Polishing pad conditioning method, apparatus, device, controller, and storage medium

Abstract

The present application relates to a polishing pad conditioning method, device, equipment, controller and storage medium, wherein the initial conditioning formula is used when conditioning the polishing pad in the initial conditioning cycle, and the conditioning formula adjusted by the self-learning adjustment method is used when conditioning in the subsequent conditioning cycle. The present application realizes controllable conditioning of the polishing pad through the polishing pad conditioning method, adjusts the conditioning parameters according to the wear state of the polishing pad, optimizes the conditioning parameters, prolongs the service life of the polishing pad, and reduces the replacement frequency and cost of the polishing pad.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of chemical mechanical planarization (CMP) of semiconductor wafers, and more particularly, to a polishing pad conditioning method, device, apparatus, controller and storage medium of a chemical mechanical polishing apparatus. BACKGROUND

[0002] Integrated circuits (ICs) are the core and lifeblood of the information technology industry. An integrated circuit is generally formed by successively depositing conductive, semiconductive or insulating layers on a silicon wafer. The wafer surface is thus deposited with a thin film of fill material. In the manufacturing process, the fill material needs to be continuously planarized until the top surface is patterned to form conductive paths between raised patterns.

[0003] Chemical mechanical polishing (CMP) is the preferred planarization process in IC manufacturing. In CMP, over-polishing can damage the material properties and affect device performance for semiconductor device manufacturing processes, and under-polishing can require additional processing steps, increasing costs.

[0004] The polishing pad plays a crucial role in the chemical mechanical polishing process, responsible for storing and delivering polishing liquid, removing processing residues, transmitting mechanical load, and maintaining the polishing environment. During the polishing process, the polishing pad surface gradually wears, reducing polishing efficiency and quality, so the polishing pad needs to be conditioned during use. The commonly used conditioning methods, such as diamond conditioner conditioning, have good mechanical grinding effect when conditioning the polishing pad due to the high hardness of diamond. However, current conditioning methods mainly rely on empirical formulas, lack comprehensive theoretical models, and have shortcomings such as long conditioning time, short polishing pad life, uneven removal efficiency, poor accuracy, etc. SUMMARY

[0005] The present application aims to provide a polishing pad conditioning method, device, apparatus, controller and storage medium, which uses intelligent adjustment of conditioning parameters to improve conditioning accuracy and efficiency, and improves the uniformity of the polishing pad and prolongs the service life.

[0006] To achieve the above-mentioned purpose, the first aspect of the present application provides a polishing pad conditioning method, which comprises the following steps:

[0007] determining a conditioning cycle of the polishing pad;

[0008] updating the conditioning recipe of the polishing pad once every conditioning cycle after the polishing pad is used to perform a polishing job,

[0009] Wherein, an initial conditioning recipe is used to condition the polishing pad in the first conditioning cycle, and a conditioning recipe adjusted by a self-learning adjustment method is used to condition the polishing pad in each subsequent conditioning cycle.

[0010] In the polishing pad conditioning method as described above, optionally, the conditioning cycle is an integer multiple of the time length for polishing each box of wafers to be polished, and the parameters of the conditioning recipe include conditioning time length, conditioning pressure, and the movement speed of the conditioner relative to the polishing pad.

[0011] In the polishing pad conditioning method as described above, optionally, the self-learning adjustment method comprises the following steps:

[0012] Step I: using a topography detection assembly to collect current topography data of the conditioned polishing pad;

[0013] Step II: calculating the actual conditioning removal amount of the polishing pad based on the current topography data and the historical topography data of the previous cycle of the polishing pad, collecting the conditioning time length for conditioning the polishing pad this time, and fitting a polishing pad conditioning removal amount calculation formula according to the actual conditioning removal amount and the conditioning time length;

[0014] Step III: calculating the predicted conditioning removal amount of the next cycle based on the conditioning removal amount calculation formula, calculating the predicted topography data of the polishing pad after conditioning in the next cycle according to the predicted conditioning removal amount of the next cycle and the current topography data, and calculating the expected conditioning adjustment coefficient of the conditioning recipe parameters through the deviation between the predicted topography data and the target conditioning topography data;

[0015] Step IV: determining the weights of system random influence and hardware parameter influence based on historical conditioning adjustment coefficients, and performing weighted processing on the expected conditioning adjustment coefficient;

[0016] Step V: outputting the weighted expected conditioning adjustment coefficient, and outputting a new conditioning recipe based on the weighted expected conditioning adjustment coefficient for conditioning the polishing pad in the next cycle.

