US 20040209465 A1
A method and apparatus to form a refractory metal layer on a substrate features nucleating a substrate using sequential deposition techniques in which the substrate is serially exposed to first and second reactive gases followed by forming a layer, employing vapor deposition, to subject the nucleation layer to a bulk deposition of a compound contained in one of the first and second reactive gases.
1. A method for forming a nucleation layer and a bulk deposition layer on a substrate disposed in a processing chamber, said method comprising:
forming a refractory metal nucleation layer by serially exposing said substrate to first and second reactive gases; and
forming a bulk deposition layer on said nucleation layer by employing vapor deposition to bulk deposit a refractory metal contained in one of said first and second reactive gases.
2. The method as recited in
3. The method as recited in
4. The method as recited in
5. The method as recited in
6. The method as recited in
7. The method as recited in
8. The method as recited in
9. The method as recited in
10. A method for forming a nucleation layer and a bulk deposition layer on a substrate, said method comprising:
serially exposing said substrate to first and second reactive gases, wherein said second reactive gas comprises a refractory metal selected from the group consisting of titanium (Ti) and tungsten (W), while said substrate is disposed in a processing chamber, to form a nucleation layer;
removing from said processing chamber said first reactive gas before exposing said substrate to said second reactive gas;
forming said layer adjacent to said nucleation layer by chemical vapor deposition while said substrate is disposed in said processing chamber by concurrently exposing said nucleation layer to said second reactive gas and a reducing agent.
11. The method of
12. The method of
13. The method of
14. The method as recited in
 This application is a continuation of co-pending U.S. patent application Ser. No. 09/678,266, filed Oct. 3, 2000, which claims priority to U.S. patent application Ser. No. 09/605,593, filed Jun. 28, 2000.
 1. Field of the Invention
 Embodiments of the present invention generally relate to the processing of semiconductor substrates. More particularly, this invention relates to improvements in the process of depositing refractory metal layers on semiconductor substrates.
 2. Description of the Related Art
 The semiconductor processing industry continues to strive for larger production yields while increasing the uniformity of layers deposited on substrates having larger surface areas. These same factors in combination with new materials also provide higher integration of circuits per unit area of the substrate. As circuit integration increases, the need for greater uniformity and process control regarding layer thickness rises. As a result, various technologies have been developed to deposit layers on substrates in a cost-effective manner, while maintaining control over the characteristics of the layer. Chemical Vapor Deposition (CVD) is one of the most common deposition processes employed for depositing layers on a substrate. CVD is a flux-dependent deposition technique that requires precise control of the substrate temperature and precursors introduced into the processing chamber in order to produce a desired layer of uniform thickness. These requirements become more critical as substrate size increases, creating a need for more complexity in chamber design and gas flow technique to maintain adequate uniformity.
 A variant of CVD that demonstrates superior step coverage, compared to CVD, is Atomic Layer Deposition (ALD). ALD is based upon Atomic Layer Epitaxy (ALE) that was employed originally to fabricate electroluminescent displays. ALD employs chemisorption to deposit a saturated monolayer of reactive precursor molecules on a substrate surface. This is achieved by alternatingly pulsing an appropriate reactive precursor into a deposition chamber. Each injection of a reactive precursor is separated by an inert gas purge to provide a new atomic layer additive to previously deposited layers to form a uniform layer on the substrate. The cycle is repeated to form the layer to a desired thickness. A drawback with ALD techniques is that the deposition rate is much lower than typical CVD techniques by at least one order of magnitude.
 Formation of film layers at a high deposition rate while providing adequate step coverage are conflicting characteristics often necessitating sacrificing one to obtain the other. This conflict is true particularly when refractory metal layers are deposited to cover gaps or vias during formation of contacts that interconnect adjacent metallic layers separated by dielectric layers. Historically, CVD techniques have been employed to deposit conductive material such as refractory metals in order to inexpensively and quickly form contacts. Due to the increasing integration of semiconductor circuitry, tungsten has been used based upon superior step coverage. As a result, deposition of tungsten employing CVD techniques enjoys wide application in semiconductor processing due to the high throughput of the process.
