CROSS REFERENCE TO RELATED APPLICATION(S)
FIELD OF THE INVENTION
This application is a continuation-in-part of U.S. patent application Ser. No. 10/225,605, filed Aug. 22, 2002, the specification of which is incorporated herein by reference.
- BACKGROUND OF THE INVENTION
The present invention relates generally to semiconductor integrated circuits and, more particularly, to atomic layer deposition of CMOS gates with variable work functions.
Conventional n-type doped polysilicon gate electrodes in CMOS technology have two problems. Firstly, the polysilicon is conductive but there can still be a surface region which can be depleted of carriers under bias conditions. This appears as an extra gate insulator thickness and is commonly referred to as gate depletion and contributes to the equivalent oxide thickness. While this region is thin, in the order of a few angstroms (Å), it becomes appreciable as gate oxide thicknesses are reduced below 2 nm or 20 Å. Another problem is that the work function is not optimum for both n-MOS and p-MOS devices, historically this was compensated for by threshold voltage adjustment implantations. However, as the devices become smaller, with channel lengths of less than 1000 Å and consequently surface space charge regions of less than 100 Å, it becomes more and more difficult to do these implantations. Threshold voltage control becomes an important consideration as power supplies are reduced to the range of one volt. Optimum threshold voltages for both PMOS and NMOS transistors need to have a magnitude of around 0.3 Volts.
A solution to the polysilicon gate depletion problem is to replace the semiconducting gate material with a metal or highly conductive metallic nitrides. As with any new circuit material, the gate electrode must be chemically and thermally compatible with both the transistor and the process. Different metals can be employed or the properties of the conductive nitride modified to provide an optimum work function.
The work function of the gate electrode—the energy needed to extract an electron—must be compatible with the barrier height of the semiconductor material. For PMOS transistors, the required work function is about 5.0 eV. Achieving the lower work function needed by NMOS transistors, about 4.1 eV, has been more difficult. FIGS. 1A and 1B illustrate the desired energy band diagrams and work functions for NMOS and PMOS transistors respectively. Refractory metals like titanium (Ti) and tantalum (Ta) oxidize rapidly under typical process conditions. One proposed solution to the problem relies on a “tuned” ruthenium-tantalum (Ru—Ta) alloy, which is stable under process conditions. When the Ta concentration is below 20 percent, the alloy's electrical properties resemble Rhubidium (Ru), a good PMOS gate electrode. When the Ta concentration is between 40 percent and 54 percent, the alloy is a good NMOS gate electrode.
Promising candidates include metallic nitrides, such as tantalum nitride (TaN) and titanium nitride (TiN). Tantalum nitride, titanium nitride, and tungsten nitride are mid-gap work function metallic conductors commonly described for use in CMOS devices. The use of a mid-gap work function makes the threshold voltages of NMOS and PMOS devices symmetrical in that the magnitudes of the threshold voltages will be the same, but both will have a magnitude larger than that which is optimum with low power supply voltages.
Recently physical deposition, evaporation, has been used to investigate the suitability of some ternary metallic nitrides for use as gate electrodes, these included TiAlN and TaSiN. However, these were deposited by physical deposition not atomic layer deposition and only capacitor structures were fabricated, not transistors with gate structures.
- SUMMARY OF THE INVENTION
Thus, there is an ongoing need for improved CMOS transistor design.
The above mentioned problems CMOS transistor design as well as other problems are addressed by the present invention and will be understood by reading and studying the following specification. This disclosure describes the use of atomic layer deposition of ternary metallic conductors where the composition and work function are varied to control the threshold voltage of both the NMOS and PMOS transistors in CMOS technology to provide optimum performance.
In particular, an embodiment of the present invention includes a transistor having a source region a drain region and a channel therebetween. A gate is separated from the channel region by a gate insulator. The gate includes a ternary metallic conductor formed by atomic layer deposition. In one embodiment the ternary metallic conductor includes Tantalum Aluminum Nitride (TaAlN). In one embodiment the ternary metallic conductor includes Titanium Aluminum Nitride (TiAlN). In one embodiment the ternary metallic conductor includes Titanium Silicon Nitride (TiSiN). In one embodiment the ternary metallic conductor includes Tungsten Aluminum Nitride (WAlN). In some embodiments the gate further includes a refractory metal formed on the ternary metallic conductor.
