US20030183513A1

US20030183513A1 - Plating system and method for management of plating solution using plating system

Info

Publication number: US20030183513A1
Application number: US10/395,220
Authority: US
Inventors: Toshiki Shinmura
Original assignee: NEC Electronics Corp
Current assignee: NEC Electronics Corp
Priority date: 2002-03-26
Filing date: 2003-03-25
Publication date: 2003-10-02
Also published as: JP2003277998A; JP3821742B2

Abstract

In a plating system, a plating consumption coefficient per unit number of substrates subjected to a treatment is calibrated once every predetermined time interval, and a replenishing amount in response to consumption of an additive associated with a plating treatment is computed by use of a formula defined as “the plating consumption coefficient×the number of substrates subjected to the plating treatment” and the replenishing amount of the additive is replenished. Accordingly, an amount of consumption of the additive coincides with the replenishing amount thereof. As a result, a concentration of the additive contained in a plating solution can be maintained at a constant concentration.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to a plating system for providing a substrate subjected to plating with a plating treatment and to a method for management of a plating solution using the plating system. More specifically, the present invention relates to a plating system using a Cu plating solution, and to a method for management of a plating solution using the plating system.

2. Description of the Prior Art

A plating system is required to constantly maintain a concentration of a plating solution component (an additive such as an accelerator) contained in a plating solution for a long period of time including the time of plating operation.

To be more precise, for example, as shown in FIG. 1, a

plating system

100 is composed of a plating cell 10, a plating solution tank 20, a plating solution analyzation machine 30, and an additive replenishing apparatus 40. Here, additive replenishing has been performed by the plating solution analyzation machine 30 in the following process.

First, the additive replenishing

apparatus

40 receives measurement results of the concentration of the additive at a predetermined interval (although various setting is possible herein such as 3 hours, 6 hours, 12 hours, and the like, in this case the time interval is set to be 3 hours as an example) from the plating solution analyzation machine 30. Then, the additive replenishing apparatus 40 calculates a deviation from a target concentration for the additive and replenishes the additive in the plating solution tank 20 appropriately. As a matter of course, no replenishing takes place when the measurement result of the additive concentration exceeds the target concentration for the additive.

No problem involving the deviation of the additive concentration would occur if periodic centering of the additive concentration takes place e.g. every minute, based on the measurement results of the

plating solution tank

20 as described above. However, in reality, one measurement of an additive concentration requires approximately 2.5 hours at the minimum. Accordingly, such a measurement is actually performed at e.g. a 3 hour-interval. Conventionally, an effort for suppressing variation of the additive concentration within 3 hours has been attempted by calculating the predictive additive concentration once every 5 minutes during the 3 hours required for the actual measurement of the additive concentration, and by centering the predictive additive concentration once every 5 minutes based on this calculation. The additive has been replenished, based on this calculation performed once every 5 minutes, in the following manner.

An additive concentration variation coefficient Kt (ml/h) per unit time to be multiplied by elapsed time irrespective of what the plating amount is, an additive concentration variation coefficient Kq (ml/amp·h) per unit electric charge for plating (which is proportional to a plated film thickness), and a target additive concentration Ct (ml/l) are inputted to the plating

solution analyzation machine

30 in advance. In this case, a constant value obtained by an empirical assumption has been used as the additive concentration variation coefficient Kq per unit electric charge for plating.

The plating solution analyzation machine performs calculation as shown in Formula (1) once every 5 minutes:

Ck=C(k−1)−(Kt×T+Kq×Q) (1)

In this formula, Ck denotes a calculation result of the additive concentration on a recent occasion, C (k−1) denotes a previous calculation result of the additive concentration (or the target additive concentration if additive replenishing took place during the period from the previous calculation to the current calculation), T denotes elapsed time since the point of calculation of C (k−1), and Q denotes a total electric charge for plating since the point of calculation of C (k−1). Here, the term “Kq×Q” in Formula (1) denotes an amount of the additive consumed in the plating treatment (the conventional calculation procedure).

Sk=(Ct−Ck)×V (2)

In this formula, Sk denotes a replenishing amount of the additive required for adjusting the additive concentration to the target additive concentration (the calculation result), and V denotes a total amount of the plating solution in the plating system.

