WO1995002472A1

WO1995002472A1 - Post treatment of a coated substrate with a gas containing excited halogen to remove residues

Info

Publication number: WO1995002472A1
Application number: PCT/US1994/007751
Authority: WO
Inventors: Eileen R. Sparks; James C. Mitchener; Stuart N. Rounds; John C. Matthews
Original assignee: Fusion Systems Corporation
Priority date: 1993-07-16
Filing date: 1994-07-12
Publication date: 1995-01-26
Also published as: EP0710161A4; DE710161T1; EP0710161A1; JPH09502646A

Abstract

Material that has previously been subjected to a removal treatment is removed from a substrate by being contacted with a gas containing an excited halogen at a pressure of greater than 50 torr.

Description

POST TREATMENT OF A COATED SUBSTRATE WITH A GAS CONTAINING EXCITED HALOGEN TO REMOVE RESIDUES

DESCRIPTION

Technical Field The present invention is concerned with a post treatment method for the removal of material (i.e., material that has previously been subjected to a removal treatment) . In particular, the removal process of the present invention involves using a gas containing halogen in an excited state and at pressures of greater than 50 torr. The present invention is especially advantageous in removal of photoresist residues, and most especially, those difficult to remove residues created from ion implantation and from etching procedures.

Background Art

In the manufacture of integrated circuits, the technique of photolithography is frequently used. In the practice of this technique, a semiconductor wafer is coated with a photoresist, which is then exposed with ultraviolet radiation, which is passed through a mask so that a desired pattern is imaged on the photoresist. This causes changes in the solubility of the exposed areas of the photoresist, such that after development in a suitable solvent, the desired pattern is fixed on the wafer, whereupon the photoresist may be hard-baked to enable it to withstand subsequent processing.

In such subsequent processing, integrated circuit components which correspond to the desired pattern are formed by processes, including plasma etching or ion implantation.

After the integrated circuit components are formed, it is desired to strip the photoresist from the wafer, which at this point, has already served its useful purpose. The relative ease or difficulty with which the photoresist may be stripped depends on the degree to which physical and chemical changes have been induced in the resist during the specific plasma etching, or ion implantation processes. Thus, it is generally known that a significant degree of hard-baking and to an even greater extent, the processes of plasma etching and ion implantation induce physical and chemical changes in the photoresist, so that stripping is particularly difficult.

In fact, it is critically important that as much as possible of the photoresist is removed in order to produce ε high yield of useful devices. Particles attached to the wafer surface can interfere with the device structure and result in device failures. In addition, residues may be present that are independent of photoresist. Such residues sometimes result from the processes used to fabricate the devices themselves. A known source of such particles include the above mentioned residues. Two examples of the residue problem are ion implant residues and sidewall polymer residues. Ion implant residues are residues remaining after ashing a wafer that has been implanted with a dopant specie of interest, typically arsenic, boron, or phosphorus, in order to create the electronically active structure. This residue may consist of fragments of the initial resist that may have combined with the implanted specie to create a complex organometallic material that is not readily amenable to removal. The second example is that of sidewall polymer residue that may be created intentionally to aid in creating a desired etch geometry, or unintentionally as a consequence of processing, particularly in a fluorine containing plasma environment, such that the resultant polymer is not readily amenable to removal. This residue may contain perfluorinated or chlorofluorinated organic materials.

In the prior art, the most common techniques which have been used for photoresist stripping are the use of wet oxidative developers, such as sulfuric acid-hydrogen peroxide solution, and the technique of plasma ashing. However, these have not proven to be altogether satisfactory, as wet oxidative developers are difficult and dangerous to work with and have tended to result in surface contamination, while plasma ashing has characteristically been too slow, and has sometimes resulted in electrical damage to the wafer.

A further technique for photoresist stripping comprises exposing the photoresist to an ozone-containing gaseous atmosphere. In Ury, et al., U.S. Patent No. 4,885,047, entire disclosure of which is incorporated hereby by reference, a photoresist stripping method using ozone which achieves high stripping rates was disclosed. The high stripping rates are achieved by passing the ozone over the photoresist through a narrow gap, while the resist is held at an elevated temperature.

In the apparatus disclosed in U.S. Patent No. 4,885,047 (Figure 1) , an oxidizing gas, such as a mixture of ozone and oxygen, is introduced at a port central to the wafer. The gas flows radially in a narrow gap defined by the wafer, and a quartz plate to the wafer edge. As the gas flows over the wafer, the photoresist is oxidized and thereby removed.

