Disclosure of Invention
In view of the above, the present invention provides a wafer polishing method, polishing apparatus, and processing apparatus that solve or at least mitigate one or more of the above-identified problems and other problems of the prior art.
The first aspect of the present invention provides a wafer polishing method for removing a target layer, the target layer including a first layer and a second layer having a coefficient of friction close to each other, a primary polishing process of the target layer including a first polishing stage and a second polishing stage, comprising the steps of:
Determining a first time;
Determining the material removal amount and the material removal rate of the current wafer in the second polishing stage, and calculating a second time according to the material removal amount and the material removal rate of the current wafer in the second polishing stage;
Adjusting the first time according to the second time to obtain a third time, wherein the third time is equal to the first time when the second time falls into a preset threshold range, and the third time is determined according to the adjustment value and the first time when the second time does not fall into the preset threshold range;
And performing a first polishing stage according to the third time and a second polishing stage according to the second time so that the second layer reaches a target thickness.
Optionally, the determining the first time includes:
A first predetermined time is determined as the first time and/or the third time of a previous wafer is determined as the first time, the first predetermined time falling within the predetermined threshold range.
Optionally, the wafer polishing method further includes:
measuring an initial thickness of a second layer of the wafer prior to the first polishing stage as a pre-thickness value;
the remaining thickness of the second layer of the wafer is measured as a post-thickness value after the second polishing phase is completed.
Optionally, the first polishing stage removes a portion of the first layer, the second polishing stage removes a remaining portion of the first layer and a portion of the second layer, and the determining the material removal amount and the material removal rate of the wafer in the second polishing stage includes:
Determining a thickness of a remaining portion of the first layer;
Determining the material removal of the wafer in a second polishing stage based on the pre-thickness value, the thickness of the remaining portion of the first layer, and the target thickness, and/or,
Determining a material removal amount of the wafer at a second polishing stage based on the pre-thickness value, the thickness of the remaining portion of the first layer, and the post-thickness value;
And determining the material removal rate according to the material removal amount of the wafer in the second polishing stage.
Optionally, the determining the thickness of the remaining portion of the first layer includes:
measuring a thickness of a remaining portion of the first layer of a first wafer after a first polishing phase of the first wafer has ended;
The thickness of the remaining portion of the first layer of the current wafer is determined based on the thickness of the remaining portion of the first layer of the previous wafer.
Optionally, the first time and the second time of the first wafer are determined by:
determining the first predetermined time as the first time of the first wafer;
Determining a material removal amount of the first wafer at a second polishing stage based on the pre-thickness value of the first wafer, the thickness of the remaining portion of the first layer, and the target thickness;
The second time of the first wafer is determined based on the theoretical removal rate of the first layer, the theoretical removal rate of the second layer, and the amount of material removed by the first wafer during the second polishing stage.
Optionally, the determining the second time of the first wafer according to the theoretical removal rate of the first layer, the theoretical removal rate of the second layer, and the material removal amount of the first wafer in the second polishing stage includes:
The second time of the first wafer is determined based on the thickness of the remaining portion of the first layer of the first wafer and the theoretical removal rate of the first layer, and the pre-thickness value of the first wafer, the target thickness, and the theoretical removal rate of the second layer.
Optionally, the first time and the second time of the current wafer are determined by steps comprising:
determining the third time of the previous wafer as the first time of the current wafer;
determining the material removal amount of the current wafer in a second polishing stage according to the thickness front value of the current wafer, the thickness of the rest part of the first layer and the target thickness;
calculating the actual material removal amount of the previous wafer in the second polishing stage according to the thickness front value of the previous wafer, the thickness of the rest part of the first layer and the thickness rear value;
Calculating the actual material removal rate of the previous wafer in the second polishing stage according to the actual material removal amount of the previous wafer in the second polishing stage and the polishing time of the previous wafer in the second polishing stage;
and determining the second time of the current wafer according to the material removal amount of the current wafer in the second polishing stage and the actual material removal rate of the previous wafer in the second polishing stage.
Optionally, the calculating the adjustment value, determining the third time according to the adjustment value and the first time includes:
calculating a difference between the second time and the first predetermined time;
Determining an adjustment factor associated with one or more of a theoretical removal rate of the first layer, a theoretical removal rate of the second layer, a thickness of the first layer, a thickness of the second layer, a thickness of a remaining portion of the first layer, the target thickness, a time of use of the polishing pad;
Determining the adjustment value according to the adjustment coefficient and the difference value;
and adding the sum of the adjustment value and the first time as the third time.
Optionally, the determining the thickness of the remaining portion of the first layer of the current wafer according to the thickness of the remaining portion of the first layer of the previous wafer includes:
correcting the thickness of the remaining portion of the first layer of the previous wafer according to the adjustment value of the previous wafer;
and determining the thickness of the residual part of the first layer of the corrected previous wafer as the thickness of the residual part of the first layer of the current wafer.