BRIEF DESCRIPTION OF THE DRAWINGS
These and other embodiments, aspects, advantages, and features of the present invention will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art by reference to the following description of the invention and referenced drawings or by practice of the invention. The aspects, advantages, and features of the invention are realized and attained by means of the instrumentalities, procedures, and combinations particularly pointed out in the appended claims.
FIGS. 1A and 1B illustrate the desired energy band diagrams and work functions for NMOS and PMOS transistors respectively.
FIG. 2 is a graph which plots electron affinity versus the energy bandgap for various metallic nitrides employed in various embodiments of the present invention.
FIG. 3 illustrates an embodiment of a transistor structure formed according to the teachings of the present invention.
FIG. 4 illustrates an embodiment of a memory device, utilizing ternary metallic gates formed by atomic layer deposition, according to embodiments of the present invention.
FIG. 5 is a block diagram of an electrical system, or processor-based system, utilizing ternary metallic gates formed by atomic layer deposition, according to embodiments of the present invention.
In the following detailed description of the invention, reference is made to the accompanying drawings which form a part hereof, and in which is shown, by way of illustration, specific embodiments in which the invention may be practiced. In the drawings, like numerals describe substantially similar components throughout the several views. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. Other embodiments may be utilized and structural, logical, and electrical changes may be made without departing from the scope of the present invention.
The terms wafer and substrate used in the following description include any structure having an exposed surface with which to form the integrated circuit (IC) structure of the invention. The term substrate is understood to include semiconductor wafers. The term substrate is also used to refer to semiconductor structures during processing, and may include other layers that have been fabricated thereupon. Both wafer and substrate include doped and undoped semiconductors, epitaxial semiconductor layers supported by a base semiconductor or insulator, as well as other semiconductor structures well known to one skilled in the art. The term conductor is understood to include semiconductors, and the term insulator is defined to include any material that is less electrically conductive than the materials referred to as conductors. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined only by the appended claims, along with the full scope of equivalents to which such claims are entitled.
This disclosure describes the use of atomic layer deposition of ternary metallic conductors where the composition is varied and work function varied, see FIG. 2, to control the threshold voltage of both the NMOS and PMOS transistors in CMOS technology to provide optimum performance. In the several embodiments, these include the use of TaAlN, TiAlN, TiSiN, and WAlN as the ternary metallic conductors. Conventional highly doped polysilicon or refractory metals as W, Ta, Ti are deposited over the metallic conductors to give the gate structure shown in FIG. 3. As shown in FIG. 3, the transistor 301 structure includes a source region 302, a drain region 304, and a channel 306 therebetween. A gate 310 is separated from the channel region by a gate insulator 308. According to the teachings of the present invention, the gate 310 includes a ternary metallic conductor formed by atomic layer deposition. In one embodiment the ternary metallic conductor includes Tantalum Aluminum Nitride (TaAlN). In one embodiment the ternary metallic conductor includes Titanium Aluminum Nitride (TiAlN). In one embodiment the ternary metallic conductor includes Titanium Silicon Nitride (TiSiN). In one embodiment the ternary metallic conductor includes Tungsten Aluminum Nitride (WAlN). As shown in FIG. 3, in some embodiments the gate further includes a layer of highly conductive polysilicon 312, or alternatively a refractory metal layer 312, formed on the ternary metallic conductor 310. In embodiments having a refractory metal layer, the layer 312 includes for example, and not by way of limitation, refractory metals such as tantalum, titanium and tungsten.
