CA1235744A

CA1235744A - Method and device for measuring the corrosivity of liquids

Info

Publication number: CA1235744A
Application number: CA000491151A
Authority: CA
Inventors: Roger Roffey; Goran Hulten; Anders Norqvist; Goran Olofsson
Original assignee: Forsvarets Forskningsanstalt (FOA)
Current assignee: Forsvarets Forskningsanstalt (FOA)
Priority date: 1985-09-19
Filing date: 1985-09-19
Publication date: 1988-04-26

Abstract

Abstract:

According to the invention a piezoelectric crystal, e.g. a quartz crys-tal, coated with a metal or another suitable material is exposed to the action of a corrosive liquid, preferably by being immersed therein. The mass change on the crystal surface, caused by the corrosion, and the resulting change in the natural oscillation frequency of the crystal is determined and constitutes a measure of the corrosivity of the liquid.
The natural oscillation frequency of the crystal can in this way also be measured in situ in the liquid, in which way the dynamic course of the corrosion reaction can be followed continuously. According to a special embodiment of the invention the sensitivity during the measure-ment can be increased, if the crystal is pretreated by being activated with a sulphide solution.

Description

574~

The present invention relates to a method for measuring the corrosi-vity of liquids towards primarily metals and to a device for carrying out the method.
When storing petroleum products there are in some cases formed various sulphur compounds because of, for instance, bacterial influence, which compounds in turn attack metals. These corrosive compounds tend to form on the surface of the metal a film of molecules consisting of atoms of the corroding metal and also consisting of sulphur, hydrogen, etc. Said film increases on the metal and may cause big problems. This is especially relevant for oil products, e.g. crude oil, oils for heating purposes and aviation fuels, that are in contact with water. This has been observed when oil products are stored inside rock chambers on a water bed, and this is a serious problem especially when jet fuel is being stored. In the interface between water and oil there are microorganisms, e.g. sulphate-reducing bacteria, that form sulphur compounds, such as hydrogen sulphide, elementary sulphur and certain organic sulphur compounds, among others mercaptans, sulphides and disulphides. Said compounds move upwards into the oil and make it corrosive. If such corrosive oil is used as motor fuel, the motors among other things may become damaged because certain metal parts also corrode, and if it is the question of aviation fuel the corrosion constitutes a potential danger for the safety in flight.
This is why it is very important to have a possibility of determining the corrosivity in an easy but at the same time reliable way, preferably at field locations. Particularly when storing for instance jet fuel over a long period in a rock chamber, there is a great need to be able to discover at a very early stage if the fuel is getting corrosive. The method should therefore preferably be so adapted that an analysis can be carried out at the very place, p ~5-71:3 - 1 -lZ3574~

for instance at a rock chamber.
It is here not necessary to exactly know the chemical composition of the corrosive compounds. It is fully adequate to learn about for instance how rapidly the surface film, which is formed on a metal immersed in the petroleum product, increases.
The method that is used today to measure the corrosivity of a petro-leum product is to perform an ocular inspection of a metal strip, preferably made of silver, which has been kept in the product in question. A silver strip, which is corroded by sulphur compounds in the petroleum product, becomes black, and a classification from 0 to 4 is used as a graduation of the blackening extent. Even if the test method that is used today is standardized, IP 227/73, 1976, Institute of Petroleum, Standards for petroleum and its products, Methods for analysis and testing, 35th ed., Applied Science Publishers, as subjective estimation is made of the colouration and thus of the corrosiveness. Besides, the method takes a long time to perform, relatively large sample volumes (250 ml) are required and the classification with five steps is too coarse.
Furthermore, a colouration recognizable for the eye cannot be observed until the film thickness exceeds about 100 ~.
Therefore it is desirable to have a more rapid and less subjective test method that is more sensitive and more reproducible than the methods used up to now.
The present invention provides a method of measuring the corrosivity of a liquid, which method comprises the steps of:
(1) providing a piezoelectric crystal with a reproducible coating of a material which can be subjected to the corrosion reaction in question;

(2) determining the natural oscillation frequency (~1) of the thus-357~'~

coated piezoelectric crystal;