Optionally, the calculating the actual material removal of the previous wafer in the second polishing stage based on the pre-thickness value of the previous wafer, the thickness of the remaining portion of the first layer, and the post-thickness value includes calculating the actual material removal of the previous wafer in the second polishing stage based on the pre-thickness value of the previous wafer, the post-thickness value, and the corrected thickness of the remaining portion of the first layer of the previous wafer.
A second aspect of the present invention provides a wafer polishing apparatus for performing the wafer polishing method as set forth in the first aspect, comprising a first polishing unit, a second polishing unit, and a measuring unit;
The first polishing unit comprises a first polishing head and a first polishing disk, and is used for executing the first polishing stage;
The second polishing unit comprises a second polishing head and a second polishing disk for executing the second polishing stage;
The measuring unit comprises measuring means for measuring the pre-thickness value and the post-thickness value.
A third aspect of the present invention provides a wafer processing apparatus comprising a controller for storing the pre-thickness value and the post-thickness value measured by the measuring device, a memory for controlling the wafer polishing device to execute the wafer polishing method according to the first aspect, and the wafer polishing device according to the second aspect.
The wafer polishing method has the following technical effects that the large disc torque change is not used as an end point measuring means, so that the failure of the measuring means is avoided, and the risk of undergrinding or overgrinding the wafer is reduced. The polishing time of the two polishing stages is graded and adjusted in a linkage way, so that the polishing thickness can be timely and accurately controlled, and the production efficiency and the polishing precision of the wafer are improved. The problem that a plurality of wafers are thinner or thicker is avoided, the fault tolerance rate and the reaction time of the system are increased, the service life of the polishing pad is prolonged, and the production cost is greatly reduced.
Detailed Description
In order to better understand the technical solutions in the embodiments of the present invention, the following description will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which are derived by a person skilled in the art based on the embodiments of the present invention, shall fall within the scope of protection of the embodiments of the present invention.
In the description of the present invention, it should be understood that the terms "longitudinal," "transverse," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships based on the orientation or positional relationships shown in the drawings, merely to facilitate describing the present invention and simplify the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present invention.
In the description of the present invention, unless otherwise specified and defined, it should be noted that the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, mechanical or electrical, or may be in communication with each other between two elements, directly or indirectly through intermediaries, as would be understood by those skilled in the art, and the specific meaning of the terms may be understood as appropriate.
Fig. 1 is a schematic view of a wafer polishing unit in a CMP apparatus, in which a polishing pad 20 is coated on an upper surface of a polishing disk 30, a polishing liquid supply means 40 distributes a polishing liquid on the surface of the polishing pad 20, and a dressing means 50 for dressing the surface of the polishing pad to a state suitable for polishing, as shown. During polishing operation, the polishing head 10 presses the surface to be polished of the wafer to the surface of the rotating polishing pad 20, and rotates and moves, and polishing liquid is distributed between the polishing pad and the wafer, so that the removal of materials on the surface of the wafer is completed under the action of chemical machinery, and the purpose of global and local planarization is achieved.
When continuous two-stage polishing is needed to be carried out on the wafer, two wafer polishing units shown in fig. 1 can be connected in series to form a double-disc process, so that the wafer sequentially passes through the two wafer polishing units and different polishing conditions are respectively applied to achieve the specific polishing purpose, namely, the first polishing unit polishes with a material removal rate far greater than that of the second polishing unit, a thicker target layer is removed, the production efficiency is improved, the material removal rate is smaller in the second polishing unit, and the stability and accuracy of a polishing result are ensured.
Taking the fabrication of a shallow trench isolation structure as an example, the process flow generally includes depositing a silicon nitride layer on the surface of a silicon substrate, etching the shallow trench, and then depositing a silicon dioxide layer, wherein the silicon dioxide fills the shallow trench in the wafer and covers the surface of the silicon nitride layer. And finally, sequentially passing the wafer through two polishing units to remove redundant silicon dioxide and a certain amount of silicon nitride on the surface of the wafer, so that the silicon dioxide covering the surface of the silicon nitride layer is not remained after polishing, and the silicon dioxide in the shallow trench is reserved, namely, the polishing is finally and stably stopped on the silicon nitride layer, so that the shallow trench isolation is formed.
Since the silicon oxide layer and the silicon nitride layer have close friction coefficients, the end point detection means for measuring the motor torque change in the first polishing unit fails, and therefore, the polishing time can only be set to a fixed value. Meanwhile, the polishing is stopped on the silicon nitride layer, namely the silicon nitride layer is not completely removed, so that the polishing time is not accurate enough or the final thickness cannot be timely adjusted in the second polishing unit by an end point detection means for measuring the torque change of the motor, the polishing quality is seriously affected, or the polishing time is too long, the polishing time of the second unit is further caused to be severely changed, and the polishing time cannot be adapted to the production beats at the upstream and the downstream in the continuous production process, so that the production efficiency of CMP equipment and even the whole chip production line is reduced. In addition, once the change curve of the motor torque changes along with the loss of the polishing pad, the control of the polishing time is directly disabled, and the risk of less grinding or overgrinding of the wafer is further increased, so that the produced wafer cannot meet the requirements of subsequent procedures, and the yield is reduced.