In the system of the prior patent, it has been found that in order to supply a sufficient quantity of ozone to conduct the ashing process at the required speed, the gas must be introduced at a velocity such that the portion of wafer underlying the gas jet is cooled. The cooling results in lesser conversion of ozone to atomic oxygen, and so slows the ashing rate, especially in the vicinity of the jet. The foregoing phenomenon is manifested by a dip in the plot of ashing rate versus radial position at the center of the wafer.

Additionally, the treatment gas is used up, i.e., the concentration of atomic oxygen, oxygen, and ozone decreases, as the gas flows over the water surface. Since the gas is introduced at the center of the wafer, the photoresist at the edge which, as discussed above, requires the most reaction to remove, is subject to the least potent gas.

Subsequently, an improved photoresist removal method and apparatus which cleans a wafer of resist quickly was disclosed by Matthews, et al. in U.S. Patent No. 5,071,485, entire disclosure of which is incorporated herein by reference.

In accordance with the invention of U.S. Patent No. 5,071,485, an apparatus wherein there is a space above the photoresist having both a peripheral area which encompasses the periphery of the space, and a central area is provided. The oxidizing gas is introduced at the peripheral area of the space, near which the edge of the wafer is located. The gas is caused to flow from the edge to the center of the wafer, where it is exhausted.

Although the removal techniques disclosed in U.S. Patent Nos. 4,885,047 and 5,071,485 are quite effective, there still is need for improvement, particularly with respect to removal of residue that remains after such treatments, such as the ion implant residues and sidewall polymer residues discussed above.

Summary of Invention

An object of the present invention is to provide a process for removing material from the surface of a substrate, wherein the material is that which is remaining after prior removal treatment. A further object of the present invention is to remove as much as possible of photoresist residue including ion implant residues and sidewall polymer residues.

In particular, it has been found pursuant to the present invention, that effective and efficient removal of material remaining after prior removal treatment is achieved by contacting the surface with a gas that contains an excited halogen at a pressure of greater than 50 torr.

Summary of Drawings Figure 1 is a block diagram of removal apparatus suitable for carrying out the present invention.

Figure 2 is a cross-section of one portion of a photoresist removal apparatus suitable for use in the invention. Figure 3 shows the other portion of a removal apparatus partly in cross section suitable for use in the invention.

Figure 4 shows the removal apparatus having both portions.

Best and Various Modes for Carrying Out Invention

The process of the present invention employs a gaseous mixture that contains excited halogen. Examples of suitable halogens employed are fluorine containing compounds such as CF_*,, C₂F₆, CHF₃, CFH₃, C₂H₂F₄, C₂H₄F₂, CH₂F₂, CH₃CF₃, SF₆, NF₃ and the corresponding iodines, bromines and chlorines. The preferred halogens are the fluorine containing materials, such as CF .

Typically, the excited halogen containing gas is created by passing the halogen through cell 1 (see Figure 1) , which may be of the silent or corona discharge type, which have been used for producing ozone. For example, the ozonizer available from Fusion Systems Corporation made pursuant to U.S. Patent No. 4,970,056 to Wooten is suitable. The excitation is carried out at a pressure of greater than 50 torr, more usually at a pressure of at least 100 torr, preferably at a pressure of about 500 to about 800 torr, and most preferably about 600 to about 800 torr.

The flow rate of the halogen containing gas is typically about 0.001 to about 30 standard liters per minute (SLM) , to about and more typically about 0.06 - 5 SLM. The treatment time (i.e., the amount of time the gas and substrate are in contact) is typically about 10 to about 300 seconds, and more typically, about 40 seconds to about 120 seconds. The gas containing excited halogen is conveyed via conduit 2 to process chamber 3 to thereby treat the substrate. Suitable process chambers are disclosed in U.S. Patent Nos. 4,885,047 and 5,071,485. The process chamber is typically at a pressure of greater than 50 torr, more usually at a pressure of at least 100 torr, preferably at a pressure of about 500 to about 800 torr, and most preferably about 600 to about 800 torr.

The temperature of the substrate on which the material to be removed is located is typically at least about 100°C, more typically about 200°C to about 350°, and most typically about 250°C to about 350°C.

Figures 2 to 4 show a process chamber suitable for carrying out the process of the present invention. Figure 2 shows a portion of the chamber, Figure 3 shows a further portion, and Figure 4 shows the two portions in working relationship.

