• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    This invention relates to instrumentation technology. The aim of the invention is to provide control of geometrical parameters of microstructures. - The device circuit contains a central device designed to analyze a signal with Y2 fB (L
    公开号:SU1310640A1
    申请号:SU827772613
    申请日:1982-09-01
    公开日:1987-05-15
    发明作者:Плонтке Райнер；Лелле Вернер
    申请人:Феб Карл-Цейсс-Йена (Инопредприятие)；
    IPC主号:
    专利说明:

    a receiver, e.g. a CRT, and acting on the deflecting systems of a CRT to scan the image of edge 3 of structure 1. First, n measurements are taken at one-line decomposition of the same element of structure 1 to reduce the signal-to-noise ratio. To account for the roughness of the edge 3 of the structure re1
    The invention relates to a method and scheme for its implementation, a user to control the relative position, width or distance, as well as the quality of microlitographically made structures, by automatically determining the characteristic of the edges of the structures.
    The invention can be applied in microlithography in cathode-ray devices used to measure and control such structure topologies. Structures can be used as photoresist masks, in metallized optical parts made of glass or as semiconductor disks.
    It is known to use Nikon-Japan brochures related to the Micro-Pattern Inspection Station or the Lasser Interferometric x-y Measuring Machine for optical measurement devices.
    However, these methods are possible if the minimum dimensions or distances of the elements are not less than a few micrometers. At the same time, an intense light beam with a rectangular cross section, the narrow side of which has a size of up to 1 µm, and the length of the side up to several thousand h of micrometres, is directed along the edge of the structure, and the transmitted or reflected light signal is processed through photo detectors.
    The disadvantage is that the measurements are associated with the requirements regarding the geometric shapes of the structures being measured and thereby significantly limit the diversity.
    The beam is placed along the edge several times by uY, each time measurements are made, and then the value of each coordinate is averaged over all measurements. The partial result of dividing the steepness of the indicated curves by each other and taking into account the constant established by the test object is given a measure of roughness Kra 2 sec. and 3 z. p. f-ly, 8-yl.
    For the control of structures, the use of electron-beam devices (DD-WP 124 091, CL H 01 J 37/26, 1979) is also known, in which.
    According to the principle of a scanning electron microscope, a thin electronic probe is guided along the object being measured. An image of the object is presented on the screen in synchronization with it, which is compared with the designed scale grid. The measurement is carried out using a scale grid or by a certain displacement of an object using a reference elevation projected onto a luminescent screen, p.
    The disadvantage of this method is that the measurement performance is too low and the measurement accuracy is subjectively influenced.
    The purpose of the invention is the measurement and control of microlitographically made structures of various sizes and
    forms to the widths and distances of structures in the submicron limit. The method must be high-performance, flow automatically and without subjective influences.
    The invention is based on the task of creating such a method with a scheme for its implementation, which allows, on the one hand, to measure the widths of structures in the submicron limit, on the other hand, to base the processing on a sufficiently large number of samples for accuracy requirements to get a conclusion about the quality of the edges of the structures. With the way
    edge detection
    31310640
    structures in measuring devices for
    Determining the position, size, or distance (as well as quality) of predominantly microlitographically fabricated topologies of the structures arising from the electron-beam rasterization, signals are used to record the edges of the structures, and the electron beam is guided step by step along the measured edge.
    The method is carried out as follows.
    In the perpendicular size of the edge, the sweep direction along the line in constant given steps points out the structure topology by moving the electron beam, and the signal received at each point is digitized and stored in the digital computer memory; Since the diameter of the electronic probe is small, the edge property is recorded only in this small area and there is no requirement for a linear edge size.
    Step a is repeated many times (n times), and the coordinates of the sweep in the direction of the edge are not changed in order to improve the signal-to-noise ratio of the signal received in a. The obtained signal values in steps a and b are related to the same places along the sweep direction, summed into a single signal curve. For n different places at specified distances a along the edge of the structure perpendicular to the sweep direction, step a is repeated. edge roughness measurement signal. The sweep signals obtained in step-by-step r are digitized and summed for all the sweep points of the same coordinate in the direction of the sweep. The signal obtained in this way is represented in the input of the spatial average and time average through the edge region n-a. Distance a is chosen so that the sweep movement is adapted to the size of the edge, i.e. pa i, where i is the size of the edge. When signals are formed in a step d, in the case of a finite roughness of the edge, signals are displaced towards each other in the direction of sweep of the edge roughness. The resulting edge signal is generally flatter than the one at which a 0.