The invention provides a wafer polishing method for removing a target layer on the surface of a wafer, wherein the target layer comprises a double-layer film structure and the first layer and the second layer have close friction coefficients. The method comprises a first polishing stage and a second polishing stage in a single polishing process of the target layer, and is used for continuously polishing the wafer so that the polished wafer has the second layer reaching the target thickness.
In one embodiment, as shown in FIG. 2, the steps include:
s1, determining a first time;
s2, determining the material removal amount and the material removal rate of the wafer in the second polishing stage;
S3, calculating a second time according to the material removal amount and the material removal rate of the wafer in a second polishing stage;
s4, adjusting the first time according to the second time to obtain a third time, wherein when the second time falls into a preset threshold range, the third time is equal to the first time, and when the second time does not fall into the preset threshold range, an adjustment value is calculated, and the third time is determined according to the adjustment value and the first time;
S5, executing the first polishing stage according to the third time, and executing the second polishing stage according to the second time so that the second layer reaches the target thickness.
The polishing method of the embodiment does not depend on end point detection, and after the second time is determined through calculation, the first time is adjusted in a linked and graded mode according to the second time, so that the polishing time of the first polishing stage is adjustable, the polishing precision can be timely and accurately controlled, and the stability and the continuity of the first polishing stage are ensured as much as possible. According to the polishing method, when the target layer comprises a double-layer film structure with similar friction coefficient, the primary polishing process can be stopped when the second layer reaches the target thickness, the control complexity is low, the polishing precision is high, and the production efficiency is high.
Optionally, in step S1, the first predetermined time is determined as the first time, or the third time of the previous wafer is determined as the first time. Specifically, the current wafer may be the first wafer or the mth (m > 1) wafer. For the first wafer, the first time is set to a first predetermined time, which may be an empirical value determined according to an actual processing situation. For the mth wafer, the first time may be set to a first predetermined time, but is preferably set to a third time of the mth-1 wafer, so that the influence of the adjustment of the polishing time of the previous wafer is continued when the current wafer is polished, which is advantageous for uniformity of the wafers in continuous production.
Optionally, as shown in fig. 3, step S2 and step S3 include the following steps:
S21, determining the material removal amount of the current wafer in the second polishing stage according to the thickness of the wafer. The material removal amount means the volume of the wafer removed during polishing, and since the diameter of the wafer is known at the time of processing, the material removal amount can be determined by measuring the thickness of the wafer.
In one embodiment, the wafer polishing method of the present invention further comprises measuring the initial thickness of the second layer of the current wafer prior to the first polishing stage as a pre-thickness value PreThk, and subtracting the target thickness TarThk of the post-CMP second layer film determined by the wafer processing target from the pre-thickness value PreThk, where the result is the thickness of the material removed by the polishing in the second polishing stage and the corresponding volume is the material removal in the second polishing stage.
In one embodiment, the wafer polishing method of the present invention further comprises measuring the remaining thickness of the second layer of the current wafer as a post-thickness value PostThk after the second polishing stage is completed. In determining the material removal of the previous wafer in the second polishing stage, the material removal calculated by subtracting the post-thickness value PostThk from the pre-thickness value PreThk is the actual material removal of the previous wafer in the second polishing stage, since the previous wafer has completed polishing, i.e., the post-thickness value PostThk has been measured.
S22, determining the actual material removal rate of the previous wafer in the second polishing stage, and taking the actual material removal rate as the material removal rate of the current wafer in the second polishing stage.
The units of material removal rate (Material Removal Rate, MRR) are typically expressed as volume units divided by time units, and are used to describe the volume of material that can be removed per unit time, calculated as the amount of material removed divided by the polishing time.
Specifically, for the first wafer, since there is no preceding wafer, the theoretical material removal rate of the second layer may be directly used as the actual material removal rate of the preceding wafer in the second polishing stage, and the theoretical material removal rate may be obtained by calculation using a model describing the material removal rate or may be measured and calculated by means of trial polishing.
For the m-1 wafer, since the m-1 wafer has finished polishing, the actual material removal rate of the m-1 wafer in the second polishing stage can be calculated according to the material removal amount or the actual material removal amount of the m-1 wafer in the second stage and the polishing time, compared with the theoretical removal rate, the actual material removal rate of the m-1 wafer in the second polishing stage is more accurate, and the accurate second time which cannot be calculated in the next step when the material removal rate changes due to the change of the polishing conditions can be reduced, so that the polishing precision is reduced.
