Prevention of Pipes Bursting by Using a Novel Deicing Technology

Abstract

Freezing of water inside above-ground steel pipes is an unwelcome phenomenon that leads to internal pressurization, bulging, and bursting of pipes and can cause noticeable financial losses, environmental pollution due to the resulted leakages, and damage to the property, equipment, and workers in the field. Thus, a practical deicing/antiicing system that is highly efficient must be developed to minimize the detrimental impacts caused by freezing of liquids inside pipes. First, numerous tests were carried out in a relatively large cold room in which the actual working conditions of bare pipes exposed to cold weather were simulated to comprehend the freezing mechanism of the pressurized water. In the second phase of the project, the performance of the novel heating system was assessed by conducting deicing tests in the cold room. It was concluded that the freezing of the enclosed water was heavily dependent on the pressurization extent of water that itself was a function of pipe size and material properties. It was also found that the novel heating system that was produced by using thermal spraying means was able to eliminate the ice that was formed inside the pipe even under harsh conditions that may not be experienced in the field.

1. Introduction

1.1 Steel pipe bursting

Ice formation inside steel pipes is a widespread incident in both residential and industrial sectors at locations where the ambient air temperature is below the freezing temperature of the water. When pipes are exposed to cold air for a prolonged duration, the ice that was formed initially at the inner surface of the pipe gradually accumulates and grows inward. In extreme cases, this causes a partial blockage inside a pipe or a full blockage at the thinner sections near connections and fittings. Once a portion of water has converted to ice, this causes an expansion in the volume that it occupies and when there is no further room in the pipe that can accommodate this volume growth, the pressure of the entrapped water starts building up. It is well-established that the pipe bursting occurs at a location where pressurized unfrozen water exists [1]. This incident is not only limited to the steel pipes, but it can also occur for plastic piping systems [2].

It is well-known that the negative consequences of bursting pipes are both costly and perilous. Furthermore, it can cause noticeable nonproductive downtime in industrial setups that results in loss of production. The financial losses because of this undesirable phenomenon are of great importance. According to the examination carried out by Insurance Information Institute [3], an average loss of $1.2 billion annually resulted from freezing issues only in the United States during the past 20 years and freezing of water inside pipes has always been one of the major problems.

In addition to the financial losses, the tremendous amount of energy that is stored in pipes is released after bursting of pipes. These explosive ruptures accompanied by a possible projection of chunks of steel from the damaged portions of pipes and pressure equipment can pose a threat to the well-being of the workers. Considering all the negative impacts of this undesirable phenomenon, it stands to reason that developing a novel and functional deicing system that can minimize or eliminate accumulation of ice inside pipes and pressure equipment is of great importance.

The common practice to minimize formation of ice is the installation of insulation around pipes. However, this cost-effective approach can only be practical in reducing the heat loss rate when pipes are exposed to cold ambient air for a short period of time [4]. That said, an efficient deicing system should accompany the thermal insulation to bring about satisfactory results when pipes are located in frigid environments for a prolonged period.

1.2 Usage of coating heating systems

It is well known that electric heat tracing is a practical method for deicing or temperature control of pipes that are used for water distribution, especially when waste heat and process steam are not obtainable [5]. In addition, heat tracers are employed where decent flow of high-viscosity fluids is needed. Another benefit of using electric heat tracing compared to steam-based tracing is its cleanliness [6]. Furthermore, they offer accurate temperature control and environmental and economic advantages in comparison to steam tracing [7].

One of the great features of electric heat tracing is the ability to control the amount of power that is supplied to the tracers. This can simply be done by input voltage adjustment. Thus, the heating or deicing operation can be adjusted at any given moment according to requirements in the field or the environmental conditions such as wind gust and air temperature. The system can become adaptable to climatic conditions and be operated automatically by using a controller [8]. The advantage of using controllers is twofold. Apart from the reduction in energy consumption, it can also increase the efficiency of the electrically resistive heating systems. Another feature of electric heat tracing is the wide range of power that can be supplied to tracers. In this regard, operating voltages that start from 24 VAC up to 750 VDC and power from 4 Watt per foot (W/ft) up to 30 W/ft at 10°C were reported [9].

