Radiant energy – Inspection of solids or liquids by charged particles – Electron microscope type
Reexamination Certificate
2000-05-22
2003-11-25
Lee, John R. (Department: 2881)
Radiant energy
Inspection of solids or liquids by charged particles
Electron microscope type
C250S252100, C250S306000, C250S307000, C250S310000
Reexamination Certificate
active
06653634
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method of measuring length with a scanning type electron microscope (hereinbelow, called as SEM), and in particular, relates to a method of measuring length with an SEM which is suitable for length measurement and detection for a semiconductor device.
2. Conventional Art
Recent SEMs, in particular, SEMs for length measurement are required that a measured length value shows a true value and does not includes a variation and offset depending on such as constitution and film type of a specimen to which length measurement has been performed. Conventionally, as a size calibration measure for the length measurement SEM a pitch pattern is prepared by making use of a material such as Al and W which generates less charges on a Si, and through matching the measured length value to the prepared pitch pattern a length measurement calibration value for the length measurement SEM is obtained.
Further, for recent length measurement SEMs, in order to enhance secondary electron resolution a retarding electric field method (hereinbelow, simply called as a retarding method) is employed in which an acceleration voltage of incident electrons is increased and a negative voltage is applied to the side of a specimen to retard the acceleration voltage of the incident electrons. More specifically, in comparison with a conventional method wherein zero voltage is applied to the side of a specimen, the retarding method is constituted in such a manner that an acceleration voltage for electrons emitted from an electron gun unit is raised by +Vr and a voltage −Vr is applied to the specimen, thereby, electrons discharged from the electron gun unit with acceleration voltage of V+Vr pass an objective lens, thereafter, are retarded by the voltage −Vr applied to the side of the specimen and finally make incident to the specimen with an acceleration voltage V. As features of the retarding method, the followings are enumerated, in that since the incident electrons at the primary side pass below to the objective lens with the acceleration voltage V+Vr, a chromatic aberration of lenses can be reduced, and since secondary electrons generated from the specimen are accelerated toward a detector by the electric field due to −Vr, a collection efficiency of the secondary electrons is enhanced.
However, in the SEM employing the above explained retarding method, since the acceleration voltage (hereinafter, sometimes called as a landing voltage) of the electron beams incident onto the specimen varies depending on the voltage applied to the side of the specimen, there arises a problem that in case when a specimen to be observed is covered by an insulation film or an insulation film is deposited on the surface thereof the landing voltage varies and the magnification (or field of view) can deviate. Since the size calibration for the length measurement SEM is performed under a constant landing voltage and a constant magnification, therefore, if one of the two varies, there causes a problem that the measured size value varies. Namely, when the landing voltage varies, a size error is caused and there arised a problem that the measured size value is not a true value.
The above problem was likely caused in the conventional length measurement SEM which does not employ the retarding method. Namely, there arose the problem likely that the landing voltage thereof varies because of a variation in surface potential of a specimen, as a result a size error is caused, and therefore, the measured size is not a true value. Table 1 below shows results of pitch length measurement in a Si pattern formed on a Si wafer, a pattern on B-PSG film having film thickness 400 nm, a pattern on SiO
2
film having film thickness 10 nm, a pattern on SiO
2
film having film thickness 100 nm, a pattern on SiN film, a pattern on HTO+SiN film (35 nm) and a pattern on SiON film having film thickness 25 nm formed on a Si wafer, and a pattern on actual devices A and B.
Design values of length measured pitch are 350 nm, 400 nm, 500 nm, 600 nm, 700 nm and 800 nm. The unit of all numeral values in the Table is nm.
TABLE 1
Pitch design value (nm)
350.00
400.00
500.00
600.00
700.00
800.00
Si Pattern
350.32
400.18
500.01
600.20
700.52
800.85
B-PSG
358.74
410.33
512.67
615.32
717.93
820.23
(400 nm)
SiO
2
(10 nm)
350.35
400.42
500.47
600.42
700.55
800.45
SiO
2
(100 nm)
354.89
405.47
506.83
608.32
709.46
811.21
SiN (10 nm)
355.75
406.29
507.78
609.52
711.30
812.55
HTO + SiN
362.76
414.33
518.64
622.05
725.89
829.35
(35 nm)
SiON (25 nm)
359.33
410.53
513.73
616.23
718.74
821.55
Device A
367.54
420.34
525.06
630.25
735.35
840.21
Device B
365.00
417.29
521.19
625.59
729.66
834.33
FIG. 18
illustrates a relationship between pitch design values (in abscissa) and SEM length measurement values (in ordinate) obtained from data in connection with Si pattern, pattern on B-PSG film and pattern of actual device A among the length measurement data as shown in Table 1. It is observed that although the pitch length measurement values of the Si pattern substantially coincide with the designed values, the pitch length measurement values of the patterns on the B-PGS insulation film and the actual device A greatly deviate from the respective designed values. It is considered that such errors in the measured values are caused due to the change-up of the specimen surface irradiated by electron beams during the pattern length measurement on the B-PSG insulation film and the actual device A.
SUMMARY OF THE INVENTION
The present invention is achieved in view of the problem encountered during the length measurement with the conventional SEMs, and an object of the present invention is to provide a method of measuring length with an SEM which prevents a conventional length measurement error depending on such as specimen constitutions and film types.
A method of measuring length with an SEM according to the present invention which resolves the above problem, is characterized in that the method comprises the steps of: performing length measurement with the SEM an already know pattern provided in advance in a predetermined region on a specimen; obtaining a magnification correction coefficient through comparison of the length measurement result with the designed value of the already known pattern; and determining a true size by multiplying a measured length value of a measurement point performed by the SEM by the obtained magnification correction coefficient.
In the above method, a specimen is used in which a particular test pattern such as lines and spaces and contact holes for obtaining correction data is in advance provided, for example, in a scribe in a semiconductor wafer specimen and in a free space in a chip. Then, the length measurement of the test pattern is performed and the magnification correction coefficient for correcting a length measurement error is determined, thereby, a size measurement error by the SEM, which is caused due to the charge up of the specimen surface (in that the surface potential variation of the place where incident electrons are irradiated), is compensated.
Further, a method of measuring length with an SEM according to the present invention, is characterized in that the method comprises the steps of: performing length measurement with the SEM a pattern on a specimen; obtaining a magnification correction coefficient through comparison of the length measurement. result with the designed value of the pattern; and determining a true size by multiplying a measured length value of a measurement point performed by the SEM by the obtained magnification correction coefficient.
In the above method, without providing on the specimen a test pattern used exclusively for determining the magnification correction coefficient and by making use of a pattern which permits measurement of a line width, a space width or a pitch among patterns in an actual device, the magnification correction coefficient is determined which compensates errors due to the charge-u
Nagai Kouichi
Otaka Tadashi
Hitachi , Ltd.
Kenyon & Kenyon
Lee John R.
Souw Bernard E.
LandOfFree
Method of measuring length with scanning type electron... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of measuring length with scanning type electron..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of measuring length with scanning type electron... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3172792