 Depositing tungsten by traditional CVD methods, however, is attendant with several disadvantages. For example, blanket deposition of a tungsten layer on a semiconductor wafer is time-consuming at temperatures below 400° C. The deposition rate of tungsten may be improved by increasing the deposition temperature to, e.g., about 500° C. to about 550° C.; however, temperatures in this higher range may compromise the structural and operational integrity of the underlying portions of the integrated circuit being formed. Use of tungsten has also frustrated photolithography steps during the manufacturing process as it results in a relatively rough surface having a reflectivity of 20% or less than that of a silicon substrate. Finally, tungsten has proven difficult to deposit uniformly. Variance in film thickness of greater than 1% has been shown with tungsten, thereby frustrating control of the resistivity of the layer. Several prior attempts to overcome the aforementioned drawbacks have been attempted.
 For example, in U.S. Pat. No. 5,028,565 to Chang et al., which is assigned to the assignee of the present invention, a method is disclosed to improve, inter alia, uniformity of tungsten layers by varying the deposition chemistry. The method includes, in pertinent part, formation of a nucleation layer over an intermediate, barrier layer before depositing the tungsten layer via bulk deposition. The nucleation layer is formed from a gaseous mixture of tungsten hexafluoride, hydrogen, silane and argon. The nucleation layer is described as providing a layer of growth sites to promote uniform deposition of a tungsten layer thereon. The benefits provided by the nucleation layer are described as being dependent upon the barrier layer present. For example, were the barrier layer formed from titanium nitride, the tungsten layer's thickness uniformity is improved as much as 15%. Were the barrier layer formed from sputtered tungsten or sputtered titanium tungsten, the benefits provided by the nucleation layer are not as pronounced.
 U.S. Pat. No. 5,879,459 to Gadgil et al. discloses an apparatus that takes advantage of ALD. To that end, the apparatus, a low profile, compact atomic layer deposition reactor (LP-CAR), has a body with a substrate processing region adapted to serve a single substrate or a planar array of substrates, as well as a valve, and a port for substrate loading and unloading. In some embodiments multiple reactors are stacked vertically and share a common robotic handler interface with a CVD system. In this manner, the robotic handler may manipulate substrates associated with both the CVD system and the LP-CAR. The compact reactor is distinguished by having individual injectors, each of which comprises a charge tube formed between a charge valve and an injection valve. The charge valve connects the charge tube to a pressure regulated supply, and the injection valve opens the charge tube into the compact reactor. Rapidly cycling the valves injects fixed mass-charges of gas or vapor into the compact reactor.
 What is needed, however, is a technique to deposit conductive layers having a deposition rate comparable to CVD techniques while providing the step coverage associated with ALD techniques.
 The present invention generally a method and system to form a refractory metal layer on a substrate features nucleating a substrate using sequential deposition techniques in which the substrate is serially exposed to first and second reactive gases followed by forming a layer, employing vapor deposition, to subject the nucleation layer to a bulk deposition of a compound contained in one of the first and second reactive gases. To that end, the system includes a processing chamber that has a holder disposed therein to support the substrate. A gas delivery system and a pressure control system is in fluid communication with the processing chamber. A temperature control system is in thermal communication therewith. A controller is in electrical communication with gas delivery systems, temperature control systems and pressure control systems. A memory is in data communication with the controller. The memory comprises a computer-readable medium having a computer-readable program embodied therein. The computer-readable program includes instructions for controlling the operation of the processing chamber.