A carrier (the additive having a plating accelerator effect) is replenished immediately when the replenishing amount of the additive Sk required for adjusting the additive concentration thus calculated to the target additive concentration reaches e.g. 10 ml or more. If the amount is below 10 ml, the plating system waits for a result of the next calculation (performed approximately 5 minutes later).

However, the amount of the carrier consumption in the plating treatment is proportional to the number of substrates subjected to the plating treatment as shown in FIG. 2B, rather than the total plated film thickness (∝ the total electric charge for plating) as shown in FIG. 2A. The reason for a loose correlation between the total plated film thickness and the amount of the carrier consumption is attributable to a loose correlation between the number of substrates subjected to the plating treatment and the total plated film thickness as shown in FIG. 3.

FIG. 4A is a graph showing the number of substrates subjected to a plating treatment and replenishing amounts of the carrier under the conventional plating system. As shown in Formula (1), the carrier consumption in the plating treatment was calculated in the conventional method by:

Kq (the empirically-determined constant value)×total electric charge for plating.

Accordingly, the carrier was replenished based on the calculated. When plating is performed for a constant film thickness such as 550 nm or 3000 nm, the relation “the total electric charge for plating=the electric charge per substrate×the number of substrates subjected to the treatment” is defined. Therefore, the following equation holds true:

A conventional replenishing amount in the plating treatment=the amount of the carrier consumption in the plating treatment based on the conventional calculation=Kq (the empirically-determined constant value)×the total electric charge for plating=Kq (the empirically-determined constant value)×the electric charge per substrate×the number of substrates subjected to the treatment

As a consequence, the electric charge per substrate is also constant in the case of plating only for the constant film thickness; therefore, the replenishing amount and the number of substrates subjected to the treatment are in proportion as shown in the graph. It can be understood readily from FIG. 4A that in the conventional calculation of the replenishing amount, the replenishing amount varies considerably as the plated film thickness is made different, even if the number of substrates subjected to the treatment is unchanged.

Meanwhile, as described above, the carrier consumption in the plating treatment is proportional to the number of substrates subjected to the plating treatment. When the number of substrates subjected to the plating treatment is unchanged, then an actual amount of consumption is constant. However, in the conventional calculation of the replenishing amount, when the plated film thickness is different, the carrier is replenished according to the inappropriately computed replenishing amount e.g. a small replenishing amount of the carrier in the event of thin-film plating or a large replenishing amount of the carrier in the event of thick-film plating even if the number of substrates subjected to the treatment is unchanged. In the conventional method, a mismatch between the amount of consumption and the replenishing amount incurs variation of the concentration.

FIG. 4B shows the number of substrates subjected to the plating treatments and variations in the additive concentrations from the target concentration when the conventional replenishing method is used. Certain constant values obtained by empirical assumptions have been conventionally applied to Kq. FIG. 4B shows that in the conventional method such variations of the concentrations cannot be corrected in the case of plating with various film thicknesses even if the constant value is selected properly.

In a practical plating treatment by lots, it is rather unusual to perform either one of thick-film plating or thin-film plating continuously. Instead, both thick-film plating treatment and thin-film plating treatment are performed depending on a situation. As a result, a rising effect of the concentration upon thick-film plating and a declining effect of the concentrations upon thin-film plating as illustrated in FIG. 4B cancel each other. Therefore, it is not always true that the deviation of the concentration grows larger as the number of substrates subjected to the treatment is increased. However, it is inevitable that either thick-film plating or thin-film plating occurs more frequently than the other in an actual plating stage (in terms of shipment of products). In this case, the additive concentration fluctuates as a dot indicated with an arrow in FIG. 5. As a consequence, products may be discarded due to deviations of the additive concentration or may be suspended for shipment due to incapability of performing the plating treatment.

