JP3981350B2 - Particle concentration evaluation method in clean room or mini-environment atmosphere - Google Patents

Description

  The present invention relates to a method for evaluating the concentration of particles in a clean room or mechanical mini-environment atmosphere, particularly surface contamination or atomic contamination of boron and phosphorus.

  The clean room is a work area in which temperature, humidity, and particle concentration are controlled in order to protect highly sensitive devices and products from contamination. High quality clean rooms are required in various industrial fields such as medical facilities and integrated circuit manufacturing.

  Therefore, clean rooms are equipped with a variety of expensive and complex air purification systems for supplying filtered clean air to the clean room.

Nevertheless, certain amounts of boron and phosphorus may remain or be mixed in the clean room atmosphere due to system leakage and air filter density. They can cause wafer contamination. For example, boron or phosphorus contamination on the surface of a treated silicon wafer can be 10 ¹² atoms per cm ² . These particles can diffuse into the surface of the silicon wafer during the heat treatment, thus changing the dopant concentration in that region and affecting the properties of the treated chip. Therefore, it is necessary to know the particle concentration in the clean room.

  One conventional technique for monitoring boron and phosphorus in a clean room environment is to pass clean room air through a liquid and dissolve a part of the boron or phosphorus in the formed bubbles in the liquid. After a relatively long period of time, for example one day, the liquid has collected a sufficient amount of boron or phosphorus for analysis. The content of boron or phosphorus in the liquid is then evaluated by atomic spectroscopy or spectroscopic based analytical methods such as ICPMS.

  The above method reflects the actual concentration of boron or phosphorus in the air relatively ambiguously. This is because most of the air containing phosphorus or boron is dissipated from the liquid with bubbles before it dissolves, and the liquid has only an unspecified content of actual boron or phosphorus. . The method also has the disadvantage that it takes a relatively long time to evaluate only the average value of boron or phosphorus in the clean room.

  It is an object of the present invention to provide a method for evaluating in a short time the particle concentration in the clean room atmosphere that results in a more actual and accurate particle concentration in the clean room.

  The object is a method for evaluating the concentration of particles in an atmosphere of a clean room, the step of exposing a test substrate, particularly a test surface of a silicon wafer, to the atmosphere during the test time, and the end of the test time. Sometimes obtaining and storing the amount of particles on the test substrate, analyzing the amount of particles, and comparing the analyzed amount of particles with the amount of particles analyzed on a reference substrate Achieved by the method.

  According to the method, a certain amount of particles closely related to the particle concentration in the clean room atmosphere on the surface of the test substrate can be stored and fixed on the test substrate after a predetermined time for analysis. The amount of particles can indicate the state of the test substrate at a certain time of contamination of the test substrate. Subsequent comparison of the analyzed particle amount on the test wafer with the analyzed particle amount on the reference substrate exhibits a relative particle concentration in the clean room atmosphere over a predetermined period of time. This gives the final measurement of atmospheric contamination.

  The method of the present invention can further be used to monitor particle concentrations within a special mini-environment of the machine (eg, boron or phosphorus contamination of the test substrate caused by a filter within the machine mini-environment). Also good.

  In one embodiment of the invention, the method further includes cleaning the test substrate prior to exposing the test substrate to the atmosphere. In this cleaning step, the concentration of particles on the surface of the test substrate can be reduced, with the result that almost no particles are present on the test surface at time t = 0 when exposure to the atmosphere of the test substrate begins. It becomes. Thereby, the analyzed amount of particles directly corresponds to the concentration of particles in the atmosphere and reflects the period from the start to the end of exposure of the test substrate to the atmosphere.

  In a further embodiment of the invention, the step of obtaining the amount of particles comprises covering the test surface of the test substrate with the surface of the other substrate after the test time has elapsed. The other substrate holds particles present on the test substrate, or particles that are absorbed or moved within a region near the surface of the test substrate. Then, after the test time elapses, the other substrate acquires the particles in their actual state. In this way, this amount can be accurately analyzed, and the particle concentration in the clean room atmosphere can be determined from this amount. The other coating substrate may be made of a material different from the test substrate.

