In everyday inspection work you will never see a blind or almost blind inspector visually checking the condition of a coating system. However one of the most frequent used tests methods can best be compared with this almost blind inspector. TQC took a closer look at the Soluble salt measurement with the use of Bresle patches. This study indicates there is a lot of confusion about interpretation of test results. Accompanied by many possible errors in the used procedures and equipment. These factors often lead to discussions and serious quality problems.
Everybody in the coating industry knows that salt contamination underneath a coating can cause serious problems in future years. This is due to the hygroscopic nature of salt. The tendency to attract water in combination with the permeability of a coating creates a accumulation of water molecules between the substrate and coating. The presence of these water molecules together with the entrapment and migration of oxidation agents, are ideal to create an electrochemical shift causing corrosion in conjunction with the present salt molecules. Blasting or mechanical cleaning will not remove these salt molecules completely and often causes chloride inclusion in the substrate making the situation even worse. Washing the surface with deionised water is the most used solution. A substrate free of soluble salts is critical in today’s protective coating jobs and is an issue in each professional paint specification. IMO PSPC regulations are setting the maximum concentration of soluble salts, measured as Sodium Chloride, on a surface to 20 mg·m-2.
The principle of a Bresle test
When performing the soluble salt test, water is injected in a patch that is placed on the surface. This injected water dissolves the salt present at the surface. The solubility in water depends on the type of salt. Common salt also called Sodium Chloride can be dissolved in cold water to a concentration of 357 g∙l-1. Not only solubility differs between salts but also the conductivity. When taking a measurement not only common salt is dissolved but also all other salts present on the surface. This mixture of salts is eventually measured with a conductivity meter or by other means.
Misunderstanding of what is actually measured
Because it is impossible to predict which salts are present at the surface an assumption is made in the Bresle method. The term “measured as Sodium Chloride” indicates that this mixture of salts is interpreted as Sodium Chloride. Clearly indicating how the conductivity is interpreted is essential when creating a report. At present there are several interpretations in use. Some speak about sodium Chloride, other mention mixed salts or just Cholides, each having a different calculation factor.
The nominal volume in the test chamber of the TQC Bresle patch is 2.5 cm3. Considering the volume and solubility of salt it is possible to dissolve 892,5 mg of common salt in the patch. This correlates to 7.29·105 mg·m-2sodium chloride. Comparing this to the IMO regulation of 20 mg·m-2 there is a factor of approximately 36000 between these concentrations. The solubility of salt is not an issue when conducting the test. A level of 20 mg·m-2 sodium chloride results actually only in 0.025 mg sodium chloride in the patch. Even salts that are harder to dissolve will be present in such concentrations that these will not provide any solubility problems. Possible concentrations where the solubility of these salts will cause problems correspond to contaminations on the surface that won’t pass any guideline by a factor of 100 times.
Contrary to solubility is dilution a major cause for possible errors. In order to make it possible to measure the soluble salts with an electronic conductivity meter usually a volume of 15 ml. sample liquid is required to submerge the instrument’s probe fully. Since the actual volume of sample liquid in the Bresle patch is only 2.5 ml it means that the final result has to be multiplied by a factor 6. Any errors or that were made during certain stages of the test will be multiplied by a factor 6 as well.
Affect of dilution on the test results.
The average residue of 0.15 ml testing liquid remaining in the patch and the inaccuracy and improper use of syringes are some causes for errors but the majority of the problems are caused by diluting the sample liquid as it is often done in the 15ml cup. Good analytical practice learns that the number of steps required to obtain a final test result has to be limited as much as possible. Dilution to 15ml was required in the past to create sufficient quantity of
sample solution to submerge a conductivity probe and to prevent extreme static disturbance from the plastic measuring beaker. All conductivity gauges on the market are influenced by this static disturbance. This can lead up to a difference of 5 µS·cm- 1 per measurement. Diluting the sample liquid by a factor 6 implies automatically that the test result has be multiplied by a factor 6 as well. In practice this means that each deviation or error will by multiplied by 6. The 5 µS·cm- 1 mentioned above could end-up in a 30 µS·cm- 1 error! New techniques make it possible to measure in smaller samples using the Direct Sample Procedure or DSP.
