TW201529906A - Bottom-up fill in damascene features - Google Patents

Abstract

The embodiments herein relate to methods and apparatus for filling features with copper by a bottom-up fill mechanism without the use of organic plating additives. In some cases, filling occurs directly on a semi-noble metal layer, without the deposition of a copper seed layer. In other cases, the filling occurs on a copper seed layer. Factors such as the polarization of electrolyte, the use of a complexing agent, electrolyte pH, electrolyte temperature, and the waveform used to deposit material may contribute to promoting the bottom-up fill.

Description

Bottom-up fill in the mosaic feature

The present invention relates to filling in a mosaic feature, particularly with respect to bottom-up filling in the mosaic feature.

In the damascene process, copper is deposited in features on partially fabricated semiconductor substrates. Conventional copper deposition typically occurs in two steps. First, a copper seed layer is deposited on the substrate by a physical vapor deposition (PVD) process. Next, copper is electroplated onto the seed layer to fill the features. As the critical dimension of the damascene interconnect decreases over time, it is increasingly difficult to achieve uniform copper coverage across all surfaces by depositing a seed layer by a PVD process. Inhomogeneous seed layer coverage is problematic because the thin copper seed regions are particularly susceptible to oxidation and dissolution in the electrolyte during the initial stages of the electroplating process. In other words, a thinner seed region (eg, on the sidewalls of the features) tends to dissolve in the electrolyte to create a discontinuous metal seed layer. When electroplating occurs on a discontinuous seed layer, the plating results are not uniform and defects may be introduced.

Certain techniques (eg, immersion under high voltage initial plating conditions, seed pretreatment, etc.) can be used to reduce seed dissolution. However, even with the full use of these techniques, it is expected that some amount of seed crystals will dissolve. Therefore, there is a need for a method of depositing copper in a semiconductor feature without the need to deposit a copper seed layer.

A technique has been developed to avoid the use of PVD to deposit copper seed crystals by directly plating a copper seed layer onto a semi-precious metal surface (e.g., a layer of tantalum). However, the process for plating the seed layer and the subsequent process for filling the features essentially require different electrolytes, and the copper plating process must therefore occur in a two-part process.

The electrolyte used to electroplate copper on the seed layer in the damascene interconnect typically contains copper salts, acids, halide ions, accelerators, inhibitors, and homogenizers. Copper salts are the source of copper for deposition. Acids are commonly used to control the conductivity of electroplating baths. Halide ions can act as a bridge to aid adsorption of some organic additives (eg, accelerators, inhibitors, and/or homogenizers) on the surface of the substrate, which facilitates the conventional bottom-up filling mechanism (described below).

In conventional copper plating processes, organic additives are critical to achieving the desired metallurgical, film uniformity, defect control, and filling effects. However, the concentration of organic additives may vary over time, so the electrolyte composition must be carefully tracked to ensure proper plating results. In many cases, the concentration of the additive is very low and it is difficult to accurately track the electrolyte composition within the range of relevant tolerances. Because of the above difficulties, some of the substrates may be plated in a plating bath that is not properly balanced, and may not be suitable for reuse. Accordingly, a need exists for a method of electroplating copper into a semiconductor feature without the use of conventional organic additives such as inhibitors, accelerators, or homogenizers.

Some embodiments herein relate to methods and apparatus for performing bottom-up filling in features on a substrate. In one embodiment of the embodiments herein, a method for performing a single-step electrofill process to fill features on a partially fabricated integrated circuit is provided. The method can include (a) receiving a substrate having an exposed semi-noble metal layer and a plurality of features thereon; (b) contacting the substrate with the electrolyte, and the electrolyte having (i) a copper cation of between about 1 and 100 mM; and (ii) a complexing agent that forms a complex with the copper cation, wherein the electrolyte is substantially free of inhibitors, accelerators, and homogenizers; and (c) is in contact In the case of an electrolyte, copper is electroplated into the features by a bottom-up filling mechanism with a potential of the electrodeposited substrate between about 0.03 and 0.33 V relative to the NHE reference electrode.

In various embodiments, the inhibitor, accelerator, or homogenizer does not substantially contribute to the bottom-up filling mechanism. The bottom-up filling can be performed directly on the semi-precious metal layer without first forming a seed layer. A variety of different waveforms can be used. In some cases, the copper plating step in operation (c) includes applying a modulation waveform that alternately generates a current pulse between the first level and the second level, wherein the copper is first at the first level Deposited on the substrate and copper etched on the substrate previously plated on the substrate at a second level. The second current level for copper etching can have an absolute value of less than 0.1 mA for a 300 mm diameter wafer. In some embodiments, the current pulses alternating between the first current level and the second current level have a frequency between about 100-1000 Hz. In these or other instances, the plated surface of the substrate can withstand a current density of between about 0.004 and 0.4 mA/cm^<2>.

Some different combinations of reagents can be used. In some embodiments, the wrong agent is selected from the group consisting of ethylenediaminetetraacetic acid (EDTA), nitrilotriacetic acid (NTA), citric acid, and glutamic acid. The group formed. In certain cases, the miscible is EDTA. The electrolyte can be at or about room temperature. In one embodiment, the electrolyte is maintained at a temperature of between about 20 and 80 °C, such as between about 50 and 70 °C. The pH of the electrolyte can be between about 1-5, and in some cases between about 1.5-3.5. The dissolved oxygen content of the electrolyte can be about 2 ppm or less.

The methods described herein can be used to electroplate a variety of different metals. In some cases, the semi-precious metal layer comprises a material selected from the group consisting of rhodium, tungsten, cobalt, rhodium, platinum, palladium, aluminum, gold, silver, rhodium, and ruthenium. In certain cases, the semi-precious metal layer is tantalum. In some embodiments, at least a portion of the features on the semiconductor substrate have an opening width of about 100 nm or less. For example, in some cases, the features have a width of about 20 nm or less.

In another embodiment of the disclosed embodiment, a method of depositing copper in features on a partially fabricated integrated circuit is provided. The method can include (a) receiving a substrate having a plurality of features and a copper seed layer thereon; (b) contacting the substrate with an electrolyte having a copper cation of between about 1 and 100 mM, Wherein the electrolyte is substantially free of inhibitors, accelerators, and homogenizers; and (c) electroplating the copper by a bottom-up filling mechanism with a potential of about 0.03-0.33 V relative to the NHE reference electrode To the feature.

In some embodiments, the electrolyte is maintained at a temperature between about 20-80 °C during plating, such as between about 20-50 °C. The plated surface of the substrate can withstand a current density of between about 0.004 and 0.4 mA/cm ² . In some embodiments, the pH of the electrolyte can be between about 1-5, such as between about 1.5-3.5. The disclosed method can be used to fill relatively small features. In some cases, at least a portion of the features have a width of about 100 nm or less, such as between about 20 nm or less. In some embodiments, the copper plating step in operation (c) includes applying a quiescent current controlled current to the substrate.

These and other features will be described below with reference to related figures.

In this specification, the terms "semiconductor wafer", "wafer", "substrate", "wafer substrate", and "partially fabricated integrated circuit" are used interchangeably. Those of ordinary skill in the art will appreciate that the term "partially fabricated integrated circuit" can refer to a germanium wafer during any of a number of stages of the fabrication phase of a plurality of integrated circuits. Wafers or substrates used in the semiconductor device industry typically have a diameter of 200 mm, or 300 mm, or 450 mm. In addition, the terms "electrolyte", "electroplating bath", "plating bath", and "plating solution" can be used interchangeably. The following detailed description assumes that the invention is implemented on a wafer. However, the invention is not limited thereto. The work pieces can be of various shapes, sizes, and materials. In addition to semiconductor wafers, other work items that may utilize the present invention include various items such as printed circuit boards and the like.

