SEMICONDUCTOR DEVICE PRODUCTION METHOD AND SEMICONDUCTOR PRODUCTION APPARATUS

Abstract

A semiconductor device production method of the present invention first collects data including an initial volume of plating solution, volume of replenished solution, number of wafers processed, value of current applied and volume of waste solution in a step of filling a metal plating film in a via hole or a trench formed in an insulating film on a semiconductor substrate. Then, a cumulative charge during the plating is calculated based on the obtained current value. Also, a total volume of plating solution is calculated. Furthermore, an amount of decomposition products of suppressors contained in the plating solution based on the calculated total volume of plating solution, the volume of waste solution and the calculated cumulative charge. The semiconductor substrate is plated only when the amount of decomposition products is equal to or smaller than a predetermined threshold.

Description

CROSS-REFERENCE TO RELATED APPLICATION

The present application claims the benefit of Japanese Patent Application No. 2008-060835 filed Mar. 11, 2008, the subject matter of which is incorporated herein by reference.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to a semiconductor device production method and a semiconductor production apparatus and particularly to a plating method and a plating apparatus used in a metal wiring process for semiconductor devices.

2. Description of the Related Art

FIG. 9 is a schematic diagram showing a structure of a prior art electroplating apparatus. This electroplating apparatus comprises a circulation line 101, a plating tank 102, a cathode 104, an anode 105, a wafer 106, a filter 107, an additive concentration measurement unit 108, a CPU 109, and an additive supply unit 110.

In prior art electroplating, metal-plating is performed while additives are supplied (reference is made, for example, to Japanese Laid-Open Patent Application No. 2001-152398). The concentration of additives are measured based on the current-potential curve in some cases (for example, Kimiko OYAMADA et al., “Plating Time Dependence of Filling Ability with Additives by Copper Electroplating”, Journal of Japan Institute of Electronics Packaging, Vol. 7, No. 3, pp. 261-265 (2004)).

FIG. 10 is a flowchart of the prior art plating process. As shown in FIG. 10, the prior art plating process comprises the steps of replenishing the plating solution (S101), measuring the additive concentration (S102), starting a recipe (S103), loading a wafer (S104), plating (S105), unloading the wafer (S106), and ending the recipe (S107).

In prior art plating, the concentration of constituents of additives, such as brightener, leveler and suppressor, contained in the plating solution are measured during electroplating. Each constituent is added as needed so to maintain the concentration in a specific range during electroplating.

Specifically, the concentrations of constituents of additives are measured based on the current-potential curve of the plating solution containing the additive obtained at the CV (cyclic voltammetry) electrode placed in the plating solution within the plating apparatus.

SUMMARY OF THE INVENTION

It is difficult in the prior art electroplating method and electroplating apparatus to measure and assess the concentration of all constituents present in the plating solution including by-products produced in the course of plating. Therefore, it is difficult to fill a metal in the wiring regions consisting of nanoscale microstructures in semiconductor devices without creating any voids.

The purpose of the present invention is to fill a metal in wiring regions consisting of nanoscale microstructures in a semiconductor device without creating any voids and seams so as to reduce electric defects, thereby improving the production yield of semiconductor devices.

In order to achieve the above purpose, a semiconductor device production method of the present invention first collects data including an initial volume of plating solution, volume of replenished solution, number of wafers processed, value of current applied and volume of waste solution in a step of filling a metal plating film in a via hole or a trench formed in an insulating film on a semiconductor substrate. Then, a cumulative charge during the plating is calculated based on the obtained current value. Also, a total volume of plating solution is calculated. Furthermore, an amount of decomposition products of suppressors contained in the plating solution based on the calculated total volume of plating solution, the volume of waste solution and the calculated cumulative charge.

It is preferable in the above semiconductor device production method that a predetermined volume of waste solution when the amount of decomposition products exceeds a predetermined value and fresh plating solution be replenished.