An alternative to electric heat tracers is thermally-sprayed coating heaters that have been developed recently. These heaters are composed of several layers that have favorable mechanical and electrical features for the given tasks. The usage of these coating-based heaters has been investigated and tested for a wide variety of applications with different compositions, geometries, and power requirements [10, 11, 12, 13, 14, 15].

1.3 Development and performance of coating-based resistive heating systems

It has been about a century since the thermal spraying concept was presented initially. This idea was optimized and refined over time, and it has become one of the most popular and cost-efficient means of additive manufacturing. This process employs a high-temperature flame or plasma to melt and accelerate tiny powder particles and deposit a relatively dense coating onto given surfaces. These coatings have been used successfully in different fields such as automotive industry and electricity production so much so that it has become a reliable and essential element in today’s industry. Fabrication of electric resistive heaters by using thermal spraying processes has received an increasing attention recently.

The application of thermal spraying methods in production of meso-electronics was studied by Sampath, et al. [16]. It was concluded that different electrical elements including resistors, insulators, and conductors can be manufactured by using thermal spraying processes to deposit proper ceramic and metal feedstock materials that have favorable electrical properties. Nickel-chromium alloy was deposited onto alumina by using high-velocity oxygen fuel and plasma processes to develop resistors. The fabricated resistors had a sheet resistance in the range of 17–54 KΩ/sq. The advantages of this production procedure including high production rate, low cost of production, flexibility on thermal spray method, and ability to deposit millimeter-thick coatings made this method an interesting means of production of electrical components [16]. Similarly, high-quality conductors and dielectrics were developed by using both high-velocity oxygen fuel and cold spray processes [17].

1.3.1 Deposition of the electrically insulating layer

Thanks to its unique dielectric properties, alumina has been used widely as electrical insulating layer. Capability of thermal spraying processes in easy and quick deposition of alumina on large surfaces made this method appealing to experts in electronics industry. The volume resistivity of alumina coating fabricated by plasma spraying was reported to be very high in the order of 10⁹–10¹⁰ Ω cm [18]. It was found that the electrical resistivity of alumina depends on several factors, namely humidity, microstructural characteristics, phase composition of alumina, and applied pressure [19, 20, 21, 22, 23].

1.3.2 Deposition of the heating element

The heating element was made by depositing a conductive metallic alloy that possesses relatively high electrical resistivity onto electrical insulating layer (alumina). Several researchers reported successful deposition and operation of several materials, namely molybdenum, nickel, nickel-20 wt.% chromium, nickel-5 wt.% aluminum, iron-13 wt.% chromium, and iron-chromium-aluminum that were manufactured by using different thermal spraying processes such as wire arc, VPS, APS, HVOF, and wire flame spray for usage as the heating element [10, 11, 12, 13, 14, 15, 24]. As an example, three thermal spraying methods, namely vacuum plasma spray, air plasma spray, and high-velocity oxygen fuel, were used to develop heating elements to produce an easily controlled and uniform heat flux. The maximum heat fluxes that were generated from the developed samples before they fail were relatively high in the range of 10.6–17.2 MW/m² [10].

2. Bursting of steel pipes due to ice formation

Given solidification of water inside pipes and pressure equipment is a widespread phenomenon in both industrial and residential sectors, and it brings about negative consequences every year, placing emphasis on further study of the freezing mechanism of pressurized enclosed water seems reasonable. In this regard, numerous experiments were conducted on two well-known pipe materials, namely ASTM A106-B and ASTM A333-6, to study the freezing of pressurized enclosed water and examine the damage exerted on the integrity of pipes that led to their ultimate failure. Then, the fracture surfaces of the pipes were cut and removed from the pipes for further analysis by using a scanning electron microscope.

2.1 Experimental method

The pipe assemblies that were used for the freezing tests were composed of 254-mm long and 51-mm diameter carbon steel pipes. The samples were made from both ASTM A333-6 and ASTM A106-B materials so that the impact of pipe material on the freezing behavior of the enclosed water can be investigated. Schedule 40 pipes that are used extensively in industry were selected for these experiments. The geometry of the pipe assembly completed with fittings and sensors is shown in Figure 1.

Figure 1.

Pipe assembly and the installed thermowells, pressure transmitter, and thermocouple used for freezing tests [25].