Referring to Figure 2, it is desired to pass the processing gas through space 34 over semiconductor wafer 30 from the edge of the wafer towards the center. Space 34 has a peripheral area which encompasses the periphery of the space (adjacent the wafer edge) , and a central area. An annular orifice 32 is provided, and the gas is fed through the orifice to the peripheral area of space 34. The space 34 is preferably a narrow gap of about 4 millimeters or less, more preferably less than 2 millimeters, and most preferably less than about 0.6 mm. The annular orifice may be of such diameter and angular disposition that the gas is directed so that it impinges the heated platform (Figure 3) just beyond the edge of the wafer, and is then directed to the edge of the wafer, and in turn towards the center of the wafer. The annular orifice 32 is defined by an outer distribution plate 40 and an inner distribution plate 41. The distribution plates 40, 41 may be exchanged for those of different sizes, so that different size wafers can be accommodated. Additionally, the angular orientation of the orifice may be changed in those instances where it is desirable to impinge the edge of the wafer directly without first impinging the heated platform. Treatment gas enters the treatment chamber through a gas fitting 44. It flows around an annular distribution ring 46 and then through the annular orifice 32.

Instead of an annular orifice, the chamber may have a plurality of local orifices which are disposed in an annular ring around the chamber. Thus, the annular distribution ring 46 may be as shown, but instead of opening an annular orifice, may open into a plurality of discrete, local orifices.

A water jacket 50 is provided to allow control of the inlet gas temperature. In the central area of space 34 is the treatment gas outlet conduit 54. After the gas moves across the wafer, it flows into orifice 56 which is at the mouth of conduit 54. Plate 58 is secured to the bottom of the chamber portion shown in Figure 2, and along with the wafer defines a narrow gap over through which the treatment gases flow as they react with the material to be removed.

Referring next to Figure 3, the other part of the treatment chamber is shown. The wafer 30 is held on a wafer support platform 50, which is heated, for example, by a resistance heater. The wafer support platform may be operated at temperatures up to above 350°C, high temperature being desirable to optimize the process. The wafer support platform is supported on flexible support rods 52, and lift mechanism 54 is arranged to raise and lower the platform via support rods 52. Water is circulated in jacket 58 to effect temperature control.

Figure 4 depicts the two portions of the treatment chamber in operative relationship. In an actual system, the bottom portion would be mounted on a chassis, while the top portion would be connected to a power mechanism for lifting

and lowering it on to the bottom portion, each time a new wafer is inserted in the chamber for processing.

In the operation of the unit, after the wafer is inserted and the top portion is lowered so as to create a gap between the wafer and plate 58, processing gas is fed to inlet 44. It flows to annular distribution ring 46, and out annular orifice 32 to the heated platform 50, and the edge of the water (not shown) , then across the wafer to centrally located outlet orifice 56.

The material to be removed by the present invention is material that has already been subjected to removal treatment technique, including wet stripping, plasma ashing and preferably a non-plasma ozone treatment, such as that disclosed in U.S. Patent Nos. 4,885,047 and 5,071,485. In other words, the process of the present invention is a post treatment. When the substrate has been processed pursuant to U.S. Patent No. 4,885,047 or U.S. Patent No. 5,071,485, the same chamber can be used for both treatment steps of the process, if desired.

In the case of semiconductor processing, photoresists typically composed of novolak resin based compositions containing a photoactive compound, such as a naphthoquinone, are used. Typically, such are applied to the desired substrate at thicknesses of about 0.5 to 5 microns. Although the above detailed description has been directed mainly to photoresists and residue remaining therefrom, the present invention is just as applicable to removal of other undesired materials and contaminants. The underlying substrates on which the material to be removed is located, include silicon wafers, polycrystalline silicon, silicon dioxide, silicon nitride, glass, Group IV- VI semiconductor substrates, such as gallium arsenide, and ceramics, such as aluminum oxide, aluminium suicide, aluminum nitride, and silicon carbide.

It was quite surprising that chemically active species using the halogen containing gas could be achieved at the pressures required by the present invention. Also, it was not predictable that materials that would not deleteriously effect the underlying substrate or equipment used would be achieved. In addition, it could not be predicted that the path length between the excitation source and reaction chamber, which in the present invention is typically about three to about four feet, would be such that the active species could reach the substrate surface without significant recombination.

The following non-limiting examples are presented to further illustrate the present invention.