  In a useful example of the invention, the other substrate is a further test substrate, in particular a silicon wafer. In that case, the two substrates form a pair of test substrates. Covering test substrates with the same or similar substrates has the advantage that these substrates have the same or comparable particle acquisition behavior. Furthermore, the substrates can be easily combined, for example, to form a pair of bonded substrates.

  According to an advantageous variant of the invention, the step of storing the amount of particles comprises a heat treatment of the test substrate after obtaining the amount of particles. By the heat treatment step, the obtained particles can diffuse into the test substrate. There, the particles are stored or fixed for further analysis of the test substrate.

  In another variation of the invention, the heat treatment is performed for a time ranging from about 1 hour to about 3 hours, preferably about 2 hours. During that time, the particles diffuse into the test substrate and a specific particle distribution can be generated on the test substrate. The distribution can be analyzed to evaluate the acquired amount of particles. With a heat treatment time of 2 hours, the penetration depth of the particles becomes particularly suitable.

  In a further embodiment of the invention, the heat treatment is performed at a temperature between 800 ° C. and 1050 ° C., preferably about 950 ° C. At that temperature, the particle diffusion activity is high enough to obtain a good particle distribution on the test substrate that is easy to analyze. Particularly at 950 ° C., a good relationship between the particle distribution produced and the energy required for diffusion is achieved. The upper temperature limit of 1050 ° C. prevents excessive homogeneity of the particle distribution and / or excessive bonding of the substrates.

  According to a particular embodiment of the invention, a pair of test substrates joined together at the end of the test time is subsequently separated for analysis. In this method, the test surface exposed to the atmosphere and preserved in the bonding step during the test time is in a state where it is easy to analyze the amount of particles stored on the substrate at the end of the test time.

  In a further example of the invention, the method further includes eroding at least one of the back surfaces of the pair of test substrates joined together. By eroding material from the back of the test substrate pair, a stepwise back approach to the test surface is possible. The test surface is preserved by bonding the substrates so that at least the amount of particles can be analyzed on this surface.

  According to a further embodiment of the present invention, the analysis of the amount of particles includes evaluating an atomic concentration profile of the test substrate. The atomic concentration profile directly displays the amount of particles contained in and on the test substrate.

  In a more effective embodiment of the invention, the atomic concentration profile is evaluated by secondary ion mass spectrometry. Secondary ion mass spectrometry is a very accurate and effective analytical technique that provides an atomic concentration profile of a test substrate.

  In yet another embodiment of the invention, the atomic concentration profile is evaluated at the test surface of the test substrate. Since the test surface is exposed to a clean room atmosphere, the atomic concentration profile well reflects the particle concentration in the atmosphere.

  In another preferred embodiment of the invention, the test surface is analyzed over a thickness of about 100 to 500 nanometers, preferably about 200 nanometers. The thickness provides appropriate information about the amount of particles in the test substrate. Analysis at a thickness of 200 nm in particular shows an effective relationship between the energy consumption for analysis and the amount of particles evaluated over that thickness.

  According to another preferred variant of the invention, a series of test substrates are formed at suitable intervals, preferably about every 30 minutes. This makes it possible to monitor changes in particle concentration in the atmosphere in a clean room or mini-environment at given intervals. In particular, the time interval of every 30 minutes is very suitable for obtaining an overview of the particle concentration in the atmosphere in the clean room or mini-environment throughout the time interval. In this way, it is possible to evaluate the degree of contamination in a clean room or mini-environment.

  According to another more preferred variant of the invention, the method further comprises cleaning the reference substrate, preferably together with the test substrate. The cleaned reference substrate exhibits no particles on the substrate at time t = 0 and can be easily compared to the test substrate. When the reference substrate is cleaned with the test substrate, both substrates have the same initial state on the surface. It makes the results of subsequent comparisons of these substrates more accurate.

  In a further embodiment of the invention, the method further comprises covering the reference surface of the reference substrate with the reference surface of another reference substrate. Thereby, the initial or original state on the reference surface of the reference substrate is retained for analysis and acquisition of the reference value of the initial or original particle amount. The initial or original particle amount can be compared to the particle amount of other test substrates.

  According to another preferred embodiment of the invention, the method further comprises a heat treatment of the reference substrate. In the heat treatment step, the final particle amount on the reference substrate diffuses into the substrate, and the particle amount can be stored in a region near the surface of the reference substrate.