(3) exposing the crystal to the corrosive liquid which is in the liquid state by immersing the crystal therein; and ~ 4) determining the natural oscillation frequency (v2) of the crystal after being exposed to the corrosion reaction for a certain period of time;
whereby the change in the natural oscillation frequency (vl - v2) of the crystal is a measure of the mass change on the surface of the crystal and thus of the content of corroding compounds in the liquid.
The present invention also provides a device for measuring the cor-rosivity of a liquid, which comprises a crystal detector comprising a piezo-electric crystal, encapsulated in an inert environment and provided with a coating layer, thereon of a metal which is capable of functioning as an electrode and can be subjected to corrosion, wherein the coating layer has been pretreated for cleaning and activation and, the crystal can be decapsulated before measuring the corrosivity of a liquid.
The method of the invention is a simple, reliable and objective measuring method that is easily converted to automatic working. It is more sensitive than the standard method used up to now, and as thin films as 1 A
can be detected.
According to one preferred embodiment of the device of the present invention, the detector is a silver-coated quartz crystal encapsulated in an ampoule made of glass and filled with an inert gas, wherein the silver-coating is, before the encapsulation, freed from sulphide and oxide contaminations by means of an etching cleaning process with a cyanide solution saturated with oxygen and the crystal can be decapsulated before measuring the corrosivity of a liquid.

i~S744 According to another preferred embodiment of the device of the present invention, the detector is a silver-coated quartz crystal encapsulated in a glass ampoule filled with an inert gas, wherein the silver-coating is, before the encapsulation, activated with a sulphide solution and in this way given a thin, reproducible sulphide layer and the crystal can be decapsulated before measuring the corrosivity of a liquid.
A piezoelectric crystal, for example quartz, has a constant natural oscillation at rest and oscillates at a frequency that is dependent on for instance the mass on its surface. If the crystal is given a coating that interacts or reacts with a component in its environment, the mass on the cry-stal surface changes and thus the oscillation frequency of the crystal also changes. Such coated piezoelectric crystals are known from gas analysis techniques. Swedish Patent No. 434,438 for instance relates to a gas detector, in which a gas component of a gas mixture is adsorbed to a thin organic film applied on the surface of a piezoelectric crystal, and the change in oscil-lation frequency caused by the increase in mass is a measure of the content of the gas component in question.
It is also known for instance from United States Patent No. 3,164,004 to coat a piezoelectric crystal with a thin substrate of a solid substance, e.g.silver and copper, and to measure the change in frequency that is caused by chem-ical or physical interaction between the substrate and a gas streaming passed.
On the other hand, it is not previously known to immerse a piezo-electric crystal coated with a substrate into a liquid to determine the con-tent of those components in the liquid that react with the substrate. When determining the corrosivity of a petroleum product, it is valuable to be able to carry out measurements in the liquid phase, as an evaporation may change the components in the liquid that are to be determined.

~357'~4 When the crystal coated with a metal, e.g. silver, is corroded in a petroleum product, the mass on the crystal surface increases because of the supply of mainly sulphur to the surface. This gives a decrease in the crystal natural oscillation frequency. Said decrease gives a measure of the mass iD-crease, which in turn is proportional to the amount of corrosive compounds in the petroleum product. As an alternative the corroded layer can be washed away, and the resulting decrease in mass gives an increase in the natural oscil-lation frequency which is registered.
If the product of the metal and the corrosive substance is easily dissolvable in the surrounding medium, a decrease in mass on the crystal sur-face can be registered.
The method and the device of the invention can also be used to prove sulphur compounds in other liquids than petroleum products, such as in alcohols and aqueous solutions. Also, the content of corrosive materials, such as sul-phur compounds in a gaseous material, such as in air, e.g. in sulphur discharges from combustion installations and the like, can be determined after taking up the air in a suitable liquid which by itself is non-corrosive for analysis according to the invention.
The method and the device of the invention are not limited to measur-ing the reaction between sulphur compounds and metal. Other compounds can also be analy~ed with a suitable choice of material on the crystal. An example thereof is chloride.
The method and the device of the invention can also preferably be used to study interacting and counteracting effects respectively from for in-stance different sulphur compounds in fuel or an aqueous solution. The invention can here be used as an analytical aid.