Methods of Formation
Atomic Layer Deposition developed in the early 70s is a modification of CVD and can also be called as “alternately pulsed-CVD”. In this technique, gaseous precursors are introduced one at a time to the substrate surface, and between the pulses the reactor is purged with an inert gas or evacuated. In the first reaction step the precursor is saturatively chemisorbed at the substrate surface, and during the subsequent purging the precursor is removed from the reactor. In the second step, other precursor is introduced on the substrate and the desired films growth reaction takes place. After that the reaction byproducts and the precursor excess are purged out from the reactor. When the precursor chemistry is favorable, i.e. the precursor adsorb and react with each other aggressively, one ALD cycle can be preformed in less than one second in the properly designed flow type reactors.
The striking feature of ALD is the saturation of all the reaction and purging steps which makes the growth self-limiting. This brings the large area uniformity and conformality, the most important properties of ALD, as shown in very different cases, viz. planar substrates, deep trenches, and in the extreme cases of porous silicon and high surface area silica and alumina powers. Also the control of the film thickness is straightforward and can be made by simply calculating the growth cycles. ALD was originally developed to manufacture luminescent and dielectric films needed in electroluminescent displays, and a lot of effort has been put to the growth of doped zinc sulfide and alkaline earth metal sulfide films. Later ALD has been studied for the growth of different epitaxial II-V and II-VI films, nonepitaxial crystalline or amorphous oxide and nitride films are their multilayer structures.
There has been considerable interest towards the ALD growth of silicon and germanium films but due to the difficult precursor chemistry the results have not been very successful.
Reaction sequence ALD (RS-ALD) films have several unique and unmatched advantages:
Continuity at the interface avoiding poorly defined nucleating regions that are typical for CVD (<20 Å) and PVD (<50 Å) films. To achieve this continuity, the substrate surface must be activated to react directly with the first exposure of RS-ALD precursor.
Unmatched conformality over toughest substrate topologies with robust processes that can only be achieved with a layer-by-layer deposition technique.
Typically, low temperature and mildly oxidizing processes. This is thought to be a major advantage for gate insulator processing where deposition of non-silicon based dielectrics without oxidizing the substrate (with the oxidation-precursor) is a major concern.
RS-ALD ability to engineer multilayer laminate films, possibly down to monolayer resolution, as well as alloy composite films appear to be unique. This ability comes from the combination of being able to control deposition with monolayer precision and the ability to deposit continuous monolayers of amorphous films (that is unique to RS-ALD).
Unprecedented process robustness. RS-ALD processes are free of first wafer effects and the chamber dependence. Accordingly, RS-ALD processes will be easier to transfer from development to production and from 200 to 300 mm wafer size.
Thickness depends solely on the number of cycles. Thickness can be “dialed in” as a simple recipe change bearing no need for additional process development upon technology generation advance.
Atomic Layer Deposition of Nitrides
Ta—N: Plasma-enhanced atomic layer deposition (PEALD) of tantalum nitride (Ta—N) thin films at a deposition temperature of 260° C. using hydrogen radicals as a reducing agent for Tertbutylimidotris(diethylamido) tantalum have been described. The PEALD yields superior Ta—N films with an electric resistivity of 400 μΩcm and no aging effect under exposure to air. The film density is higher than that of Ta—N films formed by typical ALD, in which NH3 is used instead of hydrogen radicals. In addition, the as-deposited films are not amorphous, but rather polycrystalline structure of cubit TaN. The density and crystallinity of the films increases with the pulse time of hydrogen plasma. The films are Ta-rich in composition and contain around 15 atomic % of carbon impurity. In the PEALD of Ta—N films, hydrogen radicals are used a reducing agent instead of NH3, which is used as a reactant gas in typical Ta—N ALD. Films are deposited on SiO2 (100 nm)/Si wafers at a deposition temperature of 260° C. and a deposition pressure of 133 Pa in a cold-walled reactor using (Net2)3 Ta=Nbut [tertbutylimidotris(diethylamido)tantalum, TBTDET] as a precursor of Ta. The liquid precursor is contained in a bubbler heated at 70° C. and carried by 35 sccm argon. One deposition cycle consist of an exposure to a metallorganic precursor of TBTDET, a purge period with Ar, and an exposure to hydrogen plasma, followed by another purge period with Ar. The Ar purge period of 15 seconds instead between each reactant gas pulse isolates the reactant gases from each other. To ignite and maintain the hydrogen plasma synchronized with the deposition cycle, a rectangular shaped electrical power is applied between the upper and lower electrode. The showerhead for uniform distribution of the reactant gases in the reactor, capacitively coupled with an rf (13.56 MHz) plasma source operated at a power of 100 W, is used as the upper electrode. The lower electrode, on which a wafer resides, is grounded. Film thickness and morphology were analyzed by field emission scanning electron microscopy.