S23, dividing the material removal amount of the current wafer in the second polishing stage by the material removal rate to calculate second time.
Optionally, in a polishing recipe (recipe) of the dual-disk process, corresponding polishing conditions are set so that polishing times of the first polishing stage and the second polishing stage are not greatly different, so that waiting time of the wafer can be reduced in continuous production, thereby improving production efficiency. Accordingly, in step S4, the predetermined threshold range may be set according to the polishing time of the first polishing stage or the second polishing stage in the polishing recipe. The predetermined threshold range is set around the first predetermined time in this embodiment, that is, such that the first predetermined time falls within the predetermined threshold range. Preferably, the first predetermined time is set to be the median of the predetermined threshold range. Further, a polishing time limit value of the second polishing stage is determined by means of trial polishing, and a predetermined threshold value range is set according to the polishing time limit value. Further, the second time of the first wafer falls within a predetermined threshold range, such that the first time of the first wafer is not adjusted. And S4, adjusting the first time according to the magnitude relation between the second time and the preset threshold range, and adjusting the first time only after the preset threshold range is exceeded, so that the grading and linkage adjustment of the first time and the second time are realized.
When the second time is less than the lower limit of the predetermined threshold range, the wafer is overground, the desired adjustment direction of the polishing time of the second polishing stage is reduced, and when the second time is greater than the upper limit of the predetermined threshold range, the wafer is underground, and the desired adjustment direction of the polishing time of the second polishing stage is increased. However, only adjusting the second time has not only a slow response speed, but also an adjustment limit for the second time, so in this embodiment, the adjustment direction of the polishing time in the first polishing stage is the same as the desired adjustment direction of the polishing time in the second polishing stage, so as to reduce the adjustment amount of the second time, thereby realizing control of the polishing precision of the wafer with a faster response speed, and also avoiding that an unsatisfactory wafer is produced after the second time reaches the adjustment limit. The magnitude of the polishing time for the first polishing stage can be measured in terms of the difference between the second time and the first predetermined time, and the degree of the different effects of the first polishing stage and the second polishing stage on the polishing process.
Specifically, as shown in fig. 4, in step S4, when the second time does not fall within the predetermined threshold range, an adjustment value is calculated, and a third time is determined according to the adjustment value and the first time, including the following steps:
s41, calculating a difference value between the second time and the first preset time. The first predetermined time is a base time of the first polishing stage, and a difference of the second time minus the first predetermined time may represent an adjustment direction of the polishing time of the first polishing stage and a portion represents an adjustment amplitude of the polishing time of the first polishing stage.
S42, determining an adjustment coefficient. The adjustment factor is used to measure the different effects of the first polishing stage and the second polishing stage on the polishing process, and thus represents in part the magnitude of the adjustment of the polishing time of the first polishing stage. Optionally, the adjustment coefficient is determined according to the material removal rate of the first polishing stage and the second polishing stage when the corresponding material layer is removed, so as to balance the different effects of the same polishing time between the first polishing stage and the second polishing stage on the polishing result by the adjustment coefficient. Preferably, the adjustment factor is an empirical value determined from the trial polishing and is associated with one or more of a theoretical removal rate of the first layer, a theoretical removal rate of the second layer, an actual removal rate of material during the first polishing stage, an actual removal rate of material during the second polishing stage, a thickness of the first layer, a thickness of the second layer, a target thickness, and a time of use of the polishing pad. Further, the adjustment coefficient is a ratio of a theoretical removal rate of the second layer to a theoretical removal rate of the first layer, or the adjustment coefficient is a ratio of an actual material removal rate of the second polishing stage to an actual material removal rate of the first polishing stage. Further, since the polishing pad is worn out with the use time and the polishing pad is replaced after the service life is reached, the adjustment coefficient is also changed according to the use time of the polishing pad in each of the first polishing stage and the second polishing stage. In a preferred embodiment, since the first polishing stage wears more severely on the polishing pad, the adjustment factor can be expressed as a decreasing function of polishing time, for example, by multiplying the ratio of the theoretical removal rate of the second layer to the theoretical removal rate of the first layer by a gradually decreasing constant in stages over time as the adjustment factor.
S43, determining an adjustment value according to the adjustment coefficient and the difference value. Alternatively, the product or quotient of the adjustment coefficient and the difference is determined as an adjustment value to indicate the extent to which adjustment is desired for the first polishing stage.
S44. the sum of the adjustment value and the first time is added as a third time, i.e. third time = first time + adjustment value.