FIG. 1 is a perspective view of a semiconductor processing system in accordance with the present invention;
FIG. 2 is a detailed view of the processing chambers shown above in FIG. 1;
FIG. 3 is a schematic view showing deposition of a first molecule onto a substrate during ALD;
FIG. 4 is a schematic view showing deposition of second molecule onto a substrate during ALD to form a refractory metal layer;
FIG. 5 is a graphical representation showing the concentration of gases, introduced into the processing chamber shown above in FIG. 2, and the time in which the gases are present in the processing chamber, in accordance with the present invention;
FIG. 6 is a graphical representation showing the relationship between the number of ALD cycles and the thickness of a layer formed on a substrate employing ALD, in accordance with the present invention;
FIG. 7 is a graphical representation showing the relationship between the number of ALD cycles and the resistivity of a layer formed on a substrate employing ALD, in accordance with the present invention;
FIG. 8 is a graphical representation showing the relationship between the deposition rate of a layer formed on a substrate employing ALD and the temperature of the substrate;
FIG. 9 is a graphical representation showing the relationship between the resistivity of a layer formed on a substrate employing ALD and the temperature of the substrate, in accordance with the present invention;
FIG. 10 is a cross-sectional view of a patterned substrate having a nucleation layer formed thereon employing ALD, in accordance with the present invention;
FIG. 11 is a partial cross-sectional view of the substrate, shown above in FIG. 10, with a refractory metal layer formed atop of the nucleation layer employing CVD, in accordance with the present invention; and
FIG. 12 is a graphical representation showing the concentration of gases shown above in FIG. 3 in accordance with a first alternate embodiment of the present invention.
 Referring to FIG. 1, an exemplary wafer processing system includes one or more processing chambers 12 and 14 disposed in a common work area 16 surrounded by a wall 18. The processing chambers 12 and 14 are in data communication with a controller 22 that is connected to one or more monitors, shown as 24 and 26. The monitors typically display common information concerning the process associated with the processing chamber 12 and 14. One of the monitors 26 is mounted to the wall 18, with the remaining monitor 24 being disposed in the work area 16. Operational control of the processing chambers 12 and 14 may be achieved by the use of a light pen, associated with one of the monitors 24 and 26, to communicate with the controller 22. For example, light pen 28 is associated with monitor 24 and facilitates communication with the controller 22 through monitor 24. Light pen 29 facilitates communication with the controller 22 through monitor 26.
 Referring both to FIGS. 1 and 2, each of the processing chambers 12 and 14 includes a housing 30 having a base wall 32, a cover 34, disposed opposite to the base wall 32, and a sidewall 36, extending therebetween. The housing 30 defines a chamber 37, and a pedestal 38 is disposed within the processing chamber 37 to support a substrate 42, such as a semiconductor wafer. The pedestal 38 may be mounted to move between the cover 34 and the base wall 32, using a displacement mechanism (not shown), but the position thereof is typically fixed. Supplies of processing gases 39 a, 39 b and 39 c are in fluid communication with the processing chamber 37 via a showerhead 40. Regulation of the flow of gases from the supplies 39 a, 39 b and 39 c is effectuated via flow valves 41.
 Depending on the specific process, the substrate 42 may be heated to a desired temperature prior to layer deposition via a heater embedded within the pedestal 38. For example, the pedestal 38 may be resistively heated by applying an electric current from an AC power supply 43 to the heater element 44. The substrate 42 is, in turn, heated by the pedestal 38, and can be maintained within a desired process temperature range of, for example, about 20° C. to about 750° C. A temperature sensor 46, such as a thermocouple, is also embedded in the wafer support pedestal 38 to monitor the temperature of the pedestal 38 in a conventional manner. For example, the measured temperature may be used in a feedback loop to control the electrical current applied to the heater element 44 by the power supply 43 such that the substrate temperature can be maintained or controlled at a desired temperature that is suitable for the particular process application. Optionally, the pedestal 38 may be heated using radiant heat (not shown). A vacuum pump 48 is used to evacuate the processing chamber 37 and to help maintain the proper gas flows and pressure inside the processing chamber 37.
 Referring to FIGS. 1 and 3, one or both of the processing chambers 12 and 14, discussed above may operate to deposit refractory metal layers on the substrate employing sequential deposition techniques. One example of sequential deposition techniques in accordance with the present invention includes atomic layer deposition. Depending on the specific stage of processing, the refractory metal layer may be deposited on the material from which the substrate 42 is fabricated, e.g., SiO2. The refractory metal layer may also be deposited on a layer previously formed on the substrate 42, e.g., titanium, titanium nitride and the like.
