METHOD FOR FORMING A FLOATING GATE USING CHEMICAL MECHANICAL PLANARIZATION

Abstract

An improved process forming a floating gate region of a semiconductor memory device. The process includes using a ceria slurry for chemical mechanical planarization to provide “stop on polysilicon” capabilities, allowing a thin nitride layer, or in the alternative no nitride layer, to be used and reducing the number of processing steps required to form the floating gate region.

Description

FIELD OF THE INVENTION

The present invention relates to the field of semiconductor devices and, in particular, to a method for forming floating gate regions.

BACKGROUND OF THE INVENTION

A nonvolatile memory is a type of memory that retains stored data when power is removed. There are various types of nonvolatile memories including e.g., read only memories (ROMs), erasable programmable read only memories (EPROMs), and electrically erasable programmable read only memories (EEPROMs). One type of EEPROM device is a flash EEPROM device (also referred to as “flash memory”).

Each nonvolatile memory device has its own unique characteristics. For example, the memory cells of an EPROM device are erased using an ultraviolet light, while the memory cells of an EEPROM device are erased using an electrical signal. In a conventional flash memory device blocks of memory cells are simultaneously erased (what has been described in the art as a “flash-erasure”). The memory cells in a ROM device, on the other hand, cannot be erased at all. EPROMs, and EEPROMs, including flash memory, are commonly used in computer systems that require reprogrammable nonvolatile memory.

Flash memory devices have developed into a popular source of non-volatile memory for a wide range of electronic applications. Flash memory devices typically use a one-transistor memory cell that allows for high memory densities, high reliability, and low power consumption. Common uses for flash memory include, e.g., portable computers, personal digital assistants (PDAs), digital cameras, portable music players, and cellular telephones. Program code, system data such as a basic input/output system (BIOS), and other firmware can typically be stored in flash memory devices.

FIGS. 1A-1G depict one conventional process of forming floating gate regions for one-transistor storage cells of non-volatile memory devices.

As shown in FIG. 1A, a pad oxide layer 12 is formed on a silicon substrate 11. A nitride layer 13 is then formed on top of the pad oxide 12. Trenches 14 are formed in the resulting structure, as shown in FIG. 1B. An oxide layer 15 is deposited within the trenches 14 and on top of the nitride layer 13. The resulting structure is shown in FIG. 1C.

Standard STI chemical mechanical planarization (CMP) is used to isolate the active regions of the device. The conventional STI CMP process uses the nitride layer 13 as a stop layer. The structure resulting from the STI CMP process is illustrated in FIG. 1D. As shown in FIG. 1E, after the STI CMP process, the nitride layer 13 and the pad oxide layer 12 are stripped, thus exposing the active areas 18.

After the nitride layer 13 and pad oxide layer 12 are stripped, a gate oxide layer 16 and a polysilicon layer 17 are deposited (FIG. 1F). The polysilicon layer 17 will form the floating gate of completed one-transistor flash memory cells. As is shown in FIG. 1F, polysilicon 17 is deposited over the active areas and the oxide 15. A self aligned floating gate (SAFG) CMP process is then implemented to remove excess polysilicon 17 and to isolate the polysilicon 17 in the active areas 18.

The SAFG CMP process is very demanding. The amount of polysilicon 17 left behind over the active areas 18 depends on the field leveling in the array and periphery, oxide dishing in the periphery, array center to edge doming, and the amount of nitride remaining. Dishing refers to the thinning of a structure, caused by uneven polishing based on the selectivity of the slurry being used, resulting in a dish-like profile when measured in reference to the surrounding material. For example, when polishing oxide and stopping on nitride, the slurry used is typically selective to nitride, so when the polish hits nitride, it polishes oxide faster than nitride, resulting in a dish-like profile in the oxide at the level of the nitride. Doming is the opposite of dishing, resulting in a dome-like profile. Doming is usually caused by fill pattern issues. The SAFG CMP needs to be highly selective to the oxide 15 with good polysilicon 17 polishing rate; however, the CMP should not cause dishing in the polysilicon 17.

It is desirable to integrate the above described STI CMP and SAFG CMP steps into a single process flow to overcome the above-noted shortcomings. Accordingly, a simplified process for forming a floating gate region of the transistor storage cells of a non-volatile memory device is needed and desired.

BRIEF DESCRIPTION OF THE DRAWINGS

The above described features of the invention will be more clearly understood from the following detailed description, which is provided with reference to the accompanying drawings.