[0017] In the polishing pad conditioning method as described above, optionally, the steps I to V are performed on the polishing pad in a region-by-region manner.

[0018] In the polishing pad conditioning method as described above, optionally, in Step I, the topography detection assembly collects the current topography data by using a contact sensor and / or a non-contact sensor.

[0019] In the polishing pad conditioning method as described above, optionally, the step of fitting the polishing pad conditioning removal amount calculation formula comprises:

[0020] establishing a linear regression model y=kx+b, wherein y is the dressing removal amount of the polishing pad, x is the dressing time length of the polishing pad, k is the dressing removal rate of the polishing pad, and b is the intercept term of the linear model;

[0021] collecting actual dressing removal amounts y of data points of each region of the polishing pad i and dressing time lengths x i ;

[0022] defining an error function, and selecting initial k and b values, and setting a tolerance;

[0023] using an iterative method to approximate k and b, and stopping iteration and outputting k and b when the value of the error function is less than or equal to the tolerance.

[0024] In the polishing pad dressing method as described above, optionally, the error function is a mean square error function:

[0025]

[0026] wherein m is the number of data points.

[0027] In the polishing pad dressing method as described above, optionally, the iterative equation is:

[0028]

[0029] wherein, is the partial derivative of the error function M with respect to k, is the partial derivative of the error function with respect to b, and a is a step size for controlling the size of each step update.

[0030] In the polishing pad dressing method as described above, optionally, the initial k and b values are randomly selected values, or the initial k and b values are calculated by the following formula:

[0031]

[0032]

[0033] Σx is the sum of all x i , Σy is the sum of all y i , Σxy is the sum of all x i y i , Σx 2 is the sum of all x i 2 , and n is the number of regions of the polishing pad.

[0034] In order to achieve the above-mentioned purpose, the second aspect of the present invention provides a polishing pad dressing device for chemical mechanical polishing equipment, wherein the polishing pad dressing device includes a fixed seat, a swing arm and a dresser, wherein the dresser is connected to the fixed seat through the swing arm, and the swing arm can swing around the fixed seat, and the dresser located at the end of the swing arm dresses the polishing pad by a dressing method as described in any one of the first aspects above.

[0035] In order to achieve the above-mentioned object, a third aspect of the present invention provides a chemical mechanical polishing device, which includes the polishing pad dressing device as described in the second aspect.

[0036] In order to achieve the above object, a fourth aspect of the present invention provides an intelligent controller, comprising:

[0037] a memory for storing computer-executable instructions or computer programs;

[0038] A processor, wherein when the processor is configured to execute computer executable instructions or computer programs stored in the memory, the processor implements the dressing method described in any one of the first aspects to dress the polishing pad used in the chemical mechanical polishing equipment.

[0039] In order to achieve the above-mentioned purpose, the fifth aspect of the present invention provides a computer-readable storage medium, which stores computer-executable instructions or computer programs. When the computer-executable instructions or computer programs are executed by the processor, the polishing pad used for the chemical mechanical polishing equipment is trimmed by the trimming method described in any one of the first aspects above.

[0040] The present invention provides a polishing pad conditioning method, apparatus, device, controller, and storage medium. This conditioning method utilizes an initial conditioning formula during an initial conditioning cycle, and utilizes a conditioning formula adjusted using a self-learning adjustment method during subsequent conditioning cycles. The present invention utilizes a controllable conditioning method for polishing pads, adjusting conditioning parameters based on the wear state of the polishing pad to optimize the conditioning effect, extend the service life of the polishing pad, and reduce the frequency and cost of polishing pad replacement.

[0041] Other features and advantages of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description only represent some of the embodiments of the present application, and those skilled in the art can also obtain other drawings from these drawings.

