How razor degradation and corrosion occur

Corrosion in closed water circuits

Circuits carrying water for the provision of heat and cold in buildings are subject to the risk of corrosion. There is a risk of functional restrictions and losses in efficiency up to total failure. For a better understanding of the damage cases, 64 systems in 21 buildings were examined in terms of water chemistry in the EnOB research project “EQM Hydraulics” 1). In addition, a newly developed sensor system is being tested, which indicates corrosion processes before damage occurs. More than half of the systems examined were affected by corrosion and in some cases also damaged. The cooling systems showed a higher frequency of damage than heating systems.


Circuits carrying water for the provision of heat and cold in buildings often show signs of corrosion. Typical signs are black or brown colored circulating water, sedimenting constituents in the circulating water, jammed valves, pumps failing prematurely or blockages of heat exchangers or other system parts. Costly mechanical or chemical rinsing is required to repair the damage, which companies often offer as a product-service combination with subsequent water treatment with a corrosion inhibitor. The track record of such treatment after the appearance of corrosion is mixed. If the system has already been damaged, inhibitor treatment does not always lead to the desired minimization of the rate of corrosion and deposit formation. How corrosion damage could have occurred in individual cases or how this could have been avoided remains unclear. Existing technical rules do not adequately cover all applications. For planning, commissioning and operation, standards and regulations are available to avoid corrosion in water cycles. These include

  • VDI 2035 Part 2 Avoidance of damage in hot water systems, [1]
  • DIN EN 14868 (2005) Corrosion protection of metallic materials - guidelines for determining the probability of corrosion in closed water circulation systems, [2] and
  • since April 2017 the BTGA rule 3.003 closed water-borne cold / or. Cooling water circuits in buildings - Reliable operation under water technology aspects [3].


In all of the rules, special attention is paid to avoiding the introduction of oxygen and limiting the salt content - especially when using copper and steel in combination. VDI 2035 Part 1 [4] also limits the content of hardness components to prevent stone formation in heating systems. Although VDI 2035 Part 2 specifies the most extensive requirements for preventing corrosion in water circuits, its validity is limited to hot water systems. For cooling systems, the BTGA rule 3.003 has only recently come into existence that are much more detailed than those of VDI 2035 Part 2.

The BTGA rule 3.003 prescribes a water analysis as part of the annual inspection to monitor the circulating water in order to guarantee the requirements for water quality and to identify acute corrosion processes. Continuous monitoring of the pH is recommended. A corrosion coupon can be used as a further monitoring method. The corrosion in this part of the system is shown on a workpiece brought into the water cycle. The coupon reacts in particular to oxygen, the pH value and corrosion-promoting water components. With the coupon method, local corrosion attacks at other points in the system, e.g. B. under biofilms, however, not recognized. Possible causes of corrosion cannot be determined by the coupon alone.

Therefore, it is also the aim of the research project to develop tools and methods for quality management to prevent corrosion. The focus is on the FeQuan sensor system, which continuously collects water-chemical parameters and also calculates the current corrosion rate [5] [6]. The dissolved iron (II) content is calculated from the combination of four individual measurement parameters. The method has so far only been tested under laboratory conditions [5]. In the research project, the application in the building sector is being tested for the first time. Open questions relate to long-term stability (contamination) and malfunctions when the sensor measuring system is permanently installed in the real system.

Different types of heating and cooling systems as well as combined systems in large non-residential buildings from the class of office and administration buildings are examined. The focus of the research project is the field study on 21 buildings with 64 systems. The field study is divided into three phases:


I. 64 circuits were subjected to a punctual corrosion investigation (recording of the system technology, on-site water chemistry measurements and water chemistry laboratory analyzes).

II. Eight circuits for energetic and corrosion monitoring were selected from the system pool and the "FeQuan sensor" was permanently installed in the system and connected to an online evaluation. In addition to the field test of the sensor method, this serves to prepare for the investigation of remedial measures.

III. Three already damaged systems in the buildings equipped with a FeQuan sensor are being renovated under scientific supervision. Different remedial measures are tested and the effectiveness of various measures is assessed.

Subject of investigation and analysis methods

To investigate the corrosion processes in water cycles, on-site analyzes and water-chemical and biochemical laboratory analyzes were carried out.