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the embodiments. The disclosed embodiments may be practiced without some or all of these specific details. In other instances, well-known processing operations are not described in detail in order to unnecessarily obscure the disclosed embodiments. While the disclosed embodiments are described with respect to the specific embodiments, it is understood that this is not intended to limit the disclosed embodiments.

As noted above, conventional copper deposition processes typically employ organic additives such as inhibitors, accelerators, and homogenizers to achieve bottom-up filling. While the embodiments herein do not require the use of these additives and often benefit from the absence of such additives, these additives will be discussed below in order to be able to follow the disclosed embodiments. Inhibitor

While not wishing to be bound by any theory or mechanism of action, it is believed that the inhibitor (either alone or in combination with other plating bath additives) is a surface kinetic energy polarizing compound that causes a significant increase in voltage across the substrate-electrolyte interface. , especially when combined with surface chemisorption of halides (eg chloride or bromide). The halide acts as a bridge between the inhibitor molecule and the surface of the wafer. The inhibitor not only (1) increases local polarization of the substrate surface at the region where the inhibitor is present (relative to the region without the inhibitor), but (2) generally increases the polarization of the substrate surface. The increased polarization (local and/or general) corresponds to the increased resistivity/impedance and therefore corresponds to a slower plating process at the particular potential applied.

Although the inhibitor will adsorb on the surface of the substrate, it is believed that it does not bind to the deposited film and will degrade over time. Compounds that do not function primarily by adsorption on the surface of the substrate are not considered inhibitors. The inhibitor is typically a relatively large molecule and, in many instances, is essentially a polymer (e.g., polyethylene oxide, polypropylene oxide, polyethylene glycol, polypropylene glycol, etc.). Other examples of the inhibitor include a polyethylene having a functional group containing S and/or N, a bulk polymer of a polypropylene oxide, a polyethylene oxide, and a polypropylene oxide, and the like. The inhibitor may have a linear structure or a branched structure. Typically, inhibitor molecules of various molecular weights are co-present in commercial inhibitor solutions. To some extent due to the large size of the inhibitor, these compounds diffuse relatively slowly into the concave features. Accelerator

While not wishing to be bound by any theory or mechanism of action, it is believed that the accelerator (whether alone or in combination with other plating bath additives) tends to locally reduce the polarization associated with the presence of the inhibitor, and thus locally increase the electricity. Deposition rate. The reduced polarization is most pronounced in the region where the adsorbed adsorbent is most concentrated (i.e., the polarization decreases as a function of the local surface concentration of the adsorbed adsorbent). Examples of accelerators include, but are not limited to, dimercaptopropane sulfonic acid, dimercaptoethane sulfonic acid, mercaptopropane sulfonic acid, mercaptoethane sulfonic acid ( Mercaptoethane sulfonic acid), bis-(3-sulfopropyl) disulfide, and derivatives thereof. Although the accelerator becomes firmly adsorbed on the surface of the substrate due to the plating reaction and generally cannot move laterally on the surface, the accelerator usually does not bond into the film. Therefore, the accelerator remains on the surface while depositing metal. When the recess is filled, the local accelerator concentration on the surface inside the recess rises. The accelerator tends to be smaller molecules and exhibits a faster diffusion into the concave features than the inhibitor. A compound that does not function primarily by adsorption on the surface of the substrate is not considered an accelerator. Homogenizer

While not wishing to be bound by any theory or mechanism of action, it is believed that the homogenizing agent (either alone or in combination with other plating bath additives) will act as an inhibitor to counteract the accelerator-related polarization, especially in the field ( In the field region) and at the side walls of the feature. The homogenizer can locally increase the polarization/surface resistance of the substrate, thereby slowing down the local electrodeposition reaction in the region where the homogenizer is adsorbed. The local concentration of the homogenizer is determined to some extent by mass transfer. Thus, the homogenizing agent acts primarily on surface structures having a geometry that protrudes away from the surface. This action makes the surface of the electrodeposited layer "smooth". It is believed that the homogenizing agent reacts (or consumes) at the diffusion limited rate (or near the diffusion limiting rate) at the surface of the substrate, and thus continuous supply of the homogenizer generally facilitates maintaining uniform plating conditions over time.

The homogenizer compound is generally classified as a homogenizer based on its electrochemical function and influence, and does not require a specific chemical structure or chemical formula. However, the homogenizing agent usually contains one or more nitrogen, amine, sulfoxide, or imidazole, and may also contain a sulfur functional group. Compounds that do not function primarily by adsorption on the surface of the substrate are not considered to be homogeneous agents. Some homogenizing agents include: one or more rings of five and six constituent molecules, and/or conjugated organic compound derivatives. The nitrogen group can form part of the ring structure. In the amine-containing homogenizer, the amines may be primary, secondary or tertiary alkylamines. Further, the amine may be an aryl amine or a heterocyclic amine. Examples of amines include, but are not limited to, dialkylamines, trialkylamines, arylalkylamines, triazoles, imidazole, tetrazole. Tetrazole, benzimidazole, benzotriazole, piperididine, morpholines, piperazine, pyridine, oxazole, Benzoxazole, pyrimidine, quinoline, and isoquinoline. Imidazole and pyridine are particularly useful. The homogenizer compound may also include an ethoxide group. For example, the homogenizing agent can include a general backbone (e.g., Janus Green B) similar to that found in polyethylene glycol or polyethylene oxide having an amine moiety functionally inserted into the chain. Examples of epoxides include, but are not limited to, epihalohydrins (e.g., epichlorohydrin and epibromohydrin) and polyepoxide compounds. Polyepoxide compounds having two or more epoxide moieties bonded together by an ether-containing chain are particularly useful. Part of the homogenizer compound is polymerizable, but the rest is not. Examples of polymerizable homogenizer compounds include, but are not limited to, polyethylenimine, polyamidoamines, and reaction products of amines with various oxyepoxides or sulfides. An example of a non-polymerizable homogenizer is 6-mercapto-hexanol. Another example of a homogenizer is polyvinylpyrrolidone (PVP, polyvinylpyrrolidone). Promote bottom-up filling by organic additives

In the bottom-up filling mechanism, the recessed features on the plated surface tend to be plated with metal from the bottom to the top of the feature and from the sidewall toward the center of the feature. In order to achieve uniform filling and to avoid inclusion of voids in the features, it is important to control the rate of deposition in the features and in the field regions. In conventional applications, in order to achieve bottom-up filling, the three types of additives described above, each for selectively increasing or decreasing the polarization at the surface of the substrate, are necessary.

After immersing the substrate in the electrolyte, the inhibitor is adsorbed on the surface of the substrate, especially in an exposed area such as a field area. There is a substantial difference in inhibitor concentration between the top and bottom of the recessed feature during the initial plating phase. This difference exists due to the relatively large size of the inhibitor molecules and their corresponding slow transfer characteristics. During this initial plating, it is believed that the accelerator accumulates over the entire plating surface (including the bottom and sidewalls of the features) at a low and substantially uniform concentration. Since the accelerator diffuses into the feature faster than the inhibitor, the initial ratio of accelerator:inhibitor in the feature (especially at the bottom of the feature) is relatively high. The initial ratio of accelerator to inhibitor in this relatively high feature promotes rapid plating from the bottom of the feature up and from the sidewall. At the same time, due to the lower ratio of accelerator:inhibitor, the initial plating rate in the field region is relatively low. Thus, during the initial plating phase, the plating process occurs relatively quickly within the features and relatively slower in the field regions.

As the electroplating process continues, the features are filled with metal and the surface area within the features is reduced. Since the surface area is reduced and the accelerator remains substantially on the surface, the local surface concentration of the accelerator within the feature rises as the plating process continues. The accelerator concentration within this rising feature helps maintain a difference in plating rate that facilitates bottom-up filling.