[0043] Figure 1 is a flow chart of the polishing pad trimming method provided by the embodiments of the present application;

[0044] Figure 2 is a schematic diagram of the implementation effect of the conventional trimming method provided by the embodiments of the present application;

[0045] Figure 3 is a schematic diagram of the implementation effect of the trimming method example provided by the embodiments of the present application;

[0046] Figure 4 is a structural schematic diagram of the trimming device for chemical mechanical polishing provided by the embodiments of the present application;

[0047] Figure 5 is a structural schematic diagram of the chemical mechanical polishing equipment provided by the embodiments of the present application. DETAILED DESCRIPTION

[0048] In order for those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the embodiments of the present application, all other embodiments obtained by those skilled in the art should belong to the scope of protection of the embodiments of the present application.

[0049] It should be understood that each step described in the method embodiments of the present application can be executed in different order and / or in parallel. In addition, the method embodiments can include additional steps and / or omit the execution of the steps shown. The scope of the present application is not limited in this respect.

[0050] The term "comprising" and its variants as used herein are open-ended, that is "including but not limited to". The term "based on" is "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". In the description of the specification, the illustrative representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the described specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner. The related definitions of other terms will be given in the following description.

[0051] During actual use, the polishing pad will be worn and deformed, and the polishing debris generated during polishing will fill the micro-pores on the surface of the polishing pad, resulting in a glazing phenomenon, which reduces the polishing pad's ability to store and deliver abrasive materials, and leads to a decrease in material removal rate. Therefore, the glazing layer on the surface of the polishing pad needs to be removed to increase the roughness of the polishing pad. During use, the polishing pad will also be deformed, so the polishing pad also needs to be trimmed to make the polishing pad more flat. Proper trimming can restore the excessive wear and glazing of the polishing pad surface and extend the service life of the polishing pad. However, after trimming, the thickness of the polishing pad is reduced, and excessive trimming can lead to excessive wear of the polishing pad and reduce the service life of the polishing pad. If the polishing pad is worn too quickly, the loss of the polishing pad can be reduced by optimizing the parameters of the trimming recipe.

[0052] Figure 1 is a flowchart of the polishing pad trimming method provided by the embodiments of the present application.

[0053] In this embodiment, the trimming period of the polishing pad is determined. According to the use of the polishing pad and the polishing effect, the polishing pad needs to be trimmed using a trimming recipe, and the time interval for adjusting the trimming recipe is the trimming period. The trimming recipe of the polishing pad is updated once for each polishing job performed using the polishing pad.

[0054] In optional embodiments, the frequency of polishing pad trimming in a suitable trimming period can be selected according to the type of polishing pad and the polishing process requirements, for example, the polishing pad is trimmed once for each polishing job performed using the polishing pad.

[0055] In this embodiment, the trimming period is an integer multiple of the time required to complete the polishing of each batch of wafers to be polished. The advantage of this implementation is that it can ensure that the same trimming recipe is used for the polishing pad during the polishing of each batch of wafers, thereby maintaining stable polishing efficiency.

[0056] In some embodiments, the parameters of the trimming recipe of the polishing pad include trimming duration, trimming pressure, and the speed of the trimmer relative to the polishing pad. The trimming duration refers to the length of time for which the trimmer trims the polishing pad. It can be understood that a trimming duration that is too long can cause excessive wear of the polishing pad, and a trimming duration that is too short can not be able to sufficiently trim the polishing pad to achieve appropriate performance. The speed of the trimmer relative to the polishing pad affects the uniformity and efficiency of the trimming effect. The trimming pressure, i.e., the pressure applied by the trimmer to the polishing pad, ensures that the surface of the polishing pad is effectively trimmed with appropriate trimming pressure. In optional embodiments, the parameters of the trimming recipe of the polishing pad can also include trimming depth and trimming temperature, and other parameters that affect the trimming effect of the polishing pad.