 During the sequential deposition technique in accordance with the present invention, a batch of a first processing gas, in this case Aax, results in a layer of A being deposited on the substrate 42 having a surface of ligand x exposed to the processing chamber 37. Thereafter, a purge gas enters the processing chamber 37 to purge the gas Aax. After purging gas Aax from the processing chamber 37, a second batch of processing gas, Bby, is introduced into the processing chamber 37. The a ligand present on the substrate surface reacts with the b ligand and B atom on the, releasing molecules ab and Ba, that move away from the substrate 42 and are subsequently pumped from the processing chamber 37. In this manner, a surface comprising a layer of A compound remains upon the substrate 42 and exposed to the processing chamber 37, shown in FIG. 4. The composition of the layer of A compound may be a monolayer of atoms typically formed employing ATD techniques. Alternatively, the layer of compound A may include a layer of multiple atoms. In such as case, the first processing gases may include a mixture of process gases each of which has atoms that would adhere to the substrate 42. The process proceeds cycle after cycle, until the desired thickness is achieved.
 Referring to both FIGS. 2 and 5, although any type of processing gas may be employed, in the present example, the processing gas Aax includes WF6 and the processing gas Bby is B2H6. Two purge gases are employed: Ar and N2. Each of the processing gases was flowed into the processing chamber 37 with a carrier gas, which in this example were one of the purge gases: WF6 is introduced with Ar and B2H6 is introduced with N2. It should be understood, however, that the purge gas may differ from the carrier gas, discussed more fully below. One cycle of the ALD technique in accordance with the present invention includes flowing the purge gas, N2, into the processing chamber 37 during time t1, which is approximately 0.01 to 15 seconds before B2H6 is flowed into the processing chamber 37. During time t2, the processing gas B2H6 is flowed into the processing chamber 37 for a time in the range of 0.01 to 15 seconds, along with a carrier gas, which in this example is N2. After 0.01 to 15 seconds have lapsed, the flow of B2H6 terminates and the flow of N2 continues during time t3 for an additional time in the range of 0.01 to 15 seconds, purging the processing chamber of B2H6. During time t4, the processing chamber 37 is pumped so as to remove most, if not all, gases. After pumping of the process chamber 37, the carrier gas Ar is introduced for a time in the range of 0.01 to 15 seconds during time t5, after which time the process gas WF6 is introduced into the processing chamber 37, along with the carrier gas Ar during time t6. The time t6 lasts between 0.01 to 15 seconds. The flow of the processing gas WF6 into the processing chamber 37 is terminated approximately 0.01 to 15 seconds after it commenced. After the flow of WF6 into the processing chamber 37 terminates, the flow of Ar continues for an additional time in the range of 0.01 to 15 seconds, during time t7. Thereafter, the processing chamber 37 is pumped so as to remove most, if not all, gases therein, during time t8. As before, the pumping process lasts approximately thirty seconds, thereby concluding one cycle of the sequential deposition technique in accordance with the present invention.
 The benefits of employing the sequential deposition technique are manifold, including flux-independence of layer formation that provides uniformity of deposition independent of the size of a substrate. For example, the measured difference of the layer uniformity and thickness measured between a 200 mm substrate and a 32 mm substrate deposited in the same chamber is negligible. This is due to the self-limiting characteristics of the sequential deposition techniques. Further, this technique contributes to a near-perfect step coverage over complex topography.
 In addition, the thickness of the layer B, shown in FIG. 4, may be easily controlled while minimizing the resistance of the same by employing sequential deposition techniques. With reference to FIG. 6 it is seen in the slope of line 50 that the thickness of the tungsten layer B is proportional to the number of cycles employed to form the same. The resistivity of the tungsten layer, however, is relatively independent of the thickness of the layer, as shown by the slope of line 52 in FIG. 7. Thus, employing sequential deposition techniques, the thickness of a refractory metal layer may be easily controlled as a function of the cycling of the process gases introduced into the processing chamber with a negligible effect on the resistivity.