In the later stages of the electroplating process (especially like overburden deposition), the accelerator may accumulate unintentionally in certain areas (eg, above the fill features), which results in a locally faster than expected plating process. Conventionally, a homogeneous agent is used to counteract this effect. The surface concentration of the homogenizer is greatest at the exposed area of the surface (i.e., not within the concave features), and convection is maximized here. It is believed that the homogenizing agent replaces the accelerator at some areas of the surface, increases local polarization, and reduces the local plating rate that would otherwise be plated at a greater rate than other locations on the deposit. In other words, the homogenizing agent tends to reduce, at least to some extent, the effects of accelerating compounds at the exposed areas of the surface, particularly the protruding structures. In conventional applications, in the absence of a homogenizing agent, the features may tend to overfill to create protrusions. Thus, in the later stages of the conventional bottom-up filling plating process, the homogenizer is helpful in producing relatively flat deposits.

The combined use of inhibitors, accelerators, and homogenizers allows the bottom to top and fill features from the sidewalls without voids while creating a relatively flat deposition surface. The exact nature/composition of the added compound is usually the business secret held by the additive supplier; therefore, information about the exact nature of these compounds is not publicly available. Electroplating without organic additives

One embodiment of the disclosed embodiment is a method of electroplating copper into features on a semiconductor substrate, and the substrate has an exposed semi-precious metal liner. In this embodiment, copper is electroplated directly onto the semi-precious metal liner, rather than on the copper seed layer. Although the electrolyte in this embodiment may include a complexing agent that is mismatched with copper in solution, the electrolyte is substantially free of organic additives such as inhibitors, accelerators, and homogenizers. When there are some small amounts of organic additives, this may be the case where the organic additive does not substantially contribute to the bottom-up filling mechanism. In other words, the bottom-up filling may even occur in the absence of an organic additive (when electroplating is performed under the same plating conditions other than this). Another embodiment of the disclosed embodiment is a method of electroplating copper into features on a semiconductor substrate, and the substrate has an exposed copper seed layer. As with the previous examples, this method can be practiced with an electrolyte that is substantially free of inhibitors, accelerators, and homogenizers. Although no organic additives are present, the method disclosed herein still achieves a bottom-up filling mechanism to fill the features. Method for electroplating on a semi-precious metal layer

In one embodiment, copper is electroplated onto the exposed semi-precious metal liner layer. The semi-precious metal backing layer can be tantalum, cobalt, tungsten, rhenium, platinum, palladium, aluminum, gold, silver, rhodium, ruthenium, or combinations thereof. The substrate with the exposed semi-precious metal layer is placed in a plating bath and immersed in an electrolyte having specific characteristics (as discussed later). A current is applied to the substrate to promote nucleation, followed by Volmer-Weber growth to form a three-dimensional copper island. The copper island continues to grow until it combines into a continuous copper film. The applied current is dependent on the composition of the electrolyte, but is typically controlled to provide a voltage between about 0 and 4 V relative to a normal hydrogen electrode (NHE) or about 0.03 and 0.33 for NHE. The voltage between V.

The electrolyte can be designed to help promote high nucleation density. One method of promoting high nucleation density is to use conditions that result in a relatively more polarized electrolyte. Some complexing agents (eg, ethylenediaminetetraacetic acid (EDTA), nitrogen triacetic acid (NTA), citric acid, glutamic acid, etc.) can be used to achieve an increase in electrolyte polarization with low copper concentration. These complexing agents form a complex with the copper ions dissolved in the electrolyte. The miscible agent combines with copper ions by, for example, electrostatic interaction and forms a soluble complex. In various examples, the dissimilar agent is shaped to partially enclose the copper ions and partially shield the copper ions. The miscible agent does not adsorb significantly on the surface of the substrate (at least to the extent of conventional plating additives (eg, inhibitors, accelerators, and homogenizers)). Therefore, the complexing agent employed herein is not an inhibitor (or accelerator, or homogenizer) compound. The polarization of the complex as described above will be explained in the experimental section below.

The wrong agent promotes high nucleation density. Although the conjugate is not an inhibitor (because it acts primarily by forming a complex with copper in solution, rather than by adsorption on the surface of the substrate), the complexing agent does function as a similar inhibitor so that The overpotential of copper electrodeposition is increased. In some embodiments, the concentration of the complexing agent is between about 1 and 100 mM, such as between about 1 and 20 mM, or between about 5 and 10 mM. The concentration of the cross-linking agent can be substantially similar (e.g., in the range of about 30%) to the concentration of copper cations (also measured in molar concentrations). In some cases, these concentrations are substantially equimolar (eg, in the range of about 10% or about 5%). In certain cases, the concentration of the binder and the concentration of copper cations are exactly equal. Since copper and the complexing agent form a complex together in a ratio of 1:1, it is advantageous to have a molar concentration of copper cations and a complexing agent. In other cases, these concentration changes are more pronounced. In some embodiments, the concentration of the cross-linking agent can be higher than the concentration of the copper cation. In some embodiments, a complexing agent with too much chemical dose can be beneficial because it can help achieve higher fractions of mismatched copper cations that can help achieve high nucleation densities on semi-precious metal surfaces.

In some embodiments, the tweaking agent can be omitted. When electroplating is performed on a semi-precious metal layer without a miscible agent, a modulated waveform can be used to help promote bottom-up plating. The modulated waveform will be discussed further below.

The low copper concentration provides a relatively high electrolyte polarization. In some embodiments, the concentration of copper cations is between about 1 and 100 mM, such as between about 1 and 20 mM, or between about 5 and 10 mM. The effect of different copper concentrations on solution polarization will be further discussed in the experimental section below.

Another factor that affects the polarization of the electrolyte is the pH. In general, electrolytes with higher pH values are more polarized. In some embodiments, the pH of the electrolyte is between about 1-5, such as between about 1.5-3.5. The effect of different electrolyte pH values will be further discussed in the experimental section below.

The polarization of the electrolyte is also affected by the temperature of the electrolyte. In general, lower temperatures result in higher electrolyte polarization. However, lower temperatures also result in lower deposition rates and more conformal films. In the case of bottom-up filling, the conformal film is not ideal because it may result in the inclusion of slits/holes in the features. Therefore, the advantage of increased polarization obtained at low temperatures should be balanced with the increase in deposition rate at high temperatures and the advantages of less conformal films. In some embodiments, the deposition occurs at a temperature between about 20-80 °C, such as between about 50-70 °C. Conventional bottom-up filling processes typically occur at about 20-25 °C. An advantage of the disclosed embodiment is that the filling process occurs at elevated temperatures, which can be deposited at a higher rate than conventional processes (typically occurring at lower temperatures).

The waveform applied to drive electrodeposition also affects the filling mechanism. In some embodiments, a DC current is used (eg, using galvanostatic or galvanodynamic control). In other embodiments, a modulated waveform is used (eg, using a current that alternates between a deposition current and an etch current). The use of a modulated waveform can result in a less conformal film, which is beneficial in the case of bottom-up filling.

As is known to those of ordinary skill in the art, the maximum current (limited current) for deposition is affected by the availability of copper at the substrate-electrolyte interface. If the current becomes higher than the acceptable level, the electrolyte may suffer from copper depletion, resulting in poor deposition results. In other words, the amount of copper at the interface may not be sufficient to maintain the reduction reaction at the relevant current level. Instead, a parasitic reaction may occur to maintain the current delivered to the substrate. For example, the electrolyte itself may begin to decompose and create a gas at the plating interface resulting in uneven plating, and in some cases, the formation of nodular growth on the substrate. Although care should be taken to ensure that the current is not so high as to completely remove the previously deposited metal, the maximum current during etching is typically only limited by the hardware limit.

