1. Field of the Invention
Embodiments of the invention generally relate to a method for depositing silicon and nitride layers using a plasma enhanced chemical vapor deposition (PECVD) process.
2. Background
Substrate size expansion has been the enabler of the TFT-LCD industry. Since TFT-LCD production using substrates having a plan area of about 2000 cm2 started in 1993, the growth rate of substrate size has been almost exponential, enlarging more than 30 times in plan area in 13 years. This rapid growth of substrate size has been very challenging for display manufactures, material suppliers and equipment makers; and a driving force for everyone to improve. Many challenges were faced in scaling up the plasma enhanced chemical vapor deposition (PECVD) reactors and PECVD processes to accommodate substrates having a 2160×2460 mm2 plan area. The most severe challenges were maintaining the integrity and stability of larger electrodes, maintaining substrate temperature uniformity, maintaining gas distribution uniformity, and last but not least, maintaining the same or better film quality achieved during processing 2000 cm2 substrates without sacrificing productivity.
As the substrate size has grown, thermal contraction of glass substrate has become more problematic for the photo engraving exposure process. Among TFT-LCD production processes, the most commonly used and highest process temperature is 350 degrees Celsius for the PECVD silicon nitride (SiNx) gate dielectric layers and amorphous silicon (a-Si) active layers. This relatively high temperature and other associated process conditions were arrived at for the first single-chamber PECVD for TFTs in and around the LCD industry's 2000 cm2 substrate timeframe. Reducing the process temperature even by 60 degrees Celsius can drastically reduce thermal contraction. An additional or alternative benefit of lower temperature processing is the possibility to use a less expensive glass substrate which may have a higher coefficient of thermal expansion. A likely further benefit of reducing process temperature is improved PECVD hardware reliability and system utilization. While the benefits of lower temperature are several, current PECVD processes are optimized to obtain required quality with high deposition rate at 350 degrees Celsius; therefore simply lowering process temperature without discovering solutions for process conditions would degrade film quality. Additionally, reducing deposition rate to improve film quality at lower temperatures is not a viable solution since sacrificing deposition rate would reduce throughput, thereby making such a process impractical for a production systems.
Embodiments for the invention generally relate to a method for depositing silicon and nitride layers using a plasma enhanced chemical vapor deposition (PECVD) process. In one embodiment, a method for forming a film stack suitable for transistor fabrication includes providing a substrate in a PECVD chamber, depositing a dual layer SiNx film on the substrate, depositing a dual layer amorphous silicon film on the SiNx film, and depositing a n-doped silicon film on the dual layer amorphous silicon film. The aforementioned films are deposited at a temperature less than about 300 degrees Celsius. In another embodiment, at least the SiNx and amorphous silicon films are deposited in the same PECVD chamber.
In another embodiment, a method for forming a film stack suitable for transistor fabrication includes depositing a first SiNx layer at a rate greater than 1500 Å/min on the substrate at a temperature less than about 300 degrees Celsius, depositing a second SiNx layer at a rate greater than 1500 Å/min on the first SiNx layer at a temperature less than about 300 degrees Celsius, wherein the first SiNx layer has a more SiH content than the second SiNx layer, depositing a first a-Si:H layer at a rate less than 600 Å/min on the second SiNx layer at a temperature less than about 300 degrees Celsius, depositing a second a-Si:H layer at a rate greater than 600 Å/min on the first a-Si:H layer at a temperature less than about 300 degrees Celsius, wherein the second a-Si:H layer has a higher optical bandgap as compared to the second a-Si:H layer, and depositing a n-doped silicon film on the dual layer amorphous silicon film at a temperature less than about 300 degrees Celsius.
In another embodiment, a second SiNx layer of a dual layer SiNx film is deposited by providing SiH4 at a flow rate less than that provided during deposition of the first SiNx layer, providing NH3 at a flow rate less than that provided during deposition of the first SiNx layer; providing N2 at a flow rate greater than that provided during deposition of the first SiNx layer; and providing less RF power to sustain a plasma formed from the SiH4 and NH3 gases than that provided during deposition of the first SiNx layer.
In another embodiment, a second a-Si:H layer of a dual layer a-Si:H film is deposited by providing SiH4 at a flow rate greater than that provided during deposition of the first a-Si:H layer and providing H2 at a flow rate less than that provided during deposition of the first a-Si:H layer.
