Advanced search
1 file | 25.06 MB Add to list

Strategies for the design and predictive maintenance of heat exchangers exposed to a geothermal brine

Willem Faes (UGent)
(2020)
Author
Promoter
(UGent) and (UGent)
Organization
Abstract
The increasing demand for heat and electricity during the past decades has globally caused an increase in carbon dioxide concentrations (CO2). This contributes to global warming and, if continued, can have disastrous consequences for the planet. Therefore, a larger share of renewable sources of energy is necessary. Currently, hydropower has the biggest share, followed by wind and solar power. One additional alternative is to use the energy that is available as heat contained in or discharged from the Earth’s crust. This geothermal energy flows from the interior of the Earth to the crust and has two main contributors: residual primordial heat (originating from the formation of the planet) and radiogenic heat (caused by the radioactive decay of different isotopes). People have been using geothermal energy for centuries. While, for a long time, geothermal energy was mainly used for bathing and washing, there now globally is an installed capacity of 12.7GW. In Belgium, geothermal energy is mainly limited to shallow and medium-deep geothermal applications, where buildings are being equipped with ground-source heat pumps (GSHPs). Additionally, two shallow geothermal power plants are installed in Saint-Ghislain and Douvrain that are used for district heating and biogas production. However, recently, the Flemish Institute of Technological Research (VITO) drilled three wells up to a depth 3.5 km on the Balmatt-site in Mol. The geothermal doublet was finished in 2018 and a geothermal brine with a temperature up to 130 °C can be obtained at the well head. A geothermal power plant was constructed to provide the VITO-buildings and the neighbouring cities with heat and to generate electricity with an organic Ranking cycle (ORC). In this power plant, a heat exchanger is installed to transfer the heat from the geothermal brine to a secondary fluid. Since the brine has a salinity of approximately four times the salinity of seawater, it is highly corrosive. Therefore, the heat exchanger was constructed with a highly alloyed stainless steel, having a strong impact on the profitability of the geothermal project. One alternative approach could be to construct the device of carbon steel, which has a lower material cost and a better machinability. Unfortunately, this material will be susceptible to corrosion and the heat exchanger might need to be replaced several times over its lifetime. It is however not a priori known how high the corrosion rate will be and if the degradation of the metal will have an influence on the thermohydraulic behaviour of the heat exchanger. In this dissertation, the corrosion behaviour of metals in the brine and the influence on the performance of the heat exchanger will be studied and the economic viability of employing a device, constructed of a corroding material, will be examined. First, an overview will be given of the available literature on corrosion and corrosion prevention in heat exchanger. Ample examples can be found of failure analyses where a heat transfer device suffered from (one of the various forms) of corrosion. Only a minority of the failures are caused by uniform corrosion. Localised forms, especially pitting and stress corrosion cracking, are most frequently occurring. It appears from literature that heat transfer can influence the corrosion process, mainly by creating a temperature gradient causing differences in solubility of the elements in the fluid. Subsequently, the corrosion products can influence the performance of the heat exchanger by forming a layer with a low thermal conductivity and an increased surface roughness. This can cause a decrease in the heat transfer rate and an increase in pressure drop. Several methods of preventing corrosion in heat exchangers are described as well. An adequate material selection and control of the environment are crucial. Additionally, attention should be paid to avoiding crevices and the highest heat transfer coefficient should preferably be located on the cold side. To evaluate the behaviour of metallic materials in the geothermal brine, two types of experiments are preformed: static and dynamic corrosion tests. With the static tests, several steels are exposed to the brine for an extended period of time. By measuring the mass loss after exposure, a uniform corrosion rate can be calculated. Carbon steel exhibits corrosion rates up to 0.3 mm/y, while the mass loss of the stainless steel samples is negligible. They can however still be susceptible to localised form of corrosion (like pitting or crevice corrosion). Therefore, also electrochemical techniques are applied to determine the open circuit potential, the critical pitting (or crevice) potential and the repassivation potential. From these measurements, it appears that all evaluated materials are, to a certain extent, susceptible to localised corrosion, and that, if a crevice is present, repassivation can become a problem. An interesting observation made with the exposure tests, is that the corrosion rate of the carbon steel decreases over time. Additionally, it is lower for elevated temperatures, although the Arrhenius equation would suggest otherwise. This phenomenon was investigated by an analysis of the corrosion products. Apparently, a protective layer of iron carbonates (FeCO3) is formed. Since the solubility of FeCO3 is lower at higher temperatures, this layer is more protective at these temperatures. Since in reality, the brine will not be static, an experimental setup was constructed to allow dynamic corrosion tests. During these experiments, the brine flows inside a carbon steel tube that is subject to heat transfer. The heat transfer rate and pressure drop are being monitored and small tubular samples allow to determine uniform corrosion rates. The corrosion rates obtained under flowing conditions are approximately 60% higher than those under static conditions. During the execution of the measurements, the pressure drop remained unchanged. The heat transfer rate demonstrated an initially increasing trend (with an increase of almost 6%), followed by a decrease. Although this needs to be confirmed by longer exposure periods, this could be due to the build-up and removal of a layer of corrosion products, influencing the turbulence inside the tube. To examine the economic viability of the proposed alternative, a heat exchanger design model was developed. Although models exist in literature to minimise the total cost of ownership (TCO) of heat transfer devices, no publications were encountered where the surfaces are degrading over time. Therefore, in addition to a thermohydraulic model and a cost model, a corrosion model was implemented. This model allows to specify the corrosion rate based on several parameters and to define the behaviour of the corrosion products. Additionally, also pitting corrosion was considered to be able to take stainless steel types into account. This model is finally demonstrated with the design of a heat exchanger for the geothermal plant on the Balmatt-site. The TCO of a titanium heat exchanger (both with the current design and with the optimised design) is compared to that of a heat exchanger made from carbon steel. The results indicate that the carbon steel unit can have a TCO that is 36% lower than the current device and 16% lower than the optimised titanium heat exchanger. Furthermore, since (parts of) this device need to be replaced several times over the lifetime of the plant, the investments are distributed over time and the financial risk is reduced. Care however needs to be taken when applying this approach. The obtained results are valid for the specific corrosion behaviour and the sensitivity analysis shows that an increase of the corrosion rate can undermine the cost advantage.
Keywords
Heat exchangers, corrosion, cost optimization, geothermal energy