A semiconductor production apparatus of the present invention is used to fill a metal plating film in a via hole or a trench formed in an insulating film on the semiconductor substrate. The semiconductor production apparatus of the present invention comprises a unit circulating plating solution, a unit replenishing plating solution and a unit discharging the plating solution. The apparatus further comprises a unit collecting data including an initial volume of plating solution, volume of replenished solution, number of wafers processed, value of current applied and volume of waste solution, a unit calculating a cumulative charge during plating based on the obtained current value, a unit a total volume of plating solution and a unit calculating an amount of decomposition products of suppressors contained in the plating solution based on the calculated total volume of plating solution, the volume of waste solution and the calculated cumulative charge.

It is preferable for the above semiconductor production apparatus that the unit discharging the plating solution discharges a predetermined volume of plating solution when the amount of decomposition products exceeds a predetermined value.

The present invention prevents defective filling due to fluctuations in the plating solution composition by calculating and controlling the amount of by-products (the amount of decomposition products) produced during electroplating, thereby improving the production yield of semiconductor devices.

The foregoing and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram showing a structure of an electroplating apparatus relating to a first embodiment of the present invention.

FIG. 2 is a graphical representation showing a relationship between a calculated amount of by-products in plating solution and defects occurrence.

FIG. 3 is a schematic illustration showing a defective filling of a via hole and trench.

FIG. 4 is a schematic illustration showing a mechanism of a defective filling of a via hole and trench.

FIG. 5 is a graphical representation showing liquid chromatography measurements based on the calculated amount of by-products in the plating solution.

FIG. 6 is a flowchart of an electroplating process relating to a second embodiment of the present invention.

FIG. 7 is a graphical representation showing waste solution ratios simulation results relating to a third embodiment of the present invention.

FIG. 8 is a graphical representation showing a method of determining a waste solution ratio relating to the third embodiment of the present invention.

FIG. 9 is a schematic diagram showing a structure of a prior art electroplating apparatus.

FIG. 10 is a flowchart of the prior art electroplating process.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

First Embodiment

A first embodiment of the present invention is described hereafter with reference to FIG. 1.

FIG. 1 is a schematic diagram showing the structure of an electroplating apparatus relating to the first embodiment. The electroplating apparatus comprises a plating chamber 21 having a partition 22. An anode 14 is placed at the bottom of the plating chamber 21. Facing the anode 14, a cathode 10 holds a wafer 11 in a manner such that the surface to be treated is oriented toward the anode 14. A filter 13 and a diffusion plate 12 are provided between the anode 14 and cathode 10 in this order from the anode 14. The electroplating apparatus of this embodiment further comprises a circulation line 3 and a sub-circulation line 4 as the plating solution circulation path. The circulation line 3 is a circulation path allowing the plating solution from the top end of the plating chamber 21 to re-enter between the diffusion plate 12 and filter 13 via a plating tank 2 and a filter 7. The sub-circulation line 4 is a circulation path allowing the plating solution from the bottom of the plating chamber 21 to re-enter between the filter 13 and anode 14 via a filter 8. A supply valve 9 for introducing fresh plating solution into the circulation line 3 is provided on the circulation line 3 upstream of the plating tank 2. A waste valve 5 for discharging plating solution from the circulation path is provided on the sub-circulation line 4 upstream of the filter 8. A control unit 1 controls opening/closing of the supply valve 9 and waste valve 5 based on data obtained by a monitoring unit 6, described later. In this embodiment, the control unit 1 further controls operations of the electroplating apparatus.

In FIG. 1, in the case that the electroplating apparatus is used for plating copper, the plating solution is circulated and the wafer 11 is rotated at a rotating speed of 5 to 100 rpm while a current of 1 A to 50 A is applied to the wafer 11 placed on the cathode 10 so as to form a plating film. The plating solution circulates through the main circulation line 3 including the plating tank 2 during the plating. The plating solution gathering near the anode 14 circulates through the sub-circulation line (SAC: separate anode chamber line) 4 near the anode 14. Here, the plating solution mainly consists of copper sulfate and additives.

Here, three additives are used, including a suppressor, accelerator (brightener), and leveler. The additives play an important role in filling copper in microstructures. Their role is briefly described hereafter.

