METHOD FOR MONITORING THE FILLING PERFORMANCE OF COPPER PLATING FORMULA FOR MICROVIA FILLING

Abstract

A method for monitoring the filling performance of a copper plating formula for microvia filling includes measuring a first potential value at a first rotation speed and a second potential value at a second rotation speed with a Cu-RDE of a plating solution. Then, a potential difference is obtained by subtracting the potential measured at the second rotation speed from the first rotation speed. The filling performance is defined by the potential difference; that is, a high potential difference indicates a good filling performance.

Description

BACKGROUND

1. Field of Invention

The present invention relates to a method for monitoring the filling performance of a copper plating formula for microvia filling. More particularly, the present invention relates to a galvanostatic measurement for monitoring the filling performance of a plating solution over a long period of operation.

2. Description of Related Art

Via filling by copper electroplating is mostly adopted in the interconnection metallization of printed circuit boards (PCBs) and integrated circuit (IC) chips. Within the process of via filling, several additives must be simultaneously added in the copper plating solution to perform a bottom-up deposition. Therefore, analyzing and controlling methods for these additive concentrations in the copper plating solution are becoming more and more important in this skill.

In the process of copper electroplating, several via filling results are observed, such as conformal deposition, anti-conformal deposition or subconformal deposition, and super-filling deposition. Conformal deposition occurs at the same rate as that of copper deposition at board surface and via bottom, causing a seam to form in the via core. In general, the filling results of conformal deposition and anti-conformal deposition adversely effect the reliability of PCBs. Hence, to improve filling performance of via filling, methods to control adding the additives to electroplating solution are developed for controlling the copper deposition rate at board surface and via bottom, that is, reducing the deposition rate at board surface and increasing the deposition rate at a via bottom simultaneously to reach the optimum super-filling result.

In manufacturing a via filled by copper electroplating, abnormal deposition often results at normal additive concentration. Generally, via filling performance is affected by a certain additive concentration not being sufficient, so the conventional monitoring methods involve detecting the concentration change of single additive. However, via filling behavior is not defined by only a single additive function but also by the synergistic interaction of all additives. For this reason, the synergistic interaction of all additives must be monitored.

To reach the purpose of promoting electroplating quality, there is a need for monitoring the synergistic interaction of the whole plating system, which can provide a prediction of the filling performance of a plating system before the copper electroplating process.

SUMMARY

The present invention is directed to a method for monitoring the filling performance of a copper plating formula for microvia filling that satisfies the need of monitoring the synergistic process of a whole plating system.

It is therefore an objective of the present invention to provide a method for predicting the filling performance of one plating system before the copper electroplating process.

It is another objective of the present invention to provide a method for monitoring the filling performance of a plating solution over a long period of operation.

In accordance with the foregoing and other objectives, the method of monitoring the filling performance of a copper plating formula for microvia filling is provided. Galvanostatic measurement that is able to monitor the filling performance of one copper plating system is used. The method of galvanostat is employed to predict the filling performance of a copper plating formula for microvia filling by using a copper-working electrode that was individually operated at two different rotation speeds. Then, a potential difference (Δη value) is obtained by subtracting the potential measured at the two different rotation speeds. The filling performance is defined by the Δη value, that is, a large Δη value indicates a good filling performance.

In conclusion, the invention allows predicting the filling performance of a plating formula by detecting the variations of a Δη value of a plating system, to monitor the synergistic interaction of all additives in the plating solution. Therefore, the present invention can predict filling performance more exactly and adjust the plating solution at the right moment to ensure the quality of electroplating.

Moreover, bottom-up filling behavior is caused by synergistic interactions among these additives, so the invention allows monitoring the filling performance of a plating bath during a long period of operation.

The invention provides a feasible galvanostatic measurement to accurately predict the filling performance of a copper plating formula for microvia filling. The method of this invention is an easy and rapid method to utilize in manufacturing and academic research using copper electroplating process monitoring.

These and other features, aspects, and advantages of the present invention will become better understood with reference to the following description and appended claims.

