NO156105B - PROCEDURE FOR AA MEASURING THE CONTENT OF A SUBSTANCE IN A MIXTURE OF MULTIPLE SUBSTANCES, WHICH MIXING HAS THE FORM OF A THIN MOVIE, EXAMPLE FOR AA MEASURING THE WATER CONTENT IN PAPER. - Google Patents

Description

The invention relates to a method for measuring the content of a substance in a mixture of several substances (materials) and which is of the type stated in the introduction to patent claim 1.

In German interpretation document no. 20 55 708, it is e.g. described a procedure where infrared radiation is used to measure the thickness of a thin transparent film using a so-called two-beam measurement according to the reflection principle. In accordance with this known measurement method, a two-beam infrared radiation source is used, in addition to a radiation detector and a signal amplifier and analyzer circuit, where the radiation source produces two separate beam bundles within the infrared range, X^ and as one wavelength (the measurement wavelength) X^ is chosen in such a way that, in relation to the film material, it exhibits greater absorption than the so-called relative wavelength X2 (comparison wavelength).

In addition to plastic, which is mentioned in DE-AS 20 55 708, it is also previously known for water and other substances that the properties of these substances, when it comes to the influence of infrared radiation, are highly dependent on the infrared radiation's wavelength. As it must be considered generally known that the optically anomalous properties of the materials in question in the infrared radiation range are due to the molecular structure of these materials, it seems superfluous to go into more detail on this subject in this context.

When it comes to water, it is also a fact that the strongest absorption of radiation energy takes place in the radiation range that is close to the infrared, namely in the so-called valence fluctuation range at X = 2.92y. Consequently, this wavelength should be well suited for absorption measurement in connection with very small amounts of water or very thin layers of water. To measure the water content of organic substances, e.g. paper, however, the considerably weaker combined oscillation at X - 1.94y must be used. The reason for this is that all carbohydrates

in the vicinity of X = 2.9u has a resonance point for the CH bond and that the measurement size will therefore depend on the carrier material. Under practical conditions, however, it turns out that also the use of the significantly weaker combination oscillation at X = 1.94y provides suitable grounds for drawing the necessary conclusions with regard to water content in a material on the basis of the material's absorption ratio.

In order to measure the water content of paper and similar materials, where the water is mixed in a finer distribution with the other components of the material in question, the spectral comparison method (two-beam measurement) has thus far been used, namely both as a transmission method and as a re -flex ion method in accordance with DE-AS 20 55 708. In this measurement method, radiation powder with the strongly absorbed measurement wavelength and the weakly absorbed comparison wavelength is produced alternately with the aid of an optical vibration device. These radiation pulses penetrate the actual material, after which they are picked up by a detector and transformed into measurement and comparison signals, respectively. After common amplification, the measurement signals are separated from the comparison signals. A quotient measuring device registers there^ according to the ratio between the signals and possibly indicates this. Provided that the comparison wavelengths in the spectrum are not too far from the measurement wavelengths, this measurement method will in principle provide the following advantages compared to an in and of itself usable measurement with exclusively measurement wavelength radiation:

Any distance change between radiation source and counter-

1'

taker, or measurement object which is due to the measurement object's movement in relation to the measurement location, is eliminated with regard to the measurement result;

any contaminants found in dispersed form in the water in the paper or in similar material have approximately the same effect on the radiation losses and therefore affect the produced quotients to a lesser extent;

any variations in the detector's sensitivity due primarily to temperature fluctuations become smaller; and any variations in the gain factor in the measuring amplifier are also eliminated.

For example, a wavelength A = 1.8y can be used as a comparison wavelength when measuring the water content in paper. With this short distance to the measurement wavelength A = 1.94u, namely of the order of magnitude around AA = 0.14y, the measurement wavelength and the comparison wavelength are sufficiently close to each other to guarantee that the aforementioned disturbances are reduced at least to a significant extent.

It is also previously known that cellulose, which occurs in considerable quantities in paper, exhibits strong absorption within a wavelength range of A = 2.ly.