The first predetermined time falls within a predetermined threshold range, and when the second time is less than a lower limit of the predetermined threshold range, there is a risk of overgrinding or overgrinding the wafer. The difference calculated in step S41 is a negative value, and the adjustment value is also a negative value, and the third time calculated according to the formula in step S44 is smaller than the first time, that is, the polishing time of the first polishing stage is reduced. Therefore, the material removal amount in the first polishing stage is reduced, the risk of overgrinding the wafer can be reduced at a faster response speed, and the thickness post-value of the wafer is enabled to be more quickly approaching to the target thickness. When the second time is greater than the upper limit of the predetermined threshold range, the wafer is at risk of under-grinding or under-grinding. The difference calculated in step S41 is a positive value, and the adjustment value is also a positive value, and the third time calculated according to the formula in step S34 is longer than the first time, that is, the polishing time of the first polishing stage is increased.
According to the technical scheme, the polishing time of the first stage is regulated only when the regulating amplitude of the second time is larger, namely the stage regulation is carried out, so that the stability and the continuity of the first polishing stage are ensured as much as possible, meanwhile, the polishing time of the first polishing stage can be automatically regulated according to the regulating expectation and the regulating amplitude of the second time, namely the linkage regulation is carried out, so that the wafer reaches the target thickness more quickly, the response speed is controlled more quickly, the risk of undergrinding or overgrinding the wafer is obviously reduced, and the polishing precision and the production efficiency of chemical mechanical polishing are further improved.
In the prior process, in order to meet the requirements of the subsequent process, the first layer must be completely ground during polishing and the polishing must be stably stopped at the second layer. For this reason, as shown in fig. 5, a balance of polishing efficiency and polishing effect is achieved by removing most of the first layer with a larger material removal rate in the first polishing stage, and then removing the remaining portion of the first layer and the portion of the second layer with a smaller material removal rate in the second polishing stage. This means that in the second polishing stage, not only the second layer but also the remaining part of the first layer Offset is removed first. Therefore, when the material removal amount in the second polishing stage is calculated, only the portion of the second layer to be removed is considered, and errors exist, which affect the polishing effect.
To eliminate the adverse effect of the above-described errors on the accuracy of polishing, in one embodiment, the first time and the second time of the wafer are determined by the following steps, as shown in fig. 6.
S61, determining the first preset time as the first time of the first wafer.
S62, determining the material removal amount of the first wafer in the second polishing stage according to the pre-thickness value of the first wafer, the thickness of the rest part of the first layer and the target thickness.
For the first wafer, a first polishing stage is performed according to a first predetermined time, and a thickness value of the remaining first layer film after the end of the first polishing stage is measured as a median thickness MidThk of the first wafer, and a material removal amount=a pre-thickness value+a median thickness-target thickness of the first wafer in the second polishing stage.
S63, determining a second time of the first wafer according to the theoretical removal rate of the first layer, the theoretical removal rate of the second layer and the material removal amount of the first wafer in the second polishing stage.
Alternatively, the estimated value of the material removal rate in the second polishing stage is obtained by weighted averaging of the theoretical removal rate of the first layer and the theoretical removal rate of the second layer, and then the calculated material removal amount of the first wafer in the second polishing stage is divided by the estimated value of the material removal rate in the second polishing stage by S62, thereby calculating the second time.
Preferably, the polishing time of the remaining portion of the first layer and the polishing time of the removed portion of the second layer may be calculated separately, and then the two polishing times may be added to determine the second time of the first wafer. Namely, the polishing time of the remaining part of the first layer is calculated by dividing the median thickness by the theoretical removal rate of the first layer, the polishing time of the removed part of the second layer is calculated by dividing the difference of the target thickness subtracted from the pre-thickness value by the theoretical removal rate of the second layer, and then the second time is calculated by summing the polishing time of the remaining part of the first layer and the polishing time of the removed part of the second layer.
S64, determining the third time of the previous wafer as the first time of the current wafer.
For the mth wafer (m > 1), the third time of the mth-1 wafer is determined as the first time of the mth wafer.
S65, determining the material removal amount of the current wafer in the second polishing stage according to the thickness front value of the current wafer, the thickness of the rest part of the first layer and the target thickness.
S66, calculating the actual material removal amount of the previous wafer in the second polishing stage according to the thickness front value of the previous wafer, the thickness of the rest part of the first layer and the thickness rear value.
For the mth wafer (m > 1), when calculating the material removal amount of the current wafer in the second polishing stage and the material removal amount of the previous wafer in the second polishing stage, the material removal amounts of the second stage are corrected based on the remaining portions of the respective first layers, and the actual material removal amounts of the previous wafer in the second polishing stage are calculated. Preferably, because each wafer has a median thickness measured between the two polishing stages that can greatly impact production efficiency, the median thickness of the first wafer can be used, or measured once with several wafers apart, and the thicknesses of the previous wafer and the remainder of the first layer of the current wafer can be determined based on the median thickness.
S67, calculating the actual material removal rate of the previous wafer in the second polishing stage according to the actual material removal amount of the previous wafer in the second polishing stage and the polishing time of the previous wafer in the second polishing stage.