~L~3579~4 As the substrate disposed on the ~piezoelectric crystal, any metals can be used that are attacked by corrosion of the kind in question, e.g. copper, silver, aluminium etc. ~on-metallic materials can also be used if the reactiv-ity or corrosivity of the liquid towards these materials is to be determined.
The determination of the natural oscillation frequency of the crystal is carried out after a certain period of incubation in the corrosive liquid.
Here the measurement can also be performed in situ, i.e. with the crystal immersed in the liquid. This makes it possible to follow continuously the course of corrosion if so is desired.
The contact electrode of a piezoelectric crystal is made of a thin coating of metal, e.g. gold, silver, aluminium etc, that has been evaporated directly onto the crystal surface. In order to be able to use a piezoelectric crystal, which is found on the market already eoated with a metal, for corrosion measurements according to the invention, the surface must be reproducible, i.e.
the surface must initially be identical at different measuring occasions. In this case the process according to the invention also comprises an initial activation step with cleaning in order to pre-treat the crystal to make it re-producible, as well as comprising a crystal detector adapted to the method of the invention.
Since most of the commercially available piezoelectric crystal coated with a metal is encapsulated, before the crystal is cleaned, it must be decapsulated. After this is done, it should not be stored in air atmos-phere for a long period of time before the activation is carried out, as the crystal is aged when it is exposed to air for a long period. This should not exceed 3 days. The ageing probably proceeds as follows: sulphur compounds, for instance H2S, in the air atmosphere initiate a corrosion attack to the 1~357~4 metal layer on the crystal, which makes it more reactive when it is placed in the corrosive test liquid. This ageing effect has most influence on a sub-sequent measurement in highly corrosive fuels, while the effect is insignificant when measuring in fuels of low corrosivity.
For a large scale analysis it is unpractible to clean the crystal immediately before the incubation. Suitably the crystal can be coated with metal in good time before the analysis occasion, e.g. by means of thermal evaporation or cementing thereon, or be decapsulated, if a crystal existing on the market already coated with a metal is to be used, and cleaned and then encapsulated in an inert environment, e.g. in a glass ampoule filled with argon, which is later broken immediately before the incubation in the liquid to be analyzed. When the thermal evaporation is employed, it may be carried out at a lower pressure, e.g., in vacuum.
When using a piezoelectric quartz crystal coated with a thin layer of silver, the initial cleaning step consists of for instance an activation of the silver layer by means of etching with a cyanide solution, and in this way the silver surfaces on the crystal are freed from sulphide and oxide coatings.
Silver sulphide and metallic silver react with NaCN and oxygen according to the following:
2Ag(s) ~ 4CN + H20 + 1/2 2 > 2Ag(CN)2 + 20H
Ag2S(s) + 4CN + 202 ~ 2Ag(CN)2 + S04 As the metal surface is also attacked, there is probably obtained a breaking up of the surface structure, giving a more reactive surface with shorter analysis times as the result.
In a suitable embodiment of the cyanide activation the silver-coated crystal is etched in 0.01 - 1 M, preferably 0.1 M, sodium cyanide solution l~S7~14 saturated with oxygen for 10 - 30 seconds, preferably 10 seconds. If the etching time is too short, proble~s may arise with the reproductivity of the activation, and iE the etching time is too long so much silver may be etched away from the electrode that there is a risk that flakes scale off or that the silver layer on the crystal is used up.
For the activation also other compounds can be used having similar effects as cyanide, and the choice of compound is made taking into considera-tion which metal the crystal is coated with.
Another, very suitable activation process is to pretreat the previous-ly metal-coated piezoelectric crystal in a sulphide solution before the corrosivity is measured. The thin metal layer on the crystal surface by means of this pretreatment becomes further coated with a very thin sulphide layer, which makes the crystal surface reproducible but also causes the crystal to react quicker, with larger registration and at a lower content of corrosive compounds than if said sulphide coating had not been applied, i.e. the sulphide activation gives the crystal increased sensitivity during the subsequent corrosion measurement. The sulphide layer will then react with sulphur com-pounds and the like in the corrosive liquid resulting in an increase in the mass on the crystal.
The sulphide activation is suitably carried out in a 0.01 - 1 M, preferably 0.1 ~I sodium sulphide solution for 10 - 30 seconds, preferably 10 seconds. The concentration of the activation solution is, however, not critical as long as it is above the level where the crystal is saturated with a thin layer of sulphide, An activation time of 10 seconds has proved to be sufficient in most cases.

