JPH0561581B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention is directed to the treatment of ammonia nitrogen (hereinafter referred to as
This method relates to an analysis method for ammonia in water that measures the concentration (abbreviated as NH ₃ -N) online, and is especially easy to maintain and allows stable measurement over a long period of time, as well as the ability to measure low concentrations of NH ₃ -N. This invention relates to a method for analyzing ammonia in water, which can also be applied to ammonia in water.

[Conventional technology]

In biological denitrification processes, it is necessary to measure the concentration of _NH3 -N in the treated water to monitor the process. On the other hand, in water treatment plants, the importance of NH ₃ -N is known as an indicator of contamination of raw water. As described above, it has been pointed out in various fields that there is a need for a device capable of measuring NH ₃ -N concentration, especially an apparatus capable of automatic online analysis over a long period of time.

In order to meet these demands, the present applicant has proposed a gas purge type ammonia analyzer and is currently applying for a patent as Japanese Patent Application No. 59-281422.

The basic principle of this device is to make the sample water alkaline, bubble it with air, and measure the NH ₃ concentration contained in the exhaust gas, and apply the standard addition method to this.

FIG. 5 shows a configuration diagram of the analyzer according to the patent application. In FIG. 5, a certain amount of test water 1 is sampled by a sampling pump P1 and injected into a gas-liquid contact tank 2. During this water test 1, air is injected through the diffuser 3 to perform aeration. P2 is an air pump for supplying air,
4 is a valve that adjusts the air flow rate, and 5 is a back pressure valve. After sampling, sodium hydroxide is added to the test water 1 by the alkaline agent injection mechanism 6 to make the pH of the test water 1 11 or higher. NH ₃ −N in water is NH ₃
and NH ₄ ⁺ , and the relationship between them is expressed as NH ₃ +H ₂ ONH ₄ ⁺ +OH ^- (1). However, at pH > 11, the reaction shifts almost completely to the left. That is, NH ₃ −N
will exist as dissolved NH ₃ gas. If the above-mentioned aeration is performed in this state, NH ₃ in the water will be released into the rising bubbles, and the exhaust gas will contain
Contains _NH3 . Now, the concentration of NH ₃ −N in water is X _L ,
If the NH ₃ -N concentration in the exhaust gas is X _G , then X _G = αX _L (2). Here α is a proportionality constant. The exhaust gas passes through an electrode housing 8 into which a gas-permeable NH ₃ electrode 7 is inserted and is discharged into the atmosphere. Electrode 7 therefore produces an output E that corresponds to X _G . The relationship between X _G and E is E=E _{0 −Slo} X _G S=RT/nF (3). Here R is the gas constant, T is the absolute temperature, F
is the Faraday constant, n is the number of charges of ions involved in the reaction within the electrode, and E ₀ is E at X _G =1.
This equation is generally called the Nernst equation.
Substituting equation (2) into equation (3), we obtain E=E ₀ ′ _−Slo X _L E ₀ ′=E _{0 −Slo} nα (4). In other words, the relationship between X _L and E is similar to equation (3),
It can be expressed by the Nernst equation. Therefore, make sample water 1 alkaline and aerate it.
Equation (4) holds true when the electrode output reaches a steady value. The electrode output is input to the arithmetic unit 10 via the converter 9, and the arithmetic unit 10 holds the electrode output at this point in time. If this hold value is E ₁ , then E ₁ =E ₀ ′ _−Slo X _L (5). Next, the standard solution injection mechanism 11 is activated, and a fixed amount of NH ₄ Cl solution with a known concentration is injected. As a result, the NH ₃ -N concentration in the sample water 1 increases, and the electrode output also changes. If the change in NH ₃ −N concentration is ΔX _L and the power output after the change is E ₂ , then E ₂ =E ₀ ′ _−Slo ( _XL + ΔX _L ) (6). Subtracting equation (5) from equation (6) yields △E=E ₂ −E ₁ = _−Slo (1+△ _XL / _XL ) (7). ΔE and _ΔXL are known, and S can be determined by measuring the temperature inside the electrode housing 8 with the temperature sensor 12. Therefore, The desired _XL is measured. When the measurement is completed, the pinch valve 13 is opened, the test water 1 is discharged, and the next measurement is started.