    The steepness of the curve of the signal obtained by. in, i.e. for aO, is divided by the steepness of the signal curve, i.e. for a i Oh, and, this
    tt ir
    d,
    Noah
    the quotient is linked to the constant established from the test objects. This results in a value that is a measure for the roughness of the edge of the structure being measured.
    To implement the method according to the invention, a circuit in an electron-beam device, in which
    The electron beam falls on the structure being measured and, depending on this structure, the signal formed from the outgoing electrons is supplied to the receiver. The scheme is distinguished by the following features.
    The receiver signal, which with respect to magnitude can be very different and contains an interfering base level, is supplied to the normalizing link. The normalized signal in the analog-to-digital converter, in which data from the central digital computer is input through a sampling circuit, is digitized and fed to the central device.
    The movement of the electron beam for the guides of the x- and y-coordinates is provided by one counter.
    beef and end converters
    amplifiers position of the electron beam through the deflecting system. Information on the starting or ending values in the direction of the sweep is fed to one input of each position counter. Upon reaching the final value of the position counter, a message is generated to the central device. At the same time, the position counter returns to the initial value and a clock pulse is output to the position counter of the second coordinate direction. Due to this, a new sweep on a line parallel to the previous line is possible, with both lines having a given distance a. An electronic switching device is located between the two position counters for selecting sweep sweeps in the x- or y-coordinate directions. At the same time, the digitizing signals input pulses serve as clock pulses;
    J5
    A signal is supplied to an analog-to-digital converter.
    The input pulse receives the digitized signal into the central device. The falling edge of this pulse increases the position counter, and the electron beam takes the following position. After the delay time, the analog-to-digital converter is started up and a new value of the signal is digitized.
    To smooth out the noisy signals, there is a low-pass filter upstream of the normalizing element, the time constant of which is matched with the delay time of the analog-to-digital converter.
    To improve the edge signal with respect to processing capability, it is advantageous if the signals of different receivers are additively or subtractively mixed.
    To improve the positioning accuracy of the electron beam, the following features of the circuit are used.25
    The adjustment of the step size to the measured structure and to the sensitivity of the deflection of the deflecting system is carried out through a cascade of a pulse multiplier and a digital-mounted analogue damping link.
    The oblique position of the object relative to the direction of the deflecting systems is taken into account by the units of the rotational correction. A position counter with a step value setting and a fast digital-analog converter is used
    is used only for the intrinsic 40 raster lead symmetrically with respect to the estimated position of the edge, and the main deflection of the beam relative to the estimated position of the edge is established by the second digital-to-analog converter before the start of a series of measurements. This digital-to-analog converter is more accurate and slower, its output signal is additively added to the rasterized deflection voltage.
    Figure 1 presents the characteristic edges of the structures: figure 2 - scheme of registration of the edges of the structures; in FIG. 3, an edge signal enhancement circuit; Fig. 4 is a diagram of an improvement in electron beam positioning. one . .
    The solution to this registration scheme
    edges of the structures comes from the fact that
    45
    50
    55

    0
    five
    0

    0
    five
    0
    five
    the electron beam travels and the signal received by the receiver is digitized. In figure 1 and in cross section shows the structure 1 on the substrate 2.
    Figure 1b shows the edge of the structure 3, which should be measured. For the position of the edges of the structure, the y-coordinate direction was chosen. Perpendicular to this, the sweep direction is located, in which the edge of structure 3 is developed in predetermined steps, for example, 5 nm. The signal from the back-scattered electrons, the secondary electrons, or the displacements of both electrons obtained at each scanning point is digitized and accumulated. Since the diameter of the electron probe is small, i.e. usually 10 nm, the edge is recorded only in this small area, without requiring a linear edge size. The arrival time at a point is, for example, 4 µs. Due to the small number of registered electrons, signal 4 (Fig. 1c) has a small difference: signal-to-noise ratio. In order to improve this ratio (with a y post.), The scan is repeated n times (e.g., p 50) and the digitized values of the signal are summed up into a single curve of signal 5 (Fig. 1 d). Registration kra continues p 2 s
     about; 1 sec.
