In-situ 2D maps of pH shifts across brass-lead galvanic joints using microelectrodes

Abstract

Galvanic corrosion in drinking water distribution systems, such as conditions following partial lead (Pb) service line replacement, has received recent attention. In order to better understand conditions at galvanic connections that lead to enhanced metal release and provide remedial strategies, the water-metal and anodic-cathodic interfaces at these locations must be better understood. In this paper, a pH microelectrode system was used to create in-situ 2D spatial images of the pH of water across two brass coupons connected by a leaded solder joint at 100 μm above the metal’s surface under flowing and stagnation conditions. Water stagnation resulted in significant pH changes across the surfaces compared to flow condition. Under stagnation, the pH above the anode (leaded solder) was 1.5 pH units below the bulk water and as much as 2.5 units below the cathode (brass). These conditions can enhance lead release at the anode, which reflects different anodic-cathodic relationships of coupled metals primarily controlled by water flow. Most importantly, this work has demonstrated the ability to make real pH measurement at the surface of corroding metals using a novel microelectrode approach.

1. Introduction

Corrosion in drinking water distribution systems can cause serious economic, environmental and public safety problems as a result of associated pipe damage, water loss and water quality degradation. One form of corrosion--galvanic corrosion--results when two dissimilar metals are in electrical contact in water where one of the metals corrodes preferentially to the other. The impact of galvanic corrosion is an important issue that affects the quality of potable water by releasing metals of concern [1–3].

The practice of partial lead service line replacement and the associated potential increased risk of lead release into drinking water has been a concern [4–6]. Lead-containing particles can potentially be released from the lead pipe due to disturbances associated with construction to remove the partial service and galvanic corrosion at the connection between the remaining lead pipe and new service material, such as copper. Lead leaching from lead service lines and leaded solder joints can also be accelerated by the galvanic connection (table 1) with other metal materials [6–8] including brass and copper, where lead (Pb) serves as an anode (equation (1) and (2)) [9].

Table 1.

Galvanic series of selected metals and alloys and standard electrode potential (electromotive force [EMF] series)

Galvanic series of selected metals and alloys in seawater1		Electrode reaction	Standard electrode potential (V-SHE* @25 °C)2, 3
Cathodic (noble, least likely to corrode)	Gold	Au2+ + 2e− ↔Au	1.50
	Silver	Ag2+ + 2e− ↔Ag	0.80
↑	Copper	Cu2+ + 2e− ↔Cu	0.337
	Brasses
	Tin	Sn2+ + 2e− ↔Sn	−0.136
	Lead	Pb2+ + 2e− ↔Pb	−0.126
↓	Lead-tin solders
Anodic (active, most likely to corrode)	Zinc	Zn2+ + 2e− ↔Zn	−0.763
	Magnesium	Mg2+ + 2e− ↔Mg	−2.37

^*SHE: standard hydrogen electrode

$$\mathrm{Anode}\mathrm{reaction}:\mathrm{Pb}\to \mathrm{P}{\mathrm{b}}^{2+}+2{\mathrm{e}}^{-}$$ (1)

$$\mathrm{Cathode}\mathrm{reaction}:{\mathrm{O}}_2+2{\mathrm{H}}_2\mathrm{O}+4{\mathrm{e}}^{-}\to 4\mathrm{O}{\mathrm{H}}^{-}$$ (2)

Although general theories regarding galvanic interactions leading to lead release have been described, and experimental and microscopic support has been reported [6–9], most interpretations are based on bulk measurements, theory or microscopic analysis of corrosion by-products; detailed in-situ water quality dynamics at the surface of galvanic connections have not been reported. Specifically, no direct measurements of the micro-environment water chemistry at the surface of corroding metal surfaces have been conducted, and thus a practical link between bulk water chemistry and metal surface dynamics has not been fully established. Therefore, a better understanding of the surface chemical dynamics of galvanic couples leading to enhanced metal release through direct water is needed. In order to provide effective mitigating solutions for reducing metal release associated with galvanic corrosion which comply with lead and copper rule [10], in-situ water-metal interaction of galvanic joints must be better understood.

Many studies have been conducted to evaluate the impact of water quality parameters on brass corrosion, for example, pH, temperature, dissolved oxygen, free chlorine (or monochloramine), chloride, corrosion inhibitors and others [3, 11–15]. Electrochemical measurements (e.g., current density), visual observation, weight loss, metal release and surface characterization are also commonly used as analytical methods for studying corrosion in water. Arnold and Edwards [16] demonstrated a mass balance of lead release by measuring the electrochemical potential (E_corr versus Ag/AgCl) of metal and investigating lead release from galvanic connections in different water patterns, disinfectants and corrosion inhibitors. Nguyen et al [3] evaluated the relationship between the chloride to sulfate mass ratio and the associated galvanic current using an electrochemical method, surface characterization and microelectrode techniques. These methods provide a fundamental understanding of brass corrosion, however, they have been limited to bulk water quality monitoring. Direct water quality measurements in very close proximity to the metal surface may be necessary to fully understand the microenvironment and develop appropriate corrosion control strategies (e.g. lead leaching control). Although studies using electrode or ʻmicro-ʼ electrode techniques have been reported [3, 16], most have used a macroelectrode (e.g. a standard hydrogen electrode (SHE)) which is not able to provide detailed localized chemical information. pH microelectrodes (Microelectrode Inc.) were also used to investigate pH changes [3], however, the relatively large tip diameter (0.8 mm) of the pH microelectrode limited it to measuring the one-dimensional (1D) profile (i.e.1D pH and chloride profiles along with pipe distance) in a bulk of copper joint macrocell. Although the pH is one of the most important parameters for elucidating the corrosion mechanism, the measurement was limited only to the bulk due to the lack of microelectrode technique for measuring chemical profiles in the proximity of the very surface. A microelectrode is generally made of a glass micropipette which is fragile and easily broken when it touches a hard surface, limiting its application to a soft material only (e.g. biofilm, sediment and a microbial mat) [17–24]. In order to better understand conditions at galvanic connections that lead to enhanced metal release and provide remedial strategies, the localized corrosion reaction at water -metal interfaces must be better understood.

