Viscosity measurement method and device

Abstract

To provide a viscosity measurement method capable of coping with fluid in a wide viscosity range, and to provide a device for the method.SOLUTION: Pressure needs to be applied to allow fluid having fixed viscosity to flow into a flow path against viscosity. An internal pressure is generated in the flow path by application of the pressure. By detecting deformation of the flow path by the internal pressure as strain or the like of a wall within the flow path, viscosity of the fluid can be obtained. As a viscosity measuring device, a flow path in which fluid for which viscosity should be measured flows, means for measuring the amount of deformation generated in an inner wall of the flow path by allowing fluid to flow, and means for obtaining viscosity of fluid on the basis of the measured amount of deformation are provided.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a viscosity measurement method and an apparatus therefor, and more particularly to a viscosity measurement method and an apparatus that can be used with fluids having a wide range of viscosities. [Background technology]

[0002] Measuring the viscosity of a fluid is very important in determining its properties, and can be used, for example, to identify the fluid. Although there are many techniques for measuring the viscosity of either gases or liquids, it is very difficult to measure the viscosity of both gases and liquids using a single method because the properties of gases and liquids are significantly different and there are large discrepancies between the viscosity values ​​of the two.

[0003] The most common property of fluids in daily life and elsewhere is viscosity. For example, water is a fluid with a relatively low viscosity, while honey has a much higher viscosity. Gases also have viscosity, but the viscosity is much lower than that of liquids. Therefore, the difference in viscosity between different gases cannot be distinguished as easily as with liquids. Viscosity values ​​for gases are 10 -6 ~10 -5 Pa·s, and 10 for liquids -4 ~10 8 Viscosity varies over a very wide range, from 10 Pa·s to 10 Pa·s. As a result, a number of different analytical techniques are used to determine viscosity. For example, the viscosity of a liquid can be measured by stirring the sample using a rotational viscometer. This measuring device typically has a 10 -3 ~10 6Viscosity meters operate over a wide range of values, from Pa·s to s, and can therefore measure the viscosity of many different liquids. In contrast, measuring the viscosity of gases is more difficult, and many different methods are used. These methods include measuring the pressure difference in a capillary tube, measuring the viscous resistance of a falling object, measuring the damping of a disk's vibrations, measuring light scattering, measuring the change in velocity of a rotating, levitating disk, measuring light absorption, measuring the shift in the resonant frequency of a vibrating object such as a quartz crystal microbalance (QCM) or a microcantilever, and measuring the axial force exerted on the inner wall of a tube by the viscosity of a fluid flowing through it and the resulting elongation of the tube (Patent Document 1). Summary of the Invention [Problem to be solved by the invention]

[0004] However, only a few techniques can be applied to measure the viscosity of both gases and liquids using the same device. One example of measuring the viscosity of both types of fluids is the use of microcantilevers and combining this with wave propagation analysis. Nevertheless, multiple microcantilevers with significantly different dimensions are required: nanometer-scale thickness for gas measurements and micrometer-scale thickness for liquid measurements. As a result, separate devices must be used for liquids and gases. Another example is the use of glass capillaries to measure the viscosity of water in different phases, i.e., liquid and vapor. These measurements require high maximum pressures of several tens of MPa as well as a wide pressure range starting from several hundred kPa, making the use of high-pressure pumps and high-performance pressure transducers essential. Two-phase flows, such as bubbly, slug, and annular flows, can be analyzed using standard viscosity measurement techniques. However, the behavior of these flows is dominated by the continuous phase, resulting in a flow with a viscosity similar to that of a liquid. A significant challenge is measuring both extremes of the viscosity range using a single device. Gases are too diffusive to be measured with existing techniques, such as rotational viscometers. On the other hand, liquids are typically too viscous to measure the magnitude of viscosity-dependent parameter changes using existing gas techniques without significant changes or complexities in the design. A single device capable of measuring the viscosity of both gases and liquids must be able to probe a wide range of viscosities, from the very low viscosities of gases to the much higher viscosities of liquids. Such a device would enable measurements of any fluid, including gases, liquids, and their mixtures, without requiring any configuration changes, thereby enabling, for example, the identification of fluids based on viscosity. [Means for solving the problem]

[0005] According to one aspect of the present invention, there is provided a viscosity measurement method that includes flowing a fluid whose viscosity is to be measured through a flow path, measuring the amount of deformation caused on an inner wall of the flow path by the flow of the fluid, and determining the viscosity of the fluid based on the measured amount of deformation. Here, the amount of deformation may be measured as strain. The strain may also be a strain occurring on or in the inner wall. The strain may be a strain in a direction along the inner wall surface at a position on the inner wall surface where the strain is measured. The strain may be a strain in a cross section of the flow channel caused by a plane intersecting the flow channel. Furthermore, the deformation of the inner wall may be a deformation of the cross-sectional shape of the flow path due to a plane that intersects with the flow path, and the strain may be a strain that occurs when the length of the side of the cross-sectional shape is changed due to the displacement of the side caused by the deformation. The magnitude of the strain may be proportional to the square of the amount of the displacement. Furthermore, the relationship between the amount of deformation and the absolute value of viscosity may be calibrated by measuring the amount of deformation when a fluid with a known viscosity is passed through the flow path. Furthermore, the flow rate at which the known fluid flows through the flow channel may be the same as the flow rate at which the fluid whose viscosity is to be measured flows through the flow channel. According to another aspect of the present invention, there is provided a viscosity measurement device comprising a flow path through which a fluid whose viscosity is to be measured flows, means for measuring the amount of deformation caused on the inner wall of the flow path by the flow of the fluid, and means for determining the viscosity of the fluid based on the measured amount of deformation, and which measures the viscosity of the fluid by any of the viscosity measurement methods described above. Here, the means for measuring the amount of deformation may be a strain gauge provided on or in the inner wall of the flow channel. Additionally, the viscosity measuring device may measure the viscosity of both gases and liquids. [Effects of the Invention]

[0006] The viscosity measurement method and apparatus of the present invention make it possible to measure the viscosity of fluids over a very wide viscosity range. Therefore, if necessary, it is possible to provide an apparatus for measuring the viscosity of fluids with a wide range of viscosities, including both liquids and gases. Of course, the present invention can also be applied when the intended fluid to be measured is only gases or only liquids. Furthermore, the present invention does not require any moving parts other than those that undergo minute deformation based on the measurement principle, such as a fixed flow path wall. Furthermore, it is also not necessary in principle to have any parts protruding into the flow path of the fluid to be measured or to have a structure for extracting the fluid itself from the flow path. Therefore, a viscosity measurement apparatus with a simple and robust structure can be provided. [Brief explanation of the drawings]