In some embodiments, the current level used to deposit the material is between about 0.001 and 1.5 A, such as between about 0.05 and 1.4 A, or between about 0.05 and 1 A (based on 300 mm wafer). In these or other embodiments, the absolute value of the current level used to etch the material is between about 0.035 and 0.25 A, for example, between about 0.04 and 0.2 A, or less than about 0.1 A (based on a 300 mm wafer). . In some cases, the current used to etch the material is negative. The current density during electroplating can be between about 0.1 and 2 mA/cm ² . The current density during etching can be between about 0.05 and 0.3 mA/cm ² . In embodiments in which a modulated waveform (e.g., a square wave) is used, the waveform frequency can be between about 100 and 1000 Hz. In other words, this waveform can alternate between the deposition current and the etch current at the exposed frequency. The effect of different waveforms on the plating results will be further discussed in the experimental section below.

Without wishing to be bound by a particular theory or mechanism of action, it is believed that when a modulated waveform is used, it may result in redistribution of material over the features, as well as in the features. During the etched portion of the waveform, copper can be selectively etched near the top of the feature. The lower copper in the feature (near the bottom of the feature) is less likely to be etched away. This selective etching can effectively reduce the surface area of copper that can be used (and is suitable for) in the features of the plating. During the subsequent deposition portion of the waveform, since the energy required to deposit in the bottom region will be lower than the energy required to deposit the region near the top of the feature, copper will tend to deposit more toward the bottom of the feature, while remaining The copper is gathered here. Although both deposition and etching operations act on all parts of the feature, a significant amount of deposition occurs near the bottom of the feature (as compared to the top of the feature), and a larger amount of etching occurs near the top of the feature (compared to At the bottom of the feature). By repeating the cycle of deposition and etching, copper can be redistributed within the features to achieve bottom-up filling. Another factor that can contribute to the bottom-up filling mechanism is the relatively low deposition rate. This is because the plating takes place over a longer period of time, so there is more time to redistribute the copper in the features, thus providing good filling results.

In the case of DC waveforms, the mechanism of action that promotes bottom-up filling may be slightly different. When copper is combined with a complexing agent (for example, a relatively weak complexing agent such as NTA and/or glutamic acid) and plated at a low deposition rate, the filling mechanism becomes less conformal, resulting in a lower portion of the feature. And fill it up. The choice of the wrong agent, the concentration of copper in the electrolyte, the pH of the electrolyte, and the temperature of the electrolyte all affect the polarization of the solution. The bottom-up fill has proven to occur indeed when the substrate is maintained at a potential of between about 0.03 and 0.33 V relative to the NHE reference electrode. This voltage range has been shown to successfully promote bottom-up filling. If the voltage is clearly below this range, the plating current is too low and very little copper will be deposited; if the voltage is above this range, the filling is observed to be conformal rather than bottom to top. By applying a current in such a manner that the voltage falls within the above range, the bottom-up filling can be achieved. In some embodiments, this voltage is equivalent to a potential of about -0.3 to -0.6 V as compared to a relative mercury sulfate reference electrode (MSE) used in the experiments described below (eg, about -0.4 to -0.5) V). By maintaining the voltage within this range in combination with the above electrolyte conditions, bottom-up filling can be achieved without the use of organic additives such as inhibitors, accelerators, and homogenizers. In some cases, the electrolyte may contain traces of organic additives, but these additives do not substantially contribute to the bottom-up filling mechanism.

1 provides a flow chart depicting a method of filling features on a semiconductor substrate with an exposed semi-precious metal layer. Process 100 begins at block 101 where a substrate having an exposed semi-precious metal layer is received/disposed in an electrodeposition chamber. The substrate typically has features that are to be filled by an electrodeposition process. In some cases, the features can be trenches having a width between about 10 and 100 nm, such as a width of between about 50 and 100 nm. In these or other instances, the features can have a width of about 100 nm or less, such as about 20 nm or less. Next, in block 103, the substrate is contacted with an electrolyte (which is substantially free of inhibitors, accelerators, and homogenizer compounds). The electrolyte may have the characteristics described above, such as: a binder, a low concentration of copper cations, and a particular pH and/or temperature. These factors can contribute to a relatively highly polarized electrolyte. In block 105, a current is applied to the substrate. The applied current can be a direct current or a modulated current and is designed to maintain a substrate potential of between about 0.03 and 0.33 V relative to the NHE reference electrode. This substrate potential in combination with the disclosed electrolyte promotes bottom-up filling without the use of an organic plating additive. Plating on a copper seed layer

The above disclosed method for performing electrodeposition of copper on a semi-precious metal layer can be extended to electroplating on a copper seed layer. Although this embodiment fails to achieve the advantages of single-step filling (because the copper seed layer is independently deposited from the copper fill material), this embodiment does preserve copper plating using bottom-up filling without the use of organic plating additives. The advantages.

In general, the teachings disclosed above regarding electrolyte composition/pH/temperature/waveform are also applied to electroplating on a copper seed layer. However, the above considerations are less important when electroplating on copper seed layers, but other considerations are more important. For example, when electroplating occurs on a copper seed layer, the intercalating agent can be omitted in the electrolyte. This is because when electroplating is performed on a semi-precious metal layer, the degree of polarization required to achieve a proper electroplating result is higher than that of electroplating on copper, so in the case of electroplating on a semi-precious metal layer, the miscluster More important.

In addition, the use of modulated waveforms is somewhat more complicated when electroplating is performed on a copper seed layer. As with electroplating on semi-precious metals, the applied current can be quiescent current or dynamic current. This additional complexity is created because in all areas of the substrate it is possible that all of the copper (including the copper seed layer) will dissolve during the etched portion of the modulated waveform. If this occurs, there will be no suitable surface for electroplating on this area and the plating results will be poor. Although bottom-up filling can be achieved using the disclosed method along with the modulated waveform, care should be taken to avoid seed dissolution. Therefore, the etched portion of the waveform can be delayed until a sufficient amount of copper is plated in the initial portion of the plating process. Electroplating on the copper seed layer can also be achieved using a DC current waveform.

Compared to the embodiment in which the electroplating process occurs directly on the semi-precious metal layer, the optimum deposition temperature in the embodiment using the copper seed layer will be lower. In some cases, when electroplating is performed on a copper seed, the temperature is maintained at between about 20 and 80 ° C, for example, between about 20 and 50 ° C.

Without wishing to be bound by a particular theory or mechanism of action, the bottom-up filling mechanism of credit on copper seeds will be similar to that described above for electroplating on semi-precious metal layers (eg, germanium). The fill mechanism. However, in various cases of electroplating on copper seed, it is not necessary to use a dissolving agent or a modulated waveform to promote nucleation on the copper seed layer.

Figure 2 provides a flow chart of a method of electroplating copper onto a copper seed layer. Process 200 begins at block 201 where a substrate having an exposed copper seed layer is received/disposed in an electrodeposition chamber. There are typically features on the substrate that are to be filled by an electrodeposition process. In some cases, the features can be trenches having a width between about 10 and 100 nm, such as a width of between about 50 and 100 nm. Next, in block 203, the substrate is contacted with an electrolyte that is substantially free of inhibitors, accelerators, and homogenizer compounds. The electrolyte may have the characteristics described above, such as: a binder, a low concentration of copper cations, and a particular pH and/or temperature. In some embodiments employing a copper seed layer, no miscible agent is used. In block 205, a current is applied to the substrate. The applied current can be a direct current or a modulated current and is designed to maintain a substrate potential of between about 0.03 and 0.33 V relative to the NHE reference electrode. This substrate potential in combination with the disclosed electrolyte promotes bottom-up filling without the use of an organic plating additive. device

Many device configurations can be used in accordance with the embodiments described herein. An example device includes a clamshell fixture that seals the backside of the wafer away from the plating solution while allowing plating on the surface of the wafer. The clamshell clamp can support the wafer, for example, via a seal disposed over the bevel of the wafer, or by a vacuum applied to the back of the wafer and a seal applied adjacent the bevel.