     In this type of signal formation, the roughness of the edges is not taken into account. Therefore, after the sweep ... the sweep at Y (in each case, after the electron beam moves along the y axis direction, further measurements are carried out. The resulting signals are superimposed into the curve of signals 6 (Fig. 1e), which is the spatial average and average over time through the region of the edge of the n-d, and the doo is chosen so that (i is the size of the edge.) The slope of the signal 5 is divided by the slope of the signal-a 6, and this quotient is related to a constant that is determined from the test -objects. This results in a value that is a measure oh for edge roughness.
    Fig. 2 shows a scheme for carrying out the method according to which the borderline 7 in normalizing link 8 is normalized by additive level displacement and multiplicative change in gain. To digitize this signal serves as an analog-to-digital converter 9, into which through
    7 131064
    the gating circuit 10 inputs data from the central computer 11. The delay link 21 receives an input pulse E1 from the central device 11 and gives the initial signal to the analog j digital transducer 9. The digital movement in the coordinate direction is carried out by counters 12 and 13 positions, which, through digital to analogue converters 14 and 15, are terminal amplifiers 16 and 17 affect the deflection system 18 for the direction of the x axis and the deflection system 19 for the direction of the y axis.
    The initial and final values of A and
    J5
    Similarly, the step size in the x-axis direction is performed by a cascade of multiplier 32 pulses downstream of the counter 12 of the position, and the analog damping of the raster sweep by the central device 11 are set for the counters 12 and 13 of the position, they describe the estimated position of the edge. Electronic switching device 20 link 24, included after the digit-20 of choice, you can create x-or u-analogue converter 14. line. The clock pulses of the inclined position of the measured structure are the pulses of the input of the E1 structure relative to the direction
    , digitizing the signals. When reaching the deviation of your systems 18 and 19, it takes into account the final value of the line, which is located 25 blocks 36 and 37 of the rotating correction in the x-axis direction, by a counter. At the same time, the position set part 12 is given by the signal M to the central device 11i. At the same time, this position counter returns to the initial value, and is removed from the clock pulse of the position counter 13. Due to this, a new sweep is performed on a line moved parallel to the first
    thirty
    the deflecting voltage adds etc to one coordinate additively and to the other coordinate substrativo.
    A significant increase in the accuracy of positioning the electron beam is achieved by the fact that the counter 12 is positioned with a fast digital-to-analog converter 14 and the damping 35 link 34 is used only for its own sweep, which in the example is carried out in the direction of the X-axis and symmetrically relative to the position of the edge. The main deviation of the electric string is the yu distance.
    FIG. 3 shows an edge signal enhancement circuit in which two receivers 22 and 23 dp of back-scatter electrodes are provided and
    one receiver 24 for secondary electron beam relative to this
    thrones. The output signals of the receivers 22 and 23 in the summing link 25 are additively or substratumously mixed, the resulting signal, as well as the input signal of the receiver 24. for smoothing through the low-pass filters 26 and 27 are fed to the normalizing links 28 and 29, the output signals of which are later in The summation link 30 is additively or subcultureally connected and digitized in analog-digital converter 31.
    The estimated position of the edge is set by a second digital-to-analog converter 38 or 39 to direct the x- or y-coordinate to the beginning of a series of measurements. They work more accurately and slowly, for example with 16 bits.
    The digital value of this basic JQ deviation is set by the central device 11 and is implemented in two buffer memories 40 and 41, pre-connected before the digital-to-analog converter 38 and 39,. The time constant of the low-pass filter 26 as well as the filter 27 bus output digital-to-analog frequency signals are matched with the delay time 38 and 39 are summed up by the analog-to-digital conversion with the output signal of the digital analog 31 and the preamplifier 14 or 15 respectively. Fig. 4 contributes to improving the positioning of the electron beam for implementing the method according to the invention. Adjusting the distance to the measured edge of the structure of length z, as well as to
    j Oh
    five
    the deflection sensitivity of the diversion systems 18 and 19 is effected through a cascade of multiplier 32 pulses, upstream of the counter 13 of the position. In addition, for this purpose, an analog damping element 35 is included, which is connected after the digital-to-digital converter t5.
    Similarly, the step size in the x-axis direction is performed by a cascade of multiplier 32 pulses upstream of counter 12 of the position and an analog damping link 24 connected after digital-analogue converter 14. The inclined position of the structure being measured relative to the direction
    the deviation of juvid systems 18 and 19 is taken into account by blocks 36 and 37 of the rotational correction. At the same time install part
    the deflecting voltage adds etc to one coordinate additively and to the other coordinate substrativo.