The suppressor consists of PEG (poly(ethylene glycol))-PPG (poly(propylene glycol)) polymers (high molecules) and suppresses the growth of copper film. The accelerator (brightener) consists of sulfur-containing organic substances and accelerates the growth of copper film. The leveler consists of amine-based organic compounds and s the growth of copper film in an area of the wafer surface where the electric field concentrates and does not suppress the growth of copper film in an area of the wafer surface where the electric field does not concentrate. Therefore, the leveler serves to improve the smoothness of the copper film.

It is very important to control the composition of the plating solution for filling a copper film in nanoscale microstructures such as vias, contact holes, and trenches without creating gaps and defects such as voids and seams. However, it has been found that the composition of the plating solution changes as a result of electrolysis in the course of electroplating. Furthermore, it is not sufficient to simply control the concentrations of the above additives for predetermined value ranges in order to fill a metal in nanoscale microstructures. This is because the three additives are electrolyzed into different substances (by-products) including those having different molecular structures.

Speaking of the additive concentration control, the supply valve 9 on the circulation line 3 is opened to add fresh plating solution to the plating solution for controlling the additive concentration. Such concentration control has conventionally been used because of the belief that the by-products have no or little influence on filling properties. However, investigation by the inventor of the present application revealed that defective filling occurs in vias of nanoscale microstructures, as mentioned above. Moreover, it is difficult to measure an amount of the by-products during a plating processing. In the electroplating apparatus, shown in FIG. 1, invented by the inventor of the present application, the monitoring unit 6 collects data such as an initial volume of plating solution, a volume of replenished solution, a number of wafers processed, a value of current applied, a value of voltage applied, processing time, a number of rotations, a volume of waste solution and the like, and the control unit 1 calculates the amount of by-products (hereafter, also referred to as the amount of decomposition products) produced as a result of electrolysis based on a model using the data. Here, “the volume of replenished solution” is a volume of plating solution replenished on the basis of “the initial volume of plating solution”. “The volume of waste solution” is a volume of plating solution discharged on the basis of “the initial volume of plating solution”.

The monitoring unit 6 and control unit 1 can be realized for example by dedicated arithmetic circuits, or hardware including a processor and memory such as RAM (random access memory) and ROM (read only memory) and software stored in the memory and running on the processor. In this embodiment, the monitoring unit 6 and control unit 1 are realized by a computer and programs running on the computer. The monitoring unit 6 can acquire the data described above from various sensors that acquire such data directly or based on the operation states of the electroplating apparatus indirectly.

The model used by the control unit 1 for calculating the amount of decomposition products (the amount of by-products) is described hereafter. In this embodiment, as shown in FIG. 1, one wafer is processed in one plating operation. Therefore, the number of plating operations is equal to the number of wafers processed. In such a case, the amount of decomposition products N_bn contained in the plating solution after the n-th wafer processing is expressed by the model equation (1) below in which is N_b0 is the amount of decomposition products present in the initial plating solution, G_n is the amount of decomposition products produced in the n-th wafer processing; V₂ is the total volume of plating solution (shell volume); and V₁ is the volume of waste solution in each discharge.

$\begin{array}{cc}{N}_{\mathrm{bn}}=\left(N_{b0}+G_1\right){\left(\frac{V_1}{V_2}\right)}^n+{G_2\left(\frac{V_1}{V_2}\right)}^{n-1}+\dots +G_n\left(\frac{V_1}{V_2}\right)& \left(1\right)\end{array}$

The above model equation particularly obtains the amount of decomposition products produced (G_n) in each wafer processing using a cumulative charge (integrated current value) calculated from the value of current applied to the wafer processed (for example, calculated using the number of wafers processed and the value of current applied) and calculates the amount of decomposition products remaining after each discharging waste solution based on the volume of waste solution in relation to the total volume of plating solution. Then, each amount of decomposition products remaining between discharges is cumulated to obtain the amount of decomposition products contained in the plating solution after the n-th wafer processing (N_bn). The amount of decomposition products produced during the subsequent plating operation is added to the amount of decomposition products N_bn to obtain the amount of decomposition products at the time of the subsequent plating operation. The data (particularly the current value) used for calculating the amount of decomposition products N_bn should be obtained at least at a sampling rate higher than 2 Hz. This is because there is some influence on the accuracy of calculation of the amount of by-products. For example, sampling rates higher than 2 Hz increase the calculation accuracy more than 3%. In the above equation (1), the total volume of plating solution (V₂) and the volume of waste solution (V₁) are constant. However, when volume of waste solution and the volume of replenished solution in each operation are different, the amount of decomposition products can be calculated using their respective values.