In yet another embodiment, the deposition process includes depositing a dual layer SiNx film on a glass or polymer substrate, depositing a dual layer amorphous silicon film on the SiNx film, and depositing a n-doped silicon film on the dual layer amorphous silicon film, wherein has a plan area greater than about 1.0 m2.
So that the manner in which the above recited features of the invention are attained and can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments thereof which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical embodiments of this invention and are therefore not to be considered limiting of its scope, for the invention may admit to other equally effective embodiments.
To facilitate understanding, identical reference numerals have been used, wherever possible, to designate identical elements that are common to the figures. It is contemplated that elements of one embodiment may be advantageously utilized in other embodiments without further recitation.
The PECVD process chamber includes an upper electrode showerhead incorporates a proprietary hollow cathode effect which significantly enhances gas dissociation and ensures uniformly dissociated reactant gas across the entire substrate area. The bottom electrode, the susceptor, provides uniform multi-zone heating to the glass substrate at process set point temperatures up to 400 degrees Celsius. 13.5 MHz RF power is delivered to the showerhead. The process chamber is configured to deposited in-situ both dielectric films (such as SiNx, SiOx and SiOxNy, among others) and semiconductor films (such as a-Si:H and doped a-Si:H, among others) with good uniformity using the same set of hardware (i.e., without removing the substrate from the process chamber).
The process chamber is equipped with remote plasma source cleaning which effectively provides full dissociation of the cleaning gas outside of the process chamber. The active gas species flow downstream to the chamber to gently and efficiently etch away the film residues left by the CVD process. Typically NF3 is used for cleaning because this gas can be nearly 100 percent dissociated and the effluent easily abated, thereby minimizing release of global-warming potential (GWP) gas to the environment.
An exemplary PECVD cluster tool 200 is shown in
The mechanical challenges have been solved through carefully engineering to minimize stresses, deflection and creep of the electrode and electrode supporting materials. Parallelism between the two electrodes can be controlled to less than +/−1 mm over a 2160×2460 mm2 plan area substrate. Also the substrate temperature variation is maintained within less than 10 degrees Celsius over this large area.
The guiding principles used to scale up the PECVD processes are to maintain the intensive deposition parameters (pressure, electrode spacing, substrate temperature) as similar as possible to the previous generation, while somewhat proportionally increasing the extensive deposition parameters (gas flow rate, RF power). However as substrate size has increased beyond 1500×1800 mm2 in plan area, there no longer is a way to adequately mitigate surface standing wave effects in the PECVD plasma and still maintain the same generation to generation deposition rates and system productivity. As such, proportionally increasing deposition parameters will not provide solutions for depositing films on substrates in excess of 1500×1800 mm2 in plan area.
In general, lowering process temperature causes the following issues; increasing dangling-bond density and Si—H2 bond formation in a-Si film, and decreasing film density and increasing Si—H bond formation in SiNx film. These degraded film properties are associated with greater threshold voltage shift and lower electron mobility, among other transistor performance issues. In order to solve these issues, lowering deposition rate and/or applying very high frequency have been studied elsewhere. As stated earlier, lowering deposition rate strongly affects system throughput and productivity, so it is undesirable for production systems. Applying very high frequency would have a stronger surface standing wave effect, thus limiting scalability, thereby making high frequency power undesirable for depositing films on substrates having a plan area in excess of 1.0 m2. Beneficially, the process of the present invention maintains film performance at the same level as high-temperature active layers, while also maintaining system productivity and throughput.