    A significant increase in accuracy when positioning the electron beam is achieved by the fact that the counter 12 is positioned with a fast digital-to-analog converter 14 and the damping link 34 is used only for its own sweep, which in the example is carried out in the direction of the X-axis and symmetrically with respect to the edge position. The basic deviation of the electrical position of the edge is set by a second digital-to-analog converter 38 or 39 to direct the x- or y-coordinate to the beginning of a series of measurements. They work more accurately and slowly, for example with 16 bits.
    co-ordinate in summing link 42 or 43, to which the output signal of the corresponding rotation correction block 36 or 37 is connected and connected to one of the end amplifiers 16 or 17. To achieve a high degree of automation, it is advantageous to control all settable units digitally. . indicated in FIG. 3 and 4 by the letter d from the central device 11 and form additional feedback signals to the central device, for example, the maximum and minimum values of the normalized signal, the position of the position counters and the readiness of the analog-digital converter.
    J5 sweep points of the same coordinate in the sweep direction, and form signals in the form of spatial averages and time averages through a selectable area
    By appropriately setting the pulse multiplier 33, you can vary 20 edges of the structure n, and the slope of the distance between the lines or the co-signal obtained by repeatedly storing the same line for a multiple sweep. If a further increase in dy is needed, the clock pulses are written to the binary bit of the larger position counter value. The central device 11 is informed about the presumptive position of the edge by the fact that the sharpened sweep without changing the coordinate along the edge is divided by the slope of the signal curve obtained in
    25 spatial average values and average values in time through the selected region of the edges of the structures, and this quotient relates to the constant established from the test objects, and
    30, the obtained value is a measure of the roughness of the edge of the structure being measured,
    This position is inserted into the central device 11. This method serves as a measuring point for known structures at given measurement points. In addition, the message may occur in that the object is depicted on the screens of the edges of the structures in the measuring device and the measuring edge is marked with devices containing the source
    2. Scheme for characterizing in an annual manner. Dp rasterization, forming the image, while using counters 12, 13 position.
    whoa.
    an electron beam and a radiation receiver, characterized in that the receiver through a normalizing star Marking can be carried out on a 40 ° analog-to-digital converter, an example introduced by a movable metro gating circuit connected
    The central computer is connected to the central device; the building input of this gating circuit 45 is connected to the input pulse of the analog-digital converter through the delay link; the position counters in the X and Y coordinates are
    权利要求:
    Claims (5)
    [1]
    1. The method of determining the characteristics of the edges of the structures in the measuring devices, which consists in the fact that arising during scanning by electronic
    Beam signals are used by the dp registers of the initial and final values of the edges of the structures and the electronic 50 raster sweep is connected to the centrifugal step by step along the transmitter, the first outputs being measured by the edge, which differs through digital-analogue that, in perpendicular, the distributors to the final amplifiers measure the edge of the sweep direction along the line connected to the deflecting system of the concept in constant predetermined steps of sampling the electron beam, and the structure of the structure outputs position counters tours using the electron beam, connected to the central device, the scanning signals are digitized and vice between the two, multiply repeated position counters included electronic switching kaplivayut
    perpendicular to the edge of the point deployment, and the coordinates in the direction of the edge do not change, thus obtained and digitized
    Signal values relative to the same places along the sweep direction are summed in one signal curve, for n different places at given distances, and along the edge of the structure perpendicular to the sweep direction, the point sweep and accumulation of signals obtained by dp n different places are repeated. and digitized sweep signals summarize for all
    the sweep points of the same coordinate in the sweep direction, and form signals in the form of spatial averages and time averages through a selectable area
    edges of the structure of n, a, the steepness of the curve of the signal obtained by repeated
    edges of the structure of n, a, the steepness of the curve of the signal obtained by repeated
    dotted scan without changing the coordinates along the edge is divided by the steepness of the signal curve obtained as spatial average values and average values over time through the selected region of the edges of the structures, and this quotient is associated with the constant established from the test objects, and
    the value obtained is a measure of the edge roughness of the measured structure,
     the edges of the structures in the gauges, containing the source
     the edges of the structures in the gauges, containing the source
    [2]
    2. A circuit for determining the nature of the device for selecting the direction of the coordinate along the X and Y coordinates, the second outputs of each of the position counters are connected via a switching device to each other by a position counter.
    [3]
    3. The circuit according to claim 2, wherein some receivers are provided, connected through a delay link to a normalizing link acting as a low-pass filter, and the output signals of all normalizing links are supplied through an additive or substratively active summing link
    to analog-to-digital converter
    [4]
    4. The circuit according to the item 3, about tl and h and y and so that at least after part of receivers it is included
    summed link, which is connected to an additional link delay.