FIG. 6 shows results of an advanced research regarding the amount of a carrier consumption of per wafer and time elapsed since entire replacement of a plating solution, which are investigated based on the findings that the amount of the carrier consumption in the plating treatment does not depend on the total plated film thickness but on the number of substrates subjected to the plating. It was conventionally thought that the same amount of the carrier would be consumed when the same plating treatment was performed. Therefore, a constant value obtained by an empirical assumption has been used as Kq. However, as shown in FIG. 6, a consumption rate of an additive in a plating treatment is not constant. Constancy of a proportionality coefficient relevant to this consumption rate is also a cause of discordance between the amount of consumption by plating and the replenishing amount computed in the conventional method.

As described above, FIG. 6 shows that the amount of the carrier consumption per unit number of substrates subjected to the treatment varies depending on the time since the entire replacement of the plating solution. The reason FIG. 2B shows the proportional relationship between the amount of the carrier consumption and the number of substrates (the amount of the carrier consumption of per unit number of substrates subjected to the treatment is constant) is that FIG. 2B shows data taken during a certain short period of time after certain time has elapsed since the entire replacement of the plating solution. As a matter of fact, FIG. 2B shows the data obtained in a period from 110 hours to 130 hours after the replacement of the solution. Since the amount of the carrier consumption per unit number of substrates subjected to the treatment exhibits a small variation during this period, the proportional relation is obtained as shown in FIG. 2B.

As described above, in the conventional plating system, centering of the additive concentration is performed within a cycle (about 3 hours) to measure the concentration e.g. by calculating the replenishing amount of the carrier once every 5 minutes. However, such calculation of the replenishing amount is inappropriate in the conventional method. The estimated amount of consumption which constitutes a basis of calculation of the replenishing amount was defined as “the constant proportional coefficient×the total plated film thickness”. However, the actual amount of consumption in the plating treatment should be defined as “a variable coefficient×the number of substrates subjected to the plating treatment”. In other words, the conventional calculation was performed by multiplying an inappropriate coefficient by an inappropriate plating parameter. As a result, the replenishing amount of the additive disagrees with the amount of consumption. Therefore there has been the case in which the additive concentration deviates from a standard during an interval of the measurements of the concentration. In this case, many substrates have been discarded as a result of fluctuation of plating properties, and yields of substrates are considerably reduced in the plating process.

SUMMARY OF THE INVENTION

An object of the present invention is to provide a plating system capable of maintaining an additive concentration of a plating solution throughout operation for plating, and a method for management of a plating solution using the plating system.

The present invention is proposed to attain an object to solve the foregoing problems. Specifically, a plating system according to the present invention includes: a plating cell for providing a substrate with a plating treatment; a plating solution tank having a function to supply and collect a plating solution while continuously circulating the plating solution between the plating cell and the plating solution tank; a plating solution analyzation machine for measuring a concentration of an additive contained in the plating solution inside the plating solution tank and thereby computing a replenishing amount of the additive into the plating solution tank; and an additive replenishing apparatus for replenishing the additive into the plating solution tank depending on the replenishing amount of the additive into the plating solution tank. Here, the plating solution analyzation machine measures the concentration of the additive contained in the plating solution inside the plating solution tank once every predetermined time period and then computes a value by dividing an amount of the additive consumed in the plating treatment being performed in the predetermined time period by the number of substrates subjected to the plating treatment in the predetermined time period so as to define a plating treatment additive consumption coefficient. Moreover, the plating solution analyzation machine computes a plating-dependent consumption amount predictive value of the additive by multiplying the plating treatment additive consumption coefficient at every monitoring time period which is shorter than the predetermined time period by the number of substrates subjected to the plating treatment in the monitoring time period, and then defines an additive consumption predictive value by adding the plating-dependent consumption predictive value to an elapsed time-dependent consumption amount of the additive. Moreover, the plating solution analyzation machine finds the additive concentration on a recent monitoring occasion based on the aforementioned additive consumption predictive value and the additive concentration on an immediately preceding monitoring occasion. Furthermore, the additive replenishing apparatus replenishes the replenishing amount of the additive into the plating solution tank if the replenishing amount, which is obtained by multiplying a value defined as subtraction of the additive concentration on the recent monitoring occasion from a target concentration of the additive by an amount of the plating solution, reaches a predetermined critical value or more.

In the plating system of the present invention, the elapsed time-dependent consumption amount of the additive maybe computed by multiplying the monitoring time period by an additive concentration variation coefficient per unit time having a constant value.

