The impact of metal cleaning agents on components of air conditioning piping systems in EMUs

As a critical component connecting the compressor to the circulation system, the air conditioning pipeline of EMUs primarily functions to absorb compressor vibrations through flexible metal hoses (such as corrugated pipes with stainless steel mesh sleeves), thereby reducing pipeline system noise and mechanical stress. However, corrosion issues primarily arise from multiple coupled factors. On one hand, environmental influences–including sand and dust impact during high-speed operation, industrial contaminants (such as oil films and particulates), and temperature-humidity cycling–can compromise the pipeline surface protective layer, accelerating electrochemical corrosion (Liu, Xu, Chen, & Zhang, 2026). For instance, condensation accumulation beneath insulation layers can induce microbiologically influenced corrosion (MIC), leading to comprehensive pipeline failure within short timeframes. Concurrently, material factors play a role: certain sealing materials exhibit inadequate temperature resistance, while ageing of air springs (e.g. photo-oxygen degradation triggered by ultraviolet radiation) diminishes vibration damping capabilities, indirectly heightening pipeline stress corrosion risks (Abd El-Maksoud, 2018; Cheng, Meng, Wang, Pang, & Duo, 2021; Tan et al., 2017; Zheng, Gong, Li & Guo, 2018; Zhu et al., 2025).

Corrosion-induced degradation of piping performance directly jeopardises train operational safety. It diminishes mechanical properties, reducing flexibility and pressure-bearing capacity, thereby amplifying vibration transmission. This results in increased carriage noise, diminished passenger comfort, and accelerated fatigue damage to other vehicle components. It also heightens leakage risks, as corrosion perforations may cause refrigerant (e.g. Freon) leaks, resulting in air conditioning system failure. Incidents have occurred where refrigerant leakage from vibration-damping hoses triggered cooling failures, necessitating emergency shutdowns for repairs and severely compromising operational safety.

Air conditioning unit piping, comprising copper tubing (TP2 deoxidised copper) and stainless steel corrugated hoses (304 grade), exhibits corrosion during operation, as illustrated in Figure 1. 3 levels of air conditioning pipe corrosion: Minor corrosion: Small, isolated rust spots cover no more than 10mm² with no larger than 16mm² areas, and no extensive flaking or accumulation; Moderate corrosion: Extensive, patchy, or block-like rust spots and verdigris are present, but no refrigerant leakage has occurred; Severe corrosion: Beyond moderate corrosion, rust has penetrated the pipe interior, resulting in refrigerant leakage (Bai et al., 2013; Eddy, Ibok, Ebenso, El Nemr, & El Ashry, 2009; El Nemr et al., 2014a, b; El Nemr, Moneer, Khaled, Khaled, & El-Said, 2014; Li, Bae, Mishrra, Shi, & Giammar, 2020; Thombare et al., 2022).

The corrosion test specimens of air conditioning copper tube were sourced from Changsha EMU Depot, measuring 40 × 20 mm. Section A represents the tail end of the copper tube, while Section B denotes the head end, as illustrated in Figure 2. The high-pressure shock-absorbing hose for air conditioning compressors was procured from Beijing Zhuoxu Company, measuring 50 × 2 cm, as shown in Figure 3.

A 10 × 10 mm corrosion sample was excised. The surface morphology of the corrosion rust layer was observed and analysed using a scanning electron microscope. The elemental composition of the rust layer was characterised via the instrument's integrated X-ray energy dispersive spectrometer. Test equipment models: Electron microscope (TESCAN MIRA4), Energy dispersive spectrometer (Xplore 30)

A handheld portable microscope was employed to examine the macroscopic corrosion morphology at representative locations within the specimens. The microscope featured 1.3 million pixels and a resolution of 1280 × 1024 (MJPG). Instrument model: Anyth.

Following GB/T1690-2010 “Test Methods for Resistance of Vulcanised Rubber or Thermoplastic Rubber to Liquids” (National Standardization Administration, 2011), the high-pressure shock-absorbing hose of an air conditioning compressor was sectioned into 3 components: corrugated hose (304 stainless steel), pressure-bearing section (carbon steel), and copper tube (TP2 deoxidised copper). Each section underwent full immersion testing in cleaning agents, as illustrated in Figure 4 and detailed in Table 1.