 Referring to FIG. 8, control of the deposition rate was found to be dependent upon the temperature of the substrate 42. As shown by the slope of line 54, increasing the temperature of the substrate 42 increased the deposition rate of the tungsten layer B. For example, at 56, the deposition rate is shown to be approximately 2 Å/cycle at 250° C. However at point 58 the deposition rate is approximately 5 Å/cycle at a temperate of 450° C. The resistivity of the tungsten layer, however, is virtually independent of the layer thickness, as shown by the slope of curve 59, shown in FIG. 9. As a result, the deposition rate of the tungsten layer may be controlled as a function of temperature without compromising the resistivity of the same. However, it may be desired to reduce the time necessary to deposit an entire layer of a refractory metal.
 To that end, a bulk deposition of the refractory metal layer may be included in the deposition process. Typically, the bulk deposition of the refractory metal occurs after the nucleation layer is formed in a common processing chamber. Specifically, in the present example, nucleation of a tungsten layer occurs in chamber 12 employing the sequential deposition techniques discussed above, with the substrate 42 being heated in the range of 200° C. to 400° C., and the processing chamber 37 being pressurized in the range of 1 to 10 Torr. A nucleation layer 60 of approximately 12 to 20 nm is formed on a patterned substrate 42, shown in FIG. 10. As shown, the substrate 42 includes a barrier layer 61 and a patterned layer having a plurality of vias 63. The nucleation layer is formed adjacent to the patterned layer covering the vias 63. As shown, forming the nucleation layer 60 employing ALD techniques provides 100% step coverage. To decrease the time required to form a complete layer of tungsten, a bulk deposition of tungsten onto the nucleation layer 60 occurs using CVD techniques, while the substrate 42 is disposed in the same processing chamber 12, shown in FIG. 1. The bulk deposition may be performed using recipes well known in the art. In this manner, a tungsten layer 65 providing a complete plug fill is achieved on the patterned layer with vias having aspect ratios of approximately 6:1, shown in FIG. 11.
 As mentioned above, in an alternate embodiment of the present invention, the carrier gas may differ from the purge gas, as shown in FIG. 12. The purge gas, which is introduced at time intervals t1, t3, t5 and t7 comprises of Ar. The carrier gas, which is introduced at time intervals t2 and t6, comprises of N2. Thus, at time interval t2 the gases introduced into the processing chamber include a mixture of B2H6 and N2, and a time interval t6, the gas mixture includes WF6 and N2. The pump process during time intervals t4 and t8 is identical to the pump process discussed above with respect to FIG. 5.
 Referring again to FIG. 2, the process for depositing the tungsten layer may be controlled using a computer program product that is executed by the controller 22. To that end, the controller 22 includes a central processing unit (CPU) 70, a volatile memory, such as a random access memory (RAM) 72 and permanent storage media, such as a floppy disk drive for use with a floppy diskette, or hard disk drive 74. The computer program code can be written in any conventional computer readable programming language; for example, 68000 assembly language, C, C++, Pascal, Fortran and the like. Suitable program code is entered into a single file, or multiple files, using a conventional text editor and stored or embodied in a computer-readable medium, such as the hard disk drive 74. If the entered code text is in a high level language, the code is compiled and the resultant compiler code is then linked with an object code of precompiled Windows® library routines. To execute the linked and, compiled object code the system user invokes the object code, causing the CPU 70 to load the code in RAM 72. The CPU 70 then reads and executes the code to perform the tasks identified in the program.
 Although the invention has been described in terms of specific embodiments, one skilled in the art will recognize that various changes to the reaction conditions, i.e., temperature, pressure, film thickness and the like can be substituted and are meant to be included herein. Additionally, while the deposition process has been described as occurring in the same chamber, it may be bifurcated. In this manner, the nucleation layer may be deposited in one chamber and the bulk deposition occurring in a differing chamber, located within the same mainframe deposition system. However, the bulk deposition may occur in a processing chamber of a mainframe deposition system that is different from the mainframe deposition system in which the processing chamber is located that is employed to deposit the nucleation layer. Finally, other refractory metals may be deposited, in addition to tungsten, and other deposition techniques may be employed in lieu of CVD. For example, physical vapor deposition (PVD) techniques, or a combination of both CVD and PVD techniques may be employed. The scope of the invention should not be based upon the foregoing description. Rather, the scope of the invention should be determined based upon the claims recited herein, including the full scope of equivalents thereof.
 While the foregoing is directed to embodiments of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.