JPH0136054B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a monitoring device for monitoring fine particles or colloidal substances mixed in ultrapure water used in the electronics industry, pharmaceutical industry, etc. Generally, the ultrapure water used for industrial and hospital purposes has a number of microparticles with a diameter of 0.5 μm or less and microorganisms such as bacteria;
ml or less, the latter is required to be sterile. Conventionally, in order to measure these substances, a large amount of sample water, such as 5 to 10 pores, is filtered using a screen-type membrane filter with a through-vertical pore structure, such as a 0.2 μm pore size membrane filter. The fine particles (0.2 μm or more) captured on the membrane were observed and measured using an electron microscope. In this case, in order to clearly count particles larger than 0.2 μm, it is necessary to observe at least 4000 times magnification, and since one field of view is limited to 1 to 6 × 10 ^-6 cm ² , 20 to 30 The field of view was observed, the number of particles per field of view was averaged, the number of particles relative to the total filtration area was calculated, and the number of fine particles per unit volume of sample water was determined from the total filtration amount. Therefore, these measurements require the use of expensive machines, long and complicated operations, and must be performed by skilled technicians. On the other hand, recently, even integrating sphere turbidity meters are being used to measure the content of extremely small amounts of suspended matter and colloidal substances in low-turbidity water (turbidity 1 ppm or less), for which measurement is uncertain. Silting Index, which is calculated from the clogging rate or clogging rate of a membrane filter by filtering a certain amount or a certain period of sample water using a membrane filter with a certain pore size;
Measurement methods and devices such as Plugging Index and Fouling Index are used. These water quality indicators are used to prevent membrane contamination such as reverse osmosis membrane equipment.
Its effectiveness as a means of monitoring the water supplied to the device has been demonstrated, but it is also a means of monitoring the contamination of ultrapure particles in the ultrapure water used in the electronic industry, pharmaceutical industry, hospital water, etc. It could not be used as such. The reason for this is that the membrane filter used is a depth type with a mesh structure, which traps particles over the entire area of the membrane, and even if some particles are trapped on the surface, there is no mesh inside. Because the flow path is intertwined, it has a structure that does not completely block the flow.
This is because in the case of ultrapure water that contains small particles and a small number of particles, clogging does not occur in a short time and the accuracy does not reach the level required for ultrapure water. The present invention improves the drawbacks of the conventional methods as described above, and is capable of detecting the clogging rate or clogging speed even in extremely pure ultrapure water, and also detecting the size, number, etc. of the particles contained. The purpose of this study is to propose a monitoring device for ultrapure water that can be used for various purposes. The present invention provides an apparatus for measuring the time required to filter a predetermined amount of sample water using a membrane filter having a vertical hole structure, and the time required to perform the same operation after a certain period of time. a first calculation device that calculates the clogging rate or clogging speed of the filter from the obtained value; a device that measures changes in the amount of filtration of sample water and the filtration time by a membrane filter having a vertical hole structure; a second calculation device that determines the particle diameter from the slope of the relationship curve between the two determined by the obtained value;
and the value obtained from the first arithmetic device and the second
ultrapure water that includes a device that determines the number of particles by comparing the particle size obtained from the calculation device with a clogging rate or clogging speed vs. particle number relationship curve or formula created in advance for each particle size. It is a monitoring device. The ultrapure water to be monitored in the present invention generally has a specific resistance of 8 MΩ·cm (25° C.) or more. The membrane filter used in the present invention has a through-vertical hole structure, that is, a screen type membrane filter, and there are some that are specially treated thin films such as polycarbonate, but the material thereof is not limited. Generally, filtration includes standard occlusion filtration, intermediate occlusion filtration,
There are three types of completely occluded filtration: standard occluded filtration is a type of filtration that traps particles throughout the interior of the filtration layer (membrane), as described above for conventional membranes, and fully occluded filtration is a type of filtration that traps particles on the surface of the filtration layer (membrane). A filtration form in which particles are captured by a filter, and an intermediate occlusion filtration is an intermediate filtration form between the two. As mentioned above, conventional membrane filters have a mesh structure.