Contact angle measurement can reflect wettability. The smaller the contact angle, the better the wetting performance of the liquid on the solid surface. The contact angles of four cleaning agents were measured according to the method specified in GB/T 30,693–2014 “Plastics—Thin Films and Sheets—Determination of Contact Angle.” The cleaning performance was tested according to the method outlined in Q/CR468-2025 “Cleaning Agents for External Surfaces of Electric Multiple Units.” The test results are shown in Table 1.

Experiment 1: Prepare cleaning agents 1#, 4#, 5#, and 6# at a dilution concentration of 20%. After continuous immersion of bellows, pressure-bearing sections, and copper tubes in the diluted solutions for 168 hours, observe changes in specimen appearance and corrosion quality.

Experiment 2: Prepare cleaning agents 1#, 4#, 5#, and 6# at a dilution concentration of 20%. Apply the cleaning agent solution to the surfaces of corrugated pipes, pressure-bearing sections, and copper tubes. Expose the specimens to air for 24 hours. After 7 cycles, observe changes in specimen appearance and corrosion mass.

Following the aforementioned tests, specimens underwent microphotography and SEM analysis.

AFM testing: Analysis of specimen surfaces was conducted using a Bruker Dimension Icon instrument.

The morphology of corrosion products on the copper tube within the high-pressure shock-absorbing hose of an air conditioning compressor was captured using a handheld microscope (The microscope featured 1.3 million pixels and a resolution of 1280×1024 (MJPG). As shown in Figure 5, multiple corrosion patches were observed on the surface of the copper tube's tail section, with severe blackening of the patches. No extensive corrosion was observed on the head section of the copper tube, as depicted in Figure 6 (Deyab, Mohsen & Guo, 2022; Olasunkanmi, Obot, Kabanda, & Ebenso, 2015; Liu, Yong, & Guo, 2015; Li, Han, Li, & Xin, 2017; Schlegel et al., 2018; Yang, Hou, Zhang, & Li, 2017).

Surface morphology observations were conducted on corrosion products at the tails of air conditioning copper tubes, with compositional analysis performed via EDS and line scanning. Results are presented in Table 2 and Figure 7. Numerous pitting corrosion sites were observed on the tube tails, exhibiting elevated proportions of Cu, C, and O elements. This primarily stems from the copper tubes' composition of Cu, Mo, and Cr, where elevated C and O content likely contributes to the formation of CuO and other non-metallic oxides (Bai, Liu, & Tu, 2014; Si, Xue, Sun, Liu, & He, 2019; Zhang, Wu, Shi, & Wang, 2025). Trace amounts of Si and Ca may have been introduced via atmospheric dust, while K and Na likely originated from alkaline substances used during cleaning. Corrosive Cl and P elements were also detected on the surface, constituting the primary causes of copper tube corrosion (Du et al., 2025; Hemdan, Taha, Gabr, & Elkady, 2014; Liu, Qi, Wang, & Wan, 2025; Ma, Li, Xue, & Lai, 2018; Saengsod, Limmatvapirat, & Luangtana-Anan, 2012).

To investigate the effects of exterior cleaning agents on different materials of air conditioning compressor shock-absorbing hoses, the high-pressure shock-absorbing hose was dissected into 3 components: corrugated hose, pressure-bearing section, and copper tube. Continuous immersion and air exposure tests were conducted, and carry out continuous soaking (168 hours) and air exposure tests (soaking 1 hour, air exposure 23 hours; cycling 7 times), as illustrated in Figure 8. Microscopic images are shown in Figures 9 and 10.

