Device forming a manometer intended for measuring biphase fluid pressure, associated method of manufacture and fluidic network

i

DEVICE FORMING MANOMETER INTENDED TO MEASURE

PRESSURE OF DIPHASIC FLUID, EMBODIMENT PROCESS

AND ASSOCIATED FLUIDIC NETWORK

TECHNICAL AREA

The invention relates to a device forming a manometer for measuring the pressure of a two-phase fluid in a fluidic network.

It relates to an improvement in the sensitivity of a manometer by allowing a noticeable reduction in dimensions and its integration into a fluidic network.

The applications particularly targeted by the invention are heat exchangers with two-phase fluids, the fuel cell or other system involving the use of a two-phase mixture operating around atmospheric pressure.

PRIOR ART

In systems, such as a two-phase fluid heat exchanger or the fuel cell, the liquid and vapor phases of the same fluid are simultaneously present in the fluidic network supporting the thermal and mass exchanges.

The pressure of the fluid, the mass flow ratio between the vapor and liquid phases (titer) of the two-phase fluid as well as their respective distribution in the fluidic network are parameters that reveal, at all times, the operating efficiency of the system.

Thus, the precise knowledge of these parameters, in real time and in the critical points of the fluidic network, such as hot spots, drying points, congestion points, could make it possible to prevent a malfunction / deterioration or to correct / adjust the operating regime of the systems.

In other words, it would be desirable to be able to integrate devices for precise measurement of these parameters, more particularly of the pressure, directly into the fluidic networks of systems mentioned above, without harming their operation or their compactness.

It is known practice to measure the pressure of a liquid using a manometer.

In particular, it is known to measure the pressure of a liquid by measuring the displacement of a meniscus of the liquid in a capillary tube.

US Patent 4,404,855 proposes the measurement of a pressure differential by measuring the displacement of a liquid meniscus at the interface with an air bubble injected by a syringe in order to increase by a factor of 10⁴ the measurement sensitivity. The dimensions of the measuring device thus proposed are macroscopic and do not allow its integration into a fluidic network of a compact system. In addition, the minimum movement of the meniscus which can be observed is of the order of 0.1 mm and requires the use of optical means external to the sensitive element of the manometer itself.

The aim of the invention is therefore to provide a solution making it possible to integrate a precise device for measuring the pressure of a two-phase fluid in a fluidic network of a system, such as a heat exchanger or a fuel cell without harm the operation of this one or its compactness.

DISCLOSURE OF THE INVENTION

To do this, the subject of the invention is a device forming a pressure gauge, intended for measuring the pressure of a two-phase fluid in a fluidic network, comprising:

- a first channel inside which a two-phase fluid is able to circulate;

- a second channel opening onto the first channel, the second channel being blind, with each of its dimensions less than the capillary length of the liquid phase of the fluid, and with its longitudinal wall having a surface energy gradient decreasing from its inlet towards the bottom , the surface energy gradient making it possible to increase the wetting angle of the meniscus of the liquid phase of the fluid in the blind channel from its entry to the bottom.

The capillary length is a characteristic liquid dimension on which the capillary forces and the gravitational forces are of the same magnitude.

For water, the capillary length

<td>and</td><td>volume</td><td colspan="2">capillary Vc</td><td>have for value</td>

<td colspan="2">respectively :</td><td></td><td></td><td></td>

<td></td><td></td><td>1 σ</td><td></td><td></td>

<td></td><td></td><td>_lc Jp-g</td><td> ~ 2.7</td><td>mm</td>

<td>and</td><td></td><td></td><td></td><td></td>

<td></td><td></td><td> 4</td><td></td><td></td>

<td></td><td></td><td>—.Π. -</td><td></td><td></td>

<td></td><td></td><td>vc ~ 3 1 ^ 2</td><td> ) ~ 10</td><td>pL.</td>

<td></td><td>So,</td><td>in the</td><td>frame</td><td>of the invention, for</td>

<td>a</td><td>given liquid,</td><td>each</td><td colspan="2">made up dimensions</td>

by the hydraulic diameter and the length of the blind channel is less than the capillary length of the liquid.