Under constant pressure filtration conditions, standard occlusion filtration or intermediate occlusion filtration is achieved, whereas with the membrane filter used in the present invention, complete occlusion filtration is achieved. As shown in FIG. 1, the screen type membrane filter used in the present invention is a filter in which a large number of through holes 3 are formed perpendicular to the filtration surface 2 of a membrane 1, and particles larger than the through holes 3 are filtered. 4 is a screen type filter that is captured by the filter surface 2; When filtration is performed using such a membrane filter, particles larger than the pore size are captured only on the surface, so clogging occurs in a short time, making it possible to measure the clogging rate or clogging rate of ultrapure water. It is. Use such a membrane filter to filter a fixed amount or a fixed ratio of water under a constant pressure (3.5 to 2.1 Kg/cm ² G), and measure the time or time ratio required for this. Calculate the clogging rate or clogging speed and calculate the SI (Silting) from this.
Index), PI (Plugging Index), FI (Fouling Index)
By calculating the index) and using it as an index, the quality of ultrapure water can be judged to a certain extent, and the degree of contamination of ultrapure water can thereby be monitored. To explain the monitoring method using FI as an example, the above membrane filter is used to measure the time T ₀ required to filter a predetermined amount of sample water at a certain point,
After a certain period of time n minutes (for example, 15 minutes), measure the time T _o required to filter the sample water by the same operation, and use the following formula.
Calculate FI. FI=1-T ₀ /Tn/n×100 Since the FI value calculated from the above increases according to the degree of contamination of the sample water, it serves as an index when monitoring the degree of contamination of ultrapure water. The pore size of the membrane filter can be arbitrarily selected depending on the purpose of monitoring, but it is 0.5 μm, such as the ultrapure water used in the electronics and pharmaceutical industries.
pore size smaller than the particle size, e.g. 0.1 μm or
Use 0.2 μm. Next, a method of examining the particle diameter, number of particles, type, etc. using the membrane filter will be explained. First, to measure the particle size, filtration is performed using a membrane filter having a specific pore size, and a determination is made based on whether or not complete occlusion filtration is achieved. In other words, screen-type membrane filters have a penetrating vertical pore structure, so particles larger than the pore diameter are captured on the surface, resulting in completely blocked filtration, while particles smaller than the pore diameter pass through the surface and are captured inside, which is standard. This is occlusion filtration, and those containing both are intermediate occlusion filtration. For this reason, by filtering with a screen-type membrane filter having a certain pore size and examining the above filtration mode, it is possible to determine which particles are larger or smaller based on a certain particle size that is slightly larger than the pore size of the membrane filter used. You can find out whether you are taking the lead. At this time, if the same operation is performed using membrane filters with different pore sizes, each particle size and abundance ratio can be determined. In order to check the filtration mode to determine whether or not it is completely blocked filtration, it is determined by the slope of the relationship curve between the amount of filtrated water and the filtration time. If the filtration time is θ and the amount of filtrated water is V, then θ/V is proportional to θ in standard occlusion filtration, and dV/dθ is proportional to V in complete occlusion filtration (Maruzen Co., Ltd., Water and Wastewater Handbook, Second Revised Edition) (See page 206). Therefore, the slope of the relationship curve between V (vertical axis) and θ (horizontal axis) is smaller in completely occlusive filtration. Figure 2 is a graph showing this relationship, where V ₁ , V ₂ , V ₃ , V ₄
The filtration time required to obtain the amount of filtrated water is θ ₁ , θ ₂ ,
By measuring θ ₃ and θ ₄ , a relationship curve A between the two can be obtained. In the case of standard occlusion filtration, this relationship curve becomes a curve close to straight line B, which is an extension of the origin and _A1 , and the slope becomes smaller as the filtration mode becomes completely occlusive.