The corrugated hose, pressure-bearing section, and copper tube were continuously immersed in 4 different cleaning agents. Solutions 1# and 5# turned blue upon immersion of the copper tube, indicating the generation of Cu²⁺. The copper tube reacted with the cleaning agents, forming dense pale yellow corrosion crystals on its surface. The quality changes of the copper tube were as follows: Cleaning agent 1# > Cleaning agent 6# > Cleaning agent 5# > Cleaning agent 4#. The compressed section exhibited more yellowish-brown corrosion products, with quality changes: Cleaning agent 1# > Cleaning agent 4# > Cleaning agent 5# > Cleaning agent 6#. The corrugated hose surface showed negligible changes, indicating that 304 stainless steel exhibits superior corrosion resistance compared to the other 2 materials. Consequently, it is evident that metal specimens undergo corrosion during prolonged immersion. The corrosion sequence of the metals is: Solution 1# > Solution 5# > Solution 4# > Solution 6#. Evidently, acidic and alkaline cleaning agents exert greater corrosive effects on metal specimens. Therefore, in subsequent component cleaning agent selection, priority should be given to neutral or weakly alkaline cleaning agents, avoiding acidic or strongly alkaline cleaning agent products. After cleaning, thorough water rinsing to neutrality (pH = 7) is mandatory, with residual residue inspection to prevent secondary corrosion.

Air exposure tests on metal specimens, corrugated hoses, pressure sections, and copper tubes indicate minimal mass change in metal specimens. Visual alterations to hoses, pressure sections, and copper tubes remain negligible, suggesting residual cleaning agent on metal surfaces. This is unlikely to adversely affect specimens in the short term.

It is also evident that the cleaning agent exerts a significantly greater impact on the pressure section (carbon steel) than on the corrugated hose (304 stainless steel) and copper tubing, as illustrated in Figure 11. Due to the higher chromium content in 304 stainless steel, the corrosion resistance of the stainless steel surface relies on the passivation film Cr₂O₃ formed by chromium. This film layer effectively isolates the corrosive medium from the base metal through physical barrier action and electrochemical protection (Abdullah, El Nemr, El-Sakka, El-Hashash, & Soliman, 2021; El Nemr, Elhebshi, El-Deab, Ashour, & Ragab, 2022; Wen, Wu, Zhang, Zhang, & Xiao, 2021; Wang, Zhang, Ni, & Kong, 2026). Carbon steel, lacking corrosion-resistant elements, is susceptible to disruption by H⁺ and OH⁻ ionsin acidic or alkaline cleaning agents. These ions destabilise the environment, causing Fe dissolution to form Fe²⁺. The reduction process of H⁺ produces highly reactive hydrogen gas, further promoting grain detachment and resulting in surface cracks of varying depths.

After continuous immersion of the pressure section (carbon steel) and copper tube (TP2 deoxidised copper) in diluted solutions (1#, 4#, 5#, 6#, diluted to 20% concentration) for 168 hours, SEM examination of both specimens revealed: Acidic Cleaner 1# caused pronounced corrosion on both the pressure section and copper tube specimens. The pressure section specimen surface exhibited numerous yellowish-brown corrosion products that were loose, multi-layered, and prone to flaking. The outer surface of the copper tube specimen formed crystalline corrosion products. Cleaners 4#, 5#, and 6# had a lesser effect on the specimens, with the post-corrosion morphology being largely similar, as shown in Figure 12.

As indicated by the elemental analysis in Tables 3 and 4, the surface concentrations of S and Cl in samples immersed in Cleaning agent 1# were markedly higher than those in Cleaning agents 4#, 5#, and 6#. The H⁺ ions in acidic cleaning agents constitute the primary driving force for corrosion (Galatis, Boyatzis, & Theodorakopoulos, 2012; Usman, Umoren, & Gasem, 2017; Zhang, Lin, & Yu, 2009; Zhao, 2024). Within an acidic environment, the porous oxide layer on carbon steel surfaces is rapidly dissolved by H⁺ ions, exposing underlying Fe atoms directly to the acidic solution.