[0057] In the polishing pad conditioning method of the present invention, an initial conditioning recipe must be pre-determined based on historical data and experience. This initial conditioning recipe refers to the conditioning recipe used during the initial conditioning of the polishing pad. Conditioning parameters in this initial conditioning recipe, such as conditioning pressure, conditioning speed, and conditioning time, are determined based on the type of polishing pad and the requirements of the polishing process. For more precise calculations, the surface of the polishing pad is divided into multiple regions based on the shape and size of the polishing pad, and initial polishing parameters are set for each region.

[0058] During subsequent trimming, the trimming formula adjusted by the self-learning adjustment method is used, and the trimming parameters are automatically optimized through the algorithm. The self-learning adjustment method of the trimming formula includes the following steps:

[0059] Step I: using a topography detection component to collect current topography data of the trimmed polishing pad;

[0060] Step II: Calculate the actual dressing removal amount of the polishing pad based on the current topography data and the historical topography data of the polishing pad in the previous cycle, collect the dressing time used for the polishing pad in this dressing process, and fit the polishing pad dressing removal amount calculation formula based on the actual dressing removal amount and the dressing time;

[0061] Step III: Calculating the predicted dressing removal amount for the next cycle based on the dressing removal amount calculation formula, and calculating the predicted topography data of the polishing pad after dressing for the next cycle based on the predicted dressing removal amount for the next cycle and the current topography data. Calculating the expected dressing adjustment coefficient of the dressing recipe parameters based on the deviation between the predicted topography data and the target topography data.

[0062] Step IV: Determine the weights of system random effects and hardware parameter effects based on historical trim adjustment coefficients, and perform weighted processing on the expected trim adjustment coefficients;

[0063] Step V: Outputting the weighted expected dressing adjustment coefficient, and outputting a new dressing recipe based on the weighted expected dressing adjustment coefficient for the polishing pad dressing of the next cycle.

[0064] In step I, the topography data of the polishing pad refers to the profile of the polishing pad, that is, the collection of height information of various areas on the surface of the polishing pad. In different embodiments, the topography detection component can be a contact sensor that measures the surface topography of the polishing pad by directly contacting the polishing pad. Other detection devices, such as non-contact sensors such as acoustic sensors and optical sensors, all fall within the scope of protection of the present invention. In some embodiments, a contact sensor can be used to measure the roughness and pressure distribution of the polishing pad, while a non-contact sensor can also be used to measure the surface profile and thickness of the polishing pad. In an optional embodiment, the topography of the polishing pad can be obtained by having a sensor abut the surface of the polishing pad and follow the undulations of the polishing pad, and measuring the distance between the sensor and the polishing disk below the polishing pad.

[0065] In step II, the actual removal amount of the polishing pad is the amount of material removed from the polishing pad during the polishing process, and the value of the actual removal amount of the polishing pad is calculated according to the difference between the current topography data of the polishing pad and the topography data of the polishing pad in the last cycle.

[0066] In optional embodiments, the calculation formula of the trimming removal amount of the polishing pad can include:

[0067] A linear regression model y = kx + b is established, where y is the trimming removal amount of the polishing pad, x is the trimming duration of the polishing pad, k is the trimming removal rate of the polishing pad, and b is the intercept term of the linear model, and then the actual trimming removal amount y of the data points of each region of the polishing pad is collected i and the trimming duration x i . The linear regression model assumes that there is a linear relationship between the trimming removal amount y of the polishing pad and the trimming duration x of the polishing pad, and by finding the best k value and b value, the difference between the polishing pad trimming removal amount predicted by the model and the actual polishing pad trimming removal amount is minimized. By establishing a linear regression model, the best match of the trimming removal amount function of each region of the polishing pad is found, so that the data points of each region of the polishing pad are as close as possible to the trimming removal amount calculation formula of each region. In optional embodiments, other linear or nonlinear regression models can also be selected.

[0068] In optional embodiments, an error function is defined to measure the difference between the predicted polishing pad trimming removal amount and the actual polishing pad trimming removal amount, and an initial k value and b value are selected, and a tolerance is set. It can be understood that the tolerance is the maximum error range that the model can accept.

[0069] In some embodiments, the error function can be a mean square error function, and the equation is:

[0070]

[0071] Where m is the number of data points.