Thus, according to the invention, the production of a blind channel connected directly to a channel (main channel) of a fluidic network and the addition of a surface energy gradient along its longitudinal wall makes it possible to precisely locate a liquid meniscus at the interface with the vapor phase.

The displacement of this meniscus is dependent on the pressure of the fluid entering the blind channel.

The inventors in fact started from the observation that in a liquid manometer the displacement of the meniscus at the interface with an air bubble under the effect of pressure was influenced by the wetting of the liquid in contact with the walls of the tube. capillary.

They therefore sought to study this influence more precisely from a blind channel in which a fluid in liquid phase and in vapor phase are simultaneously present. Air can also be present in the gas phase.

This configuration is shown in Figure 1.

The relation of balance of force can thus be written as follows:

P -P liquid gas liquid

2 (H + W) = σ cos0_e

HW in which:

Pliquid is the pressure of the liquid

P_gas is the gas pressure;

the width of the blind channel;

is the height of the blind channel is the surface tension constant of θ_Θ is the wetting angle of the water on the longitudinal wall 11 of the blind channel 1.

This equation is written in a different way:

ΔΡ = —σ cos0 (2)

D_k where ΔΡ = Piiq_Uidea ^_ Pqaz and D_h is the hydraulic diameter which is defined by:

2HW (H + W)

The pressure of the gas trapped in the cavity 10 can be determined by the ideal gas equation:

(3) p = _p ^gas_L__{y atm} where Patm is the atmospheric pressure,

L1 is the distance between a flat meniscus (cos0_e= l) and the bottom of the cavity 10;

and y the position of the liquid meniscus (water) at the interface with the

We can consider that the wetting effect is significant if they are of the same order of magnitude.

The figure shows the curve representing the pressure difference as a function of the hydraulic diameter. The area above this curve is the area of ​​low wetting sensitivity, while the area below is of high wetting sensitivity. For example, at a hydraulic diameter of µm, the effect of wetting is significant if the pressure difference is less than 1.1 bar.

The inventors then studied the influence of the contact angle (wetting angle) on the displacement y of the liquid meniscus at the interface with the gas.

By considering that the contact angle is constant, i.e. by writing the relation cos0_e = K

We have the relation:

^Li ~ y ^D_h

From where :

<img file="FR2959564B1_D0001.tif" />

There is illustrated in Figures 3A to 3C, three different liquid contact angle configurations of 20 °, 90 ° and 120 ° respectively and where the characteristic dimensions H, L1 and L2 are shown.

Considering a linear variation of the cosine of the contact angle along the longitudinal wall of the blind channel, we have cos0_e = K \ y with

COS0_max COS0_nnn^cos6max ^cos®min ^{Kl =}^ - A + ^ 2 in relation:

P - EA-y

P atm = —σ

D_h with = 4- ° ^L'⁺^~ ~^^K'y +

Pgtm) A _ q —σ K,

D_h y ^ + G \ y + G% - 0 v

(13)

<img file="FR2959564B1_D0002.tif" />

(14) / &lt;, &lt;Ο

So we have the relation

<img file="FR2959564B1_D0003.tif" />

(15)

Considering a variation of the second order of the cosine of the contact angle along the longitudinal wall of the blind channel, we have with / &lt;₂ =

COS0 cosO_e = K ^ y max (16) (17)

Thus, for the position y of the meniscus we obtain the relation:

V - L, O - ^ - y + ΙΪξΑΑ ₌ o —gK-, —cK-,

D_h² D_h² (18)

The inventors then carried out a comparative study by choosing the following numerical example:

- study pressure range of 1 to 2 bar for high wetting sensitivity;

- L1 = L2 = 100 pm;

channel depth and width

H = W = 1pm;

varying from 0 ° from the entrance of the blind channel to 90 ° over the length L2;

varying from 90 ° to 150 ° over the length L1 at the bottom of the blind channel.

The figure

4A shows the representative curve of the cosine of the angle 0e as a function of the y position of the liquid meniscus.

The figure

4B shows the curve representative of the position of the meniscus y as a function of the pressure

The figure

4C shows the curve representative of the sensitivity of displacement of the meniscus as a function of the pressure

P.