In an article, which appears in MELLIAND, booklet 10/. 1958, pages 1157-1160, it can be read that when it comes to measuring the water content in textiles, completely different measured values can be obtained for the same water content, while, as is known, uniform measured values for the water content are always obtained in connection with translucent materials, especially transparent cellulose (cell glass). The variable measurement values for textiles are based on a so-called "medium penetration" or on variable "propagation path length improvement" of the infrared radiation.

In addition, it appears from the aforementioned article in MELLIAND that attempts have already been made in the past to mathematically characterize materials of the type opal celluloses (paper) or textiles, which compared to transparent cellulose have structurally complicated properties, in such a way that satisfactory measurement results can also be obtained for these materials. However, these attempts have so far not been successful.

As before, there remains e.g. when measuring the water content in paper, there is still the problem that the measured value for the water content is subject to change even if the water content in the paper is unchanged, namely when physical or structural characteristics of other substances in the paper, especially the cellulose but also fillers, e.g. titanium dioxide, changes during a manufacturing process.

The invention aims to ensure satisfactory measurement results when measuring the content of a substance in a thin material web or in a thin material film, where a measurement method based on infrared radiation is used, also when the physical or structural characteristics of the relevant material's other substances, especially cellulose in the case of paper, changes during the manufacturing process.

This is achieved by proceeding in the manner indicated in the characterizing part of patent claim 1.

For example, when measuring the water content of paper, it is possible by means of the measurement method according to the invention to obtain measurement results that are independent of physical or structural characteristics or changes in such characteristics of other substances in the paper. The design or changes in the design for cellulose fibers and/or filler particles no longer give false measurement values.

The method according to the invention takes its starting point in the experimentally established fact that the depth of the absorption bands in the material to be measured and in a further substance, e.g. water and cellulose, changes proportionally with the surface weight of the material, respectively the cellulose, and especially with the physical or structural properties of the material's constituents. This deviation can be explained on the basis of the theory that the length of the path that the infrared radiation must cover between its entry into and exit from the material depends on the material's structure and its surface weight. In the least common cases, the radiation's path through a material corresponds approximately to the radiation's path through an equally thick, almost glass-clear medium, for example cellular glass. The longer the path that the infrared radiation has to travel, at the same material thickness straight through the material, the greater the absorption band for a material with often repeated deflection of the radiation towards small particles, such as fibres, pigments and the like , considerably more pronounced (deeper) than for a material which does indeed contain the same amount of the substance in question, but which, in optical terms, still allows the radiation to pass unhindered, as is the case with cellular glass. A more exact description of this effect is hardly possible due to the complicated and highly variable shape of the radiation-refracting and/or dispersing particles; for example in the case of paper, the shape of the cellulose fibers which are surrounded by water.

The method according to the invention can of course be advantageously used not only for measuring the water content of paper, but in general for all measurements that are carried out on thin material films or webs where the content of a substance in the material is to be measured, and where the substance is mixed with other substances in the material, as well as where a further substance, which can give rise to misleading measurement values, similar to the substance to be measured, has a distinct absorption band.

In the aforementioned two-beam measurement method, the measurement values at the measurement wavelength are set in relation to a comparison value. This comparison value can be obtained from a separate infrared radiation measurement, which has also been carried out at the measurement wavelength, but with the material removed from the measurement gap between radiator and detector, i.e. without material. With continuous measuring methods, however, it is not possible to remove the material to be measured from the measuring gap. Therefore, the comparison value is formed by a measurement value using a comparison wavelength close in the spectrum, because such a measurement value in many cases (e.g. measurement carried out on cell glass) hardly deviates from the measurement value that would have been obtained when the material had been removed from the measuring column.

With a problem solution in accordance with patent claim 1, the two measuring wavelengths and A3 are set equal in relation to a comparison wavelength A2, which is located in a spectral neighborhood of the two measuring wavelengths X-^ and X3.

To utilize the procedure in the manner indicated

in patent claim 1, however, when measurement is carried out on materials, which exert strong dispersion of infrared radiation,

to misleading results. As is known, the radiation losses due to dispersion vary with the fourth power of the wavelength. Spectrograms of such infrared radiation in the case of dispersive materials are therefore, depending on the dispersion intensity of the radiation, more or less strongly deflected just above a spectrogram for a material with the same chemical and quantitative substance composition, but nevertheless a material which optically practically does not affect the radiation,

such as cell glass. When the measurement value obtained at the comparison wavelength is used as a constant reference value, respectively to form a baseline, which is dependent on the wavelength, for the measurements at the measurement wavelength, the result must be misleading measurement results in such cases.