The clamshell fixture should enter the plating bath in a manner that allows the plated surface of the wafer to be well wetted. The quality of substrate wetting is affected by a number of variables including, but not limited to, housing rotational speed, vertical entry speed, and angle of the housing relative to the plating bath surface. These variations and their effects are discussed further in U.S. Patent No. 6,551, 487, incorporated herein by reference. In some embodiments, the electrode rotation rate is between about 5 and 125 RPM, the vertical entry speed is between about 5 and 300 mm/s, and the angle of the housing relative to the electroplating bath surface is between about 1-10 degrees. One of the goals of optimizing these variables for a particular application is to achieve good wetness by completely venting air from the wafer surface.

The electrodeposition method disclosed herein can be described with reference to various plating tool devices, and can be applied to various plating tool devices. An example of an electroplating apparatus that can be used in accordance with embodiments herein is the Sabre tool of Lam Research. Electrodeposition (including substrate immersion) and other methods disclosed herein can be performed in the elements that form larger electrodeposition equipment. Figure 3 shows a top plan view of an exemplary electrodeposition apparatus. Electrodeposition apparatus 900 can include three separate plating modules 902, 904, and 906. Electrodeposition apparatus 900 can also include three separate modules 912, 914, and 916 configured for various processing operations. For example, in some embodiments, one or more of the modules 912, 914, and 916 can be a spin rinse drying (SRD) module. In other embodiments, one or more of the modules 912, 914, and 916 may be a post-electrofill module (PEM), and each is configured to pass through the electroplating modules 902, 904 on the substrate. And one of the processes 906 performs functions such as edge beveling of the substrate, backside etching, and acid cleaning.

Electrodeposition apparatus 900 includes a central electrodeposition chamber 924. The central electrodeposition chamber 924 houses a chamber that is a chemical solution for the plating solution in the plating modules 902, 904, and 906. Electrodeposition apparatus 900 also includes a dosing system 926 that can store and deliver the electrolyte composition of the plating solution. The chemical dilution module 922 can store and mix the chemicals to be used as an etchant. The filtration and delivery unit 928 can filter the plating solution for the central electrodeposition chamber 924 and pump it to the plating module.

System controller 930 provides the electronics and interface controls required to operate electrodeposition apparatus 900. System controller 930 (which may include one or more physical or logical controllers) controls the characteristics of some or all of plating apparatus 900. System controller 930 typically includes one or more memory devices and one or more processors. The processor may include a central processing unit (CPU) or computer, analog and/or digital input/output connections, stepper motor control boards, and the like. The instructions described herein to implement appropriate control operations can be executed on the processor. These instructions may be stored on a memory device associated with system controller 930 or may be provided over a network. In some embodiments, system controller 930 executes system control software.

The system control software in the electrodeposition apparatus 900 can include instructions to control timing, mixture of electrolyte components (including concentration of one or more electrolyte components), inlet pressure, plating bath pressure, plating bath temperature, Substrate temperature, current and potential applied to the substrate and any other electrodes, substrate position, substrate rotation, and other parameters of the particular process performed by electrodeposition apparatus 900. The system control logic may also include instructions for electroplating under conditions that are tailored to a low copper concentration electrolyte and a relatively high overpotential associated therewith. For example, the system control logic can be configured to provide a relatively low current density during bottom-up filling. The control logic can also be configured to provide some level of mass transfer to the wafer surface during plating. For example, the control logic can be configured to control the flow of electrolyte during plating to ensure that sufficient mass is transferred to the wafer such that the substrate does not suffer from copper depletion conditions. In some embodiments, the control logic is operable to provide different degrees of mass transfer during different stages of the electroplating process (eg, during mass transfer during the bottom-up fill phase is higher than during the overlying phase, or under The mass transfer during the upper filling phase is lower than during the overlying phase). Additionally, the system control logic can be configured to maintain the concentration of one or more electrolyte compositions or the pH of the electrolyte within any of the ranges disclosed herein. As a specific example, system control logic can be designed or configured to maintain a copper cation concentration between about 1-100 mM. In another example, the system control logic can be configured to apply a current such that the substrate is maintained at a potential of between about 0.03 and 0.33 V relative to the NHE electrode. System control logic can be configured in any suitable manner. For example, various processing tool component subroutines or control objects can be programmed to control the operations of the processing tool components necessary to perform various processing tool processes. The system control software can be encoded in any suitable computer readable programming language. This logic method can also be implemented as hardware in a programmable logic device (eg, an FPGA), an ASIC, or other suitable tool.

In some embodiments, the system control logic includes an input/output control (IOC) sequencing instruction to control the various parameters described above. For example, various stages of the electroplating process can include one or more instructions executed by system controller 930. Instructions for setting process conditions during the immersion process stage may be included in the corresponding immersion recipe stage. In some embodiments, the plating recipe stages can be sequentially arranged such that all instructions of the electroplating process stage are performed simultaneously with the process stage.

In some embodiments, the control logic can be divided into a number of parts, such as multiple programs or multiple program sections. Examples of the logic portion for this purpose include a substrate positioning portion, an electrolyte composition control portion, a pressure control portion, a heater control portion, and a potential/current power supply control portion.

In some embodiments, a user interface associated with system controller 930 can be provided. The user interface can include a graphical software display that displays screens, device and/or process conditions, and user input devices (eg, indicator devices, keyboards, touch screens, microphones, etc.).

In some embodiments, the parameters adjusted by system controller 930 can be related to process conditions. Non-limiting examples include plating bath conditions (temperature, composition, pH, flow rate, etc.) at various stages, substrate position (rotation rate, linear (vertical) velocity, horizontal angle, etc.), and electrical conditions (current) , potential, etc.) and so on. These parameters can be provided in the form of recipes to users who can log in to use this user interface.

Signals for monitoring the process can be provided from various processing tool sensors and by analog and/or digital input connections of system controller 930. The signals used to control the process can be output on the analog and digital output connections of the processing tool. Non-limiting examples of processable sensor sensors that can be monitored include: mass flow controllers, pressure sensors (such as pressure gauges), thermocouples, optical position sensors, and the like. Properly programmed feedback and control algorithms can be used with data from these sensors to maintain process conditions.

In an embodiment, the instructions can include inserting the substrate into the wafer holder, tilting the substrate, biasing the substrate during immersion, and electrodepositing a copper-containing structure on the substrate.

Transfer tool 940 can select a substrate from a substrate cassette (eg, cassette 942 or cassette 944). The cassette 942 or 944 may be a front opening unified pod (FOUP). The FOUP is designed to securely and securely hold the enclosure of the substrate in a controlled environment and allows the substrate to be removed for processing or measurement by a tool equipped with a suitable loading magazine and robotic arm handling system. Transfer tool 940 can utilize a vacuum attachment or some other additional mechanism to clamp the substrate.

Transfer tool 940 can be coupled to wafer handling station 932, cassette 942 or 944, transfer station 950, or aligner 948. Transfer tool 946 can take the substrate from transfer station 950. Transfer station 950 can be a transfer slot or location for transfer tools 940 and 946 that can pass back and forth through the aligner 948. However, in some embodiments, to ensure that the substrate is properly aligned on the transfer tool 946 for accurate delivery to the plating module, the transfer tool 946 can align the substrate with the aligner 948. Transfer tool 946 can also deliver the substrate to one of plating modules 902, 904, or 906, or to one of three independent modules 912, 914, and 916 configured for various processing operations.

An example of a processing operation according to the above method may be performed as follows: (1) electrodepositing copper onto a substrate to form a copper-containing structure in the plating module 904; (2) cleaning and drying the substrate in the SRD module 912 And (3) performing edge beveling in module 914.