Thus, from this study, the inventors came to the conclusion that by varying the wetting angle (angle of one could very strongly increase the measurement sensitivity of a manometer whose measurement principle is the position d 'a liquid meniscus.

In other words, the inventors have concluded

<td colspan="3">than the use of a</td><td colspan="2">energy gradient</td><td>of surface</td>

<td>allows</td><td>of</td><td>increase the</td><td>sensitivity of</td><td colspan="2">measure of a</td>

<td colspan="2">manometer</td><td>compared</td><td>to a surface</td><td colspan="2">hydrophobic to</td>

<td>energy</td><td>of</td><td colspan="2">constant area. In addition,</td><td>they</td><td>concluded</td>

<td colspan="2">that this</td><td>increase</td><td>sensitivity</td><td>of</td><td>measure is</td>

<td>again</td><td colspan="2">more important</td><td colspan="2">with a gradient</td><td>cosine</td>

<td>corner</td><td>of</td><td colspan="2"> 2^eme order compared to</td><td>a</td><td>gradient</td>

<td>cosine</td><td colspan="2">linear.</td><td></td><td></td><td></td>

ίο

Thus, a device according to the invention can be, because of its high measurement sensitivity, of microscopic dimensions for relatively small pressure variations, typically of the order of 1 bar. It can therefore be integrated into a network of

<td>fluidics without</td><td>harming</td><td>her</td><td>compactness or its</td>

<td>operation.</td><td></td><td></td><td></td>

<td>For</td><td>achieve</td><td>the</td><td>energy gradient of</td>

<td>surface on at</td><td>minus one</td><td>of</td><td>longitudinal walls,</td>

the variation in the chemical composition of the material at the surface of the longitudinal wall (s) or the variation in the structuring of the longitudinal wall (s) can be used, such as playing on several parameters.

Said variation in structuring of a micro or nanometric nature relates to a variation either in the shape (depth, width) of the cavities, or in the density of said cavities or in the shape and density of the cavities forming the surface structuring of said walls. We can therefore mainly consider three solutions as follows:

- a uniform chemical composition and a surface structure with an increasing depth gradient from the entrance of the blind channel to the bottom;

- a composition with a chemical gradient from the entrance of the blind channel to the bottom and a uniform surface structure;

- a chemical gradient composition and a surface structure with a depth gradient from the entrance of the blind channel to the uniform bottom.

For a device of microscopic dimensions to be integrated into a fuel cell or two-phase fluid heat exchanger, the transverse dimensions (or hydraulic diameter) of the blind channel can be between 50 nm and 5 μm, to measure a pressure variation greater than 0.5 bar (Figure 2).

The length of the blind channel can be between 50 nm and 500 μm.

The blind channel may have a substantially rectangular cross section, that is to say a rectangular or similar shape, with / without a ridge (polygonal / ovoid).

Advantageously, provision is made to partially integrate into the blind channel the means for measuring the position of the liquid meniscus in the blind channel in order to determine the pressure of the fluid.

In an advantageous embodiment, these measuring means comprise two electrodes arranged on two parts of longitudinal walls facing one another and each extending from the inlet to the bottom of the blind channel, the electrodes forming with a two-phase fluid present in the blind channel a capacitor with variable capacitance.

The pressure measurement is determined by the position of the liquid meniscus at equilibrium in the blind channel. The meniscus defines a variable liquid / vapor volume ratio in the channel depending on its position. This volume ratio, or its variation, can be appreciated by an electrical measurement of the average capacity of the two-phase fluid in the channel.

The use of the two electrodes facing each other makes it possible to measure the capacitance of the fluid.

In FIG. 5, there is shown an implantation of two electrodes 20, 21 arranged in parallel with one another at the top and at the bottom of the blind channel 1.

The distance between the two electrodes is close to the depth of the hydraulic diameter of the pressure gauge.