Ta (Al)N(C): Technical work on thin films have been studied using TaCl5 or TaBr5 and NH3 as precursors and Al(CH3)3 as an additional reducing agent. The deposition temperature is varied between 250 and 400° C. The films contained aluminum, carbon, and chlorine impurities. The chlorine content decreased drastically as the deposition temperature is increased. The film deposited at 400° C. contained less than 4 atomic % chlorine and also had the lowest resistivity, 1300 μΩcm. Five different deposition processes with the pulsing orders TaCl5-TMA-NH3, TMA-TaCl5—NH3, TaBr5—NH3, TaBr5—Zn—NH3, and TaBr5-TMA-NH3 are used. TaCl5, TaBr5, and Zn are evaporated from open boats held inside the reactor. The evaporation temperatures for TaCl4, TaBr5, and Zn are 90, 140, 380° C., respectively. Ammonia is introduced into the reactor through a mass flowmeter, a needle valve, and a solenoid valve. The flow rate is adjusted to 14 sccm during a continuous flow. TMA is kept at a constant temperature of 16° C. and pulsed through the needle and solenoid valve. Pulse times are 0.5 s for TaCl5, TaBr5, NH3, and Zn whereas the pulse length of TMA is varied between 0.2 and 0.8 s. The length of the purge pulse is always 0.3 s. Nitrogen gas is used for the transportation of the precursor and as a purging gas. The flow rate of nitrogen is 400 sccm.
TiN: Atomic layer deposition (ALD) of amorphous TiN films on SiO2 between 170° C. and 210° C. has been achieved by the alternate supply of reactant sources, Ti[N(C2H5CH3)2]4 [tetrakis(ethylmethylamino)titanium: TEMAT] and NH3. These reactant sources are injected into the reactor in the following order: TEMAT vapor pulse, Ar gas pulse, NH3 gas pulse and Ar gas pulse. Film thickness per cycle saturated at around 1.6 monolayers per cycle with sufficient pulse times of reactant sources at 200° C. The results suggest that film thickness per cycle could exceed 1 ML/cycle in ALD, and are explained by the rechemisorption mechanism of the reactant sources. An ideal linear relationship between number of cycles and film thickness is confirmed.
TiAlN: Koo et al published paper on the study of the characteristics of TiAlN thin film deposited by atomic layer deposition method. The series of metal-Si—N barriers have high resistivity above 1000 μΩcm. They proposed another ternary diffusion barrier of TiAlN. TiAlN film exhibited a NaCl structure in spite of considerable Al contents. TiAlN films are deposited using the TiCl4 and dimethylaluminum hydride ethypiperdine (DMAH-EPP) as the titanium and aluminum precursors, respectively. TiCl4 is vaporized from the liquid at 13-15° C. and introduced into the ALD chamber, which is supplied by a bubbler using the Ar carrier gas with a flow rate of 30 sccm. The DMAH-EPP precursor is evaporated at 60° C. and introduced into the ALD chamber with the same flow rate of TiCl4. The NH3 gas is also used as a reactant gas and its flow rate is about 60 sccm. Ar purging gas is introduced for the complete separation of the source and reactant gases. TiAlN films are deposited at the temperatures between 350 and 400° C. and total pressure is kept constant to be two torr.