Downloads

  • (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 25.06 MB

Citation

Please use this url to cite or link to this publication:

MLA
Faes, Willem. Strategies for the Design and Predictive Maintenance of Heat Exchangers Exposed to a Geothermal Brine. Universiteit Gent. Faculteit Ingenieurswetenschappen en Architectuur, 2020.
APA
Faes, W. (2020). Strategies for the design and predictive maintenance of heat exchangers exposed to a geothermal brine. Universiteit Gent. Faculteit Ingenieurswetenschappen en Architectuur, Ghent.
Chicago author-date
Faes, Willem. 2020. “Strategies for the Design and Predictive Maintenance of Heat Exchangers Exposed to a Geothermal Brine.” Ghent: Universiteit Gent. Faculteit Ingenieurswetenschappen en Architectuur.
Chicago author-date (all authors)
Faes, Willem. 2020. “Strategies for the Design and Predictive Maintenance of Heat Exchangers Exposed to a Geothermal Brine.” Ghent: Universiteit Gent. Faculteit Ingenieurswetenschappen en Architectuur.
Vancouver
1.
Faes W. Strategies for the design and predictive maintenance of heat exchangers exposed to a geothermal brine. [Ghent]: Universiteit Gent. Faculteit Ingenieurswetenschappen en Architectuur; 2020.
IEEE
[1]
W. Faes, “Strategies for the design and predictive maintenance of heat exchangers exposed to a geothermal brine,” Universiteit Gent. Faculteit Ingenieurswetenschappen en Architectuur, Ghent, 2020.
@phdthesis{8681164,
  abstract     = {The increasing demand for heat and electricity during the past decades has globally caused an increase in carbon dioxide concentrations (CO2). This contributes to global warming and, if continued, can have disastrous consequences for the planet. Therefore, a larger share of renewable sources of energy is necessary. Currently, hydropower has the biggest share, followed by wind and solar power. One additional alternative is to use the energy that is available as heat contained in or discharged from the Earth’s crust. This geothermal energy flows from the interior of the Earth to the crust and has two main contributors: residual primordial heat (originating from the formation of the planet) and radiogenic heat (caused by the radioactive decay of different isotopes).
People have been using geothermal energy for centuries. While, for a long time, geothermal energy was mainly used for bathing and washing, there now globally is an installed capacity of 12.7GW. In Belgium, geothermal energy is mainly limited to shallow and medium-deep geothermal applications, where buildings are being equipped with ground-source heat pumps (GSHPs). Additionally, two shallow geothermal power plants are installed in Saint-Ghislain and Douvrain that are used for district heating and biogas production. However, recently, the Flemish Institute of Technological Research (VITO) drilled three wells up to a depth 3.5 km on the Balmatt-site in Mol. The geothermal doublet was finished in 2018 and a geothermal brine with a temperature up to 130 °C can be obtained at the well head. A geothermal power plant was constructed to provide the VITO-buildings and the neighbouring cities with heat and to generate electricity with an organic Ranking cycle (ORC).
In this power plant, a heat exchanger is installed to transfer the heat from the geothermal brine to a secondary fluid. Since the brine has a salinity of approximately four times the salinity of seawater, it is highly corrosive. Therefore, the heat exchanger was constructed with a highly alloyed stainless steel, having a strong impact on the profitability of the geothermal project. One alternative approach could be to construct the device of carbon steel, which has a lower material cost and a better machinability. Unfortunately, this material will be susceptible to corrosion and the heat exchanger might need to be replaced several times over its lifetime. It is however not a priori known how high the corrosion rate will be and if the degradation of the metal will have an influence on the thermohydraulic behaviour of the heat exchanger. In this dissertation, the corrosion behaviour of metals in the brine and the influence on the performance of the heat exchanger will be studied and the economic viability of employing a device, constructed of a corroding material, will be examined.