Devices configured to allow efficient cycling of substrates through continuous plating, cleaning, drying, and PEM processing operations can be useful for embodiments that are used in a manufacturing environment. To achieve this, the module 912 can be configured to rotate the wash dryer and the edge bevel cut chamber. With such a module 912, the substrate will only need to be transported between the plating module 904 and the module 912 for copper plating and EBR operation.

An alternate embodiment of an electrodeposition apparatus 1000 is schematically illustrated in FIG. In this embodiment, the electrodeposition apparatus 1000 has a set of electroplating cells 1007 (each comprising a plating bath) in a paired configuration or a plurality of "duet" configurations. In addition to electroplating itself, electrodeposition apparatus 1000 can perform various other electroplating related processes and sub-steps such as spin cleaning, spin drying, metal and wet etching, electroless deposition, pre-wetting and pre-chemical processing, reduction, tempering, Photoresist stripping and surface pre-activation. The electrodeposition apparatus 1000 is schematically shown in FIG. 4 in a manner viewed from above, and only a single layer or "floor" is shown in the drawing, but those having ordinary knowledge in the art should immediately understand such a class. devices (e.g.: Novellus Sabre ^TM 3D tool) has two or more layers may be "stacked" (stacked) on each other, and each may be the same or different types of processing stations.

Referring again to FIG. 4, the substrate 1006 to be plated is typically fed to the electrodeposition apparatus 1000 via the front end loading FOUP 1001, and in this example from the FOUP to the main substrate processing area of the electrodeposition apparatus 1000 via the front end robotic arm 1002; The front end arm 1002 can be driven by the shaft 1003 to contract and move the substrate 1006 from one station to another station in the multi-dimensional direction (in this example, two front-end access stations 1004 and two front-end access stations 1008 are shown). The front end access stations 1004 and 1008 may include, for example, a pretreatment station and a spin rinse drying station (SRD). The front end arm 1002 is moved laterally from one side to the other by the robot arm track 1002a. Each of the substrates 1006 can be held by a cup/conical assembly (not shown) driven by a shaft 1003 coupled to a motor (not shown), and the motor can be attached to the mounting frame 1009. Four "duet" plating baths 1007 (a total of eight plating baths 1007) are also shown in this example. Electroplating bath 1007 can be used to electroplate copper (for copper-containing structures) and electroplated solder materials (for welded structures). A system controller (not shown) may be coupled to the electrodeposition apparatus 1000 to control some or all of the characteristics of the electrodeposition apparatus 1000. The system controller can be programmed, or otherwise configured to execute instructions in accordance with the processes previously described herein.

The lithographic patterning of the film typically comprises part or all of the following steps, and each step is performed using some suitable tool: (1) applying a photoresist to the workpiece using a spin coating or spray tool (eg, formed thereon) (2) using a hot plate, or a furnace, or other suitable curing tool to cure the photoresist; (3) using a tool such as a wafer stepper to expose the photoresist to visible light, Or UV light, or X-ray; (4) developing the photoresist with a tool such as a wet bench or a spray processor to selectively remove the photoresist and pattern the photoresist; (5) using dry or plasma An auxiliary etch tool transfers the photoresist pattern into the underlying film or workpiece; and (6) removes the photoresist using a tool such as an RF or microwave plasma photoresist stripper. In some embodiments, an ashable hard mask layer (eg, an amorphous carbon layer) and another suitable hard mask (eg, an anti-reflective layer) may be deposited prior to application of the photoresist.

It is to be understood that the configurations and/or methods described herein are exemplary in nature and that the particular embodiments or examples are not considered in a limiting sense. The particular routine work or method described herein can represent one or more of any number of processing strategies. Thus, the various actions illustrated can be performed in the following order, in the order illustrated, in other sequences, in parallel, or in part. Similarly, the order of the processes described above can be changed.

The subject matter of the disclosure includes various processes, systems and configurations, and other combinations of features and functions, actions, and/or features disclosed herein, and any and all of its equivalents. combination. experiment

Some experimental studies have shown that the methods disclosed herein can be used to achieve bottom-up filling without organic plating additives. The preliminary results provided in this section relate to cyclic voltammetry (CV) scans that show different parameters (eg, characteristics of the complex, copper cation concentration, solution pH, and solution temperature) for polarization. Have the impact. The final results provided in this section show the results of the filling of features filled according to different plating conditions. All results presented in this section were generated without the use of organic plating additives. The test pieces used for electroplating all have a plating area of about 1 cm ² .

Figure 5 shows the CV results illustrating the relative effects of different combinations of agents on the polarization of the electrolyte. The electrolyte tested contained 5 mM copper cations and 5 mM of associated complexing agent. These CVs are platinum rotating disk electrodes (RDE, collected in a beaker at a rotational speed of 200 RPM, at a scan rate of 10 mV/s, and with a mercury sulfate reference electrode (MSE). Rotating disk electrode). The dissolved oxygen was controlled at about 1 ppm, and the pH was adjusted to a pH of about 3 by tetramethylammonium hydroxide (TMAH) or sulfuric acid. Ethylenediaminetetraacetic acid (EDTA) was most strongly polarized, and sulfate (sulfate) (SO ₄₎ solution is not the best polarization. Since copper forms a complex with water only in this case, the sulfate solution is the least polarized.

Figure 6 shows CV results illustrating the relative effects of different copper ion concentrations and pH levels on the polarization of solutions containing EDTA as a binder. For each of these solutions, the concentration of copper cations and the concentration of EDTA are equal molar amounts. These results were collected on platinum RDE at a rotational speed of 200 RPM, at a scan rate of 10 mV/s, and with pH adjustment to a specified level using TMAH or sulfuric acid. The reference electrode is an MSE electrode. Lower copper concentrations and higher pH levels result in more polarized solutions.

Figure 7 shows CV results illustrating the effect of electrolyte temperature on the polarization of a solution containing 10 mM copper cations and 10 mM EDTA. These data were collected at a rotational speed of 200 RPM at a scan rate of 10 mV/s and at different temperatures on a PVD copper seed test strip attached to the RDE electrode. The reference electrode is MSE in this case, the dissolved oxygen level is about 1 ppm, and the pH is adjusted to about 2.3 using TMAH or sulfuric acid. Scanning shows that lower temperatures result in a higher degree of polarized solution.

8A-8C show scanning electron microscope (SEM) images showing electroplating of a seed crystal groove after electroplating in an electrolyte containing 10 mM copper cation and 10 mM EDTA (attachment) The result of the filling produced by the RDE electrode). The pH of each electrolyte was adjusted to a pH of about 2.3 using TMAH or sulfuric acid. The dissolved oxygen level of each electrolyte was about 1 ppm. The temperature of each electrolyte was about 70 °C. In this case, the trenches have a width of about 80 nm, however this technique can also be applied to narrower trenches (e.g., trenches about 20 nm wide). The rotational speed of the RDE is about 200 RPM, and the reference electrode is the MSE electrode. Each of the test pieces shown in Figs. 8A-8C was subjected to electroplating under static current conditions. The test piece shown in Fig. 8A was plated at 0.4 mA, the test piece shown in Fig. 8B was plated at 0.6 mA, and the test piece shown in Fig. 8C was plated at 1 mA.

The test piece plated at 0.4 and 0.6 mA reached a void free bottom-up fill in the 80 nm trench. However, when the DC current rises to 1 mA, a gap is observed, as shown by the white arrow in Fig. 8C. The filling quality was checked after the tempering operation, but no holes were found in the test pieces (i.e., the test pieces shown in Figs. 8A and 8B) which were plated at 0.4 and 0.6 mA. Experiments conducted under a wide range of conditions have shown that maintaining the applied voltage between about -0.3 and -0.6 V (e.g., -0.4 to -0.5 V) relative to the MSE reference electrode can be achieved without holes. And fill it up. Since the MSE electrode is non-standard and can produce different potential readings as the particular electrode is filled, its associated potential results relative to the standard NHE electrode are also recorded. Compared to the NHE electrode, the hole-free filling can be achieved from the bottom up while maintaining the voltage in the range of about 0.03-0.33 V. When the voltage is outside this range, a gap is observed.