[0007] [Figure 1] (a) Optical microscope image of the strain gauge used in the example. (b) Photograph of a microfluidic device made of PDMS and a partially enlarged photograph of the same. The partially enlarged photograph on the right shows the strain gauge embedded in the device. (c) 3D structural diagram of the microchannel device (shown in (b)). (d, e) Cross-sectional structural diagrams of the strain gauge placed just above the microchannel so as to intersect with it. The strain gauge used here has a strain-sensing component (specifically, an alloy) (labeled "Gauge" in the figure) at the tip of one side of the epoxy film, and a copper electrode at the tip of the opposite side on the same surface. The partially enlarged cross-sectional structural diagram shown in (e) shows the deformation of the microchannel when there is no flow (left, [Without flow]) and when there is flow (right, [With flow]) (dimensions omitted). [Figure 2](a) Graph showing the time variation of the output voltage of the microfluidic device of the embodiment when the flow of five types of gases, namely CO2, N2, air, He, and Ar, is interrupted at a flow rate of 10 mL / min. (b) Graph showing the output voltage of the microfluidic device of the embodiment when the flow of five types of gases, namely CO2, N2, air, He, and Ar, is interrupted at a flow rate of 10 mL / min as a function of viscosity. In this graph, the right vertical axis also shows the magnitude of strain corresponding to the output voltage (the output voltage and strain are proportional). These five data points are fitted with a power law function, which is shown by the dashed line in the graph. The equation for the fitting result is also shown in the lower left region of the graph. [Figure 3] (a) FEA results for the cross-sectional profile of a strain gauge in a microfluidic device of an example embodiment, where 10 gases with different viscosities, namely, C3H8, H2, C2H6, CH4, SO2, CO2, N2, air, He, and Ar, were flowed at a flow rate of 10 mL / min (underflow). The strain gauge deformation is plotted as a function of position along the microchannel width. The light gray shaded area from 2.8 mm to 4.3 mm indicates the microchannel located 1 μm below the strain gauge. For comparison, the figure also shows the cross-sectional profile without gas flow (a straight line with a deformation magnitude of 0.000 mm). (b) Deformation for the above 10 gases as a function of viscosity. The dashed line shows the results of fitting these 10 data points with a power law function. The equation for the fitting result is shown in the lower left area of ​​the figure. [Figure 4] (a) and (b) Schematic diagram for relating the strain on the strain gauge (strain along the channel wall) ε to Δh. [Figure 5]The figure shows the expected output voltage from the FEA plotted as a function of viscosity for 10 gases, namely, C3H8, H2, C2H6, CH4, SO2, CO2, N2, air, He, and Ar, at a flow rate of 10 mL / min, as shown by the open squares. These data points are fitted to a power law, shown by the dashed line, and the equation for this fit is shown in the lower left area of ​​the figure. In addition, the measured experimental data for five gases, namely, CO2, N2, air, He, and Ar, are shown by the solid squares. [Figure 6] 1 shows predicted viscosities for various gases with known viscosities as a function of known viscosities (plots with solid downward triangles: Present Invention). For comparison, similar data from prior literature is also plotted in this figure (plots with solid squares: Igarashi et al., solid upward triangles: Zhang et al., solid circles: Meijer et al.). [Figure 7] This figure shows deformation by FEA as a function of viscosity for 10 gases, namely, C3H8, H2, C2H6, CH4, SO2, CO2, N2, air, He, and Ar, at flow rates ranging from 0.1 mL / min to 150 mL / min. The plots in the figure correspond to flow rates of 0.1, 0.5, 1.0, 2.5, 5, 6, 7, 8, 9, 10, 15, 20, 25, 50, 100, and 150 mL / min, respectively. All data points were fitted to a power law for each flow rate, and the results are shown as solid lines in the figure. [Figure 8] Exponent n as a function of flow rate. The value of n is obtained by power law fitting to the Δh-μ plot shown in FIG. 7. [Figure 9]A monochrome version of the color map showing the deformation distribution of a microchannel when Ar is flowed at a flow rate of 10 mL / min. [Figure 10] Figure 1 shows the measured output voltage of the strain gauge of the example microfluidic device as a function of time for five gases, namely CO2, N2, air, He, and Ar, at flow rates of 6, 7, 8, and 9 mL / min, respectively, from left to right. [Figure 11] Figure 2(b) shows the strain gauge output voltage of the microfluidic device of the example as a function of viscosity for five gases, namely CO2, N2, air, He, and Ar, at flow rates of 6, 7, 8, 9, and 10 mL / min. The data points obtained for a flow rate of 10 mL / min are the same as those shown in Figure 2(b). [Figure 12] This figure shows the time change in output voltage from the strain gauge in the microfluidic device of the example for seven types of liquids: water, MeOH aqueous solution (MeOH concentrations of 20, 40, 60, and 80 vol%), methanol, and IPA (the notation in the upper left corner of each figure (e.g., "HO (M0H10), etc.") is used to identify these liquids). Each liquid was flowed at a flow rate of 0.25 mL / min for approximately 4 seconds, and then the flow was stopped. [Figure 13] (a) Output voltage from the strain gauge in the microfluidic device of the example as a function of viscosity for seven liquids, namely MeOH, water, IPA, and four different concentrations of MeOH in water (see the description of Figure 12 above), at a flow rate of 0.25 mL / min. (b) Plot of the change in eout 1 / m (■) and the reported viscosity values ​​(light gray ●) against the volume fraction of MeOH in HO. It can be seen that the two values ​​are in good agreement. DETAILED DESCRIPTION OF THE INVENTION

[0008] This application provides a method for easily measuring the viscosity of a fluid, whether it is a gas or a liquid, and a viscosity measurement method and device for that purpose. To this end, a fluid is first flowed through a deformable channel. The deformation of this channel depends on the pressure required to induce fluid flow, which in turn depends on the viscosity of the fluid. For example, a strain gauge embedded immediately above a microchannel and intersecting the microchannel can measure the deformation of the channel caused by the fluid flow as strain. In addition to the conventional strain gauges described in the examples, other means for measuring strain include non-contact methods such as optical interference, and these can be selected as needed. This strain is proportional to the square of the deformation of the channel caused by the fluid flow. This makes it possible to precisely distinguish between gases and liquids using the same device, for example, based on the fluid's viscosity.