First, an overview will be given of the available literature on corrosion and corrosion prevention in heat exchanger. Ample examples can be found of failure analyses where a heat transfer device suffered from (one of the various forms) of corrosion. Only a minority of the failures are caused by uniform corrosion. Localised forms, especially pitting and stress corrosion cracking, are most frequently occurring. It appears from literature that heat transfer can influence the corrosion process, mainly by creating a temperature gradient causing differences in solubility of the elements in the fluid. Subsequently, the corrosion products can influence the performance of the heat exchanger by forming a layer with a low thermal conductivity and an increased surface roughness. This can cause a decrease in the heat transfer rate and an increase in pressure drop. Several methods of preventing corrosion in heat exchangers are described as well. An adequate material selection and control of the environment are crucial. Additionally, attention should be paid to avoiding crevices and the highest heat transfer coefficient should preferably be located on the cold side.
To evaluate the behaviour of metallic materials in the geothermal brine, two types of experiments are preformed: static and dynamic corrosion tests. With the static tests, several steels are exposed to the brine for an extended period of time. By measuring the mass loss after exposure, a uniform corrosion rate can be calculated. Carbon steel exhibits corrosion rates up to 0.3 mm/y, while the mass loss of the stainless steel samples is negligible. They can however still be susceptible to localised form of corrosion (like pitting or crevice corrosion). Therefore, also electrochemical techniques are applied to determine the open circuit potential, the critical pitting (or crevice) potential and the repassivation potential. From these measurements, it appears that all evaluated materials are, to a certain extent, susceptible to localised corrosion, and that, if a crevice is present, repassivation can become a problem.
An interesting observation made with the exposure tests, is that the corrosion rate of the carbon steel decreases over time. Additionally, it is lower for elevated temperatures, although the Arrhenius equation would suggest otherwise. This phenomenon was investigated by an analysis of the corrosion products. Apparently, a protective layer of iron carbonates (FeCO3) is formed. Since the solubility of FeCO3 is lower at higher temperatures, this layer is more protective at these temperatures. 
Since in reality, the brine will not be static, an experimental setup was constructed to allow dynamic corrosion tests. During these experiments, the brine flows inside a carbon steel tube that is subject to heat transfer. The heat transfer rate and pressure drop are being monitored and small tubular samples allow to determine uniform corrosion rates. The corrosion rates obtained under flowing conditions are approximately 60% higher than those under static conditions. During the execution of the measurements, the pressure drop remained unchanged. The heat transfer rate demonstrated an initially increasing trend (with an increase of almost 6%), followed by a decrease. Although this needs to be confirmed by longer exposure periods, this could be due to the build-up and removal of a layer of corrosion products, influencing the turbulence inside the tube. 
To examine the economic viability of the proposed alternative, a heat exchanger design model was developed. Although models exist in literature to minimise the total cost of ownership (TCO) of heat transfer devices, no publications were encountered where the surfaces are degrading over time. Therefore, in addition to a thermohydraulic model and a cost model, a corrosion model was implemented. This model allows to specify the corrosion rate based on several parameters and to define the behaviour of the corrosion products. Additionally, also pitting corrosion was considered to be able to take stainless steel types into account.
This model is finally demonstrated with the design of a heat exchanger for the geothermal plant on the Balmatt-site. The TCO of a titanium heat exchanger (both with the current design and with the optimised design) is compared to that of a heat exchanger made from carbon steel. The results indicate that the carbon steel unit can have a TCO that is 36% lower than the current device and 16% lower than the optimised titanium heat exchanger. Furthermore, since (parts of) this device need to be replaced several times over the lifetime of the plant, the investments are distributed over time and the financial risk is reduced. Care however needs to be taken when applying this approach. The obtained results are valid for the specific corrosion behaviour and the sensitivity analysis shows that an increase of the corrosion rate can undermine the cost advantage.},
  author       = {Faes, Willem},
  isbn         = {9789463554336},
  keywords     = {Heat exchangers,corrosion,cost optimization,geothermal energy},
  language     = {eng},
  pages        = {xxv, 188},
  publisher    = {Universiteit Gent. Faculteit Ingenieurswetenschappen en Architectuur},
  school       = {Ghent University},
  title        = {Strategies for the design and predictive maintenance of heat exchangers exposed to a geothermal brine},
  year         = {2020},
}