It becomes an upwardly convex curve. For this reason, a predetermined slope is used as a reference, and if the slope is smaller than that, it is occlusion filtration, and if it contains many particles with a particle size of, for example, 0.5 μm or more, and if it approximates the predetermined slope, it is intermediate occlusion filtration or standard occlusion filtration. It is determined that it contains many particles of 0.5 μm or less. In this case, since the reference slope is determined for each water system, it can be confirmed and determined experimentally. Although it is possible to judge changes in the slope of a curve by comparing the values obtained by differentiating the curve, it is also possible to judge changes in the slope of a curve by comparing the values obtained by differentiating _{the curve} , but it is also possible to judge changes in the slope of a curve by comparing the values obtained by differentiating the _curve .
A ₃ , B ₄ and A ₄ ) can also be substantially determined. In this case, the relational curve is not limited to a curve of θ versus V, but may be a curve of V versus θ. Also, this curve does not need to be drawn in reality;
Alternatively, the particle size may be determined using a numerical value that can be substantially equivalent to the slope of the curve. You can perform this operation once to determine the particle size based on a predetermined pore size, but if you perform this multiple times using screen-type membrane filters with different pore sizes, you can get a more specific particle size and its distribution. be able to. Since the number of particles mixed in a particular water system is approximately determined, it is also possible to determine the type of particles based on the size and distribution of the particles. Next, the number of particles can be calculated from the FI value and particle diameter obtained from the above two steps. That is, when particles of a specific particle size are mainly included, the FI value becomes a logarithmic function of the number of particles, and is expressed as a straight line on a semi-logarithmic graph. For this reason, a calibration curve is created in advance for each particle size for a specific aqueous system, the calibration curve to be applied is selected based on the particle size obtained in the second step, and the FI value obtained in the first step is applied to this curve. Determine the number of particles by checking against a calibration curve or formula. In this case, a calibration curve may be created for each specific particle size, but in a specific ultrapure water system, the infiltrating particles and their distribution are roughly fixed, and considering the purpose of monitoring, it is necessary to create a calibration curve for each specific particle size. Since it is not always necessary to accurately determine the number of particles in each case, standard materials with known particle sizes (for example, iron oxide as standard particles of 0.5 μm or less, microorganisms such as Escherichia coli as standard particles of 1 to 10 μm) are used. It is also possible to create a calibration curve using , and create each as a representative calibration curve for standard occlusion and complete occlusion. For example, a maximum particle size of 0.5 μm is cited as the water quality standard for ultrapure water for the electronics industry.
The objective can be achieved simply by determining the number of particles using a calibration curve of 0.5 μm or less. The above also applies to indicators other than the FI value, such as the SI value and the PI value. As described above, the degree of contamination of ultrapure water, particle size,
The number and type of particles can be determined, and ultrapure water production equipment, etc. can be monitored. FIG. 3 is a block diagram showing an example of a monitoring device. Reference numeral 11 denotes a pressurized water supply tank, which pressurizes ultrapure water to a constant pressure and sends it to an airtight filtration section 12. The airtight filtration section 12 is configured to perform airtight filtration by sandwiching a screen type membrane filter 14 supplied from a membrane supplier 13, and the filtrated water is sent to a filtration water tank 15. 16,
17 is a level sensor which sends a level signal to a counter 19 according to a command from a sequence section 18, and the counter 19 counts timing pulses from a time section 20 and sends a time signal to a calculation section 2.
Send to 1. In the calculation unit 21, calculations are performed by an FI calculation unit 22, a particle diameter calculation unit 23, and a particle number calculation unit 24, and the results are compared with a set value in a comparator 25 and output to an alarm 26 and a printer 27. . In the monitoring device configured as described above, sample water is pressurized to a constant pressure in the water supply pressure tank 11, filtered in the airtight filtration section 12, and sent to the filtered water tank 15. To measure FI, first, the time required to filter a predetermined amount of test water is measured at the initial stage of filtration. To do this, the level sensor 17 sends a signal when the water level in the filtered water tank 15 reaches a predetermined level, and the counter 19 sends a time signal T ₀ to the calculation unit 21 based on the timing pulses counted up to that point. and airtight filtration section 1
While the filtration in Step 2 continues, the filtered water in the filtered water tank 15 is discharged at some point, and after a predetermined period of time, the same operation is performed again to send the time signal T _o to the calculation section 21. In the calculation unit 21, the FI value is calculated in the FI calculation unit 22 according to the above formula. Measurement of particle diameter is performed in parallel with this FI measurement or separately. That is, from the start of filtration, the times θ ₁ , θ ₂ , θ ₃ , and θ ₄ required to filter each predetermined amount of V ₁ , V ₂ , V ₃ , and V ₄ shown in FIG. 2 are measured. For this purpose, a level sensor 1 installed in the water supply pressurized tank 11 is required.