In order to achieve satisfactory measurement results when measuring on materials that disperse infrared radiation, the method according to patent claim 1 has been further developed as stated in patent claim 2.

When applying the problem solution according to patent claim 2, reference values are calculated which are adapted to the measuring wavelengths X^ and A^. The calculation of these reference values takes place on the basis of an extrapolation of an area of the function (X) which lies in the vicinity of the measurement wavelengths X^ and X^ and includes the comparison wavelengths X-, and X^ and is located within the area of the wavelengths and These reference values thus does not represent any measurement values, but serves to accurately indicate the depth of the absorption bands in materials containing substances with dispersing properties.

By means of the procedure as stated in patent claims 1 and 2, a hitherto unachieved precision in the measurement of the material proportions in a material web or film is made possible, as it primarily makes possible a measurement that is independent of physical and structural varia--

tions in the substance in the material.

The drawings, fig. 1-9, further illustrates the method in accordance with the invention. Fig. 1 schematically shows a measuring device that can be used for carrying out the procedure, fig. 2 shows a diagram of radiation pulses as they are continuously received by the measuring device according to fig. 1, fig. 3 shows a diagram (spectrogram) where the measurement signal emitted by the detector is indicated idealized for a specific material, depending on the wavelength,

fig. 4 shows the diagram in fig. 3 together with further curves for other materials on a larger scale, fig. 5-7 show, roughly schematically, the infrared radiation course in three separate materials which are adapted to the spectrogram according to fig. 4. Fig. 8 shows, roughly schematically, the infrared radiation course in a further material,

and fig. 9 shows a spectrogram for the materials according to fig. 5, 6 and 8.

In fig. 1 shows a transmitter 1 for infrared radiation. In front of the transmitter's radiation output 2, a filter wheel 3 rotates in a known manner. In the filter wheel 3, filters 4,5,6,7 are distributed around the circumference of the filter wheel 3. In the radiation direction of the transmitter 1, a detector 8 is placed at a certain distance from the transmitter. The detector 8 transforms the radiation from the transmitter 1 (in the form of radiation pulses) into electrical measurement signals I. Between the filter wheel 3 and the detector 8 there is a relatively thin, film-like or web-shaped material 9. Like the rotation of the filter wheel 3, the radiation pulses 11 are adapted for different wavelengths corresponding to the filters 4-7 which follow one another.

After being influenced by the material 9, more or less weakened radiation pulses 12 arrive at the detector 8. The radiation pulses 12, which have been received by the detector 8 and are adapted for the used wavelength, are transformed into electrical measurement signals I (see also fig. 2) and transmitted for further processing into a not shown, per se known electrical calculation or calculating device. The measurement signals I, which are emitted by the detector 8, can have the same shape and size as the measurement signals 14,15,16 or 17, which are

illustrated in the diagram according to fig. 2.

Fig. 3 shows a diagram (spectrogram), where a measurement signal I is emitted from the detector 8 for a material A

and produced in dimension [A] depending on the wavelength in the infrared radiation dimension [u]. As regards the influence on the infrared radiation, the material A corresponds to the previously known cell glass and affects infrared radiation, which penetrates through the material A, neither deflected nor dispersed on the way through the material to the detector 8. The production of the measurement signal I dependence on the wavelength X in fig. 3 is idealized. In reality, the spectrogram of most materials does not show such extended rectilinear areas as the curve according to fig. 3, as these areas, which are depicted as rectilinear, in reality mostly have a more or less strongly curved curve shape. By choosing an appropriate benchmark, e.g. such a scale where the wavelength X is not applied to the abscissa but rather linearly.

It is clear from fig. 3 that, in addition to a largely rectilinear and approximately horizontal part of the curve shown, there are also areas with relatively deep incisions. These areas correspond to different absorption bands of the material A. At the wavelength X = 2.92y has. free water so-called valence fluctuation. Free water also exists at the wavelength X = 1.94y. a combination pigging. The absorption band, which is a result of this combination squinting, is used for measuring the water content in the material A, namely by means of a known in and of itself, on the basis of a two-beam measuring method with empirically determined correction of the adjustment of the received measurement signals obtained by surface weight and structure.