Moreover, in the plating system of the present invention, the plating solution may be a copper-plating solution, and the additive may be a thiol-based organic additive (an accelerator).

Next, a method for management of a plating solution according to the present invention uses a plating system which includes: a plating cell for providing a substrate with a plating treatment; a plating solution tank having a function to supply and collect the plating solution while continuously circulating the plating solution between the plating cell and the plating solution tank; a plating solution analyzation machine for measuring a concentration of an additive contained in the plating solution inside the plating solution tank and thereby computing a replenishing amount of the additive into the plating solution tank; and an additive replenishing apparatus for replenishing the additive into the plating solution tank depending on the replenishing amount of the additive into the plating solution tank. Here, the method manages the plating solution analyzation machine so that the plating solution analyzation machine performs the steps of: measuring the concentration of the additive contained in the plating solution inside the plating solution tank once every predetermined time period and then computing a value by dividing an amount of the additive consumed in the plating treatment performed in the predetermined time period by the number of substrates subjected to the plating treatment in the predetermined time period so as to define a plating treatment additive consumption coefficient; computing a plating-dependent consumption amount predictive value of the additive by multiplying the plating treatment additive consumption coefficient at every monitoring time period which is shorter than the predetermined time period by the number of substrates subjected to the plating treatment in the monitoring time period, and then defining an additive consumption predictive value by adding the plating-dependent consumption predictive value to an elapsed time-dependent consumption amount of the additive; finding the additive concentration on a recent monitoring occasion based on the aforementioned additive consumption predictive value and the additive concentration on an immediately preceding monitoring occasion; and allowing the additive replenishing apparatus to replenish the following amount of the additive into the plating solution tank if the replenishing amount, which is obtained by multiplying a value defined as subtraction of the additive concentration on the recent monitoring occasion from a target concentration of the additive by an amount of the plating solution, reaches a predetermined critical value or more.

In the method for management of the plating solution according to the present invention, in the step of measuring the concentration of the additive contained in the plating solution inside the plating solution tank once every predetermined time period and then computing a value by dividing the amount of the additive consumed in the plating treatment performed in the predetermined time period by the number of substrates subjected to the plating treatment so as to define a plating treatment additive consumption coefficient, the plating treatment additive consumption coefficient may be an amount of the additive consumed for one substrate to be subjected to the plating treatment, and the one substrate may correspond to a value obtained by converting an actual substrate area into a standardized unit based on a substrate having a certain area.

Moreover, in the method for management of the plating solution according to the present invention, the plating treatment performed by immersing the substrate in the plating cell of the plating system is a plating treatment to form a plurality of types of plating films with different film thicknesses on the substrate.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic illustration of a plating system to be used in an embodiment of the present invention. [0030]
FIG. 2A is a graph showing dependency of an amount of an additive (a carrier) consumption in a plating treatment on a total electric charge for plating. [0031]
FIG. 2B is a graph showing dependency of the amount of the additive (the carrier) consumption in the plating treatment on the number of substrates subjected to the plating treatment (in the case of using substrates of a constant size). [0032]
FIG. 3 is a graph showing a relation between the total electric charge for plating and the number of substrates subjected to the plating treatment (in the case of using substrates of a constant size), based on the data shown in FIG. 2A and FIG. 2B. [0033]
FIG. 4A is a graph showing relations between replenishing amounts of a carrier and the number of substrates subjected to plating treatments (related to two types of plating film thicknesses) in a case of using a conventional method. [0034]
FIG. 4B is a graph showing relations between variations of carrier concentrations and the number of substrates subjected to the plating treatments (which are calculated values related to two types of plating film thicknesses only) in the case of using the conventional method. [0035]
FIG. 5 is a graph showing relations between variations of carrier concentrations and the number of substrates subjected to the plating treatments (which are actual data related to more than two types of plating film thicknesses) in the case of using the conventional method. [0036]
FIG. 6 is a graph showing a relation between an amount of an additive (a carrier) consumed in a plating treatment per 200-nm substrate and time elapsed since entire replacement of a plating solution in a plating system. [0037]
FIG. 7A to FIG. 7C are partially enlarged schematic cross-sectional views of a surface of a substrate for schematically describing a phenomenon of strong dependency of an amount of an additive consumption in a plating treatment on the number of substrates subjected to the plating treatment.[0038]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