[0072] It can be understood that the mean square error function can be used to evaluate the accuracy of the model prediction, and in the training process of the model, the mean square error function can be used as a loss function, and by minimizing the value of the mean square error function, the model parameters are adjusted to improve the prediction performance of the model.

[0073] In some optional embodiments, an iterative method can be used to approximate k, b, and when the value of the error function is less than or equal to the tolerance, the iteration is stopped. In some embodiments, the iterative equation of k, b is:

[0074]

[0075] Where is the partial derivative of the error function M with respect to k, is the partial derivative of the error function with respect to b, a is the step size, used to control the size of the update at each step.

[0076] In some examples, the tolerance is usually set to a small value to ensure that the model can find the most suitable solution in the iteration process.

[0077] In optional embodiments, the initial k value and b value can be random values or estimated values.

[0078] As a feasible implementation, the initial k value and b value can be calculated by the least square method, wherein

[0079] The calculation equation of k is:

[0080]

[0081] The calculation equation of b is:

[0082]

[0083] wherein, Σx is the sum of all x i , Σy is the sum of all y i , Σxy is the sum of all x i y i , Σx 2 is the sum of all x i 2 , and n is the number of data points in the data set.

[0084] It can be understood that the initial k value and b value estimated by the least square method can help the above iteration algorithm to converge to the optimal solution more quickly.

[0085] In step III, the predicted trimming removal amount of the next cycle is calculated according to the trimming removal amount calculation formula, and the predicted topography data of the polishing pad after trimming in the next cycle can be preliminarily calculated according to the predicted trimming removal amount of the next cycle and the current topography data. By comparing the predicted topography data with the target trimming topography data, the deviation, i.e. the expected trimming removal amount of the polishing pad in the next cycle, is calculated, and the expected trimming adjustment coefficient of the trimming recipe parameter is calculated by the expected trimming removal amount of the polishing pad in the next cycle. The trimming adjustment coefficient is used to modify the coefficients corresponding to various parameters such as trimming time, trimming pressure and movement speed of the trimmer relative to the polishing pad in the trimming recipe. Adjusting the trimming adjustment coefficient means adjusting the coefficients of modifying various parameters. The trimming adjustment coefficient will be affected by various factors, such as random effects of hardware parameters and systems. By adjusting the weights of these factors, the trimming adjustment coefficient is adjusted to achieve the expected trimming effect.

[0086] In step IV, the hardware parameters, such as the hardness, elasticity and granularity of the polishing pad, have an impact on the polishing pad during the conditioning process. The accuracy of the model can be improved by adjusting the desired conditioning adjustment coefficient in a weighted manner.

[0087] It can be understood that the random influence of the system is formed by a variety of unpredictable reasons, such as temperature changes, material inhomogeneity and random jitter of the system, which can affect the accuracy of the data. By adjusting the weight of the random influence of the system, the desired conditioning adjustment coefficient of the polishing pad can be optimized.

[0088] In optional embodiments, the desired conditioning adjustment coefficient can be more accurately adjusted by comparing the conditioning adjustment coefficients in adjacent conditioning processes, and the influence of abnormal values can be reduced by analyzing similar data points.

[0089] In step V, a new conditioning recipe is formulated according to the weighted desired conditioning adjustment coefficient of the polishing pad, and the new conditioning recipe is implemented in the conditioning of the polishing pad in the next cycle.

[0090] Figure 2 and Figure 3 are effect diagrams of the traditional conditioning method and the polishing pad conditioning method provided by the embodiments of the present application.

[0091] In Figure 2 and Figure 3 , the horizontal axis represents the radius of the polishing pad in inches, and the vertical axis represents the groove depth of the polishing pad in mils. The trend lines in the figures show the different conditioning effects of the polishing pad in the traditional conditioning method and the polishing pad conditioning method provided by the present application. Different trend lines are marked as 0H, 5H, 10H, 15H and 20H, which are hours, representing the time period in the conditioning process of the polishing pad. From Figure 2 and / or Figure 3 It can be seen that the longer the polishing pad is in the conditioning process, the worse the conditioning effect of the polishing pad, and the more unstable the trend line.