It is to be understood that both the foregoing general description and the following detailed description are by examples and are intended to provide further explanation of the invention as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention can be more fully understood by reading the following detailed description of the preferred embodiment, with reference made to the accompanying drawings as follows:

FIG. 1 is a diagram of a galvanostatic measurement system according to the preferred embodiment of this invention.

FIG. 2 is a flowchart of steps of the galvanostatic measurement according to one preferred embodiment of this invention.

FIG. 3 is a diagram of filling a microvia.

FIG. 4 is a measured result of the galvanostatic measurement without leveler injection.

FIG. 5 is a cross-sectional view of the microvia filling using the plating formula as described in FIG. 4.

FIG. 6 is a measured result of the galvanostatic measurement with leveler injection.

FIG. 7 is a cross-sectional view of the microvia filling using the plating formula without leveler as described in FIG. 6.

FIG. 8 is another cross-sectional view of the microvia filling using the plating formula with leveler as described in FIG. 6.

FIG. 9 is a measured result of the galvanostatic measurement with leveler injection.

FIG. 10 is a cross-sectional view of the microvia filling using the plating formula as described in FIG. 9.

FIG. 11 is a measured result of the galvanostatic measurement with leveler injection.

FIG. 12 is a cross-sectional view of the microvia filling using the plating formula as described in FIG. 11.

FIG. 13 is a measured result of the galvanostatic measurement with leveler injection.

FIG. 14 is a cross-sectional view of the microvia filling using the plating formula as described in FIG. 13.

FIG. 15 is a diagram of the correlation between the Δη value and the filling performance of a microvia.

FIG. 16 is a diagram of the relationship between the additive concentration and the potential difference.

FIG. 17 is a diagram of the relationship between the additive concentration, the potential difference and the filling performance for microvia.

FIG. 18 is a diagram of the relationship between the potential difference and the filling performance for microvia of a commercial plating formula.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Reference will now be made in detail to the present preferred embodiments of the invention, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers are used in the drawings and the description to refer to the same or like parts.

Reference is made to FIG. 1, which shows a galvanostatic measurement system according to the preferred embodiment of this invention. A galvanostatic measurement system 100 comprises a computer 110, a potentiostat 120, a working electrode 130, a reference electrode 140, a counter electrode 150 and a glass vessel 160. The computer 110 was used to receive and process the data obtained in the galvanostatic measurements. The potentiostat 120 was used to control the electrical current output. The working electrode 130 was a platinum rotation disk electrode (Pt-RDE) with a diameter of 3 mm. The reference electrode 140 was a saturated mercury-mercurous sulfate electrode (SMSE). The counter electrode 150 was a small copper bar, which was placed in a small glass tube that contained a blank electrolyte (i.e., 0.88 M CuSO₄.5H₂O+0.54M H₂SO₄). The end side of the small glass tube was sealed with a porous polymer material to prevent the additives from directly contacting the counter electrode 150 during analysis, which may yield unfavorable byproducts. Moreover, before each electrochemical analysis, a thin copper layer with a thickness of 500 nm was predeposited onto the Pt-RDE in a predeposition bath, which contained only the blank electrolyte.

The galvanostatic measurements were carried out using the Cu-RDE at a current density of 10˜20 ASF. Besides, each plating formula was galvanostatically measured twice, once at 50˜200 rpm, the other at 700˜3000 rpm.

Reference is made to FIG. 2, which shows the process of the galvanostatic measurement. First, as shown in step 210, at the beginning of the galvanostatic measurement, the glass vessel contained only the blank electrolyte and 60 ppm Cl⁻. The galvanostatic measurement was to use fixed current to monitor variations of cathodic potential; that is, the reaction is more difficult to process at high cathodic potential condition.

