the same transverse locations on the wafer surface.
3.4 Measurement of saw mark height change depends on the physical setup of the measurement equipment, its opto-electronic noise level, and the calibration of measured signal to height.
3.5 Variations of the projected light line image due to optical imperfections, including those of the wafer surface (such as grain boundaries or stains), may affect the measurement result.
3.6 Wafer surface waviness may impact the saw mark measurement precision.
4 Referenced Standards and Document
4.1 SEMIStandardsand Safety Guidelines
SEMI E89 –– Guide for Measurement System Analysis (MSA)
SEMI M59 –– Terminology for Silicon Technology
SEMI MF1569–– Guide for Generation of Consensus Reference Materials for Semiconductor Technology
4.2 ISO Standards[1]
ISO 11462-1 — Guidelines for implementation of statistical process control (SPC) – Part 1: Elements of SPC
ISO 11462-2 –– Guidelines for implementation of statistical process control (SPC) – Part 2: Catalogue of tools and techniques
NOTICE: Unless otherwise indicated, all documents cited shall be the latest published versions.
5 Terminology
NOTE 1: Refer to SEMI’s Compilation of Terms (COT) for a list of the most current terms and their definitions.
5.1 Terms and acronyms relating to silicon and other semiconductor technology are defined in SEMI M59.
5.2 Other Abbreviations and Acronyms
5.2.1 AOI–– angle of incidence
5.2.2 COB –– center of brightness
5.3 Other Terms Used in this Standard
5.3.1 angle of incidence, of projected light line –– the angle between the line of sight of the projector and the surface normal (of the wafer).
5.3.2 center line, CL — an imaginary straight line bisecting a square or pseudo-square wafer surface. It is equidistant from opposing edges of the wafer surface.
5.3.3 center-of-brightness (COB) line –– a line fitted through a multiple pixel image representing the weighted average of the pixel values transverse to the projected line direction.
5.3.4 pixel width and height, effective –– the distances on the wafer surface imaged by one picture element, or pixel, of the image sensor in perpendicular directions.
5.3.5 saw mark –– a topographic step or groove along the sawing wire direction on the surface of a wafer, generated by the wire of the multiple wire saw.
5.3.6 scan line –– an imaginary line parallel to a wafer edge and parallel to the wafer transport direction along which measurements are performed.
5.3.7 waviness –– a surface height variation with spatial wavelengths of typically a few millimeters.
6 Summary of Test Method
6.1 Four patternsof light line segments are projected on the wafer surface at an AOI as the wafer, resting on belts, is moved along in the x- direction, the direction of belt travel. These segments provide illumination for image acquisition along four scan lines.
6.1.1 The four scan lines are front-left (FL), back-left (BL), front-right (FR) and back-right (BR).
6.1.2 Each of these four projected light patterns are used identically to sample wafer topography at discrete transverse locations y, the direction perpendicular to the x-direction, ultimately producing N times four surface height variation profiles: H(x,yFL), H(x,yBL), H(x,yFR) and H(x,yBR), where N is the number of line segments in a pattern.
6.2 When loaded onto the conveyor belt system the wafer is oriented such that:
6.2.1 The wafer’s saw marks are perpendicular to the direction of belt travel.
6.2.2 Each projected light pattern illuminates the wafer surface at a distance from the nearest wafer edge.
6.3 Each projected pattern’s scattered light from the wafer surface is imaged by a camera system viewing the wafer surface at normal incidence. Alternatively the positions of projector and camera may be interchanged.
6.4 Overlapping images of the projected light pattern scattered from the wafer surface are acquired during wafer transit.
6.5 Height deviations from planar, along each scan line of the wafer surface,due to steps, grooves or waviness!! result in small transverse translations of the projected light pattern on the wafer surface that are recorded by the image of the projected light line segment (see R1).
6.6 The images are processed and evaluated by an algorithm according to § 14 .
6.6.1 A calibration factor for each of the four projected light patterns image acquisition systems is determined using reference material with known step heights.
6.6.1.1 Small variations in the optical systems, such as the AOI of the projection and the actual viewing angle of the camera being slightly off-normal, make a purely geometrical calculation based on AOI less precise than is possible through calibration.
6.6.1.2 Calibration factors therefore are used to more precisely quantify surface height variation and reduce system-to-system measurement variation.
7 Interferences
7.1 Vibrations of the test specimen relative to the probe-measuring axis may produce errors. Internal system monitoring may also be used to detect non-repetitive and repetitive system mechanical translations.
7.2 Substantial variations in wafer surface properties may negatively impact the measurement result or possibly prevent measurement of some wafers.
7.2.1 Wafer surface roughness variation, along a scan line within a single wafer or between wafers, may impose signal to noise constraints on the optical system when imaging the light scatter from a nominally diffuse scattering surface.
7.2.2 Surface waviness variations may also negatively impact the measurement result.
7.3 Variation of pixel height and width within a camera system may not be corrected through the calibration.
7.4 Mechanical variations in equipment adjustment may introduce errors.
7.5 Temperature gradients across the measurement set-up may affect the measurement performance.
7.6 Large surface defects, such as pits or chips, or contamination, such as slurry residue or particles, also may impact the measurement result.