Now, an embodiment of the present invention will be described with reference to FIG. 1. FIG. 1 is a schematic cross-sectional view of a plating system. [0039]
A [0040] plating system 100 adopts a constitution mainly including a plating cell 10, a plating solution tank 20, a plating solution analyzation machine 30, and an additive replenishing apparatus 40. In the plating cell 10, a substrate 1 is placed opposite to an anode 2 and is immersed in a plating solution 3, and then voltage E is applied between the substrate 1 and the anode 2 so as to plate Cu²⁺ contained in the plating solution 3 onto a surface of the substrate 1. In this event, in addition to Cu²⁺, SO₄ ²⁻, H⁺, and Cl⁻, the plating solution contains a certain concentration of a thiol-based organic additive as a plating promoter (a carrier). Besides, the plating solution also includes another organic polymer additive (an inhibitor polymer) and the like. However, the concentration of this organic polymer additive varies virtually in proportion to time and does not depend on a plating treatment. Accordingly, the latter additive does not incur a problem regarding the additive concentration unlike the carrier.
Next, the [0041] plating solution tank 20 circulates the plating solution through a plating solution circulation line 50, and thereby supplies the plating solution to the plating cell 10 continuously and collects the plating solution from the plating cell 10. Therefore, the additive concentration values of the plating solution 3 in the plating solution tank 20 and in the plating cell 10 become equal.
Next, the plating [0042] solution analyzation machine 30 is a device for measuring the additive concentration in the plating solution 3 inside the plating solution tank 20. The plating solution analyzation machine 30 instructs the additive replenishing apparatus 40 to or not to replenish the additive in the plating solution tank based on the additive concentration data measured in the plating solution tank 20. The plating solution analyzation machine 30 also instructs the replenishing amount upon replenishing.
Next, in the case of replenishing the additive in the [0043] plating solution tank 20 in accordance with the instruction from the plating solution analyzation machine 30, the additive replenishing apparatus 40 replenishes the additive equivalent to the replenishing amount as instructed by the plating solution analyzation machine 30 into the plating solution tank 20.
The [0044] additive replenishing apparatus 40 receives a measurement result of the additive concentration from the plating solution analyzation machine 30 at a constant long time interval (which can be appropriately set to 3 hours, 6 hours, 12 hours, or the like; however, the time interval is set to be 3 hours here as an example). Then, the additive replenishing apparatus calculates a deviation of the additive concentration from a target concentration and replenishes the additive in the plating solution tank 20. Moreover, the additive replenishing apparatus 40 also performs replenishing of the additive at a shorter interval than the mentioned interval. Of course, the additive is not replenished when the measurement result of the additive concentration exceeds the target concentration. Such procedures are identical to the conventional method.
Here, description will be predetermined regarding a method of replenishing the additive, which constitutes a feature of this embodiment. In the present invention, a formula to be used for concentration monitoring and replenishing of the additive upon calculation to take place e.g. once every 5 minutes, in addition to a replenishing operation of the additive to take place once every predetermined time (3 hours, for example), is different from the corresponding formula used in the prior art. [0045]
First, the [0046] additive replenishing apparatus 40 receives the measuring result of the additive concentration from the plating solution analyzation machine 30 at a time interval (which can be appropriately set to 3 hours, 6 hours, 12 hours, or the like; however, the time interval is set to be 6 hours here as an example) for calibration relative to a predetermined proportional coefficient (equivalent to Kw, described later). Then, the additive replenishing apparatus 40 defines a value, which is obtained by dividing a quantity being equivalent to subtraction of “an amount of the additive consumption irrelevant to the plating treatment but solely dependent on elapsed time” from “an amount of decrease in the additive” by the number of substrates actually subjected to the plating treatment, as a plating treatment additive consumption coefficient Kw (which is equivalent to an amount of the carrier consumption per substrate subjected to plating during the relevant time period for calibration relative to the proportional coefficient).