~ 8 --~3S7~4 Figure 1 shows the corrosivity that has been measured at various depths in a cistern when storing jet fuel, according to the method of the invention and according to the standardized test method;
Figure 2 is a diagram that shows the proportionality between the change in the natural oscillation frequency of the crystal and the content of corrosive compounds in a jet fuel when incubating for 1 hour and 4 hours, as well as the correlation between the method of the invention and the standardized test method;
Figure 3 shows how the frequency change and thus the corrosiveness is changed with the incubation period time;
Figure 4 shows how the method can be used to study the effect by various sulphur compounds, in this case mercaptans, on the corrosivity. Figure

4 shows the relationship between type and content of mercaptan and the corrosiv-ity in synthetic fuel with 1 mg/l elementary sulphur;
Figure 5 shows the dependence of the measured corrosivity on the con-centration of the activation solution in the pretreatment step, when activation by sodium sulphide is used. For every measuring point six crystals have been incubated for 4 hours at +50C in pure fuel with an addition of 1 mg/l of ele-mentary sulphur;
Figure 6 shows how the length of the activation time in the pretreat-ment step influences the measured corrosivity, when a 0.001 M sodium sulphide solution is used to activate the crystal. For every measuring point six crystals have been incubated for 4 hours at +50 C in pure fuel with an addition of 1 mg/l of elementary sulphur;
Figure 7 shows how piezoelectric crystals can be used in order to follow the corrosivity in jet fuel when being stored on a water bed. The lZ35~4 corrosivity has been followed for a period of 43 days with crystals activated by sulphide and by cyanide and incubated for 4 hours in the fuel for analysis according to example 6;
Figure 8 gives a comparison between crystals activated by sulphide and by cyanide according to example 6, when measuring the corrosivity in fuel, which has been added elementary sulphur in order to increase the corrosivity.
Incubation period 1 hour;
Figure 9 gives a comparison between crystals activated by sulphide and by cyanide according to example 6, when measuring the corrosivity in fuel, which has been added elementary sulphur in order to increase the corrosivity.
Incubation period 4 hours.
From Figures 5 and 6 it is evident, that the reaction between the sul-phide in the aqueous solution during the activation and the crystal occurs very rapidly and is rather an adsorption phenomena. Once a monomolecule layer or thin layer of sulphide on the surface has been adsorbed on the crystal surface, the following reaction occurs very slowly in an aqueous solution. Figures 7 - 9 show that the sulphide-activated crystals react more rapidly with corro-sive compounds in a fuel, such as elementary sulphur.
It is also possible to measure by means of the crystal without activa-ting it, but the incubation period must then be prolonged.
If one should wish to use a crystal, which has electrodes of a metal that is unsuitable for corrosion measurements, the electrode contact surfaces can be coated with a metal that is more suitable for the corrosion reaction. A
gold electrode can for instance be given a silver layer of a suitable thickness by means of thermal evaporation in an evaporation set-up at a low pressure.
After an initial activation, if necessary, of the commercially avail-79~4 able, metal-coated piezoelectric crystal or after evaporation of a suitable metal onto a piezoelectric crystal surface, the crystal is placed in the liquid product, whose corrosivity or reactivity is to be determined. The incubation in the test liquid is suitably carried out at a somewhat enhanced temperature, e.g. 50 C, but a lower temperature is also possible, for instance in order to lessen the evaporation of the test liquid. The length of the incubation period is to a certain extent adapted to how corrosive or reactive the examined test liquid is. If it is very corrosive, the crystal will soon become over-loaded because of the corrosion film, which causes the crystal to stop oscillating.
When the test liquid is only weakly corrosive, a longer incubation period of up to 4 hours may be required in order to note a larger frequency change. The in-cubation period for a silver-coated quartz crystal is usually between 15 minutes and 4 hours and is suitably 1 hour.
Conversion from change in frequency to change in mass can, for gas analysis with coated piezoelectric crystals, be done according to W.H. King Jr., Anal. Chem., 36 (9), 1735 (1964) in accordance with the relationship:

~F A
2.3 10 F
where ~W = change in mass on the electrode surface in gram ~F = change in frequency in Hz F = natural frequency in MHz A = electrode area in cm This relationship is also applicable at liquid analysis with coated piezoelectric crystals.
The invention will now be described in more detail in an illustrative but not delimiting purpose by means of a number of examples with reference to the attached drawings.