The ammonia analyzer described above detects NH ₃ − in exhaust gas.
By measuring the N concentration, NH ₃ −N in water
To measure concentration, there is no contact between the sample water and the sensor.
Trouble caused by contamination of the electrode film can be prevented. Furthermore, changes in α due to changes in electrode characteristics, water quality changes, etc. appear as changes in E ₀ ', but the standard solution addition method allows measurement without being affected by changes in E ₀ '.

[Problem that the invention seeks to solve]

On the other hand, the above-mentioned NH ₃ analyzer has the following drawbacks.

First, as shown in equation (2), X _G is X _L
is proportional to , but X _G is lower than the equilibrium concentration for X _L . That is, the _NH3 partial pressure in the exhaust gas is lower than that in water. This means that when measuring near the lower limit of quantification with an NH ₃ electrode, the normal measurement method (measuring the NH ₃ partial pressure in water) involves inserting the electrode directly into the liquid.
disadvantageous compared to.

The second drawback is the problem of foaming in the gas-liquid contact tank due to uneven air. In particular, when measuring wastewater, bubbles are likely to occur, and the bubbles may accumulate on the liquid surface and reach the inside of the electrode housing 8. Furthermore, the foaming wets and contaminates the gas-liquid contact tank 2 and the inside of the piping. Since NH ₃ is easily absorbed in this contaminated area, the response of X _G to changes in X _L is significantly delayed, and eventually becomes unmeasurable.

A third drawback relates to electrode output drift compensation. Drifts in electrode output are compensated by the standard addition method, but effects appear for sudden drifts. A particular problem is the effect of daily fluctuations in temperature. Temperature compensation is basically possible by measuring the temperature of the electrode, but in reality it takes time for the temperature inside the electrode to become uniform, so it is difficult to completely eliminate the effects of temperature changes. .

Therefore, the present invention was made to solve the above-mentioned drawbacks of the conventional ammonia analyzer.The purpose of the present invention is to provide easy maintenance and stable measurement over a long period of time, as well as to make it possible to measure low concentrations of NH. An object of the present invention is to provide an on-line method for analyzing ammonia in water that can be applied to the measurement of ₃ -N concentration and can also minimize the effects of environmental changes such as temperature changes.

[Means for solving problems]

In order to achieve the aforementioned object, according to the present invention:
A standard solution with a known ammonia nitrogen concentration is mixed into the test water at an arbitrary mixing rate, and a test liquid made by mixing an alkaline agent into the mixed water is continuously introduced,
An analysis comprising a measurement cell configured to form a gas phase portion of a fixed volume between the liquid to be measured and an ammonia electrode attached to the measurement cell so as to be in contact with the gas phase portion. Using a device, while changing the mixing rate of the standard solution at a constant cycle, the moving average of the electrode output over a period of time corresponding to an integer multiple of the mixing rate change period and the past period corresponding to 1/2 of the moving average time are calculated. The method is characterized in that the ammonia nitrogen concentration in the liquid to be measured (sample water) is determined using the power output value of .

The reason why the ammonia in water analyzer according to the above-mentioned prior application uses the aeration method is based on the idea that it takes a long time to bring NH ₃ into an equilibrium state between gas and liquid. That is, in the aeration method, the NH ₃ -N concentration in the exhaust gas does not reach an equilibrium concentration with respect to the NH ₃ -N concentration in water, but the proportional relationship of equation (2) holds true. Since α is not an equilibrium relationship, α changes depending on water quality, bubble diameter, etc., but the idea is that it is sufficient to be able to measure it in a manner that does not depend on fluctuations in α using the standard addition method.

On the other hand, as a result of repeated experiments and research, the present inventors found that in the case of _NH3 , it is possible to easily achieve gas-liquid equilibrium without using means to increase the gas-liquid contact efficiency such as aeration or strong mechanical stirring. I found that I could reach it. The present invention is based on such knowledge, and will be explained in detail below with reference to Examples.

FIG. 1 is a block diagram of an apparatus used in the present invention, and FIG. 2 is an enlarged sectional view of the measurement cell in FIG. 1. In FIG. 1, solid lines represent the flow of material, and broken lines represent the flow of signals. Test water is pumped up by a water sampling pump P1 and introduced into a water sampling tank 21 installed outside the analyzer rocker. A small amount of test water is introduced into the rocker and enters the measurement cell 22 through the pinch valve V1 and tube pump P3. During this process, NH ₄ Cl standard solution and NaOH are mixed into the test water. NaOH is injected from the alkali tank 23 into the test water channel on the discharge side of the tube pump P3 by the metering pump P4. The injection flow rate is determined so that the pH of the liquid to be measured becomes 11 or higher. In this way, a liquid to be measured containing test water and an alkaline agent is continuously introduced into the measurement cell 22.