The spectrogram according to fig. 3 also shows a further absorption band which is located at a wavelength of X 2.10y and which is coordinated with the cellulose in material no. an A.

Fig. 4 shows a spectrogram, where the spectrogram according to fig. 3 in the area outlined with a dash-dotted line is reproduced on a larger scale. Fig. 4 also shows spectrographic curve progressions for additional materials B and C. While material B affects infrared radiation approximately as shown in fig. 6 in rough schematics, the material C affects infrared radiation in approximately the same way as shown in fig. 7. Compared with the course of the infrared radiation according to fig. 5, the length of the path that the infrared radiation must travel through the material to the detector increases significantly for materials B and C; this despite the fact that, with regard to materials B and C, these are those where the weight proportions of water and cellulose per unit area (surface weight) agree with material A - This is clear from the spectrograms in fig. 4 that the depth of the absorption bands at the wavelengths X-^ and X 3 for the materials A, B and C is very variable and is all the more pronounced the longer the infrared radiation has to pass when it penetrates the materials A, B and C to arrive at the detector 8. The measurement signals IX^ and IX3 also vary in

corresponding degree at the measurement wavelengths X^ and X^. As, however, the measurement signals IX2 at the comparison wavelength X2 for the three materials A, B and C practically hardly change (for easier understanding of this relationship, the rectilinear curve areas are drawn with a greater distance between them than the real ones), consequently a measurement of the water content in the materials must A, B and C on the basis of the previously known two-beam measurement lead to different measurement results despite the same water content in the relevant materials. This will appear even more clearly from the following explanation.

When determining the water content of paper and, for example, when determining a specific filler or additive in a plastic film with a separate absorption band in the infrared radiation range, the following mathematical equation is used in principle:

I. this equation 1 means I the strength of the measurement signal "i which is emitted by the detector 8 in, for example, the dimension [A] after the influence of the applied infrared radiation with the measurement wavelength (i.e. within an absorption band range) through the substance in the material to be measured. I means the measurement signal emitted by the detector 8 in the same dimension at M=Q; a means the absorption coefficient rate for the substance in question in the material at the measurement wavelength chosen for the infrared radiation, and M means the surface weight of the substance in the material.

As when measuring the water content in the paper in accordance with the spectrograms according to fig. 3 and 4, infrared radiation with the (measurement) wavelength X is used, while the (comparison) wavelength X^ is used to create a comparison value for material in the measurement gap, equation 1 is completed according to the following by introducing a corresponding index for the relevant wavelengths, so that the equation becomes:

In equation 2, the measurement signal IX2, which is received by the detector 8 at the comparison wavelength X2, corresponds to

in practice the measurement signal I according to equation 1. The absorption coefficient a is of course coordinated with the measurement wavelength X^ In the event of another possible measurement where a different suitable wavelength was used, another absorption coefficient f is.ient will have to be added, namely one which is coordinated with the aforementioned other measurement wavelength, as the basis for equations 1 and 2 respectively.

Equation 2 can be solved with regard to M, which stands for the surface weight of the substance in question in the material to be measured. This gives the following mathematical expression:

By introducing: in equation 3, the following formula is obtained:

where En denotes the natural extinction for the water's current absorption band, M^. denotes the surface weight of the water (water content) and aw the absorption coefficient for water.

While in the case of material A a correct measurement result can be obtained on the basis of equation 3, respectively 5, the measurement result in the cases of material B and material C still does not correspond to the actual values. Namely, while the measurement result M for material A (cellular glass) corresponds to a measurement result carried out gravimetrically in a laboratory and can therefore be considered exact, the measurement results M^ for materials B and C will lie more or less high above the measurement result for material A, although it in reality the same amount of water was present in the three cases. This is readily apparent from the spectrograms for the materials A, B and C in fig. 4.

By modifying equation 5, the following formula is obtained for materials of type B, C and similar:

In formula 6, M denotes the measurement result (which is erroneous) for the water content (surface weight of the water), Ew 1 denotes the extinction which is increased due to the physical or structural characteristics of the substance in question and aw is the absorption coefficient for water.