6 generates a level signal corresponding to each of the predetermined amounts, and sends the time signal counted up to that point by the counter 19 to the calculation section 21. In the calculation unit 21, the particle size calculation unit 23 calculates the difference from the differential values of A ₂ , A ₃ , and A ₄ , or the deviations of A ₂ −B ₂ , A ₃ −B ₃ , and A ₄ −B ₄ , or the ratio of the deviations. The slope of curve A is determined, and the particle diameter is calculated by comparing this with a set value. The number of particles is calculated by the particle number calculation section 24. That is, by inputting the particle diameter, a corresponding formula is selected, and the number of particles is calculated by comparing the FI value with this formula. The FI value, particle diameter, and number of particles calculated above are compared with the set value in the comparator 25,
When the set value is exceeded, the alarm device 26 issues an alarm. Also, all data is stored on printer 2.
7 and printed out. In the above explanation, FI has been mainly explained as an index of the degree of contamination of test water, but other indexes such as PI, SI, etc. may also be used. Further, the measurement of the degree of contamination and the measurement of the particle diameter may be performed simultaneously or separately, and in this case, the order of measurement is not limited. Further, the measurement may be performed with the same device or with different devices. Furthermore, particle diameter measurement is essentially based on V and θ.
The determination may be made from the slope of the relationship curve between V and θ, and the determination may also be made from a factor that indirectly represents the slope of the relationship curve between V and θ, such as the above-mentioned time shift. Further, the particle diameter and the number of particles may be determined based on a relational curve, or may be based on a formula expressing it. As described above, ultrapure water is monitored, and when the FI value exceeds a predetermined value or when the number of particles with a predetermined particle size or more exceeds a predetermined number, an alarm is issued and countermeasures are taken. Ultrapure water production equipment used in the electronics and pharmaceutical industries is usually composed of a combination of precision filtration equipment, reverse osmosis membrane equipment, ion exchange equipment, sterilization equipment, polishing filters, etc. As long as it is properly managed, the ultrapure water at the outlet of the device will meet the above-mentioned water quality requirements, and many of the problems at the point of use are due to inadequate microbial control of piping, storage tanks, faucets, etc. up to the point of use. be. In other words, when these parts are carelessly exposed to the air, bacteria, mold, or their spores are introduced from the air, grow in the system, and contaminate the ultrapure water at the point of use. There are many. In such cases, the FI value,
Since it is measured as a number of particles, countermeasures can be taken immediately. The same applies if crushed resin from the polishing filter leaks, and the source of contamination can be searched for by particle size distribution, number of particles, etc. The present invention is applicable not only to ultrapure water used in the electronics industry and pharmaceutical industry, but also to monitoring ultrapure water for other uses. Example 1 Raw water was treated with a filtration device, a reverse osmosis device, a mixed bed ion exchange device, an ultraviolet sterilizer, a cartridge type mixed bed polisher, and a precision filtration (Millipore filter) to achieve a specific resistance of 8 to 15 MΩ・cm (25 ℃) ultrapure water was produced. The degree of contamination of this ultrapure water was measured using the monitoring device shown in FIG. 3 using a Nucleipore membrane (manufactured by Nucleipore), which is a screen-type membrane filter with a vertical hole structure having a pore diameter of 0.2 μm. That is, when ultrapure water was supplied from the pressurized water supply tank 11 at a constant pressure (2.1 Kg/cm ² ) to the airtight filtration section 12 using the membrane filter, the time T ₀ until the amount of filtered water reached 500 ml was determined. It was hot in 79 seconds. Next, water was passed under the same conditions for 15 minutes, and T ₁₅ was determined by the same operation after 15 minutes, and it was 98 seconds, and the degree of fouling (FI ₁₅ ) was 1.29. Comparative Example A cellulose-based membrane filter, which is a network structure filter with a nominal pore diameter of 0.1 μm, was used instead of the above-mentioned Nyclypore membrane, and T ₀ and T ₁₅ were measured under the same conditions as in Example 1, each being 141 seconds. ,
The time was 147 seconds, and the FI ₁₅ was 0.27. As can be seen from these results, in Example 1
There is a difference of 19 seconds between T ₁₅ and T ₀ , but in the comparative example, the time difference is only 6 seconds even though the pore size is 1/2.