The length and width of the electrodes are those of the pressure gauge. The total capacitance is the sum of the capacitance of the vapor phase and the liquid phase of the two-phase fluid

C, = C_/+CV (20)

<td>with</td><td>wfy₂+ y) Q = e_oe / _H</td><td> (21)</td>

<td></td><td>IT ^ -y) Cv = ε _not εν ---------⁰ H</td><td> (22)</td>

<td>relations</td><td>in which ^εθ</td><td>is the permittivity</td>

<td>absolute of</td><td>. , ε₀ = 8.854xl0^-12vrde, ^u</td><td>F / m; and εν are</td>

<td colspan="3">respectively the relative permittivity in water and</td>

<td>in the air.</td><td>Thus, the capacitance</td><td>total is given by</td>

the relationship :

(19) „W. W.

^VSt ^{+ e}a) ^ + -^£o (^£the ~^£a) y

<td>The</td><td colspan="3">total capacitance, depending</td><td>of</td>

<td>ratio</td><td>the</td><td>dimensions</td><td>transverse (W / H)</td><td>of</td>

<td>blind channel,</td><td>is</td><td>represented</td><td>in figures 12A</td><td>and</td>

12B. FIGS. 12C and 12D show that the measurement sensitivity will be all the greater as this ratio is important. In the practical case, this ratio is greater than 1.

Obviously, those skilled in the art will take into account the change in the relative permittivity of the liquid phase, such as water, as a function of temperature.

The invention also relates to a method for producing a device forming a pressure gauge described above, according to which the following steps are carried out:

a / production of a first longitudinal part of a blind channel in a substrate plate comprising a first portion of a main channel inside which a two-phase fluid is able to circulate, b / production of a gradient of surface energy on the first part of the blind channel, c / production of a second longitudinal part of the blind channel in another substrate plate comprising a second portion of a main channel inside which a two-phase fluid is able to circulate, the first and the second longitudinal parts of the blind channel being of identical length, d / transfer from one substrate plate to the other with alignment of the two longitudinal parts of the blind channel and of the two portions of the main channel, e / closing the two substrate plates together.

To produce a chemical gradient composition according to steps b / and d /, it can be done by a process called SAM process (self-assembly process for molecules, in English "SELF-ALIGNEDMOLECULES") and first deposit a layer of hydrophobic molecules on the surface of the channel. The hydrophobic layer is then partially removed by plasma or laser ablation to locally uncover a hydrophilic substrate. The reverse is also possible: a layer of hydrophilic molecule on a hydrophobic substrate.

The control of the etching / ablation design thus makes it possible to produce a surface chemical gradient. The gradient can also be obtained by evaporation of a functionalized molecule.

According to an alternative embodiment, prior to step d /, the same surface energy gradient is produced on the second longitudinal part of the blind channel.

In order to integrate the electrodes as measuring means, prior to each of the steps b / and d /, a deposit of a thin metallic layer constituting an electrode is advantageously carried out in each longitudinal part of the blind channel. The electrode can be deposited in a thin layer by a PVD type vacuum deposition technique of a metal such as Ti, Cu, etc.

In the case where the plate is conductive, the electrical connection element with the electrodes can be made on the rear face. Otherwise, a contact path can be made at the same time and according to the same process as the electrode.

Depending on the configurations of the conductive plates or not, it is also possible to provide one or more electronically insulating layers (eg SiO2, TiO2, etc.) in order to electrically insulate the capacitor formed by the electrodes and the two-phase fluid.

The invention also relates to a fluidic network integrating one or more devices forming manometers described above.

The invention also relates to a fuel cell comprising such a fluidic network, at least one of the pressure gauges being suitable for measuring the pressure of the water produced at the cathode or at the anode.

The invention finally relates to a two-phase fluid heat exchanger capable of changing from a liquid phase to a vapor phase comprising such a fluidic network, at least one of the manometers being suitable for measuring the pressure of the fluid.