EXAMPLE 1 A six-inch silicon wafer was coated with a 1.5 micron layer of KTI 820 photoresist (positive novolak type resin) . The photoresist is patterned and then subjected to ultraviolet light and heat, in order to harden it as would be done in the course of regular manufacturing. The silicon wafer is then implanted with As⁺ at 80 KeV and 5 E 15 dose.

The resist coated wafer is then treated in a first step in an asher as shown in Figures 2-4 herein, and Figures 2-4 of U.S. Patent No. 5,071,485. The flow of gases consisted of 24 slm ozone, oxygen mixture and 0.8 slm excited N₂0 from a corona discharge ozone generator. The process time is about 120 seconds, and the substrate is at a temperature of about 325°C.

The treated substrate is then post treated in a second step in the same asher used in the first step. The flow of gases consisted of 1 slm excited CF* that was activated in a silent or corona discharge cell. The treatment time is 60 seconds and the substrate is at a temperature of 325°C during the treatment.

Lightfield and darkfield optical microscope inspection shows that the second step post process appears to remove the residues remaining after the standard single one step process has been completed. Scanning electron microscope inspection shows that the residues are not totally removed, but are substantially reduced. This result represents a substantial improvement over the standard single one step process.

EXAMPLE 2 Example 1 is repeated, except that the silicon wafer is implanted with As⁺ at 80 KeV, 1 E 16 dose. The results obtained are similar to those obtained in Example 1.

EXAMPLE 3 Example 1 is repeated, except that the silicon wafer is ion implanted with BF₂ at 80 KeV 1 E 16 dose. The results obtained are at least as good as those obtained in Example 1.

EXAMPLE 4 Example 1 is repeated, except that the silicon wafer is ion implanted with P+ at 120 KeV, 1 E 16 dose. The results obtained are similar to those obtained in Example 1. EXAMPLE 5

Example 1 is repeated, except that the silicon wafer also contains silicon dioxide layer therein, and is ion implanted with P+ at 120 KeV, 1 E 16 dose. The results obtained are similar to those obtained in Example 1.

Claims

What is claimed is:

1. A process for removing material from the surface of a substrate wherein said material is that remaining after prior removal treatment, which comprises contacting the surface with a gas containing excited halogen at a pressure of greater than 50 torr.

2. The process of claim 1 wherein said halogen is a fluorine containing gas.

3. The process of claim 1 wherein said halogen is selected from the group consisting of CF₄, C₂F₆, CHF₃, CFH₃, C₂H₂F₄ , C₂H₄F₂ , CH₂F₂ , CH₃CF₃ , SF₆ , and NF₃ .

4. The process of claim 1 wherein said halogen is CF₄.

5. The process of claim 1 wherein said pressure is at least 100 torr.

6. The process of claim 1 wherein said pressure is about 500-800 torr.

7. The process of claim 1 wherein said pressure is about 600-800 torr.

8. The process of claim 1 wherein said material to be removed is from photoresist.

9. The process of claim 8 wherein said photoresist has been subjected to ion implantation or to a fluorine containing plasma.

10. The process of claim 9 wherein said photoresist is a novolak polymer composition.

2 11. The process of claim 1 wherein the substrate is at a temperature of at least 100°C during the processing.

12. The process of claim 1 wherein the substrate is at a temperature of at least 200°C during the processing.

13. The process of claim 1 wherein the substrate is at a temperature of 250°C-350° during the processing.

14. The process of claim 1 wherein said substrate is a silicon substrate.

15. The process of claim 1 wherein said gas is excited by corona or silent discharge.

16. The process of claim 1 which comprises placing said substrate on a surface to maintain the substrate during the processing and passing said gas through a space located over the material to be removed, wherein said space forms a narrow gap to cause a thin layer of said gas of four millimeters, or less to flow over the material to be removed.

17. The process of claim 1 which further comprises providing a first surface on which said substrate is placed, providing a second surface which is separated from said first surface, so as to form a space therebetween, said space having a peripheral area, which encompasses the periphery of the space, and a central area, feeding said gas to the peripheral area of said space, and removing said gas at central area of said space, whereby a flow of said gas through said space and over the material to be removed from the peripheral area of the space to the central area is established.

18. The process of claim 17 wherein said space comprises a narrow gap.

19. The process of claim 1 wherein the time of treatment is about 10 to about 300 seconds.

20. The process of claim 1 wherein said prior removal treatment is a non-plasma ozone treatment.

21. The process of claim 9 wherein said photoresist has been stabilized with UV.