Specifically, Kw is found by the following formulae, in which: variation of the concentration during the time period (6 hours, for example) for calibration relative to the proportional coefficient (=Kw) is ΔC[0047] LT; a total amount of the plating solution is V; a total replenishing amount of the additive during the time period for calibration relative to the proportional coefficient is SLT; a natural consumption rate of the additive per unit time is Kt; the time period for calibration relative to the proportional coefficient is TLT; and the number of substrates subjected to the plating treatment in the time period for calibration relative to the proportional coefficient is WLT. Now, the variation of the concentration is defined as: variation of the concentration=(the total replenishing amount of the additive−a total amount of the additive consumption)/the amount of the plating solution.
Therefore, the following formula is established: [0048]
ΔC LT=(S LT−(Kt×T LT +Kw×W LT))/V
Hence, Kw is found by the following formula: [0049]
Kw=(S LT−(ΔC LT ×V+Kt×T LT))/W LT
In this way, it is possible to obtain the proportional coefficient Kw, which is calibrated every predetermined time unlike the prior art. [0050]
Using Kw, the plating [0051] solution analyzation machine 30 calculates a current additive concentration e.g. once every 5 minutes, in order to complement management by centering the additive concentration in response to an actual measurement which is performed e.g. once every 3 hours. The formula to be used for calculating the additive concentration once every 5 minutes is the following Formula (3), in which the term “Kq×Q” in the conventional Formula (1) is replaced by the term “Kw×W”:
Ck=C _(k−1)−(Kt×T+Kw×W) (3)
Here, W is the number of substrates plated at the interval from the point of previous calculation to the point of current calculation. Here, the number of substrates corresponds to a value obtained by converting an actual substrate area into a standardized unit based on a substrate having a certain area. For example, if a 6-inch wafer is taken as the standard size, then one sheet of an 8-inch wafer is equivalent to 1.778 sheets and a 12-inch wafer is equivalent to 4 sheets based on the standard. [0052]
Now, after finding Ck, Sk is obtained by use of Formula (2) as shown in the conventional method: [0053]
Sk=(Ct−Ck)×V (2)
As a result, the [0054] additive replenishing apparatus 40 replenishes the additive into the plating solution tank 20 immediately when Sk in Formula (2) reaches or exceeds 10 ml, for example. If Sk is below 10 ml, then the result of current calculation Ck of the additive concentration is substituted for C_(k−1)in Formula (3) approximately 5 minutes later, and Ck and Sk will be calculated again. The procedures after obtaining Ck by Formula (3) are similar to the conventional method.
In this way, the amount of consumption in the plating treatment is found by multiplying the number of substrates subjected to the plating treatment by Kw being calibrated once every predetermined time period for calibration relative to the proportional coefficient. If the additive is replenished in accordance with the amount of consumption thus calculated, it is possible to replenish the additive in proportion to the amount of consumption, and the additive concentration thereby becomes stable. For this reason, it is also possible to extend a centering interval associated with actual measurements of the concentration. Accordingly, it is also possible to obtain secondary effects such as extension of an interval of scheduled maintenance, extension of the longevity of the system, or reduction in usage of analytic reagents. [0055]
Now, description will be predetermined regarding a phenomenon which attests that the amount of the additive consumption is virtually proportional to the number of substrates subjected to the plating treatment. As described previously, FIG. 2B is the graph showing dependency of the amount of the additive (the carrier) consumption in the plating treatment on the number of substrates subjected to the plating treatment (in the case of using substrates of a constant size). If this graph is compared with the graph of FIG. 2A, it is obvious that the correlation between the amount of the additive (the carrier) consumption and the number of substrates subjected to the plating treatment as shown in FIG. 2B is stronger than the correlation between the amount of the additive (the carrier) consumption and the total electric charge for plating as shown in FIG. 2A. The reason FIG. 2A exhibits the loose correlation has been discussed previously. [0056]
Here, the reason for the strong correlation shown in FIG. 2B is the following. Specifically, the additive is absorbed onto the surface of the substrate at the instant of immersion of the substrate in the plating solution. Therefore, the amount of the additive consumed in the plating treatment for the substrate is determined primarily by the surface area of the substrate and the number of substrates subjected to the plating treatment. FIG. 7A to FIG. 7C are partially enlarged schematic cross-sectional views of the surface of the substrate for schematically describing such a phenomenon by taking Cu plating as an example. [0057]