[0092] As shown in Figure 2 and Figure 3 , compared with the traditional conditioning method, the trend line of the polishing pad conditioning method provided by the present application is more stable, which may mean that the polishing pad conditioning method provided by the present application adjusts the conditioning recipe to make the polishing pad conditioning more flat, so that the polishing pad can maintain better uniformity during conditioning, and the stability and repeatability of the conditioning process are improved.

[0093] Figure 4 is a structural diagram of a conditioning device for chemical mechanical polishing provided by the embodiments of the present application.

[0094] As shown in Figure 4 , the trimming device 100 includes a fixed seat 101, a swing arm 102, and a trimmer 103. The trimmer 103 includes a trimming head 104 and a topography detection device (not shown in the figure). In an optional embodiment, the bottom of the trimmer 103 is inlaid with diamonds, which have high hardness and corrosion resistance. Through the reciprocating motion of the trimming head 104 on the surface of the polishing pad, the diamond particles cut and trim the polishing pad. The topography detection device can measure the topography of the polishing pad. In an optional embodiment, the type of the topography detection device is not limited, such as non-contact sensors such as eddy current sensors and contact sensors such as contact displacement sensors, which fall within the protection scope of the present application. In some embodiments, the topography detection device and the trimming head 104 are arranged horizontally adjacent to each other. This arrangement has the advantage that the trimming head 104 does not interfere with the measurement of the topography detection device during the trimming of the polishing pad.

[0095] In an optional embodiment, the trimmer 103 is connected to the fixed seat 101 through the swing arm 102. By swinging the swing arm 102 around the fixed seat 101, the trimmer 103 located at the end of the swing arm 102 can perform reciprocating motion to trim the polishing pad.

[0096] Figure 5 is a structural schematic diagram of a chemical mechanical polishing equipment provided by an embodiment of the present application.

[0097] Figure 5 In the embodiment shown in Figure 4 , the chemical mechanical polishing equipment includes a trimming device 100, and the structure of the trimming device 100 is as shown in Figure 5 . In the embodiment shown in Figure 5 , the chemical mechanical polishing equipment further includes a polishing disc 200, a carrier head 300, and a liquid supply device 400. In the example shown in , the polishing disc 200 covers a polishing pad to be used for polishing a wafer, the carrier head 300 presses the wafer against the polishing pad on the polishing disc 200, and the liquid supply device 400 sprays a polishing liquid onto the surface of the polishing pad. Under the chemical action of the polishing liquid, the wafer is rubbed against the polishing pad through the relative motion of the carrier head 300 and the polishing disc 200 to perform polishing. In the process of polishing, the trimming device 100 trims the polishing pad. Specifically, the trimming device 100 can trim the surface deformation of the polishing pad generated in the polishing process of the wafer, so that the topography of the polishing pad meets the process requirements, thereby stabilizing the polishing removal rate of the wafer, achieving global planarization of the wafer, and prolonging the service life of the polishing pad.

[0098] The application further provides an intelligent controller, comprising a memory and a processor.

[0099] Further, the application further provides a computer readable storage medium, which stores computer executable instructions or computer programs, and the computer executable instructions or computer programs are executed by a processor to realize the dressing method in any one of the preceding embodiments to dress the polishing pad of the chemical mechanical polishing equipment.

[0100] It should be noted that the computer readable storage medium of the application can be, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the application, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus.

[0101] The above embodiments are only used to illustrate the application, and not to limit the application. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the application. Therefore, all equivalent technical solutions belong to the scope of the application, and the patent protection scope of the application should be defined by the claims.

