JP3547554B2 - Wafer tilt detection method during exposure - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a wafer inclination detecting method during exposure light, to a method of detecting the wafer inclination and the direction that occurs during exposure at full O preparative lithography process especially.
[0002]
[Prior art]
In general, a photo reticle for manufacturing a semiconductor integrated circuit (hereinafter, referred to as an actual reticle) includes a TEG (Test Element Group) pattern used for evaluating electrical characteristics in addition to a chip pattern to be an actual product. For example, JP-A-3-197949 discloses an example of an actual reticle in which one TEG is arranged at an arbitrary position on one shot. The configuration of this type of reticle is shown in FIG.
[0003]
As shown in FIG. 6, this actual reticle 10 is one in which two product pellets are obtained in one shot, and a street 1 provided on an outer frame, a chip pattern 2 corresponding to the product pellet, It is composed of a TEG pattern 3 sandwiched between chip patterns 2. As described above, the TEG is a test pattern for determining the quality of each product pellet by measuring electrical characteristics using the TEG. Examples of the test pattern include a sheet resistance measurement pattern and a contact resistance measurement pattern. , Transistors, and the like. Then, the quality of the product pellet is determined by checking whether the device parameters based on the test patterns, for example, various sheet resistance values, the _{Vth of} the transistor, and the like are at desired values. Normally, the quality judgment is performed to confirm whether or not each manufacturing process has been reliably performed after the completion of the wafer.
[0004]
As another example of the TEG pattern arrangement, a TEG is not inserted for each shot in the same manner, but an exposure method is devised as disclosed in Japanese Patent Application Laid-Open No. 5-291106, so that a There is also a method of transferring the TEG only to an arbitrary position in the above.
[0005]
By the way, in the exposure apparatus used in the photolithography process, the occurrence of defocus is inevitable, but in order to improve the dimensional accuracy and alignment accuracy of the pattern, it is necessary to detect this defocus and minimize the degree of the defocus. Must.
Therefore, a method of detecting this defocus is disclosed in Japanese Patent Application Laid-Open No. 60-26343. In this method, instead of an actual reticle used to form an actual product pattern, a test reticle provided with minute line elements and a light receiving element for receiving light transmitted through the test reticle through a minute slit are provided on a stage. Using a projection exposure apparatus that determines the characteristics of the projection optical system, for example, image plane tilt and field curvature, based on changes in the output signal from the light receiving element obtained when the stage is moved up and down on the Z axis. It is.
[0006]
Further, Japanese Patent Application Laid-Open No. 63-306626 discloses an exposure apparatus having a reference mark provided on a stage and a test reticle having a special reference mark. A method is disclosed in which a reticle is transmitted, and an image forming characteristic of a projection optical system is measured by receiving a light beam transmitted through each reference mark on a stage and a test reticle.
[0007]
[Problems to be solved by the invention]
However, even if a photo reticle having a TEG arrangement shown in the above-mentioned JP-A-3-197949 or JP-A-5-291106 is applied to a conventional general exposure apparatus, for example, a photolithography process may be used. If the wafer is tilted during the exposure and the focus shifts, the line width abnormality caused by the focus shift, that is, information on the image plane tilt, is immediately obtained from the device parameters (various resistance values, etc.) obtained from the TEG. Can not.
[0008]
On the other hand, if the exposure apparatus disclosed in JP-A-60-26343 or JP-A-63-306626 is used, it is possible to obtain accurate information on the image plane inclination. However, since the method using these devices uses a test reticle without using an actual reticle, it is necessary to measure the imaging characteristics using the test reticle at each exposure and correct the imaging characteristics. In such a case, a decrease in throughput is inevitable. Furthermore, when a test reticle is used, it is not possible to measure the focal position of each part of the pattern of the actual reticle immediately before exposure, and it is impossible to accurately measure the imaging characteristics of the projection optical system immediately before actual exposure. It is possible.