[0009] The present invention provides a method and apparatus for measuring fluid viscosity over a wide dynamic range with a single device. Such a device can be realized, for example, by a microfluidic device. The microfluidic device uses a deformable microchannel, where the deformation is proportional to the pressure driving the fluid flow, which in turn is proportional to the fluid's viscosity. The microchannel, for example, is made of polydimethylsiloxane (PDMS), and strain gauges are embedded in the microchannel across and above the microchannel, effectively measuring the strain induced by the fluid flow as microchannel deformation. The examples described below provide non-limiting examples in which the strain induced by the fluid flow being measured is correlated to the viscosities of the gases carbon dioxide (CO2), nitrogen, air, helium (He), and argon (Ar), as well as the liquid viscosities of methanol (MeOH), water, isopropyl alcohol (IPA), and four different concentrations of methanol-water solutions, all of which have viscosities at least two orders of magnitude greater than those of the gases.

[0010] Although the terms "microfluidic device" and "microchannel" are used above, as is clear from the principles of the present invention, the device and channel used here do not necessarily need to be tiny. A small channel has the advantage that a smaller amount of fluid to be measured is required for viscosity measurement. However, a narrow channel can easily become clogged if minute foreign matter is mixed into the fluid or if components in the fluid are prone to adhere or deposit within the channel. Even if no clogging occurs, there are also disadvantages, such as the possibility of errors being introduced into the measurement results due to changes in various parameters as the fluid flows through the channel. Therefore, it is best to determine the size of the device, channel, etc., taking into account various conditions for viscosity measurement.

[0011] A standard method for determining the viscosity of a fluid is to pass it through a channel of known geometry and measure the pressure drop, p, required to make the fluid flow. The viscosity, μ, of the fluid is proportional to the pressure drop, p. This method is also applicable when the channel is deformable. The relationship is nonlinear, i.e., p ∝ μ n (n = 0.25 or less) (Non-Patent Document 1). However, since the value of n is small, the effect of a slight change in viscosity μ on the pressure drop p is not very large. In other words, this method has limited sensitivity to differences in viscosity.

[0012] Therefore, in this invention, a different measurement principle is adopted. Specifically, in this invention, a deformable flow channel is used, and the deformation of the flow channel itself due to the pressure of the fluid in the flow channel is measured, rather than the pressure drop p. To determine this flow channel deformation, for example, the strain on the inner wall of the flow channel can be measured using a strain gauge or the like. For example, in the device of the present invention (Fig. 1, details will be described later), which shows a specific example of this measurement, a flow channel with a rectangular cross section made of polydimethylsiloxane (PDMS) is used, and the deformation of the flow channel caused by the flow of the fluid to be measured, Δh, is measured using a strain gauge embedded just above this flow channel so as to intersect with the flow channel and separated from the fluid by a thin PDMS layer. In Fig. 1, the PDMS layer is not shown because it is very thin. The strain ε (here, the change in length of the detection part in the strain gauge, i.e., the vertical extension of the loop-shaped line in the upper half of Fig. 1(a)) is expressed as Δh. 2, and the deformation Δh is proportional to the pressure drop p, regardless of the fluid flow rate. This allows for highly accurate measurement of fluid viscosity over a very wide range, from gas to liquid. In the configuration shown in Figure 1, the strain gauge is installed on the inner wall surface of the flow channel, and as can be seen from the strain detection unit's loop-shaped line in the upper half of Figure 1(a), the measured strain ε is the strain in the direction along the inner wall surface of the flow channel. Even if the strain measurement means (sensor, etc.) such as a strain gauge is embedded inside the inner wall rather than on the surface of the flow channel, if the loop-shaped line of the strain detection unit extends in the direction along the inner wall surface, the strain measured will essentially be in the direction along the inner wall surface of the flow channel. Here, if the flow path is not uniform along the direction of fluid flow, for example, because the inner wall surface of the flow path is wavy or the inner diameter of the flow path varies depending on the location, the strain measured is along the inner wall surface at the installation position of a strain measuring device (sensor, etc.) such as a strain gauge (direction along the tangent plane of the inner wall surface at this installation position). Also, if the strain measuring device is embedded at a relatively shallow position inside the inner wall, the strain measured is essentially along the inner wall surface located directly above the embedding position.

[0013] In this specification and in view of the shape of the device shown in FIG. 1, the cross section of the flow channel (more precisely, the cross section of the flow channel taken along a plane perpendicular to the direction of fluid flow within the flow channel; hereinafter, simply referred to as the cross section) is described as being rectangular. However, this is merely an example for ease of understanding, and in accordance with the principles of the present invention, the cross section may have any shape. Furthermore, in the following description, the cross section of the flow channel is defined as the cross section taken along a plane perpendicular to the direction of fluid flow within the flow channel, but the cross section of the flow channel is not limited to this. More generally, it should be noted that the cross section of the flow channel is defined as the cross section obtained by cutting the flow channel along a plane perpendicular to or obliquely intersecting the direction of fluid flow within the flow channel (referred to as a plane intersecting the flow channel). Therefore, strain within the cross section of the flow channel is generally the strain within the cross section due to such a plane intersecting the flow channel.

[0014] However, when measuring flow channel deformation (here, strain occurring on the flow channel inner wall) only along a portion of the flow channel inner wall rather than the entire circumference, as shown in FIG. 1(e), it is desirable to design the flow channel and other components to minimize the adverse effect of reducing the detection sensitivity of deformation due to deformation in the non-detection portion of the flow channel inner wall absorbing deformation in the detection portion. For example, as can be seen from FIGS. 1(d) and 1(e), in this embodiment, the flow channel has a rectangular cross section, and strain gauges are installed along the upper long side of the rectangle. If deformations other than the upper long side are significant in the flow channel deformation caused by a pressure increase in the flow channel due to fluid flow, for example, the pressure increase may cause the upper long side of the flow channel cross section to rise as shown on the right side of FIG. 1(e). This tensile stress generated in the upper long side pulls the side walls (the left and right short sides) of the cross section, causing them to collapse toward the center of the flow channel, and the strain on the upper long side will also decrease due to the partial relaxation of the tensile stress. In such cases, the sensitivity to detecting deformation decreases. To suppress this decrease in sensitivity, for example, it is possible to reduce the collapse of the side walls by increasing the ratio of the long side length to the short side length (aspect ratio), to make the thickness of the flow path wall uneven by making the upper long side portion thinner than the other portions, to use a material for the upper long side portion that is less rigid than the material for the other portions, or to use the same material for the portion close to the inner wall surface but firmly attach the wall other than the upper long side portion to a member made of a more rigid material.