TiSiN: Metal-organic atomic-layer deposition (MOALD) achieves near-perfect step coverage step and control precisely the thickness and composition of grown thin films. A MOALD technique for ternary Ti—Si—N films using a sequential supply of Ti[N(CH3)2]4 [tetrakis (dimethylamido) titanium: TDMAT], silane (SiH4), and ammonia (NH3), has been developed and evaluated the Cu diffusion barrier characteristics of a 10 mn Ti—Si—N film with high-frequency C-V measurements. At 180° C. deposition temperature, silane is supplied separately in the sequence of the TDMAT pulse, silane pulse, and the ammonia pulse. The silicon content is the deposited films and the deposition thickness per cycle remained almost constant at 18 at. % and 0.22 nm/cycle, even though the silane partial pressure varied from 0.27 to 13.3 Pa. Especially, the Si content dependence is strikingly different from the conventional chemical-vapor deposition. Step coverage is approximately 100% even on the 0.3 μm diameter hole with slightly negative slope and 10:1 aspect ratio.
WN: Tungsten nitride films have been deposited with the atomic layer control using sequential surface reactions. The tungsten nitride film growth is accomplished by separating the binary reaction 2WF6+NH3→W2N+3HF+9/2F2 into two half-reactions. Successive application of the WF6 and NH3 half-reactions in an ABAB . . . sequence produced tungsten nitride deposition at substrate temperatures between 600 and 800 K. Transmission Fourier transform infrared (FTIR) spectroscopy monitored the coverage of WF6* and NHy* surface species on high surface area particles during the WF6 and NH3 half-reactions. The FTIR spectroscope results demonstrated the WF6 and NH3 half-reactions are complete and self-limiting at temperatures >600 K. In situ spectroscopic ellipsometry monitored the film growth on Si(100) substrate vs. temperature and reactant exposure. A tungsten nitride deposition rate of 2.55 Å/AB cycle is measured at 600-800 K for WF6 and NH3 reactant exposure >3000 L and 10,000 L, respectively. X-ray photoelectron spectroscopy depth-profiling experiments determined that the films had a W2N stoichiometry with low C and O impurity concentrations. X-ray diffraction investigations revealed that the tungsten nitride films are microcrystalline. Atomic force microscopy measurements of the deposited films observed remarkably flat surface indicating smooth film growth. These smooth tungsten nitride films deposited with atomic layer control should be used as diffusion control for Cu on contact and via holes.
AlN: Aluminum nitride (AlN) has been grown on porous silica by atomic layer chemical vapor deposition (ALCVD) from trimethylaluminum (TMA) and ammonia precursors. The ALCVD growth is based on alternating, separated, saturating reactions of the gaseous precursors with the solid substrates. TMA and ammonia are reacted at 423 and 623 Kelvin (K), respectively, on silica which has been dehydroxylated at 1023 K pretreated with ammonia at 823 K. The growth in three reaction cycles is investigated quantitatively by elemental analysis, and the surface reaction products are identified by IR and solid state and Si NMR measurements. Steady growth of about 2 aluminum atoms/nm2 silica/reaction cycle is obtained. The growth mainly took place through (I) the reaction of TMA which resulted in surface Al—Me and Si—Me groups, and (II) the reaction of ammonia which replaced aluminium-bonded methyl groups with amino groups. Ammonia also reacted in part with the silicon-bonded methyl groups formed in the dissociated reaction of TMA with siloxane bridges. TMA reacted with the amino groups, as it did with surface silanol groups and siloxane bridges. In general, the Al—N layer interacted strongly with the silica substrates, but in the third reaction cycle AlN-type sites may have formed.
In FIG. 4 a memory device is illustrated according to the teachings of the present invention. The memory device 440 contains a memory array 442, row and column decoders 444, 448 and a sense amplifier circuit 446. The memory array 442 consists of a number of transistor cells 400, having ternary metallic gates formed by atomic layer deposition, whose word lines 480 and bit lines 460 are commonly arranged into rows and columns, respectively. The bit lines 460 of the memory array 442 are connected to the sense amplifier circuit 446, while its word lines 480 are connected to the row decoder 444. Address and control signals are input on address/control lines 461 into the memory device 440 and connected to the column decoder 448, sense amplifier circuit 446 and row decoder 444 and are used to gain read and write access, among other things, to the memory array 442.