[0009]
Therefore, a method of measuring the imaging characteristics of the projection optical system without using a test reticle is disclosed in Japanese Patent Application Laid-Open No. 5-41344. That is, based on the output of a photoelectric detector that outputs irradiation light to a reticle, receives the reflected light as light transmitted through an opening, and outputs a photoelectric signal corresponding to the amount of received light, and the output of a focus detection system. This is a projection exposure apparatus in which an image plane state is detected by a focus position detection unit, and an imaging characteristic is accurately measured from the detected information. Another method that does not use a test reticle is disclosed in Japanese Patent Application Laid-Open No. 6-61120. The focus adjusting method of the exposure apparatus described in this publication is such that light is incident on a wafer from a light source for inspecting a surface to be exposed, via a plurality of rotatable reflectors, and the reflected light is transmitted through the reflector. The focus of the exposure apparatus is adjusted based on the detected data based on the relative height to the wafer, which is detected by a photodetector.
[0010]
However, these methods of measuring imaging characteristics do not use a test reticle, but include special light sources such as a light source for inspecting a surface to be exposed, a plurality of rotatable reflectors, a photoelectric detector, and a photodetector. There is a disadvantage that the equipment cost is enormous because a large number of parts are required. Furthermore, since the above-mentioned components are used, even when the imaging characteristics of the projection optical system can be accurately measured at the beginning of use, the installation portion of each component or the rotation portion of the component is slightly shifted with time. Occurs, and the deviation increases, so that the imaging characteristics cannot be accurately measured for a long period of time unless the minute positional deviation can be corrected.
[0011]
The present invention was made to solve the above problems, without using a test reticle or special exposure apparatus, c E for easily detecting the tilt of the wafer during exposure in the photolithography step It is an object of the present invention to provide a tilt detection method.
[0014]
[Means for Solving the Problems]
According to the method for detecting a wafer tilt at the time of exposure of the present invention, in a photolithography step, exposure is performed using a photo reticle in which TEG patterns including the same pattern are arranged at four corners on one shot of a reticle, and then transferred onto a wafer. The feature is to determine the presence or absence or the direction of the tilt of the wafer at the time of exposure by comparing the magnitude relationship of the line width of the same pattern in each of the TEG patterns at the four corners on one shot. In addition, when comparing the line width relationship, (1) measuring the sheet resistance using the sheet resistance measurement pattern in each of the TEG patterns at the four corners, and then measuring the sheet resistance and the sheet resistance measurement known in advance. A method of determining the electrical line width of each of the sheet resistance measurement patterns from the relationship between the line width of the pattern and the resistance value, and comparing the electrical line widths with each other; (2) within each TEG pattern at the four corners The method of measuring the line width of the same pattern and comparing the magnitude relation of the line width obtained by the measurement may be adopted.
[0015]
That is, according to the wafer inclination detecting method of the present invention, the distance from the photo reticle to the wafer in the exposure apparatus is reduced by one shot by comparing the line width of the same pattern located at the four corners of one shot. The four corners can be compared relatively. As a result, for example, if the distances from the photo reticle to the wafer are all the same in each of the four TEG patterns, the wafer is not tilted, and if not, the wafer is tilted. If the patterns match and there is no match between the opposing sides, the presence or absence of the wafer tilt and the direction of the tilt can be detected such that the wafer is tilted only in a direction perpendicular to those sides.
[0016]
BEST MODE FOR CARRYING OUT THE INVENTION
An embodiment of the present invention will be described below with reference to FIGS.
FIG. 1 is a diagram showing a real reticle (photo reticle) of the present embodiment. As shown in this figure, this real reticle 11 is one in which two product pellets can be obtained by one shot, a street 1 provided on an outer frame, and two chip patterns 2, 2 corresponding to the product pellets. Is provided. TEG patterns 3 (TEG-A to D) are provided on the street 1 at the four corners of the real reticle 11, respectively.