FIGS. 1A-1G illustrate a method of manufacturing a floating gate region of a non-volatile memory device according to the prior art;

FIGS. 2A-2D illustrate a method of manufacturing a floating gate region of a non-volatile memory device according to a first exemplary embodiment of the invention;

FIGS. 3A-3D illustrate a method of manufacturing a floating gate region of a non-volatile memory device according to a second exemplary embodiment of the invention;

FIG. 4 illustrates an intermediate processing step of a third exemplary embodiment of the invention, which fits between the steps of FIGS. 2C and 2D; and

FIG. 5 illustrates a flash memory device including a floating gate region formed in accordance with the invention.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a method of manufacturing non-volatile memory devices having memory cell transistors with floating gates.

In the following detailed description, reference is made to the accompanying drawings, which form a part hereof and show by way of illustration specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is to be understood that other embodiments may be utilized, and that changes may be made without departing from the spirit and scope of the present invention. The progression of processing steps described is exemplary of embodiments of the invention; however, the sequence of steps is not limited to that set forth herein and may be changed as is known in the art, with the exception of steps necessarily occurring in a certain order.

The term “substrate” is used in the following description to refer to any supporting layer suitable for fabricating an integrated circuit, typically semiconductor based, but not necessarily so. A substrate may be silicon-based, may include epitaxial layers of silicon supported by a base semiconductor or non-semiconductor foundation, can be sapphire-based, silicon-on-insulator (SOI), metal, polymer, or any other suitable materials. When reference is made to a substrate in the following description, previous process steps may have been utilized to form regions, junctions or other structures in or over a base semiconductor or other foundation.

Additionally, while exemplary embodiments of the invention are described in connection with flash memory, the invention is not so limited. The invention is applicable to other integrated circuit devices and systems, which might employ floating gate structures.

The invention relates to an improved method for forming a floating gate semiconductor device. The method uses an improved chemical mechanical planarization (CMP) process to form a self-aligned floating gate region, which requires less steps than the number of steps used in the prior art.

An first exemplary embodiment of the invention is described below in connection with FIGS. 2A-2D. A similar second exemplary embodiment is also described in connection with FIGS. 3A-3D. A third exemplary embodiment is described in connection with FIG. 4.

As shown in FIG. 2A, a gate oxide layer 116 is formed over a silicon substrate 111. A thick polysilicon layer 117 is formed over the gate oxide layer 116. Next, a thin nitride layer 113 is formed over the polysilicon layer 117. The thickness of the nitride layer 113 may be in the range of about 50 Å to about 150 Å thick. The thin nitride layer 113 enables the thicker polysilicon layer 117 to be used. The thickness of the final polysilicon layer 117 (FIG. 2D) may be in the range of about 400 Å to about 1000 Å thick.

As shown in FIG. 2B, trenches 114 are formed in the resulting structure, by any process known in the art for forming shallow trench isolation (STI) regions. An oxide layer 115 is then deposited within the trenches 114 and on top of the nitride layer 113, as is shown in FIG. 2C. The oxide layer 115 may be formed of, for example, a high density plasma (HDP) oxide, O3-TEOS oxide, spin on dielectric (SOD), or any other suitable oxide known in the art.

In a second exemplary embodiment of the invention, the nitride layer 113 may be eliminated, as shown in FIGS. 3A-D. In the case of the second embodiment, the oxide layer 115 is deposited directly on top of the polysilicon layer 117, as shown in FIG. 3C. Like reference numerals in FIGS. 3A-D refer to the same elements as in FIGS. 2A-D and are not discussed in detail herein.

In both the first and second embodiments, chemical mechanical planarization (CMP) is performed during which the oxide layer 115 is partially removed and planarized. As the oxide 115 is being removed in the first embodiment, the CMP process will reach the nitride layer 113. The slurry used in the CMP process of the invention has the capability to remove nitride as well as the oxide. Thus, the thin nitride layer 113 of the first embodiment can be easily removed. As explained below, the selectivity of the slurry allows the process to stop on polysilicon layer 117. Thus, polysilicon layer 117 acts as the stop layer for the CMP process of the invention. In the second embodiment, the CMP process removes the oxide layer 115 and stops on the polysilicon layer 117, as well. The result is a desired thickness of polysilicon 117 in between the oxide areas 115. The resulting floating gate region 110 is shown in FIGS. 2D and 3D. The floating gates 120 are the remaining portions of polysilicon layer 117, insulated from each other by the oxide areas 115. As can be seen, this process requires fewer steps than the conventional process for forming floating gate regions.