This is in seconds, which is close to the error range, and the reason why this is considered as the degree of contamination is from the viewpoint of the use of ultrapure water.
There are problems with reliability. Example 2 The slope of the relationship curve between V and θ was determined using the same ultrapure water as in Example 1, and the number of particles was measured from this. In this case FI ₁₅ =1.29. First, measure the filtration time when the amount of filtrated water is 2, set the time corresponding to the value on the straight line (B in Figure 2) that is an extension of the line connecting the origin to the standard time, and then actually measure 3, 4 , 5, and the difference from the standard time was determined as a time difference. The results are shown in the table below. Note that the ratio of the difference is expressed as θ _o /θ _o-1 .

[Table] From the above results, it is determined that the ratio of the slope (difference ratio) of filtrate water amount 4 to the slope (difference ratio) of filtrate water amount 5 in the filtration time - filtrate water amount relationship curve is 1.86, and this value Since it is clearly larger than 1, it is judged that it approximately matches the characteristic expression of standard occlusive filtration. Therefore, it is determined that the particles contained in this ultrapure water are mainly particles with a particle size of 0.5 μm or less. Next, when the above conditions were compared with a previously prepared relationship curve between FI ₁₅ and the number of particles when the particles were mainly composed of particles with a particle size of 0.5 μm or less, the number of particles was calculated to be 45 particles/ml. On the other hand, when the number of membranes used in the above filtration was counted using an electron microscope, it was found to be 60 membranes/ml. Comparing the results of electron microscopy and the above example, it can be seen that the two are similar, indicating that the device of the present invention is fully applicable to monitoring ultrapure water and has high reliability. As described above, the present invention has the following effects. (1) Be able to roughly judge the size and type of fine particles in ultrapure water. (2) Even an unskilled person can calculate the approximate number of particles without making a big mistake. (3) By using membrane filters with different pore sizes, it is possible to estimate the abundance ratio based on the size of fine particles. (4) Required water quality regarding the number of fine particles in the electronics industry, pharmaceutical industry, etc. can be easily and quickly monitored. (5) Regarding the above measurements, results can be obtained in a short time and with simple operations without using expensive equipment, and automation is also possible. (6) If the contaminated contaminant particles are microorganisms, the contamination route can be searched based on the abundance ratio according to size.

[Brief explanation of drawings]

Fig. 1 is a schematic diagram showing a closed state of a through-vertical membrane filter, Fig. 2 is a graph showing a relationship curve between the amount of filtrated water and filtration time, and Fig. 3 is a block diagram showing an example of the monitoring device of the present invention. be. In the figure, 1 is a membrane, 3 is a through hole, 4 is a particle, 11 is a water supply pressure tank, 12 is an airtight filtration part, 15
18 is a sequence section, 19 is a counter, 22 is a calculation section, and 25 is a comparator.

Claims (1)

[Claims] 1. A device that measures the time required to filter a predetermined amount of sample water using a membrane filter with a through-vertical hole structure, and the time required to perform the same operation after a certain period of time, and the values obtained from this device. A first computing device for computing the clogging rate or clogging speed of the filter, a device for measuring the amount of filtration of sample water and changes in filtration time by a membrane filter having a vertical hole structure; a second calculation device that determines the particle diameter from the slope of the relationship curve between the two determined by the value, and a value obtained from the first calculation device and the particle diameter obtained from the second calculation device in advance. An ultrapure water monitoring device that includes a device that determines the number of particles by comparing it with a clogging rate or clogging rate vs. particle number relationship curve or formula created for each particle size.