The pressure and temperature data were collected regularly at the rate of one reading per second from the pressure transmitter and thermocouples. Two separate data acquisition systems (cDAQ-9171, National Instruments, Austin, TX, USA and SCXI-1600, National Instruments, Austin, TX, USA) were used for collection and storage of entrapped water pressure and temperature data, respectively. The NI MAX software package was used for collection of the measurements.

3. Development of multi-layered coating-based heater

A multi-layered thermally-sprayed heating system was fabricated to eliminate formation of ice inside the pipes and control the temperature of the enclosed liquid. Given the steel pipe was a conductive material, alumina layer as the insulating layer was deposited onto the carbon steel pipe to prevent flow of electrons and short-circuiting between the pipe and the conductive heating element. In addition to flame spraying process, cold spraying method was also used to deposit dense copper coatings where proper electrical connections were required. The effect of various spraying parameters on the quality and heating performance of the coating heater was investigated.

After the samples were fabricated and their heating functionality was tested, they were sectioned for microstructural evaluation and elemental composition analysis. Important features of the coating system, namely continuity, homogeneity, and microstructural defects, were analyzed by using a scanning electron microscope.

3.1 Experimental method

3.1.1 Feedstock powder

Several powder feedstock materials, namely aluminum oxide (Al₂O₃, AMDRY 6060, Oerlikon Metco, Westbury, NY, USA), nickel-50 wt.% chromium (50 Nickel-50 Chromium, 1260F, Praxair, Concord, NH, USA), and copper (SST-C5003, CenterLine, Ltd., Windsor, ON, Canada), were utilized to produce the coatings by using flame spraying and cold spraying methods. The micrographs that were taken in secondary mode by using a scanning electron microscope (Zeiss Sigma 300 VP-FE, Carl Zeiss Canada Ltd., Toronto, ON, Canada) from these powders at 500x magnification are shown in Figure 2a–c. The alumina powder had angular/blocky morphology due to their manufacturing process, which was fusing and crushing, and their size distribution was between 5 and 45 μm [26]. The size distribution of Ni-50Cr powder particles was from 22 to 53 μm and their spheroidal shape was because of the gas atomization process that was used for fabrication of this powder [27]. The minimum purity of the copper powder that was used in this research was 99.7% and its size distribution was between 5 and 45 μm [28]. The dendritic structure of this powder was obtained through electrolysis process.

Figure 2.

SEM images taken at 500 X magnification from powder materials, namely (a) Al₂O₃, (b) Ni-50Cr, and (c) Cu.

3.1.2 Substrate preparation

The coating layers were sprayed onto the pipe sections so that the performance of the heating system as a deicing element can be assessed. The pipe sample that was used as the substrate in this study is shown in Figure 3.

Figure 3.

The components of the pipe assembly that were used as the substrate [29].

In order to deposit the alumina layer onto pipe sections, the surface of the pipe was cleaned and then grit blasted with #24 alumina grit (Manus Abrasive Systems Inc., Edmonton, AB, Canada) to create the roughness required to cause the adhesion between the substrate and alumina coating. An air pressure of 586 kPa (85 psig) was used to accomplish the grit blasting process. Only the central section of the pipe assemblies was grit blasted and the ends were covered by using a heat-resistant masking tape (170-10S Red, Green Belting Industries, Mississauga, ON, Canada) to ensure that coating layers would not be deposited at the welded end caps.

3.1.3 Deposition of coating layers

Once the pipe assemblies were grit blasted, they were coated with a layer of pure alumina (Al₂O₃ 99.5+ wt.%.). The powder particles were consistently fed to the oxy-acetylene torch (6P-II, Oerlikon Metco, Westbury, NY, USA). The spraying parameters for deposition of this layer were obtained through detailed trial and error procedure. Each time the microstructure of the obtained coatings was analyzed to ensure that the thickness of the coatings was uniform all over the substrate and that the coating layer was not damaged during the spraying process. Given the high melting point of alumina, relatively high flow rates for both acetylene and oxygen were selected to bring about a large high-temperature flame that is capable of melting the alumina powder particles. Furthermore, argon was selected as the carrier gas. The flow rate of argon was set at 0.56 m³/h (20 standard cubic feet per hour) to carry the powder particles to the flame spray torch that was installed on a programmable robot (HP-20, Motoman, Yaskawa Electric Corp., Waukegan, IL, USA). The robot was used to ensure the uniformity and reproducibility of the fabricated coatings.