On-site water chemistry analyzes and sampling

The parameters oxygen, pH value, conductivity, temperature and redox potential were determined with the help of sensors in a mobile, flow-through measuring unit (phase I), in phases II and III measurements were made in a permanently installed fitting. The measuring unit is connected to the system to be examined with corrugated stainless steel hoses so that there is no contact with the ambient air. The system status with regard to its tendency to corrosion is determined from the values. The content of dissolved iron (II) is calculated from the sensor values ​​[5]. From phase II, the real-time data from the sensor system of the eight building circuits was transferred to a server. The sensor values, the calculated iron (II) content and the corrosion rate derived from it are displayed online on a dashboard and can be viewed via an https address.

Dashboard for monitoring the system status Image: Zagari

For the laboratory analyzes, several water samples were taken from each district to determine the ingredients listed in the table below.

Laboratory analysis

If large quantities of particulate corrosion products are present (visual inspection), a sample is also taken. Depending on the filling volume of the system, either a different, higher-lying tap is selected, or the system water is drained until an optically homogeneous sample is reached. If microbiologically induced corrosion is suspected, samples are taken for a genetic profile analysis to determine the germs that occur.

Water chemistry laboratory analyzes

The corrosion products contained in the samples (iron, zinc, copper and other alloy metals) are determined in the laboratory by means of element-analytical methods. To differentiate between dissolved and undissolved corrosion products, filtered (0.2 µm syringe filters) and unfiltered samples are examined. To determine the risk of lime deposits, the content of hardness components is determined. In addition to stone formation, calcium and magnesium are also microbiologically relevant. [7]

In addition to the metal cations mentioned, sulfate and nitrate (SO42-, NO3) as nutrients promote microbial growth and thus microbially induced corrosion. Chloride attacks protective layers from corrosion products and thus enables further corrosion. In addition, the content of inorganic carbon and organic carbon as non-blown carbon (NPOC) is determined.

In general, high concentrations of the salts mentioned increase the conductivity of the system water, which can accelerate electrochemical corrosion. The individual analyzes are carried out according to Table 1.

Biochemical laboratory analysis

Microbiological processes have a direct influence on the risk of corrosion. So z. B. microbiological activities lower the pH. Under favorable circumstances, the circulating water should be sufficiently alkalized by initial corrosion processes to protect against further corrosion. The activity of sulfate-breathing, fermenting and nitrogen-fixing bacteria, however, creates acids that prevent the protective alkalization of their own.

The sequencing (polymerase chain reaction of the 16S rRNA gene segment) makes it possible to compare the genetic code of the bacteria from the sample with the sequences already available in special databases and thus to draw conclusions about the population composition in the respective heating or cooling system.

Assessment scheme

The results of the water chemistry analyzes are assessed on the basis of defined limit values. These are based on already known limit values ​​of VDI 2035 or an adjustment was made based on experience in the course of the project. The most important value is the content of dissolved iron, which indicates active corrosion processes. In addition, the measured on-site parameters and salt content play a decisive role in assessing the rate of corrosion and the likelihood of corrosion. If more than one parameter was slightly increased, dissolved or particulate metals, or the pH value was too low, the rating “critical system status” was assigned, while at the same time greatly increased metal contents the rating was “poor system status”. The following evaluation criteria were determined according to the following table.

Evaluation criteria for assessing the system status


The results of on-site and laboratory analyzes are presented below.

On-site analyzes

The iron (II) values ​​determined from the on-site values ​​using the FeQuan method were compared directly with the values ​​determined by laboratory analysis. Even in phase I (point measurements on 64 systems), neither false negative nor false positive results were achieved by the FeQuan sensor. That means: if there were fresh corrosion products in the circulating water, the sensor indicated this. If the system status was good, the sensor also reliably indicated this, as the subsequent laboratory analysis confirmed. Phase II provided more points for comparison.

A building heating circuit with continuous iron (II) measurement (FeQuan sensor) and laboratory-analytical point measurements (red points) as reference. Image: Zagari

In the next picture, the iron content determined by the sensor is compared to the iron content determined by laboratory analysis. The value of the FeQuan sensor shows a high level of agreement with the laboratory values ​​over a wide range.

Iron (II) content with the FeQuan sensor and through laboratory analyzes. (1) increased values ​​shortly after filling, (2) microbiologically influenced water changes, (3) work on sensors, (4) investigation of the microbiology. Image: Zagari

The increased iron content during commissioning is exaggerated due to the formation of more reactive complexes (cf. [5]), as well as water changes caused by microbiological activity. The analytical iron concentrations only increase as a result of the drop in pH value and the microbiological colonization and then agree again with the sensor values.

Water chemistry laboratory analyzes

The elemental analysis showed that some of the cooling systems were filled with softened water. Softening does not reduce the chloride content of the water. The following comparison of the content of dissolved iron and the chloride content was obtained from the measured values: From a content of 15 mg / l chloride in the circulating water, cases with high iron contents increase significantly. Below this value there is no significant corrosion.