First, as shown in FIG. 7A, the [0058] substrate 1 is provided with a groove 51, and a Cu seed layer 52 is formed on the entire surface of the substrate including the groove 51 in advance by a sputtering method. When this substrate 1 is immersed in the plating solution, then a carrier 53 (a plating accelerator) is uniformly absorbed onto the surface of the Cu seed layer 52 as shown in FIG. 7A.
Next, when Cu plating is started, the [0059] groove 51 is gradually filled with a Cu film 54 as shown in FIG. 7B, and the surface area of the Cu film 54 inside the groove becomes small. In this event, the carrier 53 absorbed onto the surface of the Cu seed layer 52 remains on the surface of the substrate even when the Cu film 54 grows. As a result, the density of the carrier becomes relatively high inside the groove where the surface area of the Cu film 54 is diminished. Therefore, plating is accelerated at this part.
Next, the density of the carrier inside the groove is further increased along with the growth of the Cu-plated film as shown in FIG. 7C, and eventually, the groove is completely filled with the [0060] Cu film 54.
As explained, it is clear that the amount of the additive consumed in the plating treatment is determined primarily by the surface area of the substrate and the number of substrates subjected to the plating treatment, and that the plated film thickness is determined by the electric charge for plating irrespective of the amount of the additive consumption. [0061]
As described above, according to the plating system and the method for management of a plating solution using the plating system of the present invention, the amount of the carrier consumed in the plating treatment is determined by calculating the amount of the additive consumption once every monitoring time period (every 5 minutes, for example) by use of the formula defined as “the coefficient (Kw) to be calibrated periodically (every 6 hours, for example)×the number of substrates subjected to the plating treatment”. Then the amount of the carrier consumption is combined with consumption attributable to a lapse of time, and replenishing of the additive is thereby determined. Therefore, management by centering the additive concentration in response to an actual measurement to be performed once every predetermined period (3 hours, for example) is completely complemented. As a result, it is possible to maintain the constant concentration of the additive contained in the plating solution throughout the operation, and yields of substrates can be significantly improved. [0062]
As described above, the present invention is based upon the novel findings that the amount of additive consumed in the plating treatment is not proportional to the total electric charge for plating but to the number of substrates subjected to the treatment, and that the coefficient therein is not constant but is variable. Accordingly, the amount of the additive consumption is calculated once every monitoring time period (5 minutes, for example) by use of the formula defined as “the proportional coefficient (Kw) to be calibrated periodically×the number of substrates subjected to the plating treatment”. Then the amount of the additive consumption is combined with consumption attributable to a lapse of time, and replenishing of the additive is thereby determined. Therefore, management by centering the additive concentration in response to an actual measurement to be performed once every predetermined period (3 hours, for example) is completely complemented. As a result, it is possible to maintain the constant concentration of the additive contained in the plating solution throughout the operation. Therefore, use of the plating system according to the present invention enables manufacturing of the substrates provided with plated films with predetermined film characteristics (such as a filling property) throughout the operation time of the plating system. As a result, the plating system of the present invention has an effect of eliminating considerable reduction in yields of substrates in the plating process due to discarding many substrates. [0063]
Moreover, according to the present invention, the additive concentration is stable because the amount of consumption and the replenishing amount in the plating treatment are the same. For this reason, it is possible to reduce a frequency of centering the additive concentration in response to the actual measurements. As a result, it is also possible to obtain secondary effects such as extension of an interval of scheduled maintenance, extension of the longevity of the system, and reduction in usage of analytic reagents. [0064]