During the evaluation of the study results the need for a higher accuracy proved to be
a hot issue. The accuracy can be increased in two ways. First by taking a closer look at the gauge. Previous available handheld or mobile conductivity gauges have a resolution of 1 µS·cm-1, with an accuracy of 1 µS·cm-1. Calculation according to ISO 8502-6 means that the final result has a resolution of 6 mg·m-2, with also an inaccuracy of 6 mg·m-1. When a measurement result is 18 mg·m-1 soluble salts measured as sodium chloride the actual value fluctuates between 12 and 24 mg·m-2. Leaving an 33% chance that the actual soluble salt
concentration is above the limit of 20 mg·m-2. Increasing the gauge’s resolution to 0.1 µS·cm-1 contributes to a higher accuracy when determining the soluble salt concentration. This however is only one part of the analysis.
Besides gauge resolution, dilution also influences the measurement. The earlier mentioned 0.15 ml of residue remaining in the patch causes an error up to 5% in the 15ml diluted solution. When this dilution is not applied and the measurement is made directly on the pure solution from the patch the 0.15ml residue will not affect the final result. New gauges can already measure in 2 ml solution with a resolution of 0.1 µS·cm-1. When measuring in a volume of 2.5ml, the same as the nominal volume of the patch, there is a significant change in calculation factor. Use of a 2.5ml sample leads to the elimination of the normal calculation factor 6. The concentration of soluble salts measured as sodium chloride is equal to the conductivity in µS·cm-1. This not only makes the determination easier but also more reliable. Results can now be given with an 1 mg·m-2 uncertainty and resolution of 0.1 mg·m-1. Increasing the accuracy 60 fold.
TQC Direct Sample Procedure
The new Direct Sample Procedure eliminates the use of the 15ml measurement solution. Measurements can now directly be made in the solution that is extracted from the
patch, eliminating the dilution step. This not only increases efficiency but also eliminates the most error sensitive part of the old procedure. To achieve this there is only 2.5 ml of deionised water injected in the patch. This also reduces the calculation factor to 1. The reading from the gauge doesn’t have to be multiplied anymore to get the soluble salt measured as sodium chloride concentration in mg·m-1. Due to the measurement in the gauges own measuring cell all static disturbance is also eliminated. Increasing the reliability of the analyses even further.
There is a great difference between the variable soluble salt tests kits on the market. Not only the gauge but also the patches differ in quality. A test patch should be as clean as possible. Any salts that remain on the patch during its production process influences the test significantly. Some of the round patches which can be found in the market contribute significantly to the final measurement. During tests these inferior patches contribute on average 0.7 mg·m-2 soluble salts measured as sodium chloride per patch. High quality patches such as the latex membrane square patches, don’t contain any salt residue. These patches pass multiple wash cycles in a clean room quality production plant to ensure that no contaminants are present.
The ISO 8502-6 standard subscribes in annex A that only certified patches may be used. This annex describes a stress test to ensure patch adhesion and wash ability. In ratio to the nominal volume of the patch it has to be injected with an excess of water. Time to leakage has to be determined and eight out of twelve patches must pass in order for the type of patch to be approved. This test must be carried out by an accredited laboratory and the producer must be able to provide a certificate of the test. The high quality patches have passed these tests. Most inferior round patches fail this test by 100%, only one third of the required volume can be injected in the patch, before leakage starts. When measurements are taken during arbitration using non certified patches, all acquired value’s will be useless. Only certified patches may be used. Some patches also face problems with poor and irreproducible adhesion making the test surface irregular. Often 20% extra surface area is exposed due to the fact the water crawls under the edges of the patch. This value is not corrected and causes even bigger errors in the final results. All errors caused by using inferior patches lead to higher results which add up to each other usually generating a significant higher and erroneous result.
Each generated soluble salts report should include climate conditions and substrate temperature. The ISO 8502-6 standard demands that the test has to be done at 23°C and a relative humidity of 50%. Any deviation from the defined parameters has to be reported and agreed upon by both inspector and customer. Surface temperature also influences the test, meaning that this parameter also needs to be recorded. During arbitration the lack of these recorded value’s also will render the acquired results invalid. Although the above show there is a lot of “science” behind proper testing for salt contamination inspectors can benefit from readily made test-kits which are available on the market making these kind of tests quite simple.
The recently updated TQC Bresle kit is the first to support this new technique and enables inspectors not only to work faster but also produce more accurate results. The combination of the new technique, high quality gauges and patches makes the kit the ultimate inspection set for arbitration.