Figures 9A-9C show SEM images illustrating the results of the filling of the seed crystal trench test strips, which are attached to the RDE and electroplated using the modulated waveform at different temperatures, currents, and plating times. Each of the electrolytes used for electroplating in Figures 9A-9C contains 10 mM copper cations and 10 mM EDTA, and is adjusted to a pH of about 2.3 with TMAH or sulfuric acid, and also contains about 1 ppm of dissolved oxygen. . The rotational speed of each example was about 200 RPM, and the reference electrode was an MSE electrode. For each example, the modulated waveform is a square wave alternating between deposition current and etch current with a frequency of about 100 Hz (a frequency between about 50-1000 Hz has been tested and presents a good fill result). For each deposition, the etch current was set to -0.05 mA and the voltage was maintained at about -0.4 to -0.5 V relative to the MSE electrode.

The test piece shown in Fig. 9A was plated at room temperature for 20 minutes at a deposition current level of 0.45 mA. The test piece shown in Fig. 9B was subjected to electroplating at 50 ° C for 20 minutes with a deposition current of 0.5 mA. The test piece shown in Fig. 9C was subjected to electroplating at 70 ° C for 8 minutes in the case of a 1.4 mA deposition current. Although the filling occurs faster at higher temperatures, each of the examples shown in Figures 9A-9C achieve a bottomless fill without voids. In fact, the filling rate is about 10 times higher at 70 ° C than in the case of room temperature.

Figures 10A-10B show SEM images of a seed crystal channel test piece attached to an RDE and plated in an electrolyte having a different compounding agent. The test piece shown in Fig. 10A was plated in an electrolytic solution containing 5 mM of copper cation and 5 mM of NTA, adjusted to a pH of about 3.1 with TMAH or sulfuric acid, and a dissolved oxygen content of about 1 ppm. The test piece shown in Fig. 10B was electroplated in an electrolyte containing 10 mM of copper cation and 10 mM of glutamic acid, and adjusted to a pH of about 3.1, and a dissolved oxygen content of about 1 ppm. For each example, the spin rate was about 200 RPM, the temperature was room temperature, the reference electrode was the MSE electrode, and the waveform used to drive the deposition was a quiescent current (Figure 10A/NTA was 0.1 mA, while Figure 10B/glutamic acid was 0.6). mA). Both examples achieve a good quality bottom-up fill.

In another experiment, the seed crystal test piece was electroplated in an electrolyte containing no wrong agent to achieve bottom-up filling. In this case, the electrolyte contained 10 mM CuSO ₄ (pH 2.3). The modulated waveform is used to electroplate copper, and the modulated waveform is similar to that used in the waveforms of Figures 9A-9C.

The rest of the experiments were related to electroplating that occurred on test pieces with a copper seed layer. Figures 11A-11C show SEM cross-sectional images of copper seed test strips filled at various temperatures, and Figures 12A-12C show SEM top views (after chemical mechanical polishing) of these same test pieces, respectively. The copper seed test strip was attached to the RDE and utilized at 10 RPM of copper cation and 10 mM of EDTA at a rotational speed of 200 RPM and with an MSE reference electrode, with a dissolved oxygen level of about 1 ppm. TMAH or sulfuric acid is electroplated by adjusting the pH to 2.3. The test pieces shown in Figures 11A-11C and 12A-12C were electroplated by a process of entering the electrolyte with a trigger potential of 0.25 seconds at an open potential of -0.5 V, followed by quiescent current deposition of a current of 0.2 mA. The trench in the test piece is about 50 nm wide. The test pieces shown in Figs. 11A and 12A were plated at room temperature, and the test pieces shown in Figs. 11B and 12B were plated at 50 ° C, and the test pieces shown in Figs. 11C and 12C were attached at 70 ° C. Perform electroplating. A bottom-up filling of good quality and no voids is achieved at room temperature. However, at higher temperatures of 70 ° C, seed dissolution and lack of growth appear to occur at relatively low current densities (0.2 mA) selected for this particular test. Therefore, the advantage of a higher plating rate at higher temperatures should be balanced with the increased likelihood of seeding at these higher temperatures.

Figure 13 shows a transmission electron microscope (TEM) image of a copper seed channel test piece which was electroplated in an electrolyte without a blocking agent. The electrolyte in this example contained 10 mM copper cations, about 1 ppm dissolved oxygen, and a pH of 2.3. The rotation speed is 200 RPM and the reference electrode is the MSE electrode. The quiescent potential was applied to the electrolyte using a relative open circuit potential of -0.5 V for 0.25 seconds, followed by a quiescent current plating of 1.2 mA current. As shown in Fig. 13, the bottom-up filling of good quality has been achieved. Thus, in some embodiments, the miscible agent can be omitted from the electrolyte.

100‧‧‧Process
101, 103, 105‧‧‧ blocks
200‧‧‧ Process
201, 203, 205‧‧‧ squares
900‧‧‧Electrodeposition equipment
902, 904, 906‧‧‧ plating modules
912, 914, 916‧‧‧ modules
922‧‧‧Chemical Dilution Module
924‧‧‧Central Electrodeposition Chamber
926‧‧‧ Dosing system
928‧‧‧Filtering and delivery unit
930‧‧‧System Controller
932‧‧‧ wafer loading and unloading station
940‧‧‧Transfer tool
942, 944‧‧‧Carmen
946‧‧‧Transfer tool
948‧‧‧ aligner
950‧‧‧Transfer station
1000‧‧‧Electrodeposition equipment
1001‧‧‧ Front-end loading FOUP
1002‧‧‧ front arm
1002a‧‧‧Mechanical arm track
1003‧‧‧Axis
1004‧‧‧ front-end access station
1006‧‧‧Substrate
1007‧‧‧ plating bath
1008‧‧‧ front-end access station
1009‧‧‧Installation rack

Figure 1 shows a flow diagram of a method of electroplating copper into features on a substrate having an exposed semi-precious metal layer.

2 shows a flow chart of a method of electroplating copper into features on a substrate having an exposed copper seed layer.

3 illustrates an exemplary multi-station device in accordance with an embodiment of the disclosure.

4 illustrates an alternate embodiment of a multi-station device in accordance with an embodiment of the disclosure.

Figure 5 is a graph showing the relative polarization of different complexing agents in an electrolyte.

Figure 6 is a graph showing the relative polarization effects of different copper cation concentrations and different pH levels in the electrolyte.

Figure 7 is a graph showing the relative polarization of different electrolyte temperatures.

8A-8C show SEM images of a seed crystal channel test piece subjected to electroplating at 0.4 mA (Fig. 8A), 0.6 mA (Fig. 8B), and 1 mA (Fig. 8C).

Figures 9A-9C show SEM images of a seed crystal channel test piece electroplated using a modulated waveform at room temperature (Figure 9A), 50 °C (Figure 9B), and 70 °C (Figure 9C).

10A and 10B show SEM images of a seed crystal channel test piece subjected to electroplating in an electrolyte containing NTA (Fig. 10A) and glutamic acid (Fig. 10B) as a binder.

Figures 11A-11C and 12A-12C show cross-sectional (Figures 11A-11C) and top (Figures 12A-12C) SEM images of copper seed channel test pieces that were plated at different temperatures.

Figure 13 shows a TEM image of a copper seed channel test piece which was electroplated in an electrolyte without a binder.