When calculating the time material removal rate of the preceding wafer, the actual material removal rate of the m-1 wafer in the second polishing stage is calculated by dividing the calculated actual material removal amount of the m-1 wafer in the second polishing stage by the actual polishing time of the m-1 wafer in the second polishing stage.
S68, determining the second time of the current wafer according to the material removal amount of the current wafer in the second polishing stage and the actual material removal rate of the previous wafer in the second polishing stage.
The second time of the current wafer is calculated by dividing the material removal amount of the mth wafer in the second polishing stage calculated in step S65 by the actual material removal rate of the mth-1 wafer in the second polishing stage calculated in step S67.
By correcting the material removal amount of the wafer in the second stage, the embodiment ensures that most of the first layer can be removed rapidly in the first polishing stage, and the wafer can be stopped stably at the target thickness of the second layer in the second polishing stage, thereby improving the accuracy of polishing control.
In one embodiment, to further improve the control accuracy, the thickness of the remaining portion of the first layer of the previous wafer may be corrected according to the adjustment value of the previous wafer, and the corrected thickness of the remaining portion of the first layer of the previous wafer may be determined as the thickness of the remaining portion of the first layer of the current wafer.
Optionally, the adjustment value is used to decrease the first time or increase the first time, so that the amount of change in the material removal amount of the first layer in the adjusted first polishing stage, that is, the amount of change in the increase or decrease in the thickness of the remaining portion of the first layer of the m-1 th wafer, can be calculated according to the adjustment value and the theoretical removal rate of the first layer, thereby correcting the thickness of the remaining portion of the first layer of the m-1 th wafer. The thickness of the rest of the first layer of the m-1 th wafer after correction is determined as the thickness of the rest of the first layer of the m-1 th wafer, and compared with the thickness median value of the first wafer, the thickness median value is closer to an actual value without repeatedly measuring the thickness median value, and meanwhile, the polishing precision and the polishing efficiency are improved. It will be appreciated that in calculating the actual material removal rate of the m-1 th wafer in the second polishing stage, the thickness of the remaining portion of the first layer of the corrected m-1 th wafer may be used instead of the value before correction, further improving the polishing accuracy and polishing efficiency. In wafer processing, the improvement of polishing precision means that the wafer can meet the requirements of the subsequent process, so that the processing yield of the wafer is improved.
The present invention also provides a wafer polishing apparatus 100 for performing the wafer polishing method of the present invention, which in one embodiment, as shown in fig. 6, includes a first polishing unit 1, a second polishing unit 2, and a measuring unit 3. The first polishing unit 1 comprises a first polishing head and a first polishing disc for performing the first polishing stage, and the second polishing unit 2 comprises a second polishing head and a second polishing disc for performing the second polishing stage. It is understood that the first polishing pad and the second polishing pad also include respective slurry supply means and dressing means thereon.
The measurement unit 3 comprises measurement components, preferably optical measurement components, to measure the thickness of the wafer placed on the measurement unit, including pre-thickness values, mid-thickness values and post-thickness values.
When the median thickness does not need to be measured, the wafer is transported in the wafer polishing apparatus 100 along the path of the measurement unit 3, the first polishing unit 1, the second polishing unit 2, and the measurement unit 3 in this order. When the median thickness needs to be measured, the wafer is transported in the wafer polishing apparatus 100 along the path sequentially including the measuring unit 3, the first polishing unit 1, the measuring unit 3, the second polishing unit 2, and the measuring unit 3. Optionally, a wafer buffer position 4 is further disposed between any two of the first polishing unit 1, the second polishing unit 2 and the measuring unit 3, so as to be used for interacting with the manipulator, the first polishing head and the second polishing head to transfer the wafer.
The present invention also provides a wafer processing apparatus 1000, as shown in fig. 7, which includes the wafer polishing device 100, the controller 200, and the memory 300, and a polishing recipe of a wafer and a thickness of the wafer measured at the measurement unit 3 are stored in the memory 300. The controller 200 is electrically connected to the wafer polishing apparatus 100 and the memory 300 for controlling the wafer polishing apparatus 100 to perform the wafer polishing method according to any one of the embodiments of the present invention.
Fig. 8 is a flow chart of a wafer polishing method for fabricating a shallow trench isolation structure according to another embodiment of the invention. The wafer polishing apparatus 100, the wafer processing device 1000, and the wafer polishing method of the present invention are specifically described below with reference to fig. 8.
In this embodiment, the first layer of the wafer to be polished is a silicon dioxide layer, and the second layer is a silicon nitride layer. Because the diameter of the wafer is a constant value in the processing of the same batch of wafers, the material removal amount is directly expressed by the removed material thickness for simplicity and clarity of description, and the material removal rate is expressed by the removed material thickness per unit time.