Then, as shown in step 220, when the galvanostatic measurement was performed for about 500 seconds, 200 ppm PEG (polyethylene glycol) was injected into the glass vessel 160. Next, as shown in step 230, an accelerator SPS (Bis(3-sulfopropy)disulfide) was subsequently injected into the glass vessel at the time of one thousand seconds. Then, as shown in step 240, a leveler with a concentration of 1 ppm was subsequently injected into the glass vessel at the time of two thousand seconds. The influence of a leveler concentration on the filling performance was examined by using the above-mentioned approach.

Millipore Direct-Q DI water (18.2 MΩ cm) was used to make all solutions that were used in the electrochemical analysis. The additives, such as SPS, JGB (Janus Green B), DB (Diazine Black), ABPV (Alcian Blue) and BTA (Benzotriazole), were individually added to the plating bath through dilution from stock solutions prepared with properly concentrated contents of these additives. The temperature of the plating solution was maintained at 20° C.˜30° C.

The filling performance was examined from pictures of microvia cross-sections obtained by an optical microscope (OM). Reference is made to FIG. 3, which shows an example of microvia filling, comprising a substrate 310, an inner conducting layer 320, a dielectric layer 330, a via 340 and a copper deposition layer 350. The copper deposition layer 350 has a height 360, which is the sum of the copper deposit over a dielectric layer 330 and the copper deposition layer 350. The copper deposit in the via has another height 370. The definition of filling performance is: (height 370 height 360)×100%.

Demonstration 1

FIG. 4 shows the results of galvanostatic measurement using the Cu-RDE with two different rotation speeds and a plating formula composed of 200 ppm PEG, 60 ppm Cl⁻ and 0.3 ppm SPS. The influence of a formula without a leveler on the filling performance was examined by using this approach. A potential difference that was an average of the Δη value, obtained by subtracting the potential measured at 1000 rpm from the potential measured at 100 rpm, was used as an indicator of the filling performance. That is, the positive Δη value (>0) indicates the formula was able to fill the via; on the other hand, a negative Δη value (<0) indicates the formula cannot carry out the filling behavior.

As shown in FIG. 4, the Δη value indicates the formula is effective in the bottom-up filling of microvia. In addition, as shown as FIG. 5, the filling performance of the formula, described in FIG. 4, was observed from pictures of microvia cross-sections obtained by an optical microscope. The dimensions of the PCB fragment were 4.5×6 cm. The diameter of the microvia was around 130 μm. The depth of the microvia was around 85 μm. The sidewall of the microvia was first metallized by electroless copper plating and subsequent deposition by copper electroplating in order to increase the thickness of the copper seed layer to 2˜3 μm before filling was conducted. The PCB fragment was plated at a current density of 18 A/ft² (ASF, ≅19.4 mA/cm²) for 70 minutes. Two phosphorus-containing copper slices were used as anodes and directly placed in the plating bath with a working volume of 700 ml. Constant agitation was performed by continuously flowing air bubbles at a flow rate of 2.5 L/hr during the electroplating to ensure a good convection. FIG. 5 shows the plating result of this formula. The filling performance of this plating formula is excellent; even a semispherical bump is formed on the top of the filled microvia. The hemisphere-like bump was formed due to the formula not containing any leveler.

The result obtained from FIG. 5 coincides with the result of galvanostatic measurement of FIG. 4. The coincidental results can demonstrate that the galvanostatic measurement of the present invention is reliable.

Demonstration 2

FIG. 7 shows the filling performance of a plating formula, which contains 200 ppm PEG, 60 ppm Cl⁻, and 1 ppm SPS. The result of galvanostatic measurement is shown in FIG. 6, demonstrating that the potential of the Cu-RDE measured at 1000 rpm was slightly lower than that measured at 100 rpm after 1 ppm SPS was injected into the plating solution.

This depolarization behavior caused by 1 ppm SPS injection is the reverse of that caused by 0.3 ppm SPS injection shown in FIG. 4. Therefore, filling performance of this formula is very poor. In addition, the corresponding result was shown in FIG. 7. Conformal depositions occurred at a board surface and via bottom were observed. This filling result is attributed to the excessive acceleration of copper deposition at the board surface caused by the high SPS concentration. When the SPS concentration was increased from 0.3 ppm to 1 ppm, the synergy of the suppressor with the accelerator was thereupon destroyed.