The samples were put into a rotating chuck to create a uniform coating. As the torch was passed in a linear manner along the pipe, the pipe was rotated at the angular speed of 600 rpm. To avoid making helical patterns for the alumina coating, relatively low linear speeds of 10 mm/s for alumina layer, 24 mm/s for nickel-chromium layer, and relatively high angular speed of 600 rpm for the rotating chuck were selected. To minimize the inconsistencies, the powder feed rate was reduced, and the number of passes was increased to achieve the most uniform microstructure for the deposited coatings. A relative term (flow meter reading parameter) was used to adjust the flow of powder particles to the torch.

After deposition of alumina, nickel-chromium alloy with a composition of Ni 53 wt.%, Cr 46 wt.%, and Fe 1.0 wt.% was deposited onto the alumina layer by using flame spraying method. The structure of the pipe assembly complete with bi-layered coating can be seen in Figure 4. The alumina coating was intentionally sprayed over a longer section to ensure that conductive nickel-chromium coating would not be deposited directly onto the substrate as it can result in short-circuiting and malfunction of the deicing system.

Figure 4.

The pipe assembly coated with flame-sprayed insulator and resistor layers and cold-sprayed connector [29].

In order to connect the developed coating system to the power source, copper coatings in the form of rings were fabricated at both sides of the pipe assembly. Copper was selected for this purpose because of its proper electrical properties, which makes it an ideal option for electrical connections. The coating fabrication was accomplished by using the low-pressure cold spray system (SST series P, CenterLine, Ltd., Windsor, ON, Canada). The dendritic copper powder particles were accelerated inside the converging–diverging nozzle, and they formed a dense coating upon impact with the substrate. The working fluid that was used for this process to preheat and spray the copper particles was compressed air. The programmable robot was used for deposition of the copper rings as well. As the pipe was held in the chuck and was rotating, the cold spray nozzle passed the pipe assembly in the radial direction and generated the contact rings that are shown in Figure 4.

3.1.4 Installation of fittings and sensors

In order to read the temperature values of the enclosed water, two long thermocouples were inserted inside the thermowell at each end of the pipe assembly. The pipe assembly coated with the heating system is shown in Figure 5. In addition to the water/ice temperature, the temperature of the heating element was also measured by using a temperature sensor. The ambient air temperature inside the cold room was also measured by using a Type-T thermocouple.

Figure 5.

The coated pipe assembly used for deicing tests completed with all fittings and thermocouples [30].

3.1.5 Deicing test

The coated samples were put inside an 18.2 m³ cold room freezer (Foster Refrigerator USA, Kinderhook, NY, USA) to simulate the harsh environmental conditions. The temperature of the cold room was set at −25°C with a ± 2°C bandwidth. To control the airflow over the pipe, the samples were placed horizontally inside a closed galvanized sheet metal duct with dimensions of 2 m × 0.66 m × 0.48 m. The impact of circulation of air inside the cold room was minimized for the free convection scenario by placing the lead of the duct over the pipe assembly. Furthermore, the forced convection was simulated by turning on the fan at the end of the duct. In this regard, a 0.25 kW (0.33 hp) direct-drive tube-axial fan (DDA-12-10033B, Leader Fan Industries, Toronto, ON, Canada) was used to simulate the harsh environmental conditions when pipes are exposed to wind gusts during cold winter days. Further details about the structure and dimensions of the apparatus that was designed and fabricated for this investigation can be found in the previous study [31]. Figure 6 shows the location at which the coated pipe assemblies were placed during the freezing and deicing tests.

Figure 6.

The apparatus that was used for simulating the harsh environmental conditions [30].