1~357~4 Example 1 __ __ _ _ __ _ Quartz crystals (Kvartselektronik A~, 20 000 ~Hz 20ppm hc18/u KVE) with contact surfaces of silver (put on by the manufacturer by means of eva-poration) are opened in order to expose the crystal. The crystal and the silver contact surfaces are etched in 0.1 molar sodiumcyanide solution for 10 seconds and rinsed in distilled water, 99% ethanol and isooctane, The natural frequency is measured, and then the crystals are incubated for 1 hour at 50 C
in 10 ml jet fuel of different degrees of corrosiveness according to the standard method. Then the cr7stals are rinsed in lsooctane and blown dry.
The natural frequency is measured again, and the change for every crystal is registered. The results from the measurements are reported in tables 1 and 2.
In the tables there are also reported the corrosiveness according to the standard method and the amount of elementary sulphur determined by means of a polarographic method for every fuel sample.

i~S7~4 TAEILE 1. Measurements of the corrosivity of jet fueL in some rock chambers installations.

Corrosion Corrosion Elementary 5Installation no. kMz (1 h) IP 227/73 sulphur -1 3.89 4.82 3.45 4.13 1 0.67 2 9.28 10.50 11.18 2+ 1.6 3 4.75 5.02 4.34 4.64 2+ 0.48 ( 10 4 2.38 2.14 2.03 2.62 0+ 0.14 TA~LE 2. Measurements of the corrosivity of stored fuel in a rock ( shamber installation.

15 Depth Corrosion Corrosion Elementary Standard sulphur rn kHz (1 h) IP 227/73 mg/l 2 0.16 0.21 0.24 0 0.2 4 0.25 0.25 0.19 0 0.3 20 6 2.92 2.77 2.86 1 0.6 8 4.87 4.79 4.84 2 1.0 4.99 5.61 5.89 2+ 1.2 12 6.64 6.51 5.89 2+ 1.2 ( 25Control 0.35 0.33 ( Fig. 1 shows graphically the result in table 2 from sampling at various depthsina cistern, when fuels are stored in rock chambers. From the diagram one can see the increased sensitivity of the method according to the invention compared with the standardized test method, where the result is expressed as an integer frorn 0 to 4.

Examele 2 In order to examine how the change in frequency depends on the corrosi-vity, a test series has been carried out, where highly corrosive fuel is 1~357~4 mixed with different contents of non-corrosive fuel. Crystals are etched and incubated according to example 1.

A mixture of corrosive fuel of classification 3 with fuel of classifi-cation 0 (according to the standard method) gives a relationship bet-ween the corrosivity and the percentage of the additive as is seen from fig. 2. It is also valuable to be able to quantitatively estimate how much of a non-corrosive fuel there is required if mixed into a corrosive fuel to make the corrosivity of the mixture acceptable from for instance a flight safety point of view. From the example is also seen the correlation between the crystal method of the invention and the standard method IP 227/73 for this fuel.
( xamele 3 Crystals according to example 1 are incubated for different lengths of time in a fuel of corrosivity classification 2 according to the stan-dard method, and the measured results are shown in the diagram of fig.
3. From the appearance of the curve is evident that a linear relation-ship between the corrosivity and the incubation time is obtained aftera certain period of time, here 1 hour.

Examele 4 ____ ____ The corrosivity of aviation fuel was determined after incubating crystals according to example 1 for 1 hour and 4 hours. The corrosivi-( ty was also determined by means of the standard method IP 227/73 and bymeasuring the content of elementary sulphur. From fig. 2 is seen how a longer incubation time can be used in order to be able to detect at an early stage tendencies of corrosivity.

Examele 5 The utility of the method as an analytical aid is seen in fig. 4, where the corrosivity has been determined after incubating crystals accordingto example 1 for 1 hour in fuel mixtures with various contents and ty-pes of mercaptans.