[0017]
TEG is a test pattern for judging the quality of each product pellet by performing electrical characteristics using the TEG. Examples of the test pattern include a sheet resistance measurement pattern, a contact resistance measurement pattern, a transistor, and the like. ing. Therefore, the TEG-A to TEG include at least the same sheet resistance measurement pattern even if all the individual patterns inside the TEG-A to DEG are not the same. FIG. 2 shows an example of the sheet resistance measurement pattern. The sheet resistance measurement pattern 12 uses various layers such as a diffusion layer region, polysilicon, tungsten, titanium, and aluminum as layers to be measured.
[0018]
FIG. 2 is a view showing a pattern actually transferred from the TEG pattern 3 on the actual reticle 11 to the wafer. The test pad 4 is provided at both ends of the layers to be measured 5, 7 and 9 via the contact 6. It is connected. The line width of the layers to be measured 5, 7, and 9 corresponds to the line width of the semiconductor integrated circuit, and has a line width of, for example, about 0.1 to 3 μm. The conventional general TEG measures the sheet resistance of the layer to be measured using these test pads, and checks whether or not the measured value has a desired value, thereby determining the quality of the product pellet. Things. In addition, in the present embodiment, the sheet resistance measurement pattern 12 is also used to determine whether or not the wafer is tilted during exposure.
[0019]
FIG. 2A is a view assuming a case where there is no defocus and the measured layer 5 on the wafer transferred from the actual reticle 11 has a desired line width. FIG. 2B shows a case where the defocus occurs and the measured layer 7 on the wafer is formed thicker than a desired line width. FIG. FIG. 4 is a view assuming a case where a measured layer 9 is formed thinner than a desired line width. On the other hand, FIG. 3 is a diagram showing the relationship between the line width of the measured layer and the sheet resistance.
[0020]
A point A in FIG. 3 indicates a sheet resistance value with respect to the line width of the layer to be measured when there is no defocus (in the case of FIG. 2A), and when the line width is about 0.1 to 3 μm. The standard value of the sheet resistance usually falls within a range of several Ω to several kΩ. Point B indicates a sheet resistance value with respect to a line width larger than a desired value (in the case of FIG. 2B), and a focus shift occurs. For example, the line width of the layer to be measured is smaller than a desired value. As the thickness increases by 10 to 20%, the sheet resistance decreases by 10 to 20% in inverse proportion thereto. A point C indicates a sheet resistance value for a line width smaller than a desired value (in the case of FIG. 2C), and a defocus occurs. For example, the line width of the layer to be measured is smaller than a desired value. As the thickness decreases by 10 to 20%, the sheet resistance increases by 10 to 20% in inverse proportion thereto. Therefore, if the resistance value of the sheet resistance measurement pattern is known, the line width of the measured layer can be known by calculation based on FIG. In this specification, the line width of the layer to be measured obtained from the sheet resistance value is referred to as “electric line width”.
[0021]
Hereinafter, a method of detecting a wafer tilt at the time of exposure using the real reticle 11 having the above configuration will be described with reference to the drawings. FIG. 4 is a flowchart illustrating the wafer tilt detection method according to the present embodiment.
FIG. 5B shows a state in which, among the TEG patterns at the four corners of the real reticle 11 shown in FIG. 5A, TEG-A to D, particularly the sheet resistance measurement pattern 12 is transferred onto an actual wafer. ing. Here, the measured layer 5 of TEG-A and TEG-B on the left side of the reticle is formed with a desired line width, and the measured layer 7 of TEG-C and TEG-D on the right side of the reticle is thicker than the desired line width. An example will be described.
[0022]
As shown in FIG. 4, first, the resistance of four sheet resistance measurement patterns is measured using an arbitrary measuring device (step S1). The resistance value obtained by the TEG-to D respectively _{r a, r b, r c} , and _{r d.} Next, from these resistance values, the electrical line widths s of the four sheet resistance measurement patterns 12 are calculated using a correlation graph of the line width and the resistance value that match the sheet resistance measurement pattern as shown in FIG. _{a, s} b, _s c, calculates a _{s d} (step S2).