FIG. 2 is a graphical representation showing a relationship between the calculated amount of by-products and the occurrence of electrically defective chips due to seams and voids. In FIG. 2, the amount of by-products (amount of decomposition products) is plotted as ordinate. FIG. 2 shows that defective chips occur when the amount of by-products exceeds a specific value. In other words, it was found that by-products are produced as the plating solution is electrolyzed in the course of the above described wafer processing and defects such as seams and voids occur depending on the amount of by-products. Such behavior of the by-products indicates that the by-products possess an ability of suppressor. The inventors of the present application have found out this new knowledge. In the electroplating apparatus of this embodiment, a maximum volume of plating solution to be filled is 200 L.

FIG. 3 is an illustration showing a result of physical analysis of an electrically defective chip (the defective part was identified and its cross-section was observed). A contact 35 has a diameter of 60 to 70 nm. An interlayer insulating film 32 corresponding to the total depth of a trench 34 and the contact 35 has a thickness of 200 to 300 nm. FIG. 3 also shows an interlayer insulating film 31 that is an under layer for the interlayer insulating film 32 and a metal plating layer filled in a trench 33 formed in the interlayer insulating film 31.

In FIG. 3, the depth of the trench 34 is smaller than 100 nm. Therefore, the contact 35 has an aspect ratio up to approximately 5. There is difficulty in filling when a contact has a diameter of 100 nm or smaller and an aspect ratio of 3 or higher. The electrically defective chip had a void 41 on the sidewall of the trench 34 as shown in FIG. 3. Furthermore, a seam 42 occurred in the contact 35. Neither the seam 42 nor the void 41 was prevented even if the copper concentration in the plating solution was increased from 10 mg/L to 100 mg/L.

The mechanism of occurrence of defective filling in vias is described hereafter.

FIG. 4 is a schematic illustration showing a mechanism of change in the via filling properties in association with alteration in the plating solution. Investigation by the inventor of the present application revealed that it is important to control by-products 52 produced as a result of electrolysis of the suppressor 51 for obtaining good via filling properties. A phenomenon was observed that indicates that molecules having molecular weights lower than the original PEG-PPG polymers are produced as the suppressor 51 is consumed through electrolysis. Molecules having lower molecular weights are the above described by-products 52. As mentioned above, the by-products 52 possess an ability suppressing the growth of copper film as does the suppressor 51. With lower molecular weights, the by-products 52 have larger diffusion coefficients and a larger amount of by-products 52 can reach the small via bottom (the bottom of the contact 35 in FIG. 4). This leads to defective filling in vias and contact holes. In other words, the by-products 52 presumably reach the via bottom earlier and suppress the growth of metal film from the via bottom (bottom filling), thereby causing defects such as seams and voids.

Therefore, even if fresh plating solution is added to control the amounts of three additives, it is impossible to control the by-products produced in the above mechanism or reduce defective filling. A fundamental solution removes the by-products from the plating solution and bring them to the outside, namely discharging waste solution. Hence, it is most important to control the volume of waste solution. In the electroplating apparatus of this embodiment, the waste valve 5 on the SAC line 4 is opened to release the plating solution when the amount of by-products calculated by the above equation (1) exceeds a predetermined value. In this structure, the by-products are discharged by releasing the plating solution and the amount of by-products present in the plating solution and responsible for voids and seams can be reduced.

FIG. 5 is a graphical representation showing the amount of by-products that was calculated by the above equation (1) and the amount of by-products in the plating solution actually measured by liquid chromatography (off line measurement). FIG. 5 shows that the amount of by-products in the plating solution increases as the calculated amount of by-products increases. In this measurement, the specific molecular structures of by-products to be detected are not identified. Therefore, the plating solution was sampled based on the calculated amount of by-products, and the molecules having molecular weights different from the suppressor were measured by liquid chromatography.