On the other hand, A in Figure 1 is the standard solution mixing mechanism,
The standard solution is transferred from the standard solution tank 24 to the pinch valve V2.
and enters the suction flow path of the tube tank P3.

The liquid to be measured is introduced into the measurement cell 22 from the liquid inlet 31 shown in the second diagram, passes through the liquid reservoir 32, and flows out from the liquid outlet 33. The liquid to be measured in the liquid reservoir 32 is slowly stirred by a stirring magnet 35 that rotates by magnetic coupling with a magnet 34 externally.

Further, in FIG. 2, reference numeral 25 is an ammonia electrode, for example, a diaphragm type ammonia electrode, which is installed in the measurement cell 22 so as to be in contact with a gas phase portion 42, which will be described later. For example, as shown in the second figure, it is inserted into an electrode holder 36 made of stainless steel and supported by a spacer 37 made of fluororesin. The electrode holder 36 is heated by a band heater 38, and the temperature of the electrode holder 36 is kept higher than the temperature of the liquid to be measured using a platinum resistance thermometer 39 and a temperature controller not shown in this figure. It's starting to be preserved. This is to prevent water vapor on the inner surface of the spacer 37 and the electrode measurement surface 40 from condensing and absorbing NH ₃ and to reduce measurement errors due to temperature drift.

Inside the measurement cell 22, a gas phase portion 41 with a constant volume is formed by the water surface of the liquid retention portion 32, the spacer 37, and the electrode measurement surface 40. When the liquid to be measured flows through this measurement cell 22, the gas phase part 41
NH ₃ −N concentration X _G is NH ₃ −N concentration X _L in the liquid to be measured
It becomes an equilibrium state. That is, X _G =1/HX _L (9). Here H is Henry's constant. Therefore. The output E of the ammonia electrode 25 is E=E ₀ ″ _−Slo X _L E ₀ ″=E _{0 −Slo} H (10), and this signal is input to the arithmetic unit 26 in FIG. The arithmetic unit 26 generates on/off signals for alternately opening the pinch valves V1 and V2, and adjusts the on/off ratio to adjust the standard solution mixing rate r.
Adjust.

In this example, the mixing rate r is set in two stages (γ ₁ and γ ₂ ,
γ ₁ <γ ₂ ) periodically (period T _n =1 hour). Here, r ₁ is determined so that the NH ₃ --N concentration of the liquid to be measured is equal to or higher than the measurement limit of the ammonia electrode.
Therefore, if the NH ₃ -N concentration of the sample water is above the measurement limit, γ ₁ =0. In the following case, we will proceed with the discussion assuming that γ ₁ ≠ 0, but it goes without saying that the case of γ ₁ =0 can also be handled in the same way by setting γ ₁ in the formula to zero.

As γ changes, the electrode output E also changes periodically, and this is used to perform measurements. The problem is that changes in E include fluctuations not due to changes in γ but also due to changes in environmental conditions (especially temperature).

Therefore, in order to solve this problem, the present inventors focused on the frequency characteristics of the moving average.

Expressing the moving average of time interval L as a transfer function, G _MAL (S)=1−G _L (S)/SL (11) Here, G _L (S) is the transfer function of the dead time element of the delay time L. , and G _L (S)=e ^-SL (12). Note that S is a Laplace transform parameter.