Before starting wafer polishing, the theoretical material removal rate of the silicon dioxide layer is determined to be RR1, the theoretical material removal rate of the silicon nitride layer is determined to be RR2, and the coefficient x=rr2/RR 1 is adjusted. In addition, given the target thickness TarThk = ThkT, a first predetermined time is set to Dt, the predetermined threshold range is [ T1, T2], dt e [ T1, T2], stored in the memory 300 together with the polishing recipe.
The first wafer is transferred to the measurement unit 3 of the wafer polishing apparatus 100 under the control of the controller 200, and the pre-thickness PreThk 1 of the first wafer, that is, the thickness of the silicon nitride layer of the first wafer is PreThk 1, is measured and stored in the memory 300.
The controller determines a first time of the first wafer as t 1 =dt and controls the first polishing head to absorb the first wafer in the first polishing disk to start a first polishing stage as a polishing time of the first stage in the polishing formulation.
After the first polishing stage is completed, the first wafer is transferred to a measurement unit, and the median thickness MidThk of the first wafer, that is, the thickness of the remaining portion of the silicon dioxide layer after the first polishing stage is completed, that is, offset 1 = MidThk, is measured. The amount of material removed RRT 1=PreThk1+Offset1 -ThkT at the second polishing stage of the first wafer is known.
The controller calculates a second time t2 1 of the first wafer by dividing the thickness of the remaining portion of the first layer by the theoretical material removal rate of the first layer, and by dividing the material removal rate of the second layer by the theoretical material removal rate of the second layer, and then calculates a second time, i.e., t2 1=Offset1/RR1+(PreThk1 -ThkT)/RR 2, by summing the first and second layer remaining polishing times. When the predetermined threshold range is set, the calculated second time of the first wafer should fall within the predetermined threshold range, i.e. T2 1 e [ T1, T2], so that the third time T3 1=t11.
The controller uses the second time t2 1 as the second stage polishing time in the polishing formulation, and controls the second polishing head to adsorb the first wafer so as to start the second polishing stage on the second polishing disk.
After the second polishing stage is completed, the first wafer is transferred to the measuring unit again under the control of the controller to measure the thickness, and the measured thickness is a thickness post value PostThk 1 of the first wafer.
To this end, a polishing process of the first wafer is completed, and the polishing time and the measured value are stored in a memory. It will be appreciated that the second wafer is then transferred to the measurement unit 3 under the control of the controller and a polishing process of the second wafer is started. The polishing process of the mth wafer from the second wafer is as follows (m > 1).
When the m-th wafer is polished under the control of the controller 200, one polishing process of the m-1 st wafer has been completed. The memory stores a first time t1 m-1, a third time t3 m-1, and a second time t2 m-1 for the m-1 th wafer, and a pre-thickness value PreThk m-1, a thickness Offset m-1 for the remainder of the first layer, and a post-thickness value PostThk m-1. It is also known that the adjustment value is Δt m-1=t3m-1-t1m-1, wherein the adjustment value is positive when the first time is increased, negative when the first time is decreased, and t3 m-1 is 0 when the first time is not adjusted.
The controller sets the first time of the mth wafer to t1 m=t3m-1 and corrects the thickness Offset m-1 of the remainder of the first layer of the mth-1 wafer to determine the thickness Offset m of the remainder of the first layer of the mth wafer, specifically Offsetm=(1-Δtm-1/t1m-1)*Offsetm-1=(2-t3m-1/t1m-1)*Offsetm-1.
Under the control of the controller, the mth wafer is transmitted to the measuring unit 3, the thickness pre-value PreThk m of the mth wafer is measured, and then the controller sequentially performs the following calculation steps:
Calculating the actual material removal amount of the m-th wafer in the second polishing stage, specifically RRT m=PreThk1-ThkT+Offsetm;
the actual material removal amount of the m-1 th wafer is calculated based on the corrected thickness of the remaining portion of the first layer of the m-1 th wafer, and the actual material removal rate of the m-1 th wafer in the second polishing stage is calculated based on the second polishing time t2 m-1, and the corrected RR m-1=(PreThkm-1-PostThkm-1+Offsetm)/t2m-1 is obtained.
And calculating a second time of the mth wafer according to the actual material removal amount of the mth wafer in the second polishing stage and the actual material removal rate of the mth wafer-1, and specifically t2 m=RRTm/RRm-1.
The second time T2 m is compared with a predetermined threshold range [ T1, T2], if T2 m e [ T1, T2], the first time T1 m is not adjusted, if T2 m < T1, the adjustment value is calculated, the first time T1 m is reduced according to the adjustment value, if T2 m > T2, the adjustment value is calculated, and the first time T1 m is increased according to the adjustment value. Specifically, a difference between the second time t2 m and the first predetermined time Dt is calculated, and then an adjustment value Δt m is determined according to the adjustment coefficient X and the difference, where the calculation formula is Δt m=(t2m -Dt). Preferably, the third time is calculated by equation t3 m=t1m+Δtm=t3m-1+Δtm, and it can be seen that if t2 m<T1,Δtm is negative, the polishing time of the first polishing stage will decrease, and if t2 m>T2,Δtm is positive, the polishing time of the first polishing stage will increase.