In liquid chromatography, an intensity of refractive-index for a column retention time is observed. Molecules having such different molecular weights emerge in retention times different from those in which the suppressor is observed; therefore, they can be detected separately. In actual observation, peaks supposedly for by-products other than the suppressor were observed in times longer than the retention time (observation time) in which the suppressor was observed. This liquid chromatography was of the size exclusion mode. In this mode, molecules having smaller molecular weights enter the column deep inside and have difficulty coming out of the column, lengthening their column retention time. In measurement, peaks were observed in the longer retention times. Therefore, the detected by-products presumably had molecular weights smaller than the suppressor and the above inventor's model was verified.

Second Embodiment

A second embodiment provides a filling method creating no defects such as voids and seams based on the first embodiment of the present invention.

FIG. 6 is a flowchart of a plating process presenting the second embodiment. When the electroplating apparatus starts a recipe for plating, first, the monitoring unit 6 measures the concentrations of the three additives (S1). The monitoring unit 6 orders the control unit 1 to replenish plating solution in the case of NG in which the measurements are outside predetermined value ranges. The control unit 1 being ordered opens the supply valve 9 and replenishes a predetermined volume of plating solution (S1NG, S2). When the replenishment is completed, the monitoring unit 6 measures the concentrations of the additives once again (S1). The replenishment of plating solution is repeated until the concentrations of the additives are within the above predetermined ranges. When the replenished volume exceeds the total volume of plating solution, the control unit 1 stops the replenishment of plating solution. When the concentrations of the additives become within the above ranges (S1OK), the monitoring unit 6 collects apparatus data such as the initial volume of plating solution, volume of replenished solution, number of wafers processed, value of current applied, value of voltage applied, processing time, number of rotations, and volume of waste solution (S3). The control unit 1 calculates the amount of by-products based on the equation (1) using the collected apparatus data (S4). When the amount of by-products exceeds a predetermined value range, the control unit 1 opens the waste valve 5 and discharges the solution (S4NG, S5). After the discharging, the monitoring unit 6 collects the apparatus data, and the control unit 1 calculates the amount of by-products using the data once again (S4). The control unit 1 repeats the discharge of plating solution until the amount of by-products becomes within the predetermined value range (S4NG, S5). When the amount of by-products becomes within the predetermined value range, the control unit 1 loads a wafer into the apparatus (S6) and places it on the cathode electrode. Then, the wafer is metal-plated (S7) and unloaded after the plating is completed (S8); then, the recipe ends.

What is important here is that the steps of measuring the concentrations of additives, which is conventionally performed outside the recipe, and determining whether the concentrations are appropriate are newly provided and, furthermore, the mechanism and step of replenishing plating solution when the concentrations of additives are outside predetermined value ranges (S1NG) is provided. What is more important is that the step of collecting apparatus data, calculating the amount of by-products, and determining whether the mount is appropriate is newly provided and, furthermore, the mechanism and step of discharging the solution when the amount of by-products is outside a predetermined value range (S4NG) is provided. Because these are performed within the recipe, it is possible to feedback the calculated result using the apparatus data collected in situ. Then, the volume of replenished solution and the volume of waste solution are controlled according to the amount of by-products. Therefore, the accuracy of calculation of the amount of by-products is significantly improved.

In the above explanation, the recipe start (operation start) is the initial point. However, needless to say, the calculation accuracy will not be so different even if the initial point is when a lot is engaged in the apparatus (track-in).

Third Embodiment

A third embodiment of the present invention will be described hereafter. The third embodiment of the present invention relates to an electroplating method in which the optimum volume of waste solution is determined. The plating solution is expensive and it is preferable that the volume of waste solution is minimized. Then, in this embodiment, a technique of determining the optimum volume of waste solution based on the quantity of wafers to be processed, namely the amount of by-products to be produced, total volume of plating solution, and volume of waste solution, will be described.

FIG. 7 is a graphical representation showing results of simulating the amount of by-products in the plating solution in relation to the number of times of discharging waste solution. Here, the amounts of by-products were calculated for the cases when volume ratios of waste solution to the total volume of plating solution (hereafter, referred to as a waste solution ratio) were 8% and 29% with the initial amount of by-products being “1”. The waste solution ratios were randomly determined.