When S=jw is substituted into equation (11) to study the frequency characteristics, G _MAL (jw)=1−e ^−jwL /jwL (13) is obtained. Here w is the angular frequency and j=-√1. Transforming equation (13) using e ^-jwL = cos wL−jsin wL, G _MAL (jw) = 1/wL[sin wl−j(1−cos wL] (14), and further, sin wL= 2sin(wL/2)・cos(wL/2) 1−cos wL=2sin ² (wL/2), so G _MAL (jw) =2/wLsinwL/2(coswL/2−jsinwL/2) = 2/wLsinwL/2・e ^-jw(L/2) (15) Also, when sin(wL/2) is expanded into a series, equation (15) becomes G _MAL (jw) = [1-1/3〓 (wL/2) ² +1/5〓(wL/2) ⁴
−…] ^{-j w(L/2)} = [1-1/3〓(wL/T) ² + 1/5〓(πL/T) ⁴ −…〕e ^-jw(L/2) (16) It can be expressed as Here, T is the period of the signal. As can be seen from these, the signal has a long period compared to the average time, that is, (1/3〓)
(πL/T) ² <<1, the second term and the following in equation (16) can be ignored, and G _MAL (jw)e ^-jw(L/2) = G _L/2 (jw ) (17) becomes. In other words, the moving average is 1/2 of the average time for signals with a longer period than the average time.
It also acts as a wasted time element. Also, equation (15)
From this, |G _MAL (jw)|=T/πL|sinπL/T| (18) When expressed graphically, the gain curve of the moving average element is obtained as shown in FIG. It can be seen from FIG. 3 that the gain of the moving average decreases rapidly at L/T1, and in particular, the signal of L/T=n (n=1, 2, . . . ) completely disappears due to the moving average operation.

The present invention utilizes the above-mentioned properties of the moving average.
The cycle of the main frequency components of output fluctuations due to temperature changes is 24 hours, but shorter-period signal components may be superimposed on fluctuations that occur during sudden changes in sunlight. Now, the electrode output E is expressed as the sum of the background E _B and the fluctuation component ε (period = T _n = 1 hour) due to the change in the mixing rate, E = E _B + ε (19), where E _B has a 6-hour period fluctuation. Assume that the ingredients are included. When E is passed through a moving average element with an average time of 1 hour, the output E becomes E=G _MAL (S)E=G _MAL (S)E _B +G _MAL (S)ε (20), but 1/ 3〓(π×1/6) ² =0.046≪1(21), and since the average time is equal to the period of ε, G _MAL ε=0 (22), so EG _L (S)E _B (23 ) becomes. In other words, the output of the moving average is only E _B ,
This results in a signal approximately equal to that passed through a dead time element of delay time T _n /2. On the other hand, if E is actually passed through a dead time element of delay time T _n /2, the output E _L
Since E _L = G _L (S)E _B + G _L (S)ε (24), only the fluctuation component due to the change in the mixing rate can be obtained by E ^* = E _L −G _L (S)ε (25). become.

Considering the steady-state value of the signal E ^* obtained in this way, E ^* = E ₀ ″−−Sl _{o XL} (26), and = ₀ ″− _{o L} E ₀ ″−S _{o L} (27 ) Therefore, E ^* = S _{o L} − S l _{o XL} (28) Therefore, E ^* is not affected by changes in E ₀ ″ due to environmental changes.

The principle of drift compensation in the present invention is as described above, and the final output is obtained by processing the signal E ^* .

The contents of the calculations until the final output is obtained will be explained below with reference to the block diagram of FIG. In the figure, 51 is a moving average element, and 52 is a dead time element, the functions of which have already been described. Then the signal
E ^* is input to the divider 53 to obtain -E ^* /S. From equation (28), −E ^* /S = − _{o L} + l _o X _L (29) and through the exponential function element 54, y = e ^-E*/S = e−l _{o X L} 30) becomes. _{o L} does not include periodic fluctuations due to changes in the mixing rate r, but changes with changes in the sample water NH ₃ -H concentration. However, since this change is gradual compared to the change in _XL , it can be regarded as y= _αXL and α=constant (31).