In one embodiment, the active layers of a-Si TFT film stack 400 consist of SiNx layer 402, a-Si:H layer 404 and phosphorous doped n+a-Si:H layer 406 (n+a-Si:H). In order to achieve panel makers' requirements in both productivity and process performance, the SiNx layer 402 is comprised of a high deposition rate SiNx layer 402GH and a low deposition rate SiNx layer 402GL, while the a-Si:H layer 404 comprises a low deposition rate a-SiNx layer 404AL and a high deposition rate a-SiNx layer 404AH. Here, the SiNx layer 402GH is deposited at rate greater than about 1500 Å/min for the gate dielectric, the SiNx layer 402GL is deposited at rate less than about 1500 Å/min for the gate dielectric, the a-Si:H layer 404AL is deposited at rate less than about 600 Å/min for a portion of the a-Si:H layer 404, and the a-Si:H layer 404AH is deposited at rate greater than about 600 Å/min for a portion of the a-Si:H layer 404. In one embodiment, the SiNx layer 402GH has a thickness of about 250 nm, the SiNx layer 402GL has a thickness of about 50 nm, the a-Si:H layer 404AL has a thickness of about 50 nm, the a-Si:H layer 404AH has a thickness of about 150 nm, and the n+a-Si:H layer 406 has a thickness of about 50 nm. The electron mean free path is on the order of a few angstroms and the carrier transport region of the a-Si is mostly within about 300 angstroms of the a-Si:H channel and gate dielectric interface. Therefore, mostly the film properties of the SiNx layer 402GL and the SiNx layer 402GL determine the device mobility, threshold voltage, and other performance attributes because these low deposition rate layers (SiNx layer 402GL and a-Si:H layer 404AL) are relatively thin, the film properties of those layers can be tuned and the deposition rate reduced if necessary to maintain high TFT performance without much effect on system productivity. On the other hand, one can increase the deposition rate of SiNx layer and a-Si:H layers 402GH, 404AH to compensate for the layer deposition rates of the SiNx and a-Si:H interface (e.g., SiNx layer and a-Si:H layers 402GL, 404AL) and maintain throughput without sacrificing device performance. Finally, the dual nitride layer 402 (comprised for layers 402GL and 402GH) can be engineered to have two different dry etch rates to produce a desirable tapered profile when using dry plasma etch to open the metal contact to the gate line of the transistor.
In one embodiment, the SiNx layer 402GH may deposited at a rate greater than about 1500 Å/min by forming a plasma from a gas mixture in the process chamber, heating the substrate while providing sufficient RF power to maintain the plasma during deposition. The gas mixture may be comprised of SiH4 and NH3 gases. SiH4 may be provided at a flow rate of about 900 to about 3000 sccm/m2 of substrate plan area, for example about 1500 to about 2000 sccm/m2, or about 1800 sccm/m2. NH3 may be provided at a flow rate of about 5000 to about 15000 sccm/m2 of substrate plan area, for example about 8000 to about 11000 sccm/m2, or about 9600 sccm/m2. The gas mixture may also include a carrier gas, such as N2. In one embodiment, N2 may be provided at a flow rate of about 7000 to about 30000 sccm/m2 of substrate plan area, for example about 11000 to about 17000 sccm/m2, or about 14000 sccm/m2. Pressure is regulated within the processing chamber between about 1.0 to about 2.5 Torr, for example about 1.2 to about 1.8 Torr, or about 1.3 Torr. Power provided to maintain the plasma is may be delivered to the electrode of the processing chamber, and may be provided at about 3000 to about 7000 Watts/m2 of substrate plan area, for example about 4800 to about 5800 Watts/m2, or about 5300 Watts/m2. Spacing may be maintained between the substrate and the showerhead in the range of about 850 to about 1200 mils, for example about 900 to about 1000 mils, or about 950 mils. During deposition, the temperature of the substrate is maintained between about 250 to about 300 degrees Celsius, for example about 280 to about 290 degrees Celsius, or about 285 degrees Celsius.
In one embodiment, the SiNx layer 402GL may deposited at a rate less than about 1500 Å/min by forming a plasma from a gas mixture in the process chamber, heating the substrate while providing sufficient RF power to maintain the plasma during deposition. The gas mixture may be comprised of SiH4 and NH3 gases. SiH4 may be provided at a flow rate of about 300 to about 1000 sccm/m2 of substrate plan area, for example about 600 to about 750 sccm/m2, or about 650 sccm/m2. NH3 may be provided at a flow rate of about 1400 to about 4200 sccm/m2 of substrate plan area, for example about 2500 to about 3100 sccm/m2, or about 2800 sccm/m2. The gas mixture may also include a carrier gas, such as N2. In one embodiment, N2 may be provided at a flow rate of about 9000 to about 27000 sccm/m2 of substrate plan area, for example about 14400 to about 21600 sccm/m2, or about 18000 sccm/m2. Pressure is regulated within the processing chamber between about 0.8 to about 1.8 Torr, for example about 1.0 to about 1.3 Torr, or about 1.0 Torr. Power provided to maintain the plasma is may be delivered to the electrode of the processing chamber, and may be provided at about 1500 to about 4500 Watts/m2 of substrate plan area, for example about 2800 to about 3400 Watts/m2, or about 3100 Watts/m2. Spacing may be maintained between the substrate and the showerhead in the range of about 450 to about 800 mils, for example about 500 to about 650 mils, or about 580 mils. During deposition, the temperature of the substrate is maintained between about 250 to about 300 degrees Celsius, for example about 280 to about 290 degrees Celsius, or about 285 degrees Celsius.