Referring to again FIG. 6, when 1 ppm DB was subsequently injected into the plating solution, the CDA (convection-dependent adsorption) behavior of the additives was recovered, in which the potential of the Cu-RDE measured at 1000 rpm returned to a higher level than that measured at 100 rpm. Therefore, the bottom-up filling behavior also recovered after 1 ppm DB was added into the plating solution with 200 ppm PEG, 60 ppm Cl⁻, and 1 ppm SPS as shown in FIG. 8. However, DB is a derivative of JGB, which can interact with PEG as a composite suppressor, so the hemisphere-like bump does not appear because the composite suppressor has a strong CDA behavior.

Demonstration 3

Reference is made to FIG. 9, wherein a good filling performance was demonstrated of a plating formula with DB replaced by JGB. Its Δη value was larger than that obtained from the formula with DB of FIG. 6. Therefore, it also exhibited a good filling performance, as shown in FIG. 10. However, the hemisphere-like bump on the top of the filled microvia still did not appear. In comparison with the polarization curves plotted in FIG. 4, the reason for this is attributed to the enhanced inhibition effect after the addition of JGB or DB. If a leveler does not possess a strong inhibition effect, like that of DB or JGB, but rather, a moderate one, then the hemisphere-like bump may arise after electroplating.

Demonstration 4

Reference is made to FIG. 11, wherein an excellent filling performance was demonstrated of a plating formula, of which JGB was replaced with ABPV. FIG. 12 show that the addition of 1 ppm ABPV can effectively turn the conformal deposition into super-filling. A slightly hemisphere-like bump was observed from the image shown in FIG. 12. FIG. 11 plots the corresponding polarization curves, indicating that a significant Δη value appears in the galvanostatic measurement. Therefore, this plating formula is effective in the microvia filling.

Demonstration 5

A leveler BTA, which has been confirmed to be ineffective in via filling, was adopted to examine the evaluation method of filling performance. FIG. 13 shows the result obtained in the galvanostatic measurement, no significant Δη values between the higher rotation speed and the lower rotation speed were measured. The galvanostatic measurement is consistent with the prediction. On the other hand, the typical plating result is shown in FIG. 14, which indeed was a conformal deposition after 1 ppm BTA was added to the base electrolyte with 200 ppm PEG, 60 ppm Cl⁻, and 1 ppm SPS. Various BTA concentrations were tested, but conformal deposition was the only outcome.

Demonstration 6

FIG. 15 plots the Δη values versus the filling performance as defined in FIG. 3. The horizontal axis indicates the potential difference (Δη values); and vertical axis indicates the filling performance of each plating formula. The potential differences of the plating formula which contains 1 ppm SPS, 200 ppm PEG, and 60 ppm Cl⁻, and the plating formula which contains 1 ppm SPS and 1 ppm BTA, 200 ppm PEG, and 60 ppm Cl⁻, are both smaller than 5. Hence, their filling performances are insufficient (lower than 20%) and worse than the other plating formulae that exhibit large potential differences. Evidently, when the Δη value exceeds a certain value, the filling performance is good, whereas when the Δη value is small, or even negative, conformal deposition occurs. A negative Δη value is indicative of the absence of CDA behavior in the plating system. According to the definition of CDA, a fast rotation speed of Cu-RDE should result in a low deposition rate, namely a highly reductive over-potential. Thus, the galvanostatic measurements can provide proper prediction for filling performance of a plating formula.

Demonstration 7

The CDA behavior caused by the physiochemical interaction can be quantitatively analyzed by the galvanostatic measurement shown in Demonstrations 1˜4. The practical measurement results are shown in FIG. 16. Curve (a) and curve (b) indicate the relative potential variation of a plating formula without JGB measured at two different rotation speeds and having a negative Δη value. Curve (c) and curve (d) indicate the relative potential variation of a plating formula with 0.5 ppm JGB measured at two different rotation speeds and having a positive Δη value. Curve (e) and curve (f) indicate the relative potential variation of a plating formula with 1 ppm JGB measured at two different rotation speeds and having a positive Δη value. Curve (g) and curve (h) indicate the relative potential variation of a plating formula with 3 ppm JGB measured at two different rotation speeds and having a positive Δη value.