8 Apparatus (see Figure 1)
8.1 Wafer transport device –– consisting of two belts, which transport the wafer continuously through the measurement apparatus without obstructing the projected light line or the line of sight of the cameras. The material of the belts shall not leave traces or residue on the wafer surface.
8.2 Projector –– projecting one of the following patterns at an AOI onto the wafer surface at a distance from the wafer edge during transport.
8.2.1 Pattern A:Nclosely adjacent and equidistant parallel line segments that are oriented nominally parallel to the direction of belt travel. Typically N = 1 or N = 2, but more are possible. The distance between two adjacent line segments is denoted as .
8.2.2 Pattern B:Nadjacent light spots (to be understood as limit of short line segments). Typically N = 1 or N = 2, but more are possible. The distance between two adjacent spots is denoted as .
8.2.3 The distance is measured from the wafer edges to the outmost part of the pattern.
8.3 In total four projectors are used, two above the wafer and two below the wafer.
8.4 Sensor –– an imaging digital camera systemwith an effective pixel width sw and an effective pixel height sh. It is set-up with its line of sight normal to the wafer surface, recording images of the projected light pattern on the wafer surface while in transit. In total four cameras are used, two above the wafer and two below the wafer.
8.5 Computer –– for controlling the measurement system and equipped with software for synchronouslyacquiring and processing the camera images according to § 14.
8.6 The positions of projectors and corresponding cameras may be interchanged, meaning that the projector AOI is normal and the camera views the wafer surface at an oblique angle.
9 Test Specimen
9.1 Clean, dry Si wafers with an as-cut surface condition.
10 Safety Precautions
10.1 The entire equipment has to be placed in a closed housing and has to be secured with a safety lock that stops the belts and safely switches the tool off when the housing is opened.
10.2 If required by local, national or international safety requirements, eye protection goggles have to be used by operators and maintenance personnel.
11 Preparation of Apparatus
11.1 The suitability of the equipment is determined by performing statistically based instrument repeatability and reproducibility study to ascertain whether the equipment is operating within the manufacture’s stated specification, e.g. according to SEMI E89.
11.2 Adjust the sensitivity of the camera and the light intensity of projected light pattern so that the intensity profile of the image of the light pattern in the camera extends over at least three digital intensity levels above the noise level of the camera while avoiding saturation and image bloom.
11.3 Define the control limits for SPC for the measurement equipment with a set of selected wafers.
NOTE 2:As this test method is intended for a high throughput, high volume measurement the equipment cannot be calibrated for measuring each individual wafer. Therefore careful SPC has to be performed.
12 Calibration and Standardization
12.1 Each projection and camera imaging system is calibrated by using one or more reference wafers.
12.2 The reference wafers contain topographic surface features, called reference marks, with known step heights or groove depths.
12.2.1 Care has to be taken selecting a tool for establishing the surface feature heights or depths of the reference wafers. This tool has to be calibrated with traceable reference materials and be non-destructive, typically non-contact.
12.3 Measure the reference wafer(s) according to the procedure given in § 13.
12.4 Determine for each reference mark kthe calibration factor fkby dividing the known step height/groove depth by the corresponding tk.
12.5 Determine the mean calibration factor f for that imaging system for several surface featuresk from one or several wafers.
13 Procedure
13.1 Adjust the equipment and calibrate it according to the instructions of the supplier.
13.2 Determine the calibration factors f for eachprojection and camera imaging system.
13.3 Verify the equipment is within SPC limits.
13.4 Measure the wafer.
13.4.1 Place a wafer horizontally on the transport belts.
13.4.1.1 Align the wafer so that two of its edges are parallel to, and the saw marks are oriented perpendicular to, the direction of transport so that is within the range of 5 mm to 1/6 of wafer side length with a tolerance of ± 2 mm.
13.4.2 Move the wafer through the measurement set-up.
13.4.3 Scan the entire length of the wafer.
13.4.4 For pattern A record successive, overlapping images Hj(x, yi,)of each projected pattern during the scan so that 100% wafer coverage across the four scan lines is ensured, where i denotes the scan line (FL, FR, BL and BR),j denotes the image recorded.
13.4.5 For pattern B, combine successive images of the spot pattern so that images of line segments Hj(x, yi,) are obtainedthat cover 100% of the wafer across the four scan lines.
13.4.6 Process each Hj(x,yi)from all fourprojection and camera imaging systems according to § 14.
13.4.7 Report the heights/depths of all saw marks identified as well as the maximum saw mark height/depth of the wafer according to § 15.
13.5 Repeat with the next wafer.
14 Calculations
14.1 The following calculations are performed automatically within the instrument. An outline of the calculation structure is provided here to indicate the nature of the procedure.
14.2 The projected light pattern and its image recorded by the digital camera have a finite width extending over several pixels. Therefore the image of the light pattern consists of an array of pixels that has to be processed to generate a line representing the surface profile that can be evaluated with respect to saw marks.