Claims

1. A method of conditioning a polishing pad, characterized by, The method comprises the following steps: determining a conditioning cycle of the polishing pad; updating a conditioning recipe of the polishing pad after each conditioning cycle of the polishing pad is performed using the polishing pad, wherein an initial conditioning recipe is used when conditioning the polishing pad in a first conditioning cycle, and a conditioning recipe adjusted by a self-learning method is used when conditioning the polishing pad in each subsequent conditioning cycle; the conditioning cycle is an integer multiple of a time length for completing polishing of each box of wafers to be polished, and parameters of the conditioning recipe include a conditioning time length, a conditioning pressure, and a movement speed of a conditioner relative to the polishing pad; the self-learning method comprises the following steps: Step I: using a topography detection assembly to collect current topography data of the polishing pad after conditioning; Step II: calculating an actual conditioning removal amount of the polishing pad based on the current topography data and historical topography data of a previous cycle of the polishing pad, collecting a conditioning time length used for conditioning the polishing pad this time, and fitting a polishing pad conditioning removal amount calculation formula according to the actual conditioning removal amount and the conditioning time length; Step III: calculating a predicted conditioning removal amount of a next cycle based on the conditioning removal amount calculation formula, calculating predicted topography data of the polishing pad after conditioning in the next cycle according to the predicted conditioning removal amount of the next cycle and the current topography data, and calculating an expected conditioning adjustment coefficient of a conditioning recipe parameter through a deviation between the predicted topography data and target conditioning topography data; Step IV: determining weights of system random influence and hardware parameter influence based on historical conditioning adjustment coefficients, and performing weighted processing on the expected conditioning adjustment coefficient; Step V: outputting the weighted expected conditioning adjustment coefficient, and outputting a new conditioning recipe based on the weighted expected conditioning adjustment coefficient for conditioning of the polishing pad in the next cycle.

2. The method of claim 1, wherein the polishing pad conditioner is a diamond tip. The steps I to V are all performed on a region of the polishing pad.

3. The method of claim 2, wherein the polishing pad conditioner is a diamond tip. In the step I, the topography detection assembly collects the current topography data by using a contact sensor and / or a non-contact sensor.

4. The method of claim 2, wherein the polishing pad conditioner is a diamond tip. The step of fitting the polishing pad conditioning removal amount calculation formula comprises: establishing a linear regression model y=kx+b, where y is a conditioning removal amount of the polishing pad, x is a conditioning time length of the polishing pad, k is a conditioning removal rate of the polishing pad, and b is an intercept term of the linear model; collecting actual trim removal amounts y of data points of each region of the polishing pad i and trim length x i ; defining an error function, and selecting initial k and b values and setting a tolerance; using an iterative method to approximate k and b, and stopping iteration and outputting k and b when a value of the error function is less than or equal to the tolerance.

5. The method of claim 4, wherein the polishing pad conditioner is a diamond tip. The error function is a mean square error function: where m is a number of data points.

6. The method of claim 4, wherein the polishing pad conditioner is a diamond tip. The iterative equation is: wherein, is the partial derivative of the error function M with respect to k, is the partial derivative of the error function with respect to b, and a is a step size used to control the size of the update at each step.

7. The method of claim 4, wherein the polishing pad conditioner is a diamond tip. 5 The initial k and b values are randomly selected values, or the initial k and b values are calculated by the following formula: ∑x is the sum of all x i ∑y is the sum of all y i ∑xy is the sum of all x i y i ∑x 2 is the sum of all x i 2 is the sum of all x, n is the number of areas of the polishing pad.

8. A polishing pad conditioning apparatus for a chemical mechanical polishing apparatus, characterized by, The polishing pad conditioning device comprises a fixed seat, a swing arm, and a conditioner, the conditioner is connected to the fixed seat through the swing arm, the swing arm can swing around the fixed seat, and the conditioner at an end of the swing arm uses the conditioning method according to any one of claims 1 to 7 to condition the polishing pad.

9. A chemical mechanical polishing apparatus characterized by comprising: The chemical mechanical polishing equipment comprises the polishing pad conditioning device according to claim 8.

10. An intelligent controller characterized by, The intelligent controller comprises: a memory for storing computer executable instructions or computer programs; a processor for executing the computer executable instructions or computer programs stored in the memory to implement the dressing method according to any one of claims 1 to 7 to dress the polishing pad for a chemical mechanical polishing device.

11. A computer readable storage medium, characterized in that, The computer readable storage medium stores computer executable instructions or computer programs, and the computer executable instructions or computer programs are executed by a processor to implement the dressing method according to any one of claims 1 to 7 to dress the polishing pad for a chemical mechanical polishing device.

Images