  In another effective variant of the invention, the heat treatment temperature and heat treatment time of the reference substrate are equal to the heat treatment temperature and heat treatment time of the test substrate. By applying the same heat treatment temperature and heat treatment time for the test substrate and the reference substrate, the measurement results on the substrate can be compared better after analysis of the substrate.

  In another more effective embodiment of the invention, the method further comprises analyzing the amount of particles in the reference substrate. The particle amount can be compared with the evaluated particle amount of the test substrate.

  In another embodiment of the present invention, the analysis of the amount of particles includes an evaluation of the atomic concentration profile of the reference substrate. The atomic concentration profile can be compared to a test substrate.

  In an effective variant of the invention, the atomic concentration of the reference substrate is evaluated by secondary ion mass spectrometry. This analysis method allows an accurate assessment of the amount of particles in and on the reference substrate.

  In a preferred embodiment of the present invention, the atomic concentration profile of the reference substrate is evaluated at the reference surface of the reference substrate. The surface represents the initial or original amount of particles on the reference substrate, which is a good indicator of the amount of particles on the reference substrate at time t = 0 and analyzed on the test substrate. Can be compared to the amount of particles produced.

  According to a further embodiment of the invention, the reference surface of the reference substrate is analyzed over a thickness of about 100 to 500 nanometers, preferably about 200 nanometers.

  This thickness corresponds to the preferred thickness for analyzing the test substrate and is therefore well suited for further comparison between the reference substrate and the test substrate. In particular, a thickness of 200 nm allows analysis of the amount of particles in the vicinity of the relevant surface of the substrate.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 shows the steps of an embodiment of a method for assessing boron or phosphorus concentrations in a clean room or mini-environment atmosphere according to the present invention. In this method, silicon wafers 1, 3, 5, and 6 are handled. N-type silicon is used for the analysis of boron, and p-type silicon is used for the analysis of phosphorus. Substrate level N _S is the bulk dopant concentration of the silicon wafer is lower than analyze different levels.

  In step a) of FIG. 1, two or more silicon wafers 1 and 3 (shown in a standing or tilted state on the wafer holder 10) are immersed in a bath 8 in which a cleaning solution 9 is placed. To do. The cleaning solution 9 removes inorganic and organic contamination on the surfaces 2 and 4 of the silicon wafers 1 and 3. As the cleaning solution 9, a typical RCA cleaning solution is used. However, the cleaning solution may be any other suitable solution that can remove any contamination from the surfaces 2 and 4 of the silicon wafers 1 and 3 or a suitable solution of gas and solution. It can be supplemented or replaced with a mixture. The cleaning solution 9 is at a temperature between about 19 ° C. to 25 ° C. and about 85 ° C., and is stirred with a stirrer (not shown) so that the components of the cleaning solution have a constant concentration, or It can be circulated.

  According to step b) in FIG. 1, the reference wafers 3 and 6 are bonded immediately after cleaning of the wafers 3, 6 by bringing their cleaned surfaces close to each other and covering each other. On the other hand, in step b ′) of FIG. 1, one of the cleaned silicon test wafers 1 and a further silicon test wafer 5 (not shown) are exposed to the clean room atmosphere 7. The atmosphere 7 contains air contaminated with boron or phosphorus. Some of the boron atoms or phosphorus atoms are absorbed or deposited on the surface 2 of the silicon wafer 1.

  In step c) of FIG. 1, a bonded test wafer pair is formed by the test wafer 1 and the test wafer 5, which covers the test surface 2 with the other test surface 16 and the test surface 2 of those substrates. And 16 are joined together. The reference wafer pair 3 and 6 maintains the initial cleaned state of the surface 4 as a reference (standard) at the bonding surface by the bonding step. On the other hand, the test wafer pairs 1 and 5 acquire the contamination of boron or phosphorus deposited on the test surfaces 2 and 16 of the test wafers 1 and 5 at the bonding surface.

  In step d) of FIG. 1, the bonded wafer pairs 3, 6 and 1, 5 are heat-treated at about 950 ° C. for about 2 hours in the heat treatment apparatus 11. In this heat treatment step, at least boron or phosphorus atoms obtained on the bonding surface between the test wafers 1 and 5 diffuse into the region in the vicinity of the bonding surfaces of the wafers 1 and 5 and are thus retained.