1~35744 -Examele 6 Quartz crystals (Kvartselektronik AB, 20 000 MHz 20ppm hc18/u KVE) with contact surfaces of silver (put on by the manufacturer by means of eva-poration) are opened so that the crystal is exposed. The crystal and the silver contact surfaces are heated in one case with 0.1 M sodiurn-cyanide solution and in the other case with 0.1 M sodiumsulphide solu-tion for 10 seconds and rinsed in distilled water, 99% ethanol and iso-octane. The natural frequency is determined, and then the crystals are 10 incubated in 10 ml jet fuel for 1 hour and 4 hours at +50 C. The crys-tal is then rinsed in isooctane and blown dry. The natural frequency is measured again, and the change is registered for every crystal. In ( tables3and 4 are shown the results of the measurements. In the tables there are also shown the corrosivity according to the standard method and the amount of elementary sulphur determined by means of a polaro-graphic method for every fuel sample.

TABLE 3. Determination of the corrosivity of jet fuel in a rock chamber installation with additions of elementary sulphur utilizing crystals that have been activated by cyanide and by sulphide.
The corrosivity has also been determined by means of the stan-dard method IP 227/73 as a reference. Inc = incubation time in the fuel, n = number of crystals, x = average value and ( s = standard deviation Amount of Corro- Corrosivity kHz added ele- sion Cyanide Sulphide Sulphide r,lentary stan- activation activation activation sulphurdard Inc 4 h Inc 4 h Inc 1 h S mg/lIP 227/73 n x s n x s n x s 0 0 24 0.425 0.133 6 0.637 0.083 0.1 0 - - - 61.026 0.148 0.3 0 5 0.245 0.093 6 4.236 0.308 1.0 0 6 0.511 0.057 6 12.428 0.526 6 4.90 0~215 The result shows that by means of sulphide activation, tendencies of corro-sivity can be indicated earlier than by means of cyanide activation. No ~Z3574'~

corrosivity can be detected by means of the standard method.

TABLE 4. Measurement of the corrosivity in fuel stored in a rock cham-ber installation and fuel brought to the installation in a tank car. Determination by means of piezoelectric crystals activated by cyanide solution (0.1 M) and sulphide solution (0.1 M). The corrosivity has also been determined by means of the standard method IP 227/73, andthe content of elemen-tary sulphur has been determined polarographically.

Sample Depth Ag-corr Ag kHz Ag kHz So m IP 4 h CN 4 h HS mg/l 227/73 n x s n x s Cistern 1 0 , 0+ 4 0.401 0.025 2 1.880 0.019 <0,2 2 0 , 0+ 4 0.450 0.044 3 0+, 0+ 4 0.526 0.087 4 0+, 0+ 4 0.565 0.023 <0.2 0+, 0+ 4 0.728 0.115 6 1-, 1- 4 0.543 0.114 2 2.346 0.111 <0.2 6.75 1-, 1- 4 0.560 0.227 2 2.699 0.452 7.25 1-, 1- 4 0.658 0.366 2 2.552 0.129 <0.2 _______________________________________________________________________ Average value 0+ 0.554 2.37 <0.2 ~, 25 Tank car 1 0 , 0 4 0.422 0.129 2 0.692 0.252 <0.2 2 0 , 0 4 0.482 0.142 2 0.702 0.212 <0.2 3 0 , 0 4 0.627 0.109 2 0.779 0.049 <0.2 4 0 , 0 4 0.637 0.115 2 0.470 0.011 <0.2 0 , 0 4 0.646 0~109 2 0.570 0.12 <0.2 ____________________________________ Average value 0 0.563 0.643 <0.2 The result of the measurements shows that by means of the standard method IP 227/73 a faint trace of corrosivity can be noted in the cistern but not in the tank car. By means of the crystal method uti-lizing cyanide activation, no appreciable difference can be noted, 1~3574~

but if one utilizes sulphide activation it is observed that the fuel in the cistern is more corrosive than the fuel brought in the tank car.

Examele 7 In order to examine how piezoelectric crystals can be used to discover early signs of corrosion when jet fuel is stored on a water bed, mea-surements have been carried out in a model system with water and fuel.
In this case a comparison has been done between corrosion determina-tions by means of sulphide-activated crystals (10 seconds in 0.1 M
su~phide so~ution) and cyanide-activated crysta~s (1û seconds in 0.1 M
cyanide solution) according to example 6. The incubation time has ( been 4 hours. The result is seen in fig. 7.