Now, assuming that γ=γ ₁ and the value when _XL reaches a steady value is _XL1 , then _XL1 = (1−γ ₁ )X ₀ +γ ₁ X _S (32). Here, X ₀ is the NH ₃ −N concentration of the sample water, X _S
is the NH ₃ −N concentration of the standard solution. Moreover, the value of y at this time is y=y ₁ =αX _L1 (33). Next, the steady value X _L2 of X _L at γ=γ ₂ is X _L2 = (1−γ ₁ −Δγ)X ₀ +(γ ₁ +Δγ)X _S (34). Here, Δγ=γ ₂ −γ ₁ . Also, y=y ₂ =αX _L2 (35). From equation ( ₃₂ ), we get X _{0 = X L1 −γ 1 L1} + (1+γ/1-γ ₁ )△γX _S (37). In FIG. 4, the sample hold element 55 calculates y when γ=γ ₁ .
6 holds y when γ=γ ₂ .
That is, the respective outputs are y ₁ and y ₂ .
Next, the divider 57 calculates y ₂ /y ₁ . Letting this be R, R=y ₂ /y ₁ =X _L2 /X _L1 = (1-△γ/1-γ ₁ ) + (1+γ/1-γ ₁ )X _S /X _L1 △r(38 ), rearranging this, X _L1 = 1/(1-γ ₁ ) (R-1) + ΔγX _S (39), and X _L1 can be found from Δγ, γ ₁ , R, and X _S. Furthermore, the NH ₃ −N concentration X ₀ of the sample water is determined using equation (36). In FIG. 4, reference numeral 58 denotes an arithmetic operation section that calculates X ₀ by calculating equations (39) and (36).

[Effect]

As described above, in the present invention, when analyzing ammonia in water, the ammonia nitrogen concentration in the sample water is determined using the moving average output and the dead time element. The output of the moving average becomes a signal that is almost the same as passing only E _B through a dead time element with a delay time T _n /2, whereas if E is actually passed through a dead time element with a delay time T _n /2. , only the fluctuation component due to the change in the mixing rate can be obtained by Equation (25).

Considering the steady-state value of the signal E ^* obtained in this way, E ^* becomes Equation (28), so it is not affected by changes in E ₀ ″ due to environmental changes. Therefore, according to the present invention, By using the dead time element and the moving average, drift in the electrode output can be efficiently removed, and therefore the influence of environmental changes such as temperature changes can be minimized.

Furthermore, the present invention does not use aeration or rapid stirring, but uses the moving average of the electrode output and the dead time element while changing the mixing rate at a constant cycle, so there is no contamination of the measurement cell due to foaming, and the measurement cell is not contaminated in the low concentration region. Ammonia analysis is possible.

〔Effect of the invention〕

While conventional NH ₃ analyzers utilize the proportional relationship between the NH ₃ -N concentration in the exhaust gas and the NH ₃ -N concentration in water, that is, equation (2),
In the present invention, equation (9) of vapor-liquid equilibrium is used. Therefore, since α>1/H, the latter X _G is larger than the former X _G for the same X _L. Furthermore, in the present invention, by setting γ ₁ ≠0, it has become possible to measure low concentration regions that were impossible to measure with conventional ammonia electrodes.

Further, in the present invention, since no aeration or rapid stirring is used, there is little contamination of the inner surface of the measurement cell due to foaming or splashing, and there is almost no delay in response due to absorption of NH ₃ .

Furthermore, since the present invention can efficiently remove drift in electrode output by using a dead time element and a moving average, the influence of environmental changes such as temperature changes can be minimized.

[Brief explanation of the drawing]

FIG. 1 shows a configuration diagram of a specific example of the device used in the present invention, FIG. 2 shows an enlarged sectional view of the measurement cell in FIG. 1, and FIG. 3 shows a frequency specific curve of a moving average element. , FIG. 4 shows a block diagram representing the content of measurement calculations, and FIG. 5 shows a configuration diagram of a conventional apparatus. 22...Measurement cell, 23...Alkaline tank,
24... Standard solution tank, 25... Ammonia electrode, 26... Arithmetic device, 31... Liquid inlet, 33
...Liquid outlet, 34...External magnet, 35...
...Stirring magnet, 40... Electrode measurement surface, 41...
... Gas phase part, 51 ... Moving average element, 52 ... Dead time element, 53, 57 ... Divider, 54 ... Exponential function element, 55 ... Sample hold element, 5
6...Sample holding element, 58...Four arithmetic operation section, A...Standard solution mixing mechanism.

Claims (1)

[Claims] 1. A standard solution with a known ammonia nitrogen concentration is mixed into the test water at an arbitrary mixing rate, and a test solution made by mixing an alkaline agent into the mixed water is continuously introduced, and the distance between the test solution and the test solution is continuously introduced. The standard solution is While changing the mixing rate at a constant cycle, the moving average of the electrode output over a period of time corresponding to an integral multiple of the mixing rate change period and the past electrode output value for a period corresponding to 1/2 of the moving average time are used. A method for analyzing ammonia in water, which is characterized by determining the ammonia nitrogen concentration in a liquid to be measured.