  In step e) of FIG. 1, after heat treatment, the bonded test wafer pairs 3, 6 and 1, 5 are separated by a laser blade (razor blade) 12 which is pushed between the bonding surfaces 4, 17 and 2, 16, respectively. Is done.

  Step f) in FIG. 1 shows the wafer in a state after separation. Here, due to the diffusion of boron or phosphorus particles into the silicon wafers 1 and 5, the content of boron or phosphorus is increased at least in the vicinity of the surfaces 2 and 16 of the wafers 1 and 5.

  In step g) of FIG. 1, the atomic concentration profiles of the wafers 1 and 3 are analyzed using a secondary ion mass spectrometry (SIMS) apparatus 13. Starting from the reference wafer 3, according to the profile 14 measured over the surface about 200 nanometers, the boron or phosphorus concentration of the reference surface 4 of the reference wafer 3 is almost constant. On the other hand, the atomic concentration profile 15 evaluated on the test surface 2 of the test wafer 1 indicates that the concentration of boron or phosphorus is increasing, particularly near the surface.

According to step h) of FIG. 1, the atomic concentration profiles 14 and 15 are calculated and compared to each other for differences between the profiles. The sum of the profile and bulk level of each surface is the value of the amount deposited. This value is generally expressed as the number of atoms per cm ² . The calculated difference is a direct indicator of boron or phosphorus in the clean room atmosphere. This is because there is a direct proportional relationship between the boron or phosphorus on the test wafer and the boron or phosphorus contamination in the clean room atmosphere.

  Although the silicon wafer has been described in the method shown with reference to FIG. 1, the present method can be applied to any substrate that can acquire or hold particles in the atmosphere. Furthermore, the acquisition and retention effects are obtained in a single method step. In the above-described effective embodiment, the above effect is realized by performing bonding and heat treatment successively. It is also conceivable that the other substrate bonded to the test wafer may be made of a material different from the test wafer.

  Further, the wafer can be heat-treated at an arbitrary temperature and an arbitrary time, and particles such as boron and phosphorus can be held by the substrate to be used. The heat treatment time and the depth of diffusion as a result of the targeted species or particles depends on the heat treatment temperature and the diffusion coefficient of each species. The analyzed depth d of the test wafer and the reference wafer must be deeper than the diffusion length of the analyzed particles or contaminants. The conventional SIMS measurement for analyzing the substrate may be replaced by TOF-SIMS (Time-of-Flight-Secondary Ion Mass Spectroscopy) surface analysis.

  Although the method has been described with reference to the evaluation of boron or phosphorus particle concentrations in the atmosphere, it can be further applied to the evaluation of contamination in any other particle concentration or other type of atmosphere.

  In place of steps e) and f) of FIG. 1, the amount of particles obtained and held is reduced by wearing or eroding one of the back surfaces of at least one of the wafers 1 or 5 shown in e) of FIG. It can also make it easier to analyze. By stepwise backward erosion of at least one of the wafers, an approach can be taken to analyze the region of interest where particles from surfaces 2, 16 can be diffused.

  FIG. 2 shows the concentration profile of the bonded test wafers 1 and 5 after the heat treatment step, showing the concentration of diffused contaminants such as boron and phosphorus as a function of depth. The depth d = 0 corresponds to the surfaces 2 and 16 of the wafers 1 and 5, respectively.

  FIG. 3 shows a SIMS concentration profile 21 of the test wafer 1 in comparison with the SIMS profile 20 of the reference wafer 3, and shows the concentration at which the contaminants are diffused as the analysis depth d. The depth d = 0 corresponds to the surfaces 2 and 4 of the wafers 1 and 3, respectively. The concentration of target species such as boron and phosphorus near the surface of the test wafer 1 is higher than the concentration near the surface of the reference wafer 3. In order to assess the amount of surface contaminants, the difference 22 between the test profile 21 and the reference profile 20 is integrated.