100‧‧‧Process

101‧‧‧ squares

103‧‧‧ square

105‧‧‧ square

Claims (32)

A method of performing a single-step plating filling process for filling features on a partially fabricated integrated circuit, the method comprising: (a) receiving a substrate having an exposed semi-precious metal layer and a plurality of features thereon (b) contacting the substrate with an electrolyte comprising: (i) a copper cation between about 1 and 100 mM; and (ii) a complexing agent that forms a complex with the copper cation, wherein the electrolyte The liquid is substantially free of inhibitors, accelerators, and homogenizers; and (c) in the case of contacting the electrolyte, in the case of an electrodeposited substrate potential between about 0.03 and 0.33 V relative to the NHE reference electrode, Copper is electroplated into the features by a bottom-up filling mechanism. A method of performing a one-step electroplating filling process as in claim 1, wherein the inhibitor, accelerator, or homogenizer does not substantially contribute to the bottom-up filling mechanism. A method of performing a one-step electroplating filling process according to the first aspect of the patent application, wherein the bottom-up filling system is directly applied to the semi-precious metal layer without first forming a seed layer. The method of performing a single-step electroplating filling process according to the first aspect of the patent application, wherein the copper electroplating step in the operation (c) comprises: applying a modulation waveform, the modulation waveform is at a first level and a second level A current pulse is alternately generated, wherein copper is deposited on the substrate at the first level, and copper previously plated on the substrate is copper etched at the second level. The method of performing a single-step electroplating filling process according to claim 4, wherein the second current level of copper etching has an absolute value of about 0.05-0.3 mA/cm ² , and wherein the first current is The current pulse alternating between the level and the second current level has a frequency between about 100 and 1000 Hz. The method for performing a single-step electroplating filling process according to any one of claims 1 to 5, wherein the wrong agent is selected from ethylenediaminetetraacetic acid (EDTA), nitrogen triacetic acid (NTA, nitrilotriacetic). Group of acid), citric acid, and glutamic acid. A method of performing a one-step electroplating filling process according to item 6 of the patent application, wherein the binder is ethylenediaminetetraacetic acid (EDTA). A method of performing a one-step electroplating filling process according to any one of claims 1 to 5 wherein the electrolyte is maintained at a temperature of between about 20 and 80 ° C during electroplating. A method of performing a one-step electroplating filling process as in claim 8 wherein the electrolyte is maintained at a temperature of between about 50 and 70 ° C during electroplating. A method of performing a one-step electroplating filling process according to any one of claims 1 to 5, wherein the plating surface of the substrate is subjected to a current density of between about 0.1 and 2 mA/cm ² during electroplating. . A method of performing a one-step electroplating filling process according to any one of claims 1 to 5, wherein the pH of the electrolyte is between about 1-5. The method for performing a single-step electroplating filling process according to any one of claims 1 to 5, wherein the semi-precious metal layer comprises a selected free germanium, tungsten, cobalt, rhodium, platinum, palladium, aluminum, gold, silver, rhodium. And the materials of the group formed by the group. A method of performing a one-step electroplating filling process according to claim 12, wherein the semi-precious metal layer comprises niobium. A method of performing a one-step electroplating filling process according to any one of claims 1 to 5, wherein at least some of the features have a width of about 100 nm or less. A method of performing a one-step electroplating filling process according to claim 14 wherein at least a portion of the features have a width of about 20 nm or less. A method of performing a one-step electroplating filling process according to any one of claims 1 to 5, wherein the electrolyte contains about 2 ppm or less of dissolved oxygen. A method of performing a single-step plating filling process for filling features on a partially fabricated integrated circuit, the method comprising: (a) receiving a substrate having an exposed semi-precious metal layer and a plurality of features thereon (b) contacting the substrate with an electrolyte comprising a copper cation of between about 1 and 100 mM, wherein the electrolyte is substantially free of inhibitors, accelerators, and homogenizers; and (c) Applying a modulation waveform when the substrate is brought into contact with the electrolyte, the modulation waveform alternately generating a current pulse between the first level and the second level, wherein copper is deposited on the substrate at the first level, And copper etching the copper previously plated on the substrate at the second level, thereby filling the bottom-up by the bottom-up with a potential of between about 0.03 and 0.33 V of the electrodeposited substrate relative to the NHE reference electrode The mechanism electroplates copper into the features. The method of performing a single-step electroplating filling process according to claim 17, wherein the second current level of copper etching has an absolute value between about 0.05 and 0.3 mA/cm ² , and wherein The current pulse alternating between a current level and a second current level has a frequency between about 100 and 1000 Hz. A method of depositing copper on a feature portion of a partially fabricated integrated circuit, the method comprising: (a) receiving a substrate having a plurality of features and a copper seed layer thereon; (b) making the The substrate is in contact with an electrolyte comprising between about 1 and 100 mM of copper cations, wherein the electrolyte is substantially free of inhibitors, accelerators, and homogenizers; and (c) is in the relative NHE reference electrode With a potential between about 0.03 and 0.33 V, copper is electroplated into the features by a bottom-up filling mechanism. A method of depositing copper on a feature of a partially fabricated integrated circuit, as in claim 19, wherein the electrolyte is maintained at a temperature between about 20 and 80 ° C during electroplating. A method of depositing copper on a feature of a partially fabricated integrated circuit, as in claim 20, wherein the electrolyte is maintained at a temperature between about 20 and 50 ° C during electroplating. A method of depositing copper on a feature of a partially fabricated integrated circuit, as in claim 19, wherein the plated surface of the substrate has a current of between about 0.1 and 2 mA/cm ² during electroplating. density. A method of depositing copper on a feature of a partially fabricated integrated circuit, as in any one of claims 19 to 22, wherein the pH of the electrolyte is between about 1-5. A method of depositing copper on a feature of a partially fabricated integrated circuit, according to any one of claims 19 to 22, wherein at least a portion of the features have a width of about 100 nm or less. At least some of the features have a width of about 20 nm or less, as in the method of claim 24, wherein copper is deposited on features of the partially fabricated integrated circuit. A method of depositing copper on a feature of a partially fabricated integrated circuit according to any one of claims 19 to 22, wherein the copper plating step in operation (c) comprises: applying a quiescent current control Current is applied to the substrate. An apparatus for electroplating copper into features on a substrate, the apparatus comprising: (a) one or more electroplating baths configured to contain an electrolyte; (b) a substrate support; and (c) a control And a set of instructions comprising instructions for: accommodating an electrolyte in the one or more electroplating baths; immersing the substrate in the electrolyte; maintaining the substrate potential at a relative NHE reference electrode between about 0.03 Between 0.33 V, copper is electroplated into the features by a bottom-up filling mechanism and does not substantially rely on inhibitors, accelerators, or homogenizers. An apparatus for electroplating copper into a feature on a substrate, as in claim 27, wherein the controller further includes an instruction to apply a modulated waveform, the modulated waveform being at a first current level A current pulse is alternately generated between the second current levels, wherein copper is deposited on the substrate at the first current level, and copper is etched away from the substrate at the second current level. An apparatus for electroplating copper into a feature on a substrate, as in claim 28, wherein the second current level for copper etching has a size of between about 0.05 and 0.3 mA/cm ² , and wherein The current pulse alternating between the first current level and the second current level has a frequency between about 100 and 1000 Hz. An apparatus for electroplating copper into a feature on a substrate, as in claim 27, wherein the controller further comprises maintaining the electrolyte at a temperature between about 20 and 80 ° C during electroplating. instruction. An apparatus for electroplating copper into a feature on a substrate, as in claim 30, wherein the controller includes instructions for maintaining the electrolyte at a temperature between about 50 and 70 ° C during electroplating . An apparatus for electroplating copper into a feature on a substrate, according to any one of claims 27 to 31, wherein the controller further comprises instructions for controlling the electroplating process to cause the surface of the substrate Withstands a current density between about 0.1 and 2 mA/cm ² .