With sulphide-activated crystals/signs of beginning corrosivity can be discovered early, which is of great importance when jet fuel is stored on a water bed in a rock chamber. In this case signs of corrosion can be discovered already after 30 days.

Examele 8 In order to examine how the frequency change is dependent upon the corrosivity, two test series were carried out, and measuring by means ( of cyanide-activated crystals was compared with measuring by means of sulphide-activated crystals. The crystals were treated according to example 6. In one test series an incubation time of 1 hour was used ( (fig. 8), and in the other test series an incubation time of 4 hours was used (fig. 9).

In both test series elementary sulphur was added to non-corrosive fuel in order to give a fuel having an increasing corrosivity.

In both cases are evident that the sulphide-activated crystals are more sensitive than the cyanide-activated crystals and can prove corro-sivity at an earlier stage. The spreading of the values is also less,when sulphide-activated crystals are used.

574~4 Even if the invention has been illustrated by means of corrosion deter-minations in petroleum products, it is submitted that the invention is generally applicable for the analysis of the corrosiv-ity of a liquid.

Claims

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A method of measuring the corrosivity of a liquid, which method comprises the steps of:
(1) providing a piezoelectric crystal with an activated and reproducible coating of a material which can be subjected to the corrosion reaction in question;
(2) determining the natural oscillation frequency (v1) of the activated piezoelectric crystal;
(3) exposing the crystal to the corrosive liquid which is in the liquid state by immersing the crystal therein; and (4) determining the natural oscillation frequency (v2) of the crystal after being exposed to the corrosion reaction for a certain period of time;
whereby the change in the natural oscillation frequency (v1- v2) of the crystal is a measure of the mass change on the surface of the crystal and thus of the content of corroding compounds in the liuqid.

2. The method according to claim 1, wherein the piezoelectric crystal is quartz; the reproducible coating material on the crystal is a metal; and the liquid whose corrosivity is to be measured is a sulfur-containing liquid.

3. The method according to claim 1, wherein the piezoelectric crystal is quartz; the reproducible coating material on the crystal is silver; and the liquid whose corrosivity is to be determined is a sulfur-containing petroleum product.

4. The method according to claim 1, wherein on immersion of the crystal a corrosion product is formed which is soluble in the liquid and the decrease in mass caused by the corrosion of the coating material is determined.

5. The method according to claim 2 or 3, wherein the piezoelectric crystal is quartz; the reproducible coating material on the crystal is a metal;
and the liquid whose corrosivity is to be measured is a sulfur-containing liquid.

6. The method according to claim 1, wherein the change in the natural oscillation frequency of the piezoelectric crystal caused by the corrosion reaction is determined in situ in the liquid, whereby the dynamic course of the corrosion reaction is followed.

7. The method according to claim 2, 3 or 4, wherein the change in the natural oscillation frequency of the piezoelectric crystal caused by the cor-rosion reaction is determined in situ in the liquid, whereby the dynamic course of the corrosion reaction is followed.

8. The method according to claim 1, wherein step (1) comprises cementing the reproducible coating material onto the piezoelectric crystal or applying the reproducible coating material onto the piezoelectric crystal by means of thermal evaporation in vacuum; and then the crystal coated in this way is stored in an inert environment until the natural oscillation frequency is determined and the exposure to the liquid is carried out.

9. The method according to claim 2 or 3, wherein step (1) comprises cementing the reproducible coating material onto the piezoelectric crystal or applying the reproducible coating material onto the piezoelectric crystal by means of thermal evaporation in vacuum; and then the crystal coated in this way is stored in an inert environment until the natural oscillation frequency is determined and the exposure to the liquid is carried out.

10. The method according to claim 4 or 6, wherein step (1) comprises cementing the reproducible coating material onto the piezoelectric crystal or applying the reproducible coating material onto the piezoelectric crystal by means of thermal evaporation in vacuum; and then the crystal coated in this way is stored in an inert environment until the natural oscillation frequency is determined and the exposure to the liquid is carried out.

11. The method according to claim 1, wherein step (1) comprises activating the metal layer of a previously metal-coated piezoelectric crystal and cleaning it before the corrosion determination.