[0015] Although the above description uses the specific material PDMS as the material for forming the flow channel, this is merely an example, and any material can be used in practice. It is desirable that such a material has good processability for forming the flow channel, and that when a fluid is passed through the flow channel at a reasonable flow rate, the amount of deformation caused by the pressure change in the flow channel inner wall can be measured sufficiently using a strain gauge or the like, and that such deformation is reversible.

[0016] In addition, Figure 1(d) shows the strain sensor (labeled Gauge) of the strain gauge for measuring the amount of deformation of the channel installed on the upper surface of the channel's inner wall, via the ultrathin PDMS layer (not shown) and the 30 μm-thick epoxy film (which serves as the strain gauge substrate). However, this sensor may be placed so that it directly contacts the fluid, or conversely, it may be embedded in the inner wall material close to the inner wall surface. The sensor (the component that detects the amount of deformation of the target) can be placed anywhere as long as it can measure the amount of deformation occurring on the inner wall of the channel (in the case of strain, the strain occurring along the inner wall surface) with sufficient sensitivity. In Figure 1(d), in addition to the strain sensor, a copper electrode is also fixed on the same surface of the epoxy film, located on the opposite side of the tip where the strain sensor is fixed.

[0017] In the example of the flow channel configuration in the device shown in Figure 1(c) etc., the flow channel is linear, and the shape and size of its cross section are uniform along the direction of fluid flow. Of course, the shape and size of the cross section along the direction of fluid flow may be changed as needed.

[0018] Furthermore, as can be seen from the measurement method described above, viscosity measurement according to the present invention requires only the deformation of the flow channel through which the fluid to be measured flows, and therefore does not require any specially shaped structures inside the flow channel. While such internal flow channel structures can make the measurement device fragile, the present invention eliminates the need for such internal flow channel structures, thereby enabling the realization of a robust measurement device. Furthermore, while the prior art mentioned above measures the pressure drop across the flow channel, measurement of the amount of flow channel deformation as in the present invention does not require the pressure of the fluid flowing through the flow channel to be extracted in some way, but can be achieved by detecting strain in the surrounding areas of the flow channel (such as the inner wall of the flow channel) using simple means such as existing strain gauges. This also simplifies the structure of the viscosity measurement device, making the present invention advantageous over the prior art. Furthermore, since the present invention is based on the principle of flowing a fluid through a flow channel and determining the viscosity of the fluid based on the deformation of the flow channel caused by pressure changes within the flow channel, such deformation reaches a steady state within a short period of time, although this depends on the shape, size, and material of the device in which the flow channel is installed, and therefore the time required for viscosity measurement is short. In fact, in an experiment using the device shown in Figure 1, by appropriately setting the fluid flow rate, the deformation of the channel reached a steady state within a few seconds of starting to flow the fluid through the channel. This makes it possible to measure the viscosity of a fluid in real time or continuously.

[0019] Furthermore, the device according to the present invention has a simple structure and few moving parts, so it can be manufactured with good reproducibility using commonly used manufacturing techniques, enabling high-precision viscosity measurements without complicated settings or adjustments.

[0020] Here, the amount of deformation of the flow path and the viscosity of the fluid flowing through the flow path follow a power law, and the relationship between the strain and the amount of deformation of the flow path is as follows: strain ε is the flow path deformation Δh 2 Explain that there is a proportional relationship between

[0021] The relationship between the fluid flowing through the channel and the deformation of the channel due to the pressure applied by this fluid to the channel inner wall does not lose generality even when considered in terms of the channel configuration of the specific device shown in Figure 1, so below we will refer to the results of an experiment using the device in Figure 1 and a finite element analysis (FEA) of the device. First, in the experiment shown in Figure 2, various gas flows were applied to the device in Figure 1, and the output voltage e of the strain gauge for each gas flow was measured. out The time variation of the output voltage e was measured (Fig. 2(a)). Since the viscosity of these gases is known, out The relationship between viscosity μ and the viscosity can be calculated. This relationship is expressed as a power law e as shown in Figure 2(b). out ∝μ m (where m=0.859). Here, the strain gauge output voltage e out It should be noted that σ directly reflects the strain of the inner wall of the channel (more specifically, the elongation rate of the inner wall in the direction perpendicular to the gas flow).

[0022] Furthermore, a finite element analysis (FEA) was performed on the flow path of the specific device shown in Figure 1, in which the gases used in the above experiment and other gases were flowed, and the relationship between the amount of deformation of the flow path (upward displacement of the upper inner wall of the flow path) Δh and the viscosity of these gases was investigated. As shown in Figure 3(b), this relationship follows the power law Δh ∝ μ n (where n=0.434).

[0023] Figure 4 is a conceptual diagram showing a cross section of a channel through which a fluid whose viscosity is to be measured flows, cut by a plane perpendicular to the flow in the channel. The cross section of the channel is a rectangle with a long side length of w, and the fluid flows in a direction perpendicular to the cross section. Strain gauges are installed along the long sides of the channel to measure the strain in the long side direction of the channel inner wall due to the pressure of the fluid in the channel, i.e., the elongation of the wall in the long side direction. Figure 4(a) shows the cross section of the channel. When no fluid is flowing through this channel, that is, when no pressure is generated in the channel due to the fluid flow, the upper inner wall surface of the channel is at the position indicated by the horizontal dashed line in the figure, and the entire cross section is rectangular. When a fluid flows through a channel (in the figure, the fluid flows perpendicular to the paper), the pressure inside the channel increases, causing the channel to expand. This causes the channel's inner wall to deform into a shape resembling a semicylinder (a cylindrical object with a nearly circular cross section (hereafter simply referred to as a cylinder) cut vertically along a plane containing its central axis). This shape, when viewed in a cross section perpendicular to the central axis of the channel, appears as an upwardly convex curve, as shown in Figure 4(a). The distance traveled by the apex of this semicylinder (a curved shape with an upwardly convex cross section) in a direction perpendicular to the upper inner wall from its original position is defined as the channel deformation Δh. The width of the channel (the length of the long side of the rectangle) is defined as w. In the model shown in Figure 4(a), to simplify the analysis, only the top surface, which corresponds to the upper long side of the cross section, is depicted as deforming convexly outward, while the other surfaces (corresponding to the short sides and lower long sides of the rectangle) remain undeformed. Of course, the pressure inside the flow channel is applied to the entire inner wall surface, but if, for example, the wall thickness on the upper surface is made much thinner than other parts, the deformation (expansion) of the inner wall surface will tend to show the same tendency as shown in the figure. Furthermore, it is clear that even if the deformation of all the inner wall surfaces is taken into account in the analysis, the results obtained will be the same as when using the simplified model shown in the figure.