4. Results and discussion

4.1 Freezing behavior

Figure 7 shows the temperature and pressure traces that were obtained from the entrapped water inside Schedule 40 ASTM A333 Grade 6 pipe during the freezing test. The pipe assembly in this experiment was not fully filled with water and there was 5 vol.% of air inside the pipe. This graph demonstrates five distinct stages during the test, namely cooling of the water, the first solidification plateau, cooling of the mixture of water and ice, the second solidification plateau, and ice cooling. The water-cooling stage started when the fans inside the cold room started operating. This was followed by reduction in the temperature of ambient air inside the cold room. This stage was finished after a short super cooling period when plate-like solid crystals that are known as dendritic ice nucleated and grew [32]. Solidification of a portion of the enclosed water occurred in the next stage. The duration of this stage is directly impacted by the amount of air (gap) inside the pipe. It is well-known that the volume of water expands by 9% when it transforms into ice. Therefore, in case the vol.% of air inside the pipe is less than 9%, not all of the enclosed water can be converted to ice simply because there is not enough room for the volume expansion during the phase change. During this stage, the annular ice formed and grew inward [33]. Only a portion of water transformed into ice and once the pipe assembly was fully filled with the mix of water and ice, the pressure inside the pipe started building up. This noticeably reduced the rate of further water being transformed into ice because the pressure rise inside the pipe lowered the freezing temperature of water.

Figure 7.

Temperature and pressure traces versus time for water/ice inside the pipe assembly.

During the third stage, the pressure of the entrapped water sharply increased. This prevented the phase change of the rest of the entrapped water. A nonlinear relationship between melting temperature and pressure is obtained for hexagonal ice, which is given by Eq. (1) [34]. The purple curve in Figure 7 shows the phase change pressure values for any given temperature according to this relation. As long as the pressure of the entrapped water was greater than the value obtained from this equation, water was not able to transform into ice even at subzero temperatures. This situation was discontinued when the pressure of the water was so high that it caused plastic deformation of the pipe. Afterward, given the volume of the pipe assembly was expanded during this inelastic deformation period and the pressure and the slope of pressure trace during this stage were much lower than that of the pressure trace during the elastic deformation stage, further water was able to undergo phase change process. Therefore, the second plateau that can be seen in fourth stage was indicative of the transformation of the remaining water inside the pipe into ice. Once the pipe was fully filled with solid ice at the end of the fourth stage, further exposure to the cold ambient air only reduced the temperature of the solid ice as shown in the fifth stage.

$$P_{\mathrm{f}}=6.11657\times {10}^{-4}-414.5\times \left[{\left(\frac{T_{\mathrm{f}}}{273.16}\right)}^{8.38}-1\right]$$E1

It is well-established that the plastic deformation of the pipe occurs when the hoop stress in the pipe reaches the yield strength of the pipe material. The hoop stress itself depends on the pipe diameter. Therefore, it can be concluded that the freezing point depression and the freezing behavior of the water that depends on the plastic deformation of the pipe, itself is a function of the pipe size, wall thickness, and material. The hoop stress values can be obtained by applying the relations developed by thick-walled pressure vessel theory. It is worth mentioning that the yield strength of the pipe is not a constant value, and it changes from one freezing–thawing cycle to another cycle due to the work hardening of the pipe material that is caused by movement of dislocations.

The comparison of the pressure values obtained from the freezing experiment with the results obtained from Eq. (1) is only valid when still water is entrapped in the pipe. That being said, the increasing deviation between the pressure experimental measurements and theoretical estimations is due to the fact that only ice exists in the pipe by the end of the fourth stage, and therefore, that comparison is not valid anymore.

4.2 Plastic deformation and work hardening

A PI tape was used to measure the inelastic deformation of the pipes during the freezing experiment. In order to calculate the transverse strain of the pipe, the outer diameter of the pipes was measured before and after freezing experiments. Obviously, the maximum deformation took place in the middle of the pipe assembly where it was farthest from the circumferential welds and the reinforcing impact of the welds was at its minimum level. It was concluded that the maximum strain values that A333-6 and A106-B steel pipes could accommodate before failure were 11.3% and 13.3%, respectively. Although the minimum transverse elongation before rupture has been reported to be 16.5% for ASTM A106 Grade B steel [35], the pipe did not meet the expected value under freezing experiments. However, the obtained value for A333-6 steel was very close to the reported minimum transverse elongation, which is 11.4% [36].

The maximum pressure values inside the pipe were greatly impacted by the number of times the pipe underwent plastic deformation during the freezing and thawing cycles. The increase in the peak pressure value from one test to another test, which was because of the work hardening of A333-6 and A106-B Schedule 40 pipes, can be clearly observed in Figure 8a and b.