    [5]
    5. The circuit according to claim 2, characterized in that the devices
    step size in the direction of the sweep and the distance a in the direction of the edge to the measured structure and for correcting the oblique position of the object in the input clock pulses of the counters
    The position includes an adjustable cascade of a pulse multiplier and between the digital-to-analog converter and a final amplifier, a damping link set by aHanoj oBoe, there are rotating correction blocks provided that a second digital-to-analog converter is provided in the coordinate direction, the output of which is additively connected to the corresponding deflection
    by the system.
    1J
    Editor L.Povkhan Tehred A.Kravchuk
    1879/36
    Circulation 678 Subscription,
    VNIIPI USSR State Committee
    for inventions and discoveries 113035, Moscow, Zh-35, Raushsk nab ,, d.4 / 5
    Production and printing company, Uzhgorod, Projecto st., 4
    I
    / A
    thirty
    31
    f (/ gA
    Proofreader, M. Pojo
    类似技术:
    公开号 | 公开日 | 专利标题
    EP0110301B1|1989-11-08|Method and apparatus for measuring dimension of secondary electron emission object
    US4327292A|1982-04-27|Alignment process using serial detection of repetitively patterned alignment marks
    US3644700A|1972-02-22|Method and apparatus for controlling an electron beam
    JP3288794B2|2002-06-04|Charge beam correction method and mark detection method
    WO1985004250A1|1985-09-26|Method and apparatus for precision sem measurements
    US4431923A|1984-02-14|Alignment process using serial detection of repetitively patterned alignment marks
    SU1310640A1|1987-05-15|Method and circuit for determining structure edge characteristic in instruments
    US4578587A|1986-03-25|Error-corrected corpuscular beam lithography
    EP0198955A1|1986-10-29|Apparatus for and a method of measuring the width of a line
    JP4647977B2|2011-03-09|Drift correction method and apparatus for automatic FIB machining
    JPH067043B2|1994-01-26|Position detection method
    IE53585B1|1988-12-21|Method and apparatus for inspecting a pattern
    JP3420037B2|2003-06-23|Dimension measuring device and dimension measuring method
    US4521686A|1985-06-04|Linewidth measuring with linearity calibration of the T.V. camera tube
    US2727153A|1955-12-13|Electron diffraction camera
    US6941006B1|2005-09-06|Method and system for calibrating the scan amplitude of an electron beam lithography instrument
    GB1597203A|1981-09-03|Position setting systems using a scanning beam
    KR20020016541A|2002-03-04|Mask defect checking method and device for electron beam exposure
    JP2002071330A|2002-03-08|Method and apparatus for inspecting pattern defect
    US4095112A|1978-06-13|Device for and a method of calibrating electron-optical apparatus
    JP2571110B2|1997-01-16|Pattern dimension measuring method and apparatus using charged beam
    JP2002008972A|2002-01-11|Equipment and method for electron beam exposure
    JP3194986B2|2001-08-06|Mark detection method and device, and electron beam drawing method and device using the same
    JP2003022773A|2003-01-24|Distortion compensating method of electron-ray deflection, electron-ray exposure device which has deflection distortion compensating means, and scanning type electron microscope
    JP3241017B2|2001-12-25|Method and apparatus for measuring beam size
    同族专利:
    公开号 | 公开日
    DE3228813A1|1983-05-26|
    DD206722A3|1984-02-01|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    DE3427981C2|1984-07-28|1988-05-19|Telefunken Electronic Gmbh, 7100 Heilbronn, De|
    JPS63166228A|1986-12-27|1988-07-09|Canon Inc|Position detector|
    JPH0663758B2|1987-10-14|1994-08-22|株式会社東芝|Pattern measurement method|
    AU2579197A|1997-03-12|1998-09-29|Ilyin, Mikhail Julievich|Method for measuring linear dimensions|
    JP3960544B2|2002-10-11|2007-08-15|パイオニア株式会社|Beam adjusting specimen, beam adjusting method and beam adjusting apparatus|
    法律状态:
    优先权:
    申请号 | 申请日 | 专利标题
    DD23452681A|DD206722A3|1981-11-02|1981-11-02|METHOD AND ARRANGEMENT FOR STRUCTURING EDGE DETERMINATION IN ELECTRON BEAM INSTRUMENTS|
    [返回顶部]