The SiNx layer 402GH has a SiH content greater than the SiNx layer 402GL. In generally, there are three kinds of bonding in SiN:H, Si—N, Si—H and N-H. The SiH content equals the SiH bond density/(SiH bond density+N—H bond density+SiN bond density). In one embodiment, the SiNx layer 402GH has a SiH content greater than about 5 percent SiH bond density while the SiNx layer 402GL has a SiH content less than about 5 percent SiH bond density. Additionally, the SiNx layer 402GL has a lower wet etch compared to the SiNx layer 402GH. For example, the SiNx layer 402GL may have wet etch less then 1000 Å/min compared to a wet etch rate of greater than 1000 Å/min for the SiNx layer 402GH.
At step 506, the dual a-Si:H layer 404 is deposited using a two step process. The dual amorphous silicon layer 404 is formed by depositing the a-Si:H layer 404AL at a low deposition rate followed by depositing the a-Si:H layer 404AH at a high deposition rate. The dual silicon nitride and dual amorphous silicon layers may be deposited in-situ the process chamber (i.e., without removing the substrate from the chamber).
In one embodiment, the a-Si:H layer 404AL may deposited at a rate less than about 600 Å/min by forming a plasma from a gas mixture in the process chamber, heating the substrate while providing sufficient RF power to maintain the plasma during deposition. The gas mixture may be comprised of SiH4 and H2 gases. SiH4 may be provided at a flow rate of about 350 to about 1050 sccm/m2 of substrate plan area, for example about 600 to about 800 sccm/m2, or about 700 sccm/m2. H2 may be provided at a flow rate of about 1500 to about 6000 sccm/m2 of substrate plan area, for example about 3000 to about 5000 sccm/m2, or about 4500 sccm/m2. In one embodiment, no carrier gases are utilized. Pressure is regulated within the processing chamber between about 1.2 to about 3.5 Torr, for example about 2.0 to about 2.5 Torr, or about 2.2 Torr. Power provided to maintain the plasma is may be delivered to the electrode of the processing chamber, and may be provided at about 200 to about 700 Watts/m2 of substrate plan area, for example about 350 to about 550 Watts/m2, or about 460 Watts/m2. Spacing may be maintained between the substrate and the showerhead in the range of about 400 to about 800 mils, for example about 450 to about 550 mils, or about 500 mils. During deposition, the temperature of the substrate is maintained between about 250 to about 300 degrees Celsius, for example about 280 to about 290 degrees Celsius, or about 285 degrees Celsius.
In one embodiment, the a-Si:H layer 404AH may deposited at a rate greater than about 600 Å/min by forming a plasma from a gas mixture in the process chamber, heating the substrate while providing sufficient RF power to maintain the plasma during deposition. The gas mixture may be comprised of SiH4 and H2 gases. SiH4 may be provided at a flow rate of about 700 to about 2800 sccm/m2 of substrate plan area, for example about 1260 to about 1540 sccm/m2, or about 1400 sccm/m2. H2 may be provided at a flow rate of about 1000 to about 7000 sccm/m2 of substrate plan area, for example about 3000 to about 6000 sccm/m2, or about 5000 sccm/m2. In one embodiment, no carrier gases are utilized. Pressure is regulated within the processing chamber between about 1.2 to about 3.5 Torr, for example about 2.0 to about 2.5 Torr, or about 2.2 Torr. Power provided to maintain the plasma is may be delivered to the electrode of the processing chamber, and may be provided at about 500 to about 1800 Watts/m2 of substrate plan area, for example about 1000 to about 1300 Watts/m2, or about 1100 Watts/m2. Spacing may be maintained between the substrate and the showerhead in the range of about 400 to about 800 mils, for example about 450 to about 550 mils, or about 500 mils. During deposition, the temperature of the substrate is maintained between about 250 to about 300 degrees Celsius, for example about 280 to about 290 degrees Celsius, or about 285 degrees Celsius.
The a-Si:H layer 404AH has a higher optical bandgap (E04 and Etauc) as compared to the a-Si:H layer 404AL. In one embodiment, the a-Si:H layer 404AH has an optical bandgap E04 greater than about 1.90 eV while the a-Si:H layer 404AL has an optical bandgap E04 between about 1.86-1.89 eV. In one embodiment, the a-Si:H layer 404AH has an optical bandgap Etauc greater than about 1.88 eV while the a-Si:H layer 404AL has an optical bandgap Etauc between about 1.83-1.85 eV.