12. The method according to claim 2 or 3, wherein step (1) comprises activating the metal layer of a previously metal-coated piezoelectric crystal and cleaning it before the corrosion determination.

13. The method according to claim 11, wherein the metal coating on the crystal is silver; and the activation comprises etching the metal layer with a cyanide solution whereby the metal layer is freed from existing sulphide and oxide coatings.

14. The method according to claim 13, wherein the piezoelectric crystal is quartz and the silver layer of a silver-coated quartz crystal is cleaned by etching it for 10 - 30 seconds, with a 0.01 - 1 M, sodium cyanide solution saturated with oxygen.

15. The method according to claim 11, wherein the activation comprises treating the metal layer of the crystal surface with a sulphide solution, where-by the metal layer is coated with a thin reproducible sulphide layer.

16. The method according to claim 15, wherein the sulphide-activation is carried out with a 0.01 - 1M, sodium sulphide solution for 10 - 30 seconds.

17. The method according to claim 1, wherein the piezoelectric crystal coated with a suitable metal, is after activation placed in water or a liquid state petroleum product, whose corrosivity is to be determined.

18. The method according to claim 17 wherein the piezoelectric crystal is a quartz crystal; the coating metal is silver, copper or aluminum; the liquid state petroleum product is aviation fuel; and then the increase in mass of the metal coating caused by the corrosion reaction is determined.

19. The method according to claim 17, wherein the corrosion reaction pro-duct is cleaned away from the metal coating and then the decrease in mass which is directly proportional to the content of corroding compounds in the petroleum product and is caused by the corrosion reaction is determined.

20. The method according to claim 1, 11 or 13, wherein a corrosive gas, whose corrosivity is to be determined, is taken up in a liquid, which by itself is non-corrosive, then the corrosivity of the thus-treated liquid is determined.

21. A method of measuring the corrosivity of a liquid petroleum product that contains at least one sulfur compound, which method comprises:
(1) providing a piezoelectric quartz crystal with a reproducible coating thereon of a metal selected from the group consisting of copper, silver and aluminum;
(2) determining the natural oscillation frequency (v1) of the thus-coated piezoelectric crystal;
(3) exposing the crystal to the petroleum product which is in the liquid state by immersing the crystal therein under such conditions that a corrosive reaction takes place between the coated metal and the sulfur compound in the petroleum product resulting in a change of the mass of the coated metal layer;

(4) determining the natural oscillation frequency (v2) of the crystal after the exposure to the corrosive reaction;
whereby the change in the natural oscillation frequency (v1 - v2) of the crystal is measured which corresponds to the content of the sulfur compound.

22. The method according to claim 21, wherein step (1) comprises etching the metal coating layer of a previously silver-coated piezoelectric quartz crystal in a 0.01 - 1 M sodium cyanide solution saturated with oxygen whereby the metal layer is activated.

23. The method according to claim 21, wherein step (1) comprises treating the metal-coating layer of a previously silver-coated piezoelectric quartz crystal in a 0.01 - 1 M sodium sulphide solution whereby the metal layer is activated.

24. The method according to claim 21, 22 or 23, wherein the liquid petroleum product is fuel.

25. The method according to claim 21, 22 or 23, wherein the liquid petro-leum product is aviation fuel.

26. A device for measuring the corrosivity of a liquid, which comprises a crystal detector comprising a piezoelectric crystal, encapsulated in an inert environment and provided with a coating layer, thereon of a metal which is capable of functioning as an electrode and can be subjected to corrosion, wherein the coating layer has been pretreated for cleaning and activation and, the crystal can be decapsulated before measuring the corrosivity of a liquid.

27. The device according to claim 26, in which the detector is a silver-coated quartz crystal encapsulated in an ampoule made of glass and filled with an inert gas, wherein the silver-coating is, before the encapsulation, freed from sulphide and oxide contaminations by means of an etching cleaning process with a cyanide solution saturated with oxygen and the crystal can be decapsulat-ed before measuring the corrosivity of a liquid.

28. The device according to claim 26, in which the detector is a silver-coated quartz crystal encapsulated in a glass ampoule filled with an inert gas, wherein the silver-coating is, before the encapsulation, activated with a sulphide solution and in this way given a thin, reproducible sulphide layer and the crystal can be decapsulated before measuring the corrosivity of a liquid.