An additional benefit of the process is that the slurry used provides enough over-polish margin to clear any oxide 115 and/or nitride 113 residuals. Moreover, polysilicon residue over “dished” oxide areas is also eliminated since there is no polysilicon deposited over the oxide areas during the process.

The STI CMP process of the present invention provides “stop on polysilicon” capability. The slurry used in the present invention is a ceria slurry. A ceria slurry is a slurry comprised of cerium oxide (CeO₂) particles. The ceria slurry of the invention includes CeO₂ particles which may have a mean particle size in the range of about 0.1 μm to about 1.5 μm. The ceria slurry has a solids percentage between about 1% and about 7%. Before use, the slurry is mixed with any cationic and/or any anionic additive. The solids percentage at the point of use may be between about 1% and about 4%. An example of an appropriate additive includes cationic cetyl trimethyl ammonium bromide (CTAB) additive. Post mixing of the slurry and additive should yield a pH range between about 5 and about 8. The ceria slurry with the additive, when mixed in the above specified ratios, provides the necessary oxide and nitride rate with selectivity and planarization capability along with good selectivity to polysilicon. Controlling the solids percentage, pH, and mix ratio can alter the selectivity between oxide, nitride, and polysilicon. Higher pH tends to increase polysilicon rate and decrease nitride rate, while higher solids percentage has a low impact on polysilicon rate but increases nitride rate significantly.

In a third exemplary embodiment, the addition of the additives to the slurry may occur after the stop on nitride using traditional STI CMP is achieved. In this case, a standard STI slurry will be used to remove the oxide layer 115 and stop on nitride 113, as shown in FIG. 4. Beyond that point, a slurry of the invention formulated to have a 1:1:1 or 0.5:1:1 poly:nitride:oxide selectivity will be used to remove the nitride layer 113 and stop on polysilicon 117. This process requires an intermediate processing step between FIGS. 2C and 2D, illustrated in FIG. 4.

As shown in FIG. 5, the floating gate region formed in accordance with the invention may be included as part of a flash memory structure 130. The oxide 115 provides insulation between the floating gates 120. Oxide layer 122 is formed above the floating gates 120 and the oxide 115. Control gates 124 are formed above the floating gates 120 and the oxide layer 122. Oxide 126 provides insulation between the control gates 124. The control gates 124 and oxide 126 are formed as is known in the art.

The above description and drawings are only to be considered illustrative of exemplary embodiments which achieve the features and advantages of the invention. Although exemplary embodiments of the present invention have been described and illustrated herein, many modifications, even substitutions of materials, can be made without departing from the spirit or scope of the invention. Accordingly, the above description and accompanying drawings are only illustrative of exemplary embodiments that can achieve the features and advantages of the present invention. It is not intended that the invention be limited to the embodiments shown and described in detail herein. The invention is limited only by the scope of the appended claims.

Claims

1-33. (canceled)

34. A method of forming a memory cell structure comprising the acts of: providing a first oxide layer over a substrate;providing a polysilicon layer over the first oxide layer;providing an insulating layer over the polysilicon layer;forming a plurality of trenches extending through the insulating layer, the polysilicon layer, the first oxide layer and into the substrate;providing a second oxide layer over the insulating layer and within the plurality of trenches;removing the second oxide layer above the plurality of trenches and the insulating layer by planarization which stops at the polysilicon layer;providing a third oxide layer over the polysilicon layer and the second oxide layer; andproviding a control gate over the third oxide layer.

35. The method of claim 34, wherein the insulating layer is a nitride layer.

36. The method of claim 35, wherein the nitride layer has a thickness of approximately 50 Å to 150 Å.

37. The method of claim 34, wherein the polysilicon layer has a thickness of approximately 400 Å to 1000 Å.

38. The method of claim 34, wherein the removing act utilizes a chemical mechanical planarization process which utilizes a slurry comprising CeO2 particles and chosen to selectively remove only the second oxide and insulating layers.

39. The method of claim 38, wherein the slurry has a solids percent within a range of about 1% and about 7%.

40. The method of claim 38, wherein the slurry is mixed with an additive.

41. The method of claim 40, wherein the additive is at least one of a cationic additive and an anionic additive.

42. The method of claim 40, wherein the additive is cetyl trimethyl ammonium bromide.

43. The method of claim 40, wherein the slurry has a pH within a range of about 5 and about 8.

44. The method of claim 38, wherein the CeO2 particles have a mean particle size of between approximately 0.1 μm and 1.5 μm.