Claims

What is claimed is:

1. A plating system comprising:

a plating cell for providing a substrate with a plating treatment;

a plating solution tank having a function to supply and collect a plating solution while continuously circulating said plating solution between said plating cell and said plating solution tank;

a plating solution analyzation machine for measuring a concentration of an additive contained in said plating solution inside said plating solution tank and thereby computing a replenishing amount of said additive into said plating solution tank; and

an additive replenishing apparatus for replenishing said additive into said plating solution tank depending on said replenishing amount of said additive into said plating solution tank,

wherein said plating solution analyzation machine measures said concentration of said additive contained in said plating solution inside said plating solution tank once every predetermined time period and then computes a value by dividing an amount of said additive consumed in said plating treatment performed in the predetermined time period by the number of substrates subjected to said plating treatment so as to define a plating treatment additive consumption coefficient,

said plating solution analyzation machine computes a plating-dependent consumption amount predictive value of said additive by multiplying said plating treatment additive consumption coefficient at every monitoring time period which is shorter than the predetermined time period by the number of substrates subjected to said plating treatment in said monitoring time period, and then defines an additive consumption predictive value by adding said plating-dependent consumption predictive value to an elapsed time-dependent consumption amount of said additive,

said plating solution analyzation machine finds said additive concentration on a recent monitoring occasion based on said additive concentration on an immediately preceding monitoring occasion and said additive consumption predictive value, and

said additive replenishing apparatus replenishes said replenishing amount of said additive into said plating solution tank if said replenishing amount, which is obtained by multiplying a value defined as subtraction of said additive concentration on said recent monitoring occasion from a target concentration of said additive by an amount of said plating solution, reaches a predetermined critical value or more.

2. The plating system according to claim 1, wherein said elapsed time-dependent consumption amount of said additive is computed by multiplying said monitoring time period by an additive concentration variation coefficient per unit time having a constant value.

3. The plating system according to claim 1,

wherein said plating solution is a copper-plating solution, and

said additive is a thiol-based organic additive.

4. A method, using a plating system, for management of a plating solution, said plating system including:

a plating cell for providing a substrate with a plating treatment;

a plating solution tank having a function to supply and collect said plating solution while continuously circulating said plating solution between said plating cell and said plating solution tank;

said method comprising the steps of:

allowing said plating solution analyzation machine to measure said concentration of said additive contained in said plating solution inside said plating solution tank once every predetermined time period and then computing a value by dividing an amount of said additive consumed in said plating treatment performed in the predetermined time period by the number of substrates subjected to said plating treatment so as to define a plating treatment additive consumption coefficient;

allowing said plating solution analyzation machine to compute a plating-dependent consumption amount predictive value of said additive by multiplying said plating treatment additive consumption coefficient at every monitoring time period which is shorter than the predetermined time period by the number of substrates subjected to said plating treatment in said monitoring time period, and then defining an additive consumption predictive value by adding said plating-dependent consumption predictive value to an elapsed time-dependent consumption amount of said additive;

allowing said plating solution analyzation machine to define an additive concentration based on said additive concentration on an immediately preceding monitoring occasion and said additive consumption predictive value; and

allowing said additive replenishing apparatus to replenish a predetermined amount of said additive into said plating solution tank if said replenishing amount, which is obtained by multiplying a value defined as subtraction of said additive concentration on said recent monitoring occasion from a target concentration of said additive by an amount of said plating solution, reaches a predetermined critical value or more.

5. The method for management of a plating solution according to claim 4, wherein said elapsed time-dependent consumption amount of said additive is computed by multiplying said monitoring time period by an additive concentration variation coefficient per unit time having a constant value.

6. The method for management of a plating solution according to claim 4,

wherein said plating treatment additive consumption coefficient is an amount of said additive consumed for one substrate to be subjected to said plating treatment, in the step of measuring said concentration of said additive contained in said plating solution inside said plating solution tank once every predetermined time period and then computing a value by dividing said amount of said additive consumed in said plating treatment performed in the predetermined time period by the number of substrates subjected to said plating treatment so as to define a plating treatment additive consumption coefficient, and

said one substrate corresponds to a value obtained by converting an actual substrate area into a standardized unit based on a substrate having a certain area.

7. The method for management of a plating solution according to claim 4,

wherein said plating solution is a copper-plating solution, and

said additive is a thiol-based organic additive.

8. The method for management of a plating solution according to claim 4, wherein said plating treatment performed by immersing said substrate in said plating cell of said plating system is a plating treatment to form a plurality of types of plating films with different film thicknesses on said substrate.