The combinations of the properties of the SiNx layer 402GL and the a-Si:H layer 404AL results in an interface having improved electron mobility and threshold voltage which results in improved transistor performance. TFTs manufactured using this low temperature process have demonstrated about 20 percent higher mobility as compared to TFTs manufactured using high temperature processes. Advantageously, manufactures may interchange passivation layer PECVD (typically performed at temperatures just under 300 degrees Celsius, with an active layer PECVD. Additionally, the lower processing temperature contributes to longer periods between chamber service intervals. Moreover, as the thicknesses of these layers are selected to be just thick enough to obtain performance goals, the remaining thicknesses of there respective layers may be deposited using higher deposition rate techniques, thereby minimizing the impact on process throughput.
At step 508, the n+a-Si:H layer 406 is deposited. The n+a-Si:H layer may be deposited in the same process chamber as the dual silicon nitride and dual amorphous silicon layers described above.
In one embodiment, the n+a-Si:H layer 406 is deposited by forming a plasma from a gas mixture in the process chamber, heating the substrate while providing sufficient RF power to maintain the plasma during deposition. The gas mixture may be comprised of SiH4, H2 and PH3 gases. SiH4 may be provided at a flow rate of about 400 to about 1500 sccm/m2 of substrate plan area, for example about 800 to about 1100 sccm/m2, or about 950 sccm/m2. H2 may be provided at a flow rate of about 1000 to about 6000 sccm/m2 of substrate plan area, for example about 2500 to about 3500 sccm/m2, or about 2900 sccm/m2. PH3 may be provided at a flow rate of about 500 to about 5000 sccm/m2 of substrate plan area, for example about 1500 to about 2500 sccm/m2, or about 2200 sccm/m2. Pressure is regulated within the processing chamber between about 1.2 to about 3.5 Torr, for example about 1.5 to about 2.0 Torr, or about 1.8 Torr. Power provided to maintain the plasma is may be delivered to the electrode of the processing chamber, and may be provided at about 200 to about 800 Watts/m2 of substrate plan area, for example about 400 to about 550 Watts/m2, or about 450 Watts/m2. Spacing may be maintained between the substrate and the showerhead in the range of about 400 to about 800 mils, for example about 450 to about 550 mils, or about 500 mils. During deposition, the temperature of the substrate is maintained between about 250 to about 300 degrees Celsius, for example about 280 to about 290 degrees Celsius, or about 285 degrees Celsius.
TFT performance was confirmed using a simple back-channel type TFT with a 40 μm channel width (W) and 10 μm channel length (L), as shown in
As a final test of reliability, accelerated bias temperature stress (BTS) tests were performed at 80 degrees Celsius with 40V of gate bias voltage for 100 minutes.
Based on the trend study results, it is understood that the low-temperature, low deposition rate SiNx deposition process may be further optimized to obtain lower wet etch rate and lower Si—H concentration. The other films, high deposition rate SiNx, high deposition rate a-Si:H, low deposition rate a-Si:H and n+a-SI:H layer were satisfactory.
Substrate size expansion has so far been a fact of life in the TFT-LCD industry. This substrate size expansion has provided economies of scale for large-size LCD production. As plasma reactors have been made correspondingly larger, surface standing wave effects have seriously come into play for larger parallel-plate reactors. In order to address these issues, an innovative process chamber, which fundamentally changed the distribution of plasma power in the chamber, was developed. Fortunately the innovations the latest PECVD systems comprise are scalable to any foreseeable size that the TFT-LCD industry may need.
As substrate size increases, lowering the processing temperature is getting ever more important to improve production yield and lower LCD manufacturing costs. Reducing the process temperature even by 60 degrees Celsius is beneficial. In this invention, sub-300 degrees Celsius deposition processes were developed, with both single-layer properties and TFT characteristics confirmed to be nearly equivalent to currently used high-temperature processes in mass production lines.
One advantage of low temperature TFT formation as described herein is compatibility for use on polymer substrates. For, example it is envisioned that polymer substrates will be used to build TFT devices and then OLED devices thereover to make flexible displays to take advantage of improved brightness, less power consumption, smaller sizes and increased viewing angles as compared to LEDs.
While the foregoing is directed to the preferred embodiment of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof. The scope of the invention is determined by the claims which follow.
This application claims benefit from U.S. Provisional Patent Application Ser. No. 60/950,761, filed Jul. 19, 2007, which is incorporated by reference in it entirety.
Number | Date | Country | |
---|---|---|---|
60950761 | Jul 2007 | US |