The controller controls the wafer polishing apparatus 100 to finish a polishing process of the mth wafer on the first polishing pad and the second polishing pad with the third time and the second time as polishing times of the first polishing stage and the second polishing stage in the mth wafer polishing recipe, respectively.
Finally, the controller controls the mth wafer to be transmitted to the measuring unit again for thickness measurement, and the measured thickness is a thickness post value PostThk m of the mth wafer.
To verify the polishing effect of the polishing method of this embodiment, the same batch of wafers was divided into two parts, and the wafer of the first part was polished by the wafer polishing method of this embodiment, with a target thickness of 700 angstroms (a), a first predetermined time dt=100 s, an adjustment coefficient x=25, a lower limit t1=80 of a predetermined threshold range, and an upper limit t2=110 of the predetermined threshold range, and the polishing results are shown in table 1 below.
TABLE 1
As shown in Table 1, overgrinding occurred at the N-2 th sheet, and the polishing time of the second polishing stage was automatically adjusted starting at the second time of the N-2 th sheet calculated from the post thickness value, but since the second time of the N-2 th sheet falls within the threshold range, the first time thereof was not adjusted, i.e., the third time was equal to the first predetermined time of 100s.
And continuously and automatically reducing and adjusting the polishing time of the second polishing stage when calculating the second time of the N-1 th piece, wherein the second time of the N-1 th piece is smaller than the lower limit of the preset threshold range, so that the first time is reduced, and the third time is calculated to be 99s.
The overgrinding still exists, so that the N-th piece and the N+1th piece are similar, and when the thickness is adjusted to the N+1th piece, the thickness post-value is only 2a difference from the target thickness.
Since the conventional polishing method is not suitable for the polishing scenario aimed at in this embodiment, for the comparative example constituted by the second partial wafer, only the polishing time of the second stage was adjusted, and the value after each adjustment was set to be the same as the second time of the first partial wafer, and the polishing results are shown in table 2 below.
TABLE 2
It can be seen that when the second time is only adjusted to the n+1th wafer, the difference between the post-thickness value and the target thickness is still 10 a, which is 5 times that of the polishing method of this embodiment, and the response speed is controlled to be far higher than that of this embodiment, so that more wafers are required to adjust the post-thickness value to the same level as that of this embodiment, which seriously affects the polishing accuracy. It will be appreciated that although better polishing accuracy can be obtained by reducing the second time more greatly, there is a limit in the amount of adjustment of the second time, and this will cause the difference between the first time and the second time to be larger and larger, and still adversely affect the control accuracy and the production efficiency.
Therefore, the wafer polishing method realizes the adjustment of the polishing time of the first polishing stage and the second polishing stage respectively, and the polishing time of the two stages is adjusted in a linkage way and in a grading way. The linkage adjustment can be used for adjusting the polishing thickness more quickly and controlling the polishing thickness more accurately, so that the final thickness of the wafer has better consistency, and meanwhile, the total polishing time of the wafer reaching the target thickness is reduced, and the problem that the polishing thickness cannot be accurately controlled or the final thickness cannot be adjusted in time only by adjusting the polishing time of the second polishing stage, so that the production efficiency and the yield of the wafer are affected is avoided. The step adjustment ensures that linkage adjustment is performed only when the adjustment amplitude is large, so that the stability and continuity of the first polishing stage are ensured as much as possible, meanwhile, the control complexity of the wafer processing equipment is not remarkably improved, the production efficiency of chemical mechanical polishing is greatly improved, and meanwhile, the yield of wafers is also improved and ensured.
Meanwhile, the wafer polishing method does not use the large disc torque change as an endpoint measuring means, ensures that polishing can stably stop on the second layer, avoids measurement failure caused by polishing pad damage, and reduces the risk of less grinding or overgrinding of the wafer. Moreover, the invention uses a two-disc process to realize two polishing stages, namely linkage adjustment of polishing time of two discs. Therefore, when one polishing disc is deteriorated in polishing conditions due to the loss of the polishing pad, double-disc linkage adjustment can be automatically performed, the problem that a plurality of wafers are thinner or thicker is avoided, the fault tolerance rate and the reaction time of the system are increased, the service life of the polishing pad is prolonged, and the production cost is greatly reduced.
The above embodiments are only for illustrating the embodiments of the present invention, but not for limiting the embodiments of the present invention, and various changes and modifications may be made by one skilled in the relevant art without departing from the spirit and scope of the embodiments of the present invention, so that all equivalent technical solutions also fall within the scope of the embodiments of the present invention, and the scope of the embodiments of the present invention should be defined by the claims.