BRIEF DESCRIPTION OF THE DRAWINGS

Other advantages and features of

The invention will emerge more clearly on reading the detailed description given by way of illustration and not by way of limitation with reference to the following figures, among which:

the figure shows a schematic longitudinal sectional view of a blind channel 1 of a device according to the invention;

the figure shows the curve representing the evolution of the pressure variations as a function of the hydraulic diameter of a blind channel of a device according to the invention;

FIGS. 3A to 3C represent a longitudinal sectional view of a blind channel of a device according to

1'invention, three different contact angle configurations;

FIG. 4A shows the curve representative of the function cosine of the position y of the liquid meniscus in the blind channel;

FIG. 4B shows the curve representative of the position of the meniscus as a function of the pressure P;

FIG. 4C shows the curve representative of the sensitivity of displacement of the meniscus as a function of the pressure P;

the figure shows the implantation of two electrodes as means for measuring the position of the liquid meniscus in the blind channel of a device according to the invention;

- Figure 6 is a schematic representation of a device according to the invention;

FIGS. 7 to 11B show different stages in the production of a blind channel of a device according to the invention;

FIGS. 12A to 12D show the various curves representing the evolution of the capacitance of a two-phase fluid as a function of the pressure and of the position of the liquid meniscus respectively in a blind channel of the device according to the invention and according to the value of the ratio between dimensions of the blind channel.

DETAILED PRESENTATION OF PARTICULAR EMBODIMENTS

Figures to 5 have been commented on in the preamble, so they are not commented on here.

The device forming a pressure gauge according to the invention is intended for measuring the pressure of a two-phase fluid F.

It comprises a blind channel 1 opening onto a main channel of a fluidic network, open or closed, through which a two-phase fluid circulates.

F. This blind channel 1 is thus an integral part of the network.

Each of the dimensions (length, depth, width) of the blind channel 1 is less than the capillary length of the liquid phase of the fluid.

At least one of the longitudinal walls 11 of the blind channel 1 has a surface energy gradient that decreases from its entrance to the bottom.

Thus, according to the invention, the surface energy gradient makes it possible to increase the wetting angle of the meniscus of the liquid phase of the fluid in the blind channel 1 from its inlet towards the bottom.

There is shown in FIGS. 7 to 11B different stages of making a device.

A first longitudinal part 110 of a blind channel is produced in a substrate plate 4 comprising a first portion 30 of a main channel inside which a two-phase fluid is able to circulate.

A metallic layer is deposited to produce an electrode 20 with its contact socket 200 on the first part 110. Then, a surface energy gradient 112 is produced at the surface of the electrode layer 20 (top view of FIG. 7 and sectional views of Figures 8A to 11B).

Similarly, a second longitudinal part 111 of the blind channel is produced in another substrate plate 5 comprising a second portion of a main channel inside which a two-phase fluid is able to circulate. Likewise, a metal layer is deposited in order to produce an electrode 21 with its contact point, then an identical surface energy gradient is produced on the second part 111 of the blind channel 1.

The first 111 and the second 110 longitudinal parts of the blind channel are of identical length.

One then carries out a transfer of a substrate plate 5 on the other 4 with alignment of the two longitudinal parts 110, 111 of the blind channel and of the two portions of the main channel (FIG. 8A, 9A, 10A,

11A).

Finally, the step of closing the two substrate plates 4, 5 between them is carried out (FIG. 8B,

9B, 10B, 11B).

As regards the realization of the energy gradient, different approaches are possible such as two identical gradients facing each other (FIGS. 8A and 8B).

Alternatively, it is conceivable that one of the walls does not have a gradient but either has a constant surface energy or is hydrophilic or hydrophobic (FIGS. 9A, 9B in which the first part 110 of the blind channel is devoid of an energy gradient and FIGS. 10A and 10B in which the second part 111 which is deferred has no energy gradient).

It is also conceivable that a longitudinal wall has a given surface energy gradient and the other facing longitudinal wall has a different surface energy gradient (FIGS. 11A and 11B in which the first part 110 of the channel has a surface energy gradient 112 while the second part 111 has a different surface energy gradient 112 ').

The curves in figures 12A and 12B show the variation of the total capacitance with the pressure and the position of the water meniscus (liquid phase) at the interface with air and water vapor (gas phase) respectively in a blind channel according to the ratio between dimensions between transverse dimensions (W / H).

Although described in relation to a two-phase fluid whose liquid phase is water and the vapor phase is air and water vapor, the invention can be applied to many other two-phase fluids.

Likewise, other production methods can be envisaged for producing the blind channel of the device according to the invention. It is thus possible to provide for mechanical machining of an aluminum or steel substrate, lithography and etching of a silicon substrate or replication by injection or embossing of a polymer substrate.