[0024] In Figure 4(a), the strain gauge strain detection section is located on the upper long side of the channel cross section. This strain detection section is not located on the entire long side, but on the left half of the upward convex arc-shaped portion, as indicated by the thin black curve above it. Therefore, when no liquid is flowing through the channel and no pressure is generated within the channel, the strain detection section is linear, with a length of w / 2. Here, we assume that l0 ≡ w / 2. As is well known, when the upper surface of the channel inner wall deforms into an upward convex curve from the left end of the long side, the strain detection section of the strain gauge is also deformed and stretched in the same way. This change in length (more precisely, the ratio of the change in length to the original length) can be calculated as the strain ε.

[0025] When a flow channel is formed using ordinary materials, the change in shape and size of the channel caused by the pressure generated by flowing a fluid through it is extremely small compared to the size of the channel cross section (width, etc.). In particular, if we consider the case where the strain detection part of the strain gauge deforms into a curved shape due to the deformation of the upper surface of the channel caused by pressure, the extension of the strain detection part due to this deformation can be approximated with a sufficiently high degree of accuracy by the difference between the length of the hypotenuse l and the length of the base l0 of a right triangle with a base length l0 and a height Δh, as shown in Figure 4(b). That is, the following equation is obtained:

[0026]

number

[0027] Here, the above approximation was performed by ignoring the higher-order terms in the Taylor expansion of the 1 / 2 power, assuming that they are sufficiently small. The strain ε is expressed as follows:

[0028]

number

[0029] The following equation can be derived from equations (S1) and (S2).

[0030]

number

[0031] From this, the strain ε is Δh 2 Regarding the power law, the two power law functions e out ∝μ m (The output voltage e of the strain gauge on the left side out can be considered to be approximately proportional to the strain ε) and Δh ∝ μ n When comparing the two, the strain changes according to a power function with an exponent twice the viscosity exponent, so it can be seen that slight differences in viscosity are sensitively reflected in differences in strain.

[0032] It should be noted that the flow rate of the fluid flowing through the flow path affects the exponent of the power law function described above, and so the above measurement principle alone cannot determine the relative viscosity, i.e., whether the viscosities of two fluids are the same, or if they are different, the magnitude relationship between the viscosities and the viscosity ratio, but it does not yet determine the absolute viscosity. The value of the viscosity ratio is calculated using the above power law function e out ∝μ m From the relationship between the output voltage and viscosity shown by the index, the output voltage e out The ratio of the values ​​raised to the 1 / m power is equal to the ratio of the viscosities μ. Absolute viscosity can be determined by a more precise analysis, including hydrodynamic and other parameters of the measurement system not considered here, or by calibrating the measurement system by creating a calibration curve using the results obtained by applying one or more standard fluids with known viscosities to the same measurement system.

[0033] In calibrating the measurement device described above, the flow rate of the fluid affects the exponent of the power-law function, so the flow rate must be consistent between the measurements used to construct the calibration curve and the measurements of the fluid for which viscosity is to be calculated. Accurately matching the flow rates in both measurements can be time-consuming. Furthermore, if the flow rate of the fluid being measured cannot always be guaranteed to be the same, calibration measurements must be performed for multiple flow rates. Therefore, it is desirable to obtain calibration measurements for multiple (or continuously varying) flow rates using a simple procedure. To achieve this, for example, a separate flow path can be connected downstream of the point in the flow path where the deformation of the inner wall is measured, and a fluid of a different phase from the fluid being measured (e.g., a liquid if a gas is being measured in the previous stage, or a gas if a liquid is being measured) is mixed into the flow path from this separate flow path. This creates a gas-liquid two-phase flow downstream, in which gas and liquid domains alternate in the flow path. In this case, since the cross-sectional area of ​​the flow path is known, the flow rate can be calculated by measuring the domain lengths of the gas and liquid and the time required for each domain to pass through a certain point in the flow path. The transition between the gas and liquid domains can be detected optically, for example, by irradiating the channel with light and detecting the transmitted light. Alternatively, the transition can be detected by detecting the vibration of the fluid in the channel, or by applying mechanical vibration to the channel and detecting its propagation or reflection at the gas-liquid interface, as in the optical detection described above. This allows the amount of deformation of the channel inner wall to be measured while changing the flow rate in multiple steps or continuously.

[0034] In the above explanation, the cross section of the channel is assumed to be rectangular to simplify the analysis, but this assumption does not imply a loss of generality. Specifically, unless the cross section is extremely complex and not considered suitable for a typical channel, this can be confirmed by considering the relationship between the channel deformation Δh, which can be defined as the displacement perpendicular to the tangent plane of the channel inner wall, and the strain ε, which is perpendicular to the flow on the channel inner wall surface in the region where strain is measured. Considering that the channel deformation Δh is significantly smaller than the channel width as long as typical materials are used for the channel, approximating the shape of the part of the channel where strain is measured in the above analysis to a triangle, as shown in Figure 4(b), provides a sufficiently good approximation for the length of that part. Therefore, by performing calculations similar to those in the above analysis, even in such a general case, the dominant term for the magnitude of strain ε is Δh. 2 It can be said that:

[0035] Furthermore, the viscosity measurement method of the present invention described above is strictly applicable to Newtonian fluids, i.e., fluids in which the shear stress (tangential stress) of the flow is proportional to the flow velocity gradient (shear rate, shear velocity). Measurements were also performed on Newtonian fluids in the examples described below. However, when the relationship between the shear stress of the flow and the flow velocity gradient deviates only slightly from the proportional relationship, the viscosity measured using the viscosity measurement method of the present invention has a sufficiently high accuracy for practical use. Therefore, if the degree of non-Newtonianity of the fluid to be measured is known, the accuracy of the viscosity obtained using the viscosity measurement method of the present invention can be determined. Furthermore, since viscosity values ​​usually required are for a limited range of flow velocities, even in such cases where the non-Newtonianity is quite large, the viscosity value obtained using the present invention can be used as the viscosity value for such application conditions. Alternatively, even if it is inconvenient to treat viscosity as a constant, the viscosity can be determined using the method of the present invention for, for example, multiple flow velocities, and the functional relationship between flow velocity and viscosity can be determined with the desired accuracy. Therefore, it should be noted that the fluids to which the present invention is applicable are not limited to Newtonian fluids. [Example]

[0036] The present invention will be described in more detail below based on examples. Note that the following examples are provided as illustrations to aid in understanding the invention and are not intended to limit the present invention. Furthermore, the following description will be given assuming the use of a microchannel device as shown in FIG. 1. However, the term "micro" here simply reflects the fact that the dimensions of the specific device exemplified here are microscopic, as shown in FIGS. 1(c) and 1(d), etc. It should be noted again that the present invention does not require such a device to be microscopic, and that any dimensions may be used as long as they do not interfere with measurements.