Figure 8.

The impact of work hardening of the pipe materials on the peak pressure during the freezing test for (a) A106-B and (b) A333-6 pipes.

4.3 Pipe rupture

Figure 9 shows the failure pattern of the pipes after several freezing and bulging experiments, which is a common pattern for pipe bursting cases. The axial crack was formed in both pipes due to the developed excessive hoop stresses during the tests. It is well-established that the hoop stress is the maximum principal stress when pipes are pressurized. This pressurization was caused by gradual formation of ice accompanied by volume expansion and low compressibility of unfrozen water [1]. This pattern and location of the failure were expected as it is known that pipes fail where water freezes last. In this case, the remaining unfrozen water was entrapped in the middle of the pipe where final rupture occurred.

Figure 9.

Failure site of (a) A106-B and (b) A333-6 pipes due to overload during freezing test [25].

It can be seen in Figure 9 that the extent of the deformation of the pipes was different for these two pipes. At the failure spot, the generated gap was wider and larger for the A106-B pipe compared to the A333-6 pipe. This was because of the amount of energy that was absorbed by the pipe material during the formation and propagation of the crack upon rupture. It is believed that the toughness of these pipe materials at low temperature has an important role to play. It is well known that the A333-6 pipe is the preferred option for installation at sites that are exposed to low-temperature environments because of its acceptable low-temperature toughness. Therefore, the higher toughness of A333-6 pipe compared to A106-B pipe results in greater resistance against brittle fractured caused by internal pressurization and overload. In contrast, the degradation of absorption energy of A106-B pipe material at lower temperature was found by conducting Charpy impact test [37].

In order to observe the fracture surfaces, the pipes were cut after the failure. To study the fracture patterns at different locations of the pipes, eight different rings were cut at 1 cm intervals from both pipes as shown in Figure 9.

4.4 Failure analysis

High-quality side images were taken from the rings shown in Figure 9 to observe the macroscopic fracture features. These features, along with the geometry of the fractured surfaces are often used to determine the mechanism of the fracture and also the pipe failure mode. The slant and double-slant fractures that can be observed in Figure 10 are indicative of ductile failure due to plane stress loading conditions [38], which is a common form of failure for pressure vessels and pipes [39]. In these cases, the pipe failure begins with local thinning. Then localization of shear bands that occurs at 45°angle develops in the necked spot [40, 41]. The necked area that led to the rupture of the pipe in slant failure mode can be seen in Figure 10.

Figure 10.

Side views of the rings cut from (a) A106-B and (b) A333-6 pipes [25].

4.5 Fractography

The micrographs taken from the fracture surfaces confirm that the failure mode of both pipes was ductile tearing. This occurred because of the propagation of microcracks as a result of the coalescence of microvoids in the axial direction (along z-axis), as shown in Figure 11. Another feature that confirms the ductile failure mode of pipes is the fibrous and dull appearance of the fractographs that embodies numerous dimples.

Figure 11.

Fractographs taken at 500X magnification from (a) A106-B pipe and (b) A333-6 pipe.

4.6 Coating characterization

High-magnification images were taken in backscattered electron mode from the cross sections of the coated samples. These SEM images are shown in Figure 12. Although generation of microcracks was anticipated initially due to the brittle nature of the ceramic layer and thermal stress that can be generated at the interfaces due to the mismatch in material properties [42, 43], no major damage or delamination was observed during the microstructural evaluation of the coated samples even after conducting expedited deicing tests with high supplied power values.

Figure 12.

SEM images taken from the coated pipe assembly: (a) bi-layered coating system, (b) NiCr-Al₂O₃ interface, and (c) Al₂O₃-steel interface [29].

The material for each layer of the coating system was selected based on the required mechanical and electrical properties to accomplish the given tasks. Alumina was selected due to its unique properties. While this material is a great electrical insulator, it conducts heat readily in comparison to other ceramic materials [26]. Therefore, this material was able to prevent short-circuiting between the pipe (substrate) and the heating element (nickel-chromium alloy), and at the same time, it was capable of transferring the generated heat by the heating element to the pipe to accomplish the deicing task [44, 45].