[0023]
Then compared before electrical linewidth _s a sheet resistance measurement pattern 12 of the four positions obtained in _step, s _b, s c, the magnitude of _{s d} (step S3). For example, if the resistance values of the four sheet resistance measurement patterns 12 all match, the electrical line widths are all equal. In this case, the distance from the actual reticle to the wafer in the exposure apparatus is equal at all four TEG pattern positions, that is, the wafer is not tilted.
[0024]
In the example of this embodiment, first, the magnitude relationship between the resistance value in the TEG pattern position of the four _{locations, r a = r b> r} c = r d becomes, as shown in FIG. 5 (b), this magnitude of the electrical linewidth of the measured layers 5 and 7 of the inverse proportion to the 4 positions the sheet resistance measurement pattern 12 _{becomes s a = s b <s c} = s d. Therefore, other parameters (for example etching rate, impurity concentration, etc.) in the manufacturing process on the assumption that a standard value, the result of resistance measurement, TEG-A, the resistance value r _a in _TEG-B, is r _b When was the desired value, TEG-a, electrical linewidth _s a with TEG-B, _{s b} becomes a desired value, TEG-a, the defocus at the position of the TEG-B does not occur Will be. Conversely, it can be seen that a defocus occurs at the positions of TEG-C and TEG-D where the electrical line width is formed larger than that of TEG-A and TEG-B.
[0025]
_Furthermore, since it is _{s c = s d, TEG-} C, the distance between the actual reticle-wafer at the location of the TEG-D is, TEG-A, the distance between the actual reticle-wafer at the position of the TEG-B Are shifted in the same direction by the same distance. This indicates that the wafer is tilted only in the X-axis direction in FIG. 5A with respect to the actual reticle, and not tilted in the Y-axis direction. In this manner, the presence or absence of the wafer tilt and the direction of the tilt can be determined from the magnitude relationship of the resistance values at the four TEGs (step S4).
[0026]
As described above, according to the method of the present embodiment, the TEG patterns 3 are provided at the four corners of the actual reticle 11, and the electrical line widths of the four sheet resistance measurement patterns 12 are compared with each other. The distance from the actual reticle to the wafer at the position of the TEG pattern 3 can be relatively compared. As a result, for example, if the distances from the actual reticle to the wafer are all the same in each of the four TEG patterns 3, the wafer is not tilted, and if not, the wafer is tilted. If the TEG patterns 3 coincide with each other and do not coincide between the opposing sides, the presence / absence of the inclination of the wafer and the direction of the inclination can be detected such that the wafer is inclined only in a direction perpendicular to those sides. Therefore, when this method is used, the state of the tilt of the wafer can be known by a simple method without using a test reticle or a special exposure apparatus.
[0027]
Usually, measurement of device parameters using the TEG pattern 3 is performed by extracting several wafers from one lot. Therefore, according to this method, variations in the inclination of the wafer during exposure between wafers within the lot, or variations between lots, etc. Can be easily grasped. Therefore, it is possible to determine the quality of the wafer by measuring the device parameters, and at the same time, to constantly and effectively monitor variations such as the tilt of the wafer stage of the exposure apparatus and the displacement of the projection optical system. Become.
[0028]
Further, when the present method is used for the purpose of failure analysis, for example, by performing similar resistance measurement with various resistance measurement patterns such as a diffusion layer resistance measurement pattern in the TEG pattern 3 and a polysilicon wiring resistance measurement pattern, a large number of measurements can be performed. In the manufacturing process including the photolithography process, it is possible to know which of the exposure processes, such as the diffusion layer forming process and the polysilicon wiring forming process, causes the wafer to be tilted. As a result, for example, in the case where the exposure apparatus is properly used in each process, the defect can be reduced by feeding back which exposure apparatus has an abnormality.