Another influencing factor is the pH value. A pH value <8.5 in addition to a chloride content> 15 mg / l leads to a higher iron content than in systems with a pH value> 8.5.

A linear relationship between oxygen and iron content cannot be seen.

Dissolved iron content depending on the chloride content of the circulating water. Image: Zagari

If most systems with an oxygen content between 0.01 and 0.1 mg / l are excluded, the expected relationship between oxygen concentration and corrosion rate becomes apparent, but this only applies to very high oxygen concentrations.

Biochemical laboratory analysis

According to an online survey carried out in advance, the respondents hardly consider microbacterial processes as a cause of corrosion. Nevertheless, around 800 different species of germs have been identified. The identified groups can be subdivided based on their metabolic properties as follows:

  • Sulphate reduction and iron oxidation
  • Nitrate reduction, nitrogen fixation
  • Biofilm former
  • Fermenting, acetogenic, sometimes complex, also halogenated organic degrading organisms (degradation of inhibitors / biocides possible)


Corrosion processes

The found cooling systems filled with softened water show that in the absence of a suitable regulation for cooling systems, the specifications of VDI 2035 (for heating systems) were applied. Softening is used in heating systems to prevent stone formation. Stone formation occurs when water is heated that contains hardeners and hydrogen carbonate [4]. The advantage of this measure does not apply in cold systems, however, since no heating takes place. The conductivity and the availability of nutrients for microorganisms remain unchanged through softening.

The influence of chloride on the tendency to corrosion is clear: chloride is the clearest monocausal factor influencing the tendency to corrosion. If high chloride contents above 15 mg / l meet low pH values ​​below 8.5, there is a further increase in the tendency to corrosion. The influence of the measured oxygen content of the circulating water on the probability of corrosion is not as important in the lower concentration range as the priority setting in the relevant rules would lead one to expect. Very high oxygen contents> 1 mg / L certainly lead to corrosion, but even lower oxygen contents <0.1 mg / L can lead to equally severe corrosion.

Dissolved iron content depending on the oxygen content of the circulating water. Image: Zagari

In the range between 0.01 mg / l and 0.1 mg / l, in which most of the circulating water is found, there is no obvious (linear) relationship between the oxygen content and the iron content. The reason for the scatter in this area could be the different salt contents of the water. A low oxygen content therefore does not guarantee freedom from corrosion. On the other hand, high oxygen contents above 1 mg / l certainly promote corrosion. In the frequently encountered area in between, other influencing factors such as the salt content, presumably especially the chloride content, seem to be more important.

Influence of microbiology and case studies

Nitrate- and sulphate-reducing bacteria influence the corrosion of ferrous materials. In addition to limiting the chloride content, this speaks in favor of full demineralisation of the fill water (deionized water). Bacteria are deprived of their nutrient base due to the lack of salts. Inhibitors and residual concentrations of glycol without sufficient biocide dosage also serve as a nutrient basis for germs. The most affected systems found in the field study often contained circulating water with inhibitor or glycol residues from previous treatments.

The following figure shows the water-chemical behavior in a test system on a laboratory scale (system volume 3 L) after commissioning.

Good alkalinization with subsequent microbiological growth. Image: Zagari

The pH value rises from 7.5 to 9.2 within two months, while the oxygen content falls slightly from the originally saturated ratios to 50 µg / L due to consumption by the initial corrosion processes. In the following three months, the temperature decrease leads to a corresponding increase in the oxygen concentration from 50 to 68 µg / L within 1.5 months. After about a month the pH begins to drop and after two more months it falls below pH 9.0. The microbiological profile analysis shows the nitrogen-fixing genus Azospira including the nitrate-dependent iron oxidizer Acidovorax. Both occur in the soil and groundwater and are adapted to temperatures of 10-18 ° C [3]2).

Bad alkalization (oxygen ingress and germs). Image: Zagari

The picture above shows the water-chemical behavior in a system with constant oxygen ingress and contamination. The pH rises within a short time to pH 9.3 and then falls steadily to pH 8.4. The microbiological analysis shows the aerobic biofilm former as the dominant species Pseudomonas and the nitrogen fixer Bradyrhizobium [6].

Assessment of the corrosion status

The majority of the systems found are in at least a critical condition from a corrosion point of view, so that there the probability of corrosion damage is significantly increased. The proportions of the systems rated as good, critical and poor are roughly evenly distributed, with the cooling systems being less often in good condition than the heating systems.