FIG. 7 shows that the by-products are not sufficiently released at the waste solution ratio of 8% so that the by-products in the plating solution continue to increase. On the other hand, increase in the amount of by-products is suppressed at the waste solution ratio of 29%. This suggests that there must be an optimum waste solution ratio. Then, increase in the amount of by-products in relation to the waste solution ratio is obtained based on the above calculation results as follows.

FIG. 8 is a graphical representation for explaining increase in the amount of by-products in relation to the waste solution ratio. In FIG. 8, the waste solution ratio is plotted as abscissa and the amount of by-products is plotted as ordinate.

As predicted above, FIG. 8 shows that the increase is smaller as the waste solution ratio is raised. The original amount of by-products is “1.” The waste solution ratio to keep the amount of by-products “1” or smaller is 34%. Therefore, 34% of the total volume of plating solution should be discharged at each discharge to prevent the amount of by-products in the plating solution from increasing. Needless to say, the amount of by-products in the plating solution can be reduced as the number of times of discharging waste solution is increased when the waste solution ratio is higher than 34%.

In this embodiment, the volume of waste solution can previously be determined as described above. Defective filling due to abnormal apparatus states can be prevented by storing the obtained value in the monitoring unit 6 in FIG. 1 as a threshold and stopping the plating apparatus when the waste solution ratio is lower than the threshold. At least 34% of plating solution should be discharged at least to fill copper in microstructures having contacts of 100 nm or smaller in diameter as shown in FIG. 3 without creating any defects such as voids and seams.

In the above embodiments, copper is filled. However, needless to say, the same is true for other metal films such as silver and aluminum.

As described above, the present invention prevents defective filling due to change in the composition of plating solution and improves the production yield of semiconductor devices.

The present invention is not restricted to the above embodiments. Various modifications and applications are available without departing from the technical idea of the present invention. For example, in the above embodiments, the present invention is applied to a semiconductor production apparatus having a circulation line and a sub-circulation line. The invention of the present application is applicable to a semiconductor production apparatus having only one circulation line as shown in FIG. 9.

The semiconductor device production method and semiconductor production apparatus of the present invention is capable of obtaining the amount of by-products to be produced by calculation and can use this capability for controlling the amount of by-products for a constant value, therefore providing a useful production technique for the semiconductor device wiring process in which metals are filled in nanoscale microstructures. The present invention also has applications in plating process for MEMSs.

Claims

1. A semiconductor device production method having a step of filling a metal plating film in a via hole or a trench formed in an insulating film on a semiconductor substrate, the step of filling the metal plating film comprising the steps of:

collecting data including an initial volume of plating solution, volume of replenished solution, number of wafers processed, value of current applied and volume of waste solution;

calculating a cumulative charge during the plating based on the value of current applied;

calculating a total volume of plating solution; and

calculating an amount of decomposition products of suppressors contained in the plating solution based on the calculated total volume of plating solution, the volume of waste solution and the calculated cumulative charge.

2. The semiconductor device production method according to claim 1, further comprising the steps of:

discharging a predetermined volume of plating solution when the amount of decomposition products exceeds a predetermined value; and

replenishing plating solution.

3. A semiconductor production apparatus for filling a metal plating film in a via hole or a trench formed in an insulating film on a semiconductor substrate, comprising:

a unit circulating plating solution;

a unit replenishing plating solution;

a unit discharging the plating solution;

a unit collecting data including an initial volume of plating solution, volume of replenished solution, number of wafers processed, value of current applied and volume of waste solution;

a unit calculating a cumulative charge during the plating based on the value of current applied;

a unit calculating a total volume of plating solution; and

a unit calculating an amount of decomposition products of suppressors contained in the plating solution based on the calculated total volume of plating solution, the volume of waste solution and the calculated cumulative charge.

4. The semiconductor production apparatus according to claim 3, wherein the unit discharging the plating solution discharges a predetermined volume of plating solution when the amount of decomposition products exceeds a predetermined value.