[0037] <Gas Measurement> First, Figure 2 shows the results of measurements using a microchannel device for a set of gases, specifically CO2, N2, air, He, and Ar, at a flow rate of 10 mL / min. The output voltage of the strain gauge depends on the presence or absence of gas flow (indicated as "Gas on" and "off" in the figure) and the type of gas, as shown in Figure 2(a). The output for each gas is characterized by its peak voltage (in the graph shown in Figure 2(b), the peak voltage is plotted as the output voltage). This peak voltage is proportional to the strain, as shown in the following equation:

[0038]

number

[0039] where e B is the bridge voltage (2.5V), and K s is the gauge factor (2.11) of the strain gauge. The measured strain is proportional to the viscosity μ, as shown in Figure 2(b). This data is fitted to the power law function e out ∝μ m (where m = 0.859, see the fitted equation in the lower left region of the graph in Figure 2(b)). Therefore, this deformable microchannel realizes highly sensitive measurement of gas viscosity.

[0040] To investigate the cause of such a response of this device, finite element analysis (FEA) was performed on the microchannel device of the present application embodiment shown in FIG. 1. Through this analysis, it is possible to examine in detail how the gas flow affects the deformation of the strain gauge. Here, the gas flow rate was set to 10 mL / min, and the viscosity was changed to examine the deformation of the microchannel and the displacement Δh in the upward direction (the direction perpendicular to the surface of this inner wall) of the inner wall of the flow path (here, the inner wall corresponding to the upper long side of the rectangular flow path cross-section), which is the amount of deformation of the flow path inner wall as a result. When performing FEA, in addition to the five types of gases used in actual measurements, the viscosities of other common gases, propane (C3H8), ethane (C2H6), hydrogen (H2), methane (CH4), and sulfur dioxide (SO2), were also added. These gases cover a wide range of viscosities corresponding to most common gases. As can be seen from FIG. 3(a), the degree of deformation of the microchannel depends on the viscosity of the gas according to the FEA simulation. Therefore, by measuring the displacement Δh, the viscosity of the gas can be determined. This dependency is well described by the power law Δh∝μ n (where n = 0.434).

[0041] To compare the results of FEA with the measurement results, the output voltage obtained from the strain gauge is determined according to the simulated deformation of the microchannel. The strain detection part in this strain gauge, that is, the loop-shaped line shown in FIG. 1(a), extends on the microchannel to approximately half of its width w, as schematically shown in FIG. 1(e). Of course, the range covered by the strain detection part is not limited to this, and it is needless to say that it can be determined appropriately as required. Hereinafter, the analysis will proceed assuming that the length of this strain detection part is exactly 1 / 2 of the flow path width w. When the displacement of the inner wall is small compared to the flow path width, that is, when Δh << w holds, as already explained by referring to FIGS. 4(a) and (b), the strain ε on the strain gauge can be expressed as follows using Taylor expansion.

[0042]

Equation

[0043] Δh∝μ n Therefore, the output voltage also shows a dependency on viscosity. out ∝μ 2n and therefore m = 2n. The simulated results agree very well with the measured data, and both exhibit a power law for viscosity, with exponent m = 0.859, as shown in Figure 5.

[0044] Here, a graph comparing the experimental data obtained in this example with data reported in prior literature, assuming that the data was obtained at the same temperature (20°C), is shown in Figure 6. In this figure, Igarashi et al., Meijer et al., and Zhang et al. are Non-Patent Documents 2, 3, and 4, respectively. The viscosity determined according to the present invention is in good agreement with known viscosity values. On the other hand, the viscosity values ​​reported in Non-Patent Documents 2 to 4 sometimes show slight deviations. This suggests the reliability of the method of the present invention.

[0045] To investigate the behavior under various flow conditions, we performed FEA simulations with varying flow rates. As shown in Figure 7, a power-law relationship between deformation and viscosity was observed in all cases. Interestingly, however, the exponent in this relationship varies with flow rate. Specifically, as shown in Figure 8, n ≈ 1 at the minimum flow rate (0.1 mL / min), and as the flow rate increases, n decreases monotonically, approaching n = 0.25 at approximately 60 mL / min and beyond. At very low flow rates, the microchannel should deform very little, resulting in n ≈ 1. The asymptotic value of n at high flow rates is approximately 0.25, which is nearly equivalent to the analytical calculation of a PDMS microchannel whose bottom surface is bonded to a rigid glass substrate and whose top surface (the surface opposite to the bonded bottom surface) deforms under fluid flow (Non-Patent Document 1). For the present geometry, FEA simulations suggest that strain induced by fluid flow occurs primarily on the upper surface, even though the entire device is made of PDMS. This is due to the relatively deep bottom of the microchannel, as shown in Figure 9. In all cases, the strain (see Equation (S2) above), defined as the length of stretching of the upper inner wall due to the deformation of the channel cross-section caused by fluid pressure changes (i.e., the length change l-l0) divided by the original length l0, is proportional to the square of the deformation, and therefore the output voltage also exhibits a power-law dependence on viscosity, with the exponent 2n.

[0046] To compare the behavior predicted by FEA with the experimental results, the strain gauge output voltage was measured when the flow rate of five gases was varied between 6 mL / min and 9 mL / min. The results are shown in Figure 10. Here, the exponent was strongly dependent on the flow rate. In this experimental configuration, even though the flow rate change increment was small at 1 mL / min, the viscosity dependence of the output voltage was different, as shown in Figure 11. Nevertheless, the results for each gas were clearly distinguishable from each other. As shown in Figure 11, in each case, the data was well described by a power law dependence, and the exponent obtained from the measurements (the strain gauge output voltage e mentioned above) out The power law function e follows out ∝μ m The exponent m of the flow path is predicted by the FEA simulation (the amount of flow path deformation Δh follows the power law function Δh ∝ μ n The exponent n of

[0047] In the response obtained at a flow rate of 6 mL / min, the output voltage reached its maximum value within a few seconds after the start of gas injection and did not change over time. In contrast, at a flow rate of 10 mL / min, a slight overshoot was observed in the time waveform of the output voltage, and the short-term saturation characteristic described above was not observed. This fact indicates that by appropriately setting the measurement conditions, such as the flow rate when the fluid to be measured for viscosity is introduced into the flow path, it is possible to measure viscosity in real time and continuously detect changes in the fluid flowing through the flow path in real time.