Evaluation of the examined systems. Image: Zagari

In addition to the evaluation of the system status, the previous treatment of the fill water is included in the evaluation on the basis of water-chemical parameters. It is noticeable that half of the cooling systems were filled with softened water. A small part of the fill water is fully desalinated.

Treatment of the fill water Image: Zagari

No obvious relationship can be found between the type of water treatment and the system status. It has only been shown that the systems with fully desalinated water are without exception in good condition, which, however, cannot be considered significant due to the low number of cases (n = 4).

Conclusion and outlook

Cooling systems are more often affected than heating systems because they are filled with unsuitable water. Until recently, this is caused by unclear or missing rules for filling cooling systems. The chloride content of the filling water has proven to be a decisive influencing factor on the tendency to corrosion. Too low a pH value caused by sulfate-reducing or nitrogen-fixing bacteria in combination with an excessively high chloride content further increases the tendency to corrosion. In order to effectively limit the chloride content and other salts as a nutrient basis for germs, it is advisable to fill and top up with deionized water for both heating and cooling systems. Alternatively, site water that is low in salt, untreated or mixed with demineralized water can be used. The BTGA rule 3.003 contains the corresponding requirements for the filling water, in which, in addition to the limitation of the chloride content below 15 mg / l, further research results are included. In addition to a suitable filling water, the continuous water-chemical monitoring of the circulating water is an effective means of avoiding damage caused by corrosion.

Monitoring the circulating water is the most suitable method to uncover corrosion risks before damage occurs. Checking the pH value, as recommended by VDI 2035 and BTGA rule 3.003, is the simplest method. Instead of an annual check as recommended in the rules, a continuous check is desirable because adverse changes are recognized earlier and cause detection and elimination is easier than with an annual inspection.

Initial experiences with the FeQuan sensor in practical tests are positive. The sensor values ​​and the calculated corrosion rate are well suited for assessing a system, especially in connection with changes in the operating conditions or the monitoring of measures to avoid corrosion, such as flushing or refilling.

1) Energy and quality management - corrosion in hydraulic systems, grant number 03ET1270B,

2) The increase in conductivity is partly due to the experimental setup: with the low system volume, the slight release of KCl from the pH and redox sensors - approx. 10 mg per month, depending on the ionic strength of the water - is noticeable. This is hardly measurable even in smaller building systems.




[1] (VDI), Association of German Engineers. VDI 2035, sheet 2: Avoidance of damage in hot water systems. Berlin: Beuth Verlag, 2009.

[2] (DIN), German Institute for Standardization. DIN EN 14868 (2005): Corrosion protection of metallic materials - Guidelines for determining the probability of corrosion in closed water circulation systems. Berlin: Beuth Verlag, 2005.

[3] (BTGA), Federal Industry Association for Technical Building Equipment e. V. Closed BTGA rule 3.003: Closed water-borne cold / or. Cooling water circuits in buildings. Reliable operation in terms of water technology. Bonn: s.n., 2017.

[4] (VDI), Association of German Engineers. VDI 2035, sheet 1: Stone formation in drinking water heating and hot water heating systems. Berlin: Beuth Verlag, 2005.

[5] Opel, Oliver. Detection of ocher processes, corrosion and deposit formation - monitoring of iron oxidation in wells, buildings, pipes and technical systems using the redox potential. Saarbrücken: SVH Südwestdeutscher Verlag für Hochschulschriften, 2013.

[6] Wiegand, Marlies. Investigations into the corrosion behavior of closed hydraulic systems of modern installation systems with the help of water and electrochemical analysis of released metal ions. Lüneburg: Unpublished university publication, 2017.

[7] Fritsche, Olaf. Microbiology. Berlin Heidelberg: Springer-Verlag, 2016.

From Prof. Dr. Oliver Opel, Dr. Stefan Plesser, Marlies Wiegand and Dipl.-Ing. Mani Zagari

Prof. Dr. Oliver Opel, University of Applied Sciences West Coast, has been researching corrosion and ocher processes in water-bearing systems for the provision of cold and heat for 15 years. Stefan Plesser, Deputy Managing Director of SIZenergie + and Head of Energy and Quality Management, SIZenergie +, Energy and Quality Management, Braunschweig. MSc. Marlies Wiegand, Leuphana University Lüneburg, Faculty of Sustainability. Works in the research project EQM.Hydraulik, focus on water chemistry and microbiology. Dipl.-Ing. Mani Zargari, SIZenergie +, Energy and Quality Management, Braunschweig. Heads the research project EQM: Hydraulics with a focus on operating boundary conditions.