The column decoder 448 is connected to the sense amplifier circuit 446 via control and column select signals on column select lines 462. The sense amplifier circuit 446 receives input data destined for the memory array 442 and outputs data read from the memory array 442 over input/output (I/O) data lines 463. Data is read from the cells of the memory array 442 by activating a word line 480 (via the row decoder 444), which couples all of the memory cells corresponding to that word line to respective bit lines 460, which define the columns of the array. One or more bit lines 460 are also activated. When a particular word line 480 and bit lines 460 are activated, the sense amplifier circuit 446 connected to a bit line column detects and amplifies the conduction sensed through a given transistor cell and transferred to its bit line 460 by measuring the potential difference between the activated bit line 460 and a reference line which may be an inactive bit line. Again, in the read operation the source region of a given cell is couple to a grounded sourceline or array plate (not shown). The operation of Memory device sense amplifiers is described, for example, in U.S. Pat. Nos. 5,627,785; 5,280,205; and 5,042,011, all assigned to Micron Technology Inc., and incorporated by reference herein.
FIG. 5 is a block diagram of an electrical system, or processor-based system, 500 utilizing transistor cells having ternary metallic gates formed by atomic layer deposition according to the teachings of the present invention. For example, by way of example and not by way of limitation, memory 512 is constructed in accordance with the present invention to have transistor cells having ternary metallic gates formed by atomic layer deposition. However, the invention is not so limited and the same can apply to transistors in the CPU, etc. The processor-based system 500 may be a computer system, a process control system or any other system employing a processor and associated memory. The system 500 includes a central processing unit (CPU) 502, e.g., a microprocessor, that communicates with the memory 512 and an I/O device 508 over a bus 520. It must be noted that the bus 520 may be a series of buses and bridges commonly used in a processor-based system, but for convenience purposes only, the bus 520 has been illustrated as a single bus. A second I/O device 510 is illustrated, but is not necessary to practice the invention. The processor-based system 500 can also includes read-only memory (ROM) 514 and may include peripheral devices such as a floppy disk drive 504 and a compact disk (CD) ROM drive 506 that also communicates with the CPU 502 over the bus 520 as is well known in the art.
It will be appreciated by those skilled in the art that additional circuitry and control signals can be provided, and that the processor-based system 500 has been simplified to help focus on the invention.
It will be understood that the embodiment shown in FIG. 5 illustrates an embodiment for electronic system circuitry in which the novel ternary metallic gate transistor cells, formed by atomic layer deposition, are used. The illustration of system 500, as shown in FIG. 5, is intended to provide a general understanding of one application for the structure and circuitry of the present invention, and is not intended to serve as a complete description of all the elements and features of an electronic system using the novel ternary metallic gate transistor cells, formed by atomic layer deposition. Further, the invention is equally applicable to any size and type of system 500 using the novel ternary metallic gate transistor cells, formed by atomic layer deposition, and is not intended to be limited to that described above. As one of ordinary skill in the art will understand, such an electronic system can be fabricated in single-package processing units, or even on a single semiconductor chip, in order to reduce the communication time between the processor and the memory device.
Applications containing the novel ternary metallic gate transistor cells, formed by atomic layer deposition as described in this disclosure, include electronic systems for use in memory modules, device drivers, power modules, communication modems, processor modules, and application-specific modules, and may include multilayer, multichip modules. Such circuitry can further be a subcomponent of a variety of electronic systems, such as a clock, a television, a cell phone, a personal computer, an automobile, an industrial control system, an aircraft, and others.
This disclosure describes the use of atomic layer deposition of ternary metallic conductors as transistor gates. The composition is varied and work function varied to control the threshold voltage of both the NMOS and PMOS transistors in CMOS technology to provide optimum performance.
It is to be understood that the above description is intended to be illustrative, and not restrictive. Many other embodiments will be apparent to those of skill in the art upon reviewing the above description. The scope of the invention should, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.