Obviously, the Δη value, which is an average of potential differences between the two polarization curves shown in FIG. 16, is a function of the JGB concentration. It is evident that the polarization curves shifted from low overpotential to high one with increasing JGB concentration, indicating that JGB is able to enhance the inhibiting effect of the plating formula on a copper deposition, which has been attributed to the formation of the composite suppressor. However, a maximal Δη occurred at 1 ppm JGB with increasing JGB concentration, which is consistent with the variation of filling performance with the JGB concentration. This result means that the synergistic effect of the chemical interaction can be characterized by the proper fluid motion.

Demonstration 8

FIG. 17 is a summary of the variation trend of the filling performance and the Δη value with the JGB concentration. Indeed, as shown as FIG. 17, the variation trend of the filling performance caused by varying the JGB concentration checks exactly with that of the Δη value. These results suggest that this approach using the galvanostatic measurement can be utilized as a metrology for monitoring the filling performance of a plating solution.

Demonstration 9

Reference is made to FIG. 18, a practical case of industrial plating formula using the galvanostatic measurements. Since the synergistic interactions of additives in plating formulae were changeable within the electroplating process, monitoring can be achieved by using the galvanostatic measurements. As shown in FIG. 18, the relationship is linear. Therefore, a database of the plating formula can be calculated and a pattern can be created for evaluating filling performance of a plating formula.

According to the above-mentioned approaches, there are many advantages of the present invention over the prior art. The invention allows predicting the filling performance of a plating formula and monitoring the synergistic interaction of all additives of the plating solution.

Moreover, the present invention provides the prediction of filling performance more exactly than conventional monitoring methods and makes a real-time adjustment of the additive concentrations in a plating bath within a long period of operation.

The invention provides a feasible galvanostatic method to accurately predict the filling performance of a copper plating formula for microvia filling. The method of this invention is an easier and faster method to utilize in manufacturing and academic research of copper electroplating process monitoring.

Although the present invention has been described in considerable detail with reference to certain preferred embodiments thereof, other embodiments are possible. Therefore, the spirit and scope of the appended claims should not be limited to the description of the preferred embodiments contained herein.

It will be apparent to those skilled in the art that various modifications and variations can be made to the structure of the present invention without departing from the scope or spirit of the invention. In view of the foregoing, it is intended that the present invention cover modifications and variations of this invention provided they fall within the scope of the following claims and their equivalents.

Claims

1. A method for monitoring the filling performance of a copper plating formula for microvia filling, comprising:

measuring a first potential value of the copper plating formula by using a working electrode at a first rotation speed;

measuring a second potential value of the copper plating formula by using the working electrode at a second rotation speed that is different from the first rotation speed; and

calculating a potential difference between the first potential and the second potential, wherein a larger potential difference indicates a better filling performance for microvia filling.

2. The method of claim 1, wherein the filling performance is defined as (the height of the copper deposition in the microvia÷the height of via)×100%.

3. The method of claim 1, wherein the copper plating formula is kept at a temperature of 20˜30° C.

4. The method of claim 1, wherein the copper plating formula comprises cupric sulfate (CuSO4) and sulfuric acid (H2SO4).

5. The method of claim 1, wherein the working electrode is a copper rotation disk electrode (Cu-RDE).

6. The method of claim 5, wherein the Cu-RDE is kept at a current density of 10˜20 ASF.

7. The method of claim 1, wherein the first potential value is a potential measured by a working electrode relative to a reference electrode.

8. The method of claim 1, wherein the first rotation speed is 50˜200 rpm.

9. The method of claim 1, wherein the second potential value is a potential measured by a working electrode relative to a reference electrode.

10. The method of claim 1, wherein the second rotation speed is 700˜3000 rpm.