FIG. 3 illustrates steps of an embodiment of a method for evaluating boron or phosphorus concentration in a clean room or mini-environment atmosphere in accordance with the present invention. It is a figure of the density | concentration profile of a pair of test wafer joined. It is a figure of the SIMS density profile of a test wafer and a reference wafer.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1,5 ... Test wafer, 2,16 ... Test surface, 3,6 ... Reference wafer, 4,17 ... Reference surface, 7 ... Atmosphere, 8 ... Container, 9 ... Cleaning solution, 11 ... Heat treatment apparatus, 12 ... Laser blade , 13 ... Secondary ion mass spectrometry (SIMS) apparatus, 14, 20 ... Reference atom concentration profile, 15, 21 ... Test atom concentration profile, 22 ... Difference between test profile and reference profile.

Claims (26)

A method for evaluating particle concentration in an atmosphere (7) of a clean room or a machine mini-environment,
Exposing the test surface (2) of the test substrate (1 ) to the atmosphere (7) for a test time (t) to obtain the amount of particles ;
After the test time (t), the test surface (2) containing the acquired particle amount is joined to the surface (16) of the second test substrate (5) in order to prevent particle loss. Steps,
Analyzing the amount of particles obtained ;
To determine the particle concentration in said atmosphere (7), comparing the amount of particles for reference from the analyzed particle amount, reference substrate (3) on,
Determining the particle concentration in the atmosphere (7);
Including methods.   The method of claim 1, further comprising cleaning the test substrate (1) before exposing the test substrate (1) to the atmosphere (7).   The method according to claim 1 or 2, further comprising the step of heat treating the pair of bonded test substrates (1, 5). The heat treatment, times ranging from 1 hour to 3 hours, The method of claim 3, wherein the performing.   The method according to claim 4, wherein the heat treatment is performed for 2 hours. The method according to any one of claims 3 to 5 the heat treatment, and performing at temperature between 800 ° C. to 1050 ° C..   The method according to claim 6, wherein the heat treatment is performed at 950 ° C.   A pair of test substrates (1, 5) joined together at the end of the test time (t) are subsequently separated for analysis, according to any one of the preceding claims. The method described.   9. The method according to claim 1, further comprising eroding at least one of the back surfaces of the pair of test substrates (1, 5) joined to each other at the end of the test time (t). The method according to item.   10. A method according to any one of the preceding claims, wherein the analysis of the amount of particles comprises evaluating an atomic concentration profile of the test substrate (1).   The method according to claim 1, wherein the atomic concentration profile is evaluated by secondary ion mass spectrometry (SIMS).   12. Method according to claim 10 or 11, characterized in that the atomic concentration profile is evaluated at the test surface (2) of the test substrate (1). Said test surface (2), 1 00-500 nanometer method of claim 12, wherein the analyzing over the thickness of the data (d).   14. Method according to claim 13, characterized in that the test surface (2) is analyzed over a thickness (d) of 200 nanometers. The method according to any one of claims 1 to 14, a series of tests substrate (1, 5), and forming at appropriate time intervals.   16. Method according to claim 15, characterized in that a series of test substrates (1, 5) are formed every 30 minutes. The method according to any one of claims 1 to 16, characterized by further comprising washing the reference substrate (3).   18. The method according to any one of claims 17, further comprising cleaning the reference substrate (3) together with the test substrate (1). According to any one of claims 1 to 18, further comprising a covering said reference reference surface of the substrate (3) and (4) other reference reference surface of the substrate (6) (17) Method. The method according to any one of claims 1 to 19, further comprising a thermal treatment of the reference substrate (3). Further comprising a heat treatment of the reference substrate (3), according to claim 3 to 7 the heat treatment temperature and heat treatment time of the reference substrate (3), characterized in that equal to the heat treatment temperature and heat treatment time of the test substrate (1) The method as described in any one of . The method according to claim 21 , characterized in that the analysis of the amount of particles comprises an evaluation of the atomic concentration profile of the reference substrate (3). The method according to claim 22 , wherein the atomic concentration is evaluated by secondary ion mass spectrometry (SIMS). 24. Method according to claim 22 or 23 , characterized in that the atomic concentration profile is evaluated on a reference surface (4) of the reference substrate (3). The method of claim 24, wherein the said reference surface (4) is analyzed over 1 00-500 nanometer thickness data (d).   26. The method according to claim 25, characterized in that the reference surface (4) is analyzed over a thickness (d) of 200 nanometers.

Images