The aforementioned incorrect measurement result M^ for the water content in the materials B, C and similar material is due to the measurement signal IX^ changing at the measurement wavelength \ . This change is a result of those in relation to the material A i. r-following fig. 5 changed "conditions for the passage of infrared radiation into and through materials of type B and C, respectively.

Experiments have shown that the changed natural effect, the injection E, not only refers to the absorption band for water, but also applies to the absorption band for cellulose which in the individual case is found in the materials B, C and the like, as E is proportional to the latter absorption band.

By focusing on the measurement and calculation of the cellulose content in the relevant material (the cellulose weight), equation 5, respectively 6, when the corresponding index is introduced, takes on the following form:

where M denotes the correct and M"^ the incorrect surface weight for the cellulose content of the material, which is determined in the individual case, E denotes the normal natural extinction, while E- denotes the natural extinction which has been changed as a result of a special physical or structural structure of the material, and where ac denotes the absorption coefficient for cellulose at the used measurement wavelength A^.

Due to the aforementioned proportionality between the variations in the measurement results M^. respectively M^. for water and M respectively. for cellulose, the following equation applies:

Equation 9 is solved with respect to E , which gives:

By substitution with equations 5 and 7, equation 10 gives the following expression:

Instead of the above-mentioned formula 11, the expression can be written as follows:

While the equations Ew and Ec in the above equation 11

is previously known on the basis of the received measurement signals IX^, IXj and IX2, and then also the absorption coefficient

a^ for water and ac for cellulose are known, the surface weight Mc of the cellulose is initially unknown. The surface weight Mc of the cellulose is determined, as already stated at the outset, by means of a specially carried out measurement, in particular by means of a 3-radiation measurement known per se. In such a 3-radiation measurement, in addition to the surface weight of the cellulose, the content of water in the material in question is admittedly included, but due to the small percentage of water in the total surface weight of the material, the resulting error can still be neglected. By using mathematical repetition, especially with the aid of a computer, this in itself already insignificant error can be made as small as you like, practically down to zero. Otherwise, this measurement is not affected by peculiarities of the material's physical or structural structure.

The surface weight of the water, according to equation 11, in a material, which is to be measured, is consequently independent of the relevant material's structural structure or changes in it, and is likewise independent of the material's surface weight, or of its surface weight variation. The calculated surface weight of the water therefore corresponds to the water's actual surface weight.

On the basis of equation 11, it is now also possible to calculate the actual water content in material B, C or similar material, after infrared radiation measurement has been carried out on materials of type B, C etc. within the range of the absorption bands for water (X^) and for cellulose (X3), and also at a resonance wavelength (X2) on the basis of the measurement signals received in each individual case, respectively IX^, IX3 or IX2.

Fibres, filler pigments, etc. which is found in the RNA material to be measured, in addition to a greater or lesser extension of the radiation's path through the material, also entails a more or less strong scattering of the infrared radiation that is used for the measurements. The radiation losses that occur in this connection create further problems with the measurement. Fig. 8 shows in a roughly schematic representation the course of the infrared radiation in a material D. The radiation through the material D to the detector 8 roughly corresponds to the course of radiation through the material C according to fig. 7. In addition, a relatively strong dispersion of the infrared radiation towards fibres, filler pigments etc. occurs in the material D, with accompanying radiation loss due to dispersion. Fig. 9 also shows a spectrogram for the material A according to fig. 5 also a spectrogram for the material D according to fig. 8. It appears here that the spectrogram area, which in the spectrograms is presented as a straight line, in the vicinity of the absorption ion bands at X^ and X^ for the chemically and quantitatively similar materials A and D, in case D is bent in relation to to the corresponding area for material A. The deflection in this spectrogram area for material D originates from the aforementioned losses in infrared radiation due to dispersion in material D (fig. 8) and similar material. As mentioned, these radiation losses are proportional to the wavelength in the fourth power.

From the spectrogram for material D in fig. 8, it will immediately appear that the calculation of the actual surface weight M of the water (water content) in the material D on the basis of equation 11 with the measurement signals IXI\3 and IX2 received during the measurement must lead to incorrect results. This is caused by the fact that the measurement signal IX2 at the comparison wavelength, as a result of radiation loss due to dispersion, can no longer be equated with the measurement signal IQ that was mentioned in the introduction. This means that the measurement signal IX2, which was received at the comparison wavelength X2, can no longer be used to convey the depth of the absorption bands at the wavelength X-^, respectively the wavelength X^.