[0048] The measurements here are accurate enough to distinguish nitrogen from air; in fact, the difference in viscosity between the two is only 0.05 x 10 -5 The noise level of the measurement is less than 1 μV, so the difference in viscosity is only 0.01×10 -5 The viscosity of n-butane and isobutane is about 7.39 × 10 Pa s. -6 Pa s and 7.48 × 10 -6The measurement accuracy is such that it can even distinguish between structural isomers such as stoichiometric isomers (Pa·s). However, since the exponent n changes as a function of flow rate, the measurement device of the present invention must be calibrated before use.

[0049] <Liquid measurement> The viscosity of the gas is 10 -6 Pa·s~10 -5 The viscosity of the liquid varies in the range of Pa·s. -3 The viscosity varies from Pa·s to often much larger values. Nevertheless, the device according to the present invention responds to liquids flowing through microchannels in a manner similar to gases, and can therefore also be used to measure the viscosity of liquids.

[0050] To demonstrate this, measurements were performed on several liquids, including MeOH, water, and IPA. The viscosities of these liquids at 20°C are 0.60×10 -3 Pa·s, 1.0×10 -3 Pa s and 2.4 × 10 -3 The flow rate for each measurement was set to 0.25 mL / min. The time-dependent strain gauge output voltage was measured for each liquid when the flow was turned on and off, and the results shown in Figure 12 were obtained. As can be seen, there is a clear correlation between the signal output and viscosity. As shown in Figure 13(a), these data fit well with a power law where m(=2n) = 0.820. To further expand the viscosity range, measurements were also performed on aqueous MeOH solutions with MeOH concentrations of 20, 40, 60, and 80 vol%. The viscosity of an alcohol-water binary system does not show a simple linear dependence on the mixture ratio, but instead varies in a non-monotonic manner. Hydrogen bonding between alcohol and water molecules plays an important role in determining the viscosity of the mixture, and its effect is maximized at a specific ratio. This results in a maximum viscosity. Nevertheless, the measurement data directly follows a power law, as shown in Figure 13(a). Furthermore, e out 1 / mPlotting m as a function of MeOH volume fraction reveals a nonmonotonic trend that closely matches the reported viscosity values, as shown in Figure 13(b). The viscosity of the liquid used here is approximately 50 times greater than that of the gas used in the gas measurements, but the flow rate for the liquid is approximately 40 times lower than that for the gas, resulting in measurements where the viscosity times flow rate is the same. However, the m values ​​obtained for liquids cannot be directly compared to values ​​predicted from FEA simulations for gases. Therefore, the microfluidic device must be calibrated separately for gases and liquids beforehand. Nevertheless, the device can be used to determine the viscosity of any fluid, including both gases and liquids.

[0051] <Conclusion> We fabricated a microfluidic device made of polydimethylsiloxane and measured the viscosity of fluids, i.e., gases and liquids, by flowing them through a deformable microchannel. The deformation of the microchannel depends on the pressure required to induce fluid flow, which in turn depends on the viscosity of the fluid. To this end, we embedded strain gauges in the microchannel device, positioned immediately above and intersecting, or straddling, the microchannel. This configuration allows for efficient measurement of the deformation induced by fluid flow and conversion of this deformation into strain, which is proportional to the square of the deformation. This device allows for the measurement of the viscosity of both gases and liquids using a single device. It also achieves sufficient resolution to precisely distinguish several gases by viscosity alone, with a viscosity range of 1.47 × 10 -5 Pa·s to 2.22×10 -5 The same microfluidic device according to the present invention can also be used to measure the viscosity of several liquids, including methanol, water, isopropyl alcohol, and aqueous methanol solutions with four different alcohol concentrations. The viscosity of these liquids is 0.60×10 -3 Pa·s to 2.4×10 -3It is in the range of Pa·s and has a viscosity approximately two orders of magnitude larger than that of the above-mentioned gas. In a situation where the requirements for "portability", "point-of-care testing", "wearable", and "on-site" devices are increasing, a device such as the present invention having a very wide dynamic range, ease of fabrication, and high accuracy would be useful for many applications in which viscosity can be used to discriminate differences between specimens.

[0052] [Details of the experiment] <Fabrication of PDMS device> The following specifically describes the fabrication procedure of the PDMS device used in the examples. Negative photoresist SU-8 3035 was spin-coated (at 3000 rpm for 30 seconds) onto a clean 3-inch single-sided polished silicon wafer to form a 25-μm-thick photoresist layer. After baking at 95°C for 15 minutes, a photomask was printed on the photoresist layer and exposed to ultraviolet light. Subsequently, the photomask was removed from the wafer and the wafer was baked again. This baking was performed at 65°C for 1 minute and then at 95°C for 5 minutes. The uncured resist was washed away with propylene glycol methyl ether acetate (PGMEA) for 8 minutes to obtain a master mold with a microchannel structure formed.

[0053] The microchannel molding procedure was applied to replicate the microchannels from the master mold. A liquid PDMS mixture consisting of a base and a curing agent (weight ratio 10:1, Sylgard 182) was poured into the SU-8 master mold, and then degassing was performed. After curing at 65 °C for several hours, the PDMS was peeled off from the master mold. Using a biopsy punch, through-holes (inner diameter 0.75 mm, outer diameter 1.5 mm each) for the inlet (「Inlet」 shown in Fig. 1(c)) and the outlet (「Outlet」 shown in Fig. 1(c)) were formed. This PDMS block (「Bottom PDMS layer」 shown in Fig. 1(c)) was irreversibly bonded to a separate 1-mm-thick PDMS thin layer (「Top PDMS layer」 shown in Fig. 1(c)) in which a strain gauge (「Strain gauge」 shown in Fig. 1(c)) was embedded. At this time, to form a stable bond, plasma treatment was performed at 70 W for 20 seconds. By heating at 95 °C for several tens of minutes to increase the bond strength, the fabrication of the PDMS microchannel device was completed. A photograph of this PDMS microchannel device is shown in Fig. 1(b).