After applying the cleaning solution to the surface of the compressed section and copper tube, the specimens were exposed to air for 24 hours. This cycle was repeated 7 times. SEM observation of the 3 specimens revealed that the effects of the different cleaning agents were largely similar. Due to the relatively short contact time, the impact was minimal, as shown in Figure 13. Elemental analysis in Tables 5 and 6 indicates that the Fe content in the pressure-bearing section sample and the Cu content in the copper tube sample immersed in Cleaning agent 1# were significantly lower than those treated with Cleaning agents 4#, 5#, and 6#. Cleaning agent 1# may have exerted a dissolving effect on the samples. For the pressure-bearing section specimens, this may stem from a minor reaction between H⁺ ions in Cleaning agent 1# and Fe, causing partial Fe loss from the specimen surface in ionic form. Similarly, for the copper tube specimens, weak interactions between H⁺ ions in Cleaning agent 1# and Cu resulted in minor Cu element depletion. The core chemical reaction is:

Acidic cleaning agents containing oxidising acid radicals (Cl⁻, SO₄²⁻) further accelerate the reaction by first oxidising Fe²⁺ to Fe³⁺, forming insoluble Fe₂(SO₄)₃ precipitates that coat the surface as a corrosion product film. However, this film layer is loose and porous; it actually exacerbates localised corrosion due to acid solution retention beneath it (Du, Yi, & Shi, 2023; Kamal & Sethuraman, 2012). The core chemical reaction is:

Cu²⁺ generated during acidic corrosion may further combine with anions in solution (e.g. SO₄²⁻, Cl⁻) to form soluble salts (e.g. CuSO₄, CuCl₂). Consequently, scanning electron microscopy reveals crystals of soluble salts on the outer surface of copper specimens. Therefore, the core chemical reactions are:

As shown in Figures 14 and 15, after continuous immersion for 168 hours, the maximum roughness of the compressed section ranged from −812.2 nm to −831.2 nm, with a minimum of −373.9 nm to −360.3 nm. The copper specimen exhibited a maximum roughness of −1.2 μm to −1.3 μm and a minimum of −427.9 nm to −396.6 nm. After 168 hours of air exposure, the maximum roughness of the compressed section ranged from −244.6 nm to −243.2 nm, with a minimum of −76.2 nm to −74.1 nm. The copper specimen exhibited a maximum roughness of −238.5 nm to −275.2 nm and a minimum of −124.3 nm to −115.0 nm. Both the corrosion depth and roughness of the compressed section and copper tube specimens after 168 hours of continuous immersion were greater than those after 168 hours of air exposure.

Atomic force microscopy results indicate that continuous immersion exhibits a higher corrosion rate. This is attributed to the liquid phase providing an ionic conduction medium, accelerating electrochemical corrosion processes. Dissolved oxygen (O₂) and Cl⁻ synergistically promote metal oxidation reactions. In contrast, atmospheric exposure corrosion proceeds more slowly, involving only surface-adsorbed water films in the reaction. The resulting oxidation products may form protective passivation layers, hence the slower corrosion rate.

Through analytical methods including SEM and atomic force microscopy, combined with corrosive element detection and cleaning agent immersion experiments, the causes of corrosion in metal components and the influence of cleaning agents were systematically investigated.

1. Multiple corrosion patches and pitting were observed on the surface of the copper tube tail. The patches exhibited severe blackening, with elevated proportions of Cu, C, and O elements in the pitted areas. This is primarily because copper tubes are mainly composed of Cu, Mo, and Cr. The high C and O content likely resulted from the formation of corrosion products such as CuO and other non-metallic oxides. Additionally, corrosive Cl and P elements were detected on the surface, which are the primary causes of corrosion in copper tubes.

2. Following continuous immersion testing, solutions 1# and 5# turned blue, indicating the generation of Cu²⁺. The copper tubes reacted with the cleaning agent, forming dense, pale yellow corrosion crystals on their surfaces. The pressurised section (primarily carbon steel) exhibited significant yellowish-brown corrosion products. The corrugated hose (primarily 304 stainless steel) showed no noticeable surface changes.

3. Continuous immersion and air exposure tests demonstrate that corrosion under continuous immersion is significantly more severe than under air exposure. The liquid phase environment provides an ionic conduction medium, accelerating the electrochemical corrosion process. Dissolved oxygen (O₂) and Cl⁻ synergistically promote the oxidation reaction of the metal.

4. Acidic and alkaline cleaning agents exert significant corrosive effects on metal specimens. Consequently, neutral or weakly alkaline cleaning agents should be prioritised for subsequent component cleaning, avoiding acidic or strongly alkaline products. Post-cleaning, thorough water rinsing to neutrality (pH = 7) is mandatory, with residual residue inspection to prevent secondary corrosion.