Although penetration of the nickel-chromium alloy inside the alumina layer was observed to some extent during the microstructural evaluation, this did not result in the malfunction of the coating-based heater. The penetration of the molten particles in the alumina layer was due to the presence of a network of open pores inside the ceramic coating, which is a characteristic of coatings fabricated by flame spraying process [45]. Given the effective thickness of alumina was reduced as a consequence of this incident, alumina coatings with at least 50 microns thickness had to be deposited onto the pipe substrates to prevent any possible short-circuiting. It is well-established that much denser coatings can be fabricated by using more sophisticated thermal spray means such as plasma spraying, high-velocity oxygen fuel, or suspension plasma spray to prevent this undesirable occurrence [46, 47].

The nickel-chromium material was selected for the heating element because of its relatively high electrical resistivity. Although the electrical resistivity is a material property, the electrical resistance of the heating element is a function of its cross-sectional area, which itself can be affected by discontinuities and presence of imperfections such as cracks and pores [45]. Presence of pores in the coating increases the electrical resistance as a result of reduction in the cross-sectional area through which free electrons can move [48]. It has been observed that formation and propagation of microcracks can also bring about the same consequences [49]. Therefore, the performance of the heating element that depends on its electrical resistance can be varied by using different thermal spraying methods.

4.7 Heating performance

The performance of the coating-based heater was assessed under different power inputs. For this purpose, when 20 V was applied to the coating, 80 W power was generated, which was sufficient to heat and melt the ice inside the pipe. The temperature traces for the coating surface and enclosed ice/water temperatures can be seen in Figure 13a for the case of free convection. The functionality of the developed deicing system was evaluated based on the time that was required to accomplish the deicing test. It was observed that even under harsh conditions where the bare pipe was exposed to the forced convection due to the circulation of the cold ambient air, the fabricated coating-base heater melted the ice inside the pipe successfully as shown in Figure 13b. It was found that the voltage and current of the heating element were proportional and therefore, the nickel-chromium layer was an ohmic material [45].

Figure 13.

Temperature traces from deicing tests for (a) free convection with supplied power of 80 W and (b) forced convection with supplied power of 500 W.

4.8 Future work

In addition to the technical aspects, the economic and environmental implications of fabrication and utilization of the coating-based heating systems have also been investigated [50, 51, 52]. The promising results obtained from these studies confirmed the possibility of using the coating-based deicing systems on mass scale, however further advancement and improvement in the geometry of patterned coatings, manufacturing process, and the selection of the materials can even bring about more encouraging outcomes. Therefore, in the next phase of the project, in order to reduce the production cost, minimize the emission of exhaust gases to the environment, and achieve a more consistent and dense structure, the heating elements are fabricated by using cold spray system in helical pattern as shown in Figure 14. Other advantages of fabrication of heating element via cold spraying are prevention of penetration of the heating element inside the electrically insulating layer, on-site repairability, and enhanced bonding to the insulating layer thanks to the applied compressive stresses during the spraying process.

Figure 14.

The presentation of (a) carbon steel pipes used as substrate and (b) pipes equipped with novel cold-sprayed tin and copper heating elements.

5. Conclusions

The freezing mechanism of the pressurized water inside steel pipes and the subsequent bulging and bursting were studied. In order to overcome this undesirable issue that brings about noticeable financial losses and environmental concerns, a novel deicing system was developed by using flame spraying and cold spraying processes. Once the performance of the developed heating system was assessed in simulated environmental conditions under different supplied powers, the coated pipes were sectioned so that their microstructure can be analyzed. The outstanding findings of this study are presented hereunder:

1. The bursting of the pipe is dependent on the pressurization of the enclosed water, which itself depends on the pipe size, material, and wall thickness. It is also a function of the prior work hardening of the pipe material, which is indicative of the number of times the pipe underwent freezing-thawing cycles.

2. The developed efficient coating-based heater was able to heat and melt the ice in the pipe easily even under harsh environmental conditions where bare pipes were exposed to low temperatures air under forced convection conditions.

3. It was found that the deicing performance of the coating heater was heavily dependent on the spraying method and spraying parameters.

4. The microstructural evaluation of the samples confirmed that the structural integrity of the coating system was not compromised even after conducting numerous freezing and heating cycles. The promising outcomes of this study emphasize the feasibility of using the coating-based heating system on mass scale for industrial applications.