[0054] <Gas / Liquid Flow Measurement in PDMS Microchannel Device> This PDMS microchannel device was used to measure strain induced by the flow of various gases and liquids. All measurements were performed using strain gauges (gauge pattern: FLKB-1-11, length: 4.3 mm, width: 1.4 mm, thickness: 30 μm, gauge factor: 2.11, gauge resistance: 120 Ω) purchased from Tokyo Measuring Instruments Laboratory Co., Ltd. Figure 1(a) shows a microscope image showing the exact dimensions of the strain gauge. All output voltages from the strain gauges were measured using an NI 9237 simultaneous bridge module (National Instruments Corporation) by applying a bridge voltage of 2.5 V and recorded at a sampling rate of 20 Hz. The data acquisition program was designed using LabVIEW (National Instruments Corporation). Gas flow was regulated using a mass flow controller (MFC; HORIBA, Ltd. SEC-N112MGM). The gas flow was injected through the inlet at a flow rate of 10 mL / min for 20 seconds, followed by a 20-second interval without gas flow. Similar gas flow cycles were applied at flow rates of 6, 7, 8, and 9 mL / min. To ensure there were no leaks, the flow rate was measured at the outlet using a volumetric flow meter (ProFLOW 6000 Electronic Flowmeter, Restek Corporation). Experiments were performed using CO2, N2, air, He, and Ar.

[0055] Liquid flow measurements were performed using a syringe pump (PHD 2000 Dual Syringe Pump, Harvard Apparatus) with a flow rate set at 0.25 mL / min. MeOH, water, IPA, and four different concentrations of aqueous MeOH solutions were used. The concentrations of the aqueous MeOH solutions were set at 20, 40, 60, and 80 vol%. These aqueous solutions are represented by the symbols M2H8, M4H6, M6H4, and M8H2 in Figures 12 and 13, respectively. Similarly, water is sometimes represented by the symbol M0H10, and MeOH is sometimes represented by the symbol M10H0.

[0056] <FEAシミュレーション> We used COMSOL Multiphysics 5.6 with the Structural Mechanics Module for FEA simulation. A PDMS device with the dimensions shown in Figures 1(a)–(d) was modeled. All parameters used in the simulation are listed in the table below. The entire device structure was divided into a mesh of approximately 60,000 elements. In addition to the CO2, N2, air, He, and Ar gases used in the experiments, five other gases, namely, C3H8, H2, C2H6, CH4, and SO2, were also simulated to obtain the most accurate fitting curves for each. Regarding the parameters listed in Table 1, while the "Channel" height is shown as 25 μm in Figure 1(d), a value of 27 μm was used in the simulation to account for unavoidable errors (of the order of a few μm) that may occur during device manufacturing.

[0057] [Table 1] [Industrial Applicability]

[0058] As explained in detail above, according to the present invention, the viscosity of fluids over a wide viscosity range (gases, liquids, and other special fluids such as gas-liquid two-phase fluids) can be measured easily and with high accuracy. Furthermore, the present invention is not limited to simply measuring viscosity, and can be applied to other fields based on viscosity measurement. For example, the present invention can be applied to identifying molecules based on differences in viscosity. In this molecular identification, the high measurement accuracy provided by the present invention allows for the identification of molecules with viscosity differences of only 0.01 x 10 -5 It is also possible to distinguish between n-butane and isobutane, which have a viscosity of about Pa·s. [Prior art documents] [Patent documents]

[0059] [Patent Document 1] Japanese Patent Application Publication No. 63-153447 [Non-patent literature]

[0060] [Non-Patent Document 1] T. Gervais, J. El-Ali, A. Gunther and KF Jensen, Lab Chip, 2006, 6, 500-507. [Non-patent document 2] K. Igarashi, K. Kawashima and T. Kagawa, Sens. Actuator A-Phys., 2007, 140, 1-7. [Non-patent document 3] AS Meijer, AS de Wijn, MFE Peters, NJ Dam and W. van de Water, J. Chem. Phys., 2010, 133, 164315. [Non-patent document 4] J.T. Zhang, H. Lin and J. Che, Metrologia, 2013, 50, 377-384. [Non-patent document 5] Ogasawara, T. et al., Compos. Sci. Technol. 71, 1826-1833 (2011).

Claims

1. A fluid whose viscosity is to be measured is passed through a flow path; measuring an amount of deformation caused on an inner wall of the flow path by the fluid flowing; Determining a viscosity of the fluid based on the amount of the measured deformation. Viscosity measurement method.

2. 2. The method of claim 1, wherein the amount of deformation is measured as strain.

3. 3. The method of claim 2, wherein the strain is a strain occurring on or in the inner wall.

4. 4. The viscosity measuring method according to claim 3, wherein the strain is a strain in a direction along the inner wall surface at a position on the inner wall surface where the strain is measured.

5. 5. The viscosity measuring method according to claim 3, wherein the strain is a strain in a cross section of the flow channel caused by a plane intersecting the flow channel.

6. the deformation of the inner wall is a deformation of a cross-sectional shape of the flow passage by a plane intersecting the flow passage, 3. The viscosity measuring method according to claim 2, wherein the strain is a strain caused by a change in length of a side of the cross-sectional shape due to the displacement of the side by the deformation.

7. 7. The method of claim 6, wherein the magnitude of the strain is proportional to the square of the amount of the displacement.

8. 3. The viscosity measuring method according to claim 1, further comprising the step of measuring the amount of deformation when a fluid having a known viscosity is caused to flow through the flow path, thereby calibrating the relationship between the amount of deformation and the absolute value of the viscosity.

9. 9. The viscosity measuring method according to claim 8, wherein a flow rate of the known fluid flowing through the flow path is the same as a flow rate of the fluid whose viscosity is to be measured flowing through the flow path.

10. a flow path through which a fluid whose viscosity is to be measured flows; a means for measuring an amount of deformation caused on an inner wall of the flow path by the flow of the fluid; 2. The method of claim 1, further comprising: determining the viscosity of the fluid based on the amount of deformation measured; A viscosity measuring device having

11. 11. The viscosity measuring device of claim 10, wherein the means for measuring the amount of deformation is a strain gauge provided on or in the inner wall of the flow passage.

12. 12. The viscosity measurement device of claim 11, wherein the viscosity measurement device measures the viscosity of both gases and liquids.

Images