The expressions

thus no longer corresponds to the actual extinction E , or E1 and E respectively. b} .

w' w 6 c c

In order to solve the stated problem when determining the actual values for the water content in the material with strong dispersing properties of the substances in the material in question, reference values are calculated using infrared radiation with an additional comparison wave, which are adapted to the measurement wavelengths X^ and X^ and which serves to determine the actual depth of the absorption bands at the measurement wavelengths X^ and X^.

When measuring on dispersing material, e.g. the material D according to fig. 8, in addition to the aforementioned comparison wavelength X2, an additional comparison wave X^ is introduced. This additional comparison wave X^ lies, like the comparison wavelength X2, within the spectrogram region, which is represented by straight lines and which lies close to the absorption bands at X^ and X^, cf. fig. 9.

At the additional comparison wavelength X^ becomes

it received a measurement signal IX^. Using the measurement signals <I>X2 and rx^ at the comparison wavelengths X2 and X4, the reference values IqX-^ and IqX^ can be calculated, which are adapted to the absorption bands at the measurement wavelengths respectively. X-^ and X^ for material D and which lies on a base line that can be accepted as a reference line, where the deflection, which is caused by the dispersion of the radiation, is taken into account. The calculation of these reference values IX.. and I X takes place approximately

o 1 two o 3 J

show in accordance with the following equations:

The quantities X1 and X2 in equations 13 and 14 can be calculated as follows:

As the course of the baseline depends on the chosen scale and is rarely completely straight, the reference values calculated in this way are only approximate. By introducing certain mathematical corrections, a further approximation to the reference value which is correct in each individual case can be achieved.

To determine the water's surface weight (water content) in a material with significant dispersion of the radiation, for example in material D, equation 12 corrected with equations 13-16 will have the following wording: The expressions

corresponds practically to the actual extinction E^ or E* at the radiation wavelength in question.

A measurement and calculation of the water content in paper or of the proportion of filler with an absorption band in a plastic film or of a similar measurement problem can be solved on the basis of the latter equation 17 with an accuracy that has not been achieved until now, as all the difficulties that have hitherto arisen during the measurement are eliminated or reduced to a significant extent, where said difficulties stem from the material's surface weight or surface weight changes and of the material's structure or structural variations as well as of how the material behaves in a dispersion context.

Claims (2)

1. Procedure for non-contact measurement of the absolute content of a substance (by-substance) in a mixture (main substance and by-substance) of several substances, which has the form of a thin film, in particular for measuring the absolute content of water in paper, where the is used infrared radiation with a first measurement wavelength ({ ^ which is absorbed by the secondary substance, e.g. by the free (chemically unbound) water in the paper, and with a comparison wavelength^ which is not significantly absorbed, either by the secondary substance or the main substance, if wavelength radiation intensities, after influence from the substance to be measured, are compared with each other, characterized by) that infrared radiation with a second measurement wavelength^ is used in a manner known per se, which is absorbed by the main substance, whose radiation intensity is compared with the radiation intensity of the comparison wavelength^ t b) that a radiation is also used, e.g. beta radiation, which is not affected by the structural properties of the main substance, and c) that a correction factor is formed, which corrects for the impact of the structural properties of the mixture of main substance and secondary substance on the measurement result concerning the absolute content of secondary substance and which is obtained with infrared radiation with the first measurement wavelength^? and the comparison wavelength fl, by comparing the measurement results obtained from the absolute content of the main substance, using infrared radiation with the other measurement wavelength3 and comparison wavelength (( on the one hand and using the beta radiation on the other hand. 2. Method in accordance with claim 1, characterized in that in combination with comparison wavelengthf? a second comparison wavelength fl is used which is also not absorbed, neither of the substance to be determined, for example the free water in the paper, nor of other substances in the relevant material, which second comparison wavelength^ is used to determine a reference value corrected with regard to the dispersion of the material radiation in the material for the measurement signals I ( fl^) and I (^3> which are obtained through the two measurement wavelengths ^ and^3-