[0029]
The technical scope of the present invention is not limited to the above-described embodiment, and various changes can be made without departing from the spirit of the present invention. For example, in the above-described embodiment, the inclination of the wafer is detected using the electrical line width of the layer to be measured calculated from the measurement data of the sheet resistance value. It is also possible to adopt a method of directly measuring the line width of the same pattern using a length measuring device or the like, and detecting the inclination of the wafer by comparing the measured values at four points. Further, the work of measuring the sheet resistance is usually performed after the completion of the wafer, but is not limited thereto, and may be performed, for example, during the manufacturing process. In this case, feedback to the exposure apparatus when detecting the tilt of the wafer can be performed more quickly than when the tilt is detected after the completion of the wafer.
[0030]
【The invention's effect】
As described above in detail, in the present invention, exposure is performed using a photo reticle provided with a TEG pattern including the same pattern at the four corners on one shot, and the TEG pattern in each TEG pattern transferred onto the wafer is exposed. By comparing the line width relationship of the same pattern, it is possible to detect the presence or absence or the direction of the inclination of the wafer during exposure. Therefore, it is possible to know the state of the tilt of the wafer in the exposure apparatus by a simple method without using a test reticle or a special exposure apparatus. In the case of comparison based on an electrical line width, the quality of a wafer is determined by measuring device parameters based on a TEG pattern, and at the same time, a change in a tilt of a wafer stage of an exposure apparatus, a displacement of a projection optical system, and the like. Can be constantly and effectively monitored. Furthermore, it is effective to use the present invention for the purpose of failure analysis.
[Brief description of the drawings]
FIG. 1 is a plan view showing a configuration of a photo reticle according to an embodiment of the present invention.
FIG. 2 is a plan view showing an example of a sheet resistance measurement pattern in a TEG pattern of the photo reticle.
FIG. 3 is a graph showing a correlation between a line width of a sheet resistance measurement pattern and a resistance value.
FIG. 4 is a flowchart showing a procedure of a wafer tilt detection method according to an embodiment of the present invention.
FIG. 5 is a plan view showing TEG patterns at four corners of the photo reticle according to the embodiment and a state of each sheet resistance measurement pattern actually transferred onto a wafer.
FIG. 6 is a plan view showing an example of a conventional photo reticle.
[Explanation of symbols]
1 Street 2 Chip pattern 3 TEG pattern 4 Test pad 5, 8 Layer to be measured (when there is no defocus)
6 Contact 7 Layer to be measured (when the pattern spreads due to defocus)
9 Layer to be measured (when the pattern becomes thin due to defocus)
10,11 Real reticle (photo reticle)
12 Sheet resistance measurement pattern

Claims (3)

In the photolithography process, after performing exposure using a photo reticle in which TEG patterns including the same pattern are arranged at four corners on one shot of the reticle, the above-described TEG patterns in the four corners on one shot transferred onto the wafer A method of detecting a wafer tilt at the time of exposure, comprising determining the presence or absence or the direction of the tilt of the wafer at the time of exposure by comparing the magnitude relationship of the line width of the same pattern. The method for detecting a wafer tilt at the time of exposure according to claim 1 ,
When comparing the line width relationship, the sheet resistance is measured using the sheet resistance measurement patterns in the TEG patterns at the four corners on one shot transferred onto the wafer, and the measured resistance value is determined in advance. Determining the electrical line width of each of the sheet resistance measurement patterns from the relationship between the line width and the resistance value of the sheet resistance measurement pattern, and comparing the magnitude relationship of the electrical line widths. Method of detecting wafer tilt at the time. The method for detecting a wafer tilt at the time of exposure according to claim 1 ,
When comparing the line width relation, the line width of the same pattern in each of the TEG patterns at the four corners is measured, and the line width relation obtained by this measurement is compared. Method of detecting wafer tilt at the time.

Images