The objective of this work is to develop and demonstrate a new experimental methodology to measure pH from the bulk to the proximity to metal surfaces and construct two-dimensional (2D) pH maps across a galvanic joint, providing direct evidence of micro-environment’s chemistry dynamics. Specifically, a galvanic couple between brass and a lead-tin solder was studied. Brass is a copper-zinc alloy that often contains relatively small amounts of lead and has been widely used in drinking water distribution systems, especially in premise plumbing components, including faucets, valves and other types of connectors [13, 25, 26]. Brass coupons alone were also used for constructing a 2D pH contour map to establish a baseline. Two different flow conditions (flow versus stagnation) were compared to evaluate the effect of flow on pH at the metal surfaces.

2. Materials and Methods

2.1. Metal coupon preparation

2.1.1. Aged brass coupon.

New brass CDA 443 coupons (UNS C44300, admiralty brass,density 8.52 g cm⁻³, Metal Samples Co., Munford, AL) 1.6 mm thick×12 mm wide×140 mm long were cut into small sections (1.6 mm thick×12 mm wide×14 mm long). The sections of the brass coupons were cleaned using a combination of two American Society for Testing and Materials (ASTM) coupon wash procedures: G31–72 [27] and D2688–83 [28]. Four sections were immersed in a specifically designed Teflon flow cell (brass 1) with a flow at 2 ml min⁻¹ of a synthetic chlorinated water (figure 1). For the first 120 d, test water with a pH 7.0, 100 mg Cl⁻/L, 100 mg SO₄^2-/L, 10 mg C/L dissolved inorganic carbon (DIC), 2.0 mg Cl₂/L free chlorine and 23 ºC was fed through the flow cell. After 120 d, free chlorine concentration was increased to 4.0 mg Cl₂/L and pH was adjusted to pH 9.0 to provide an aggressive environment in order to accelerate brass corrosion and increase the likelihood of observing local water quality differ- ences (the relatively aggressive water quality is in the range of that of some waters in the United States) [2, 7, 11]. At the same time (at 120 d of brass 1), another four sections of brass coupon were immersed in a separate flow cell (brass 2) under the same water conditions (pH 9.0 and 4.0 mg Cl₂/L). Then, after 80 d of aggressive water environment, each brass coupon was taken from each flow cell and 1D pH microprofiles and 2D pH contour maps were measured for each coupon (i.e., 200 d aged coupon for brass 1 and 80 d aged coupon for brass 2). For preparation of the test water, reagent grade chemicals (NaHCO₃, NaSO₄, and NaCl) (Fisher Scientific, Fair Lawn, NJ) were added into a 20L carboy with DI water. Free chlorine solutions were prepared by diluting 6% sodium hypochlorite (Fisher Scientific, Fair Lawn, NJ) to 10000 mg Cl₂/L as stock solution, then diluted to the target concentration in the bulk test water solution for the experiments. HCl and NaOH were used for pH adjustments. The bulk water was air-saturated before the pH adjustment and free chlorine addition. All flow cells were operated in the dark by covering them with aluminum foil.

Figure 1.

Schematic diagram of the simulated water distribution system for aged brass and galvanic coupons preparation.

2.1.2. Brass-lead galvanic joint coupon.

After cleaning (see section 2.1.1 ASTM brass cleaning procedures), two small pieces of brass coupon (CDA 443, admiralty brass, Metal Samples Co., Munford, AL) were connected to each other with a 3 mm long and 8 mm wide 50:50 Pb-Sn solder (38110, Forney) to create a simulated brass-lead soldered galvanic joint coupon (8 mm in width and 25 mm in length). The fresh galvanic coupon was placed in a flow cell (figure 1) under the same aggressive water condition as brass coupons (pH 9.0, 100 mg Cl⁻/L, 100 mg SO₄^2-/L, 10 mg C/L DIC, 4.0 mg Cl₂/L free chlorine, and 23 ºC) for two weeks. The galvanic joint coupon was then taken for pH profile measurements and 2D pH contour mapping. The flow cell was also operated in the dark by covering it with aluminum foil.

2.2. Microprofile measurements and pH contour mapping

pH microprofiles were measured using a pH microsensor (pH 10, 10 μm tip diameter, UNISENSE A/S, Denmark) and an automatic 3D micromanipulator (World Precision Instruments, Sarasota, FL) (figure 2). The pH microsensor used in this work has been widely used and its performance has been previously validated [29–31]. In addition, microelectrode profile measurements, including pH measurements using a control coupon (e.g. polycarbonate slides), were conducted and validated previously [32, 33]. Coupons removed from each flow cell were moved to a transparent acrylic flow cell which was filled with the same test water (i.e., pH 9.0, 100 mg Cl⁻/L, 100 mg SO₄^2-/L, 10 mg C/L DIC, and 4.0 mg Cl₂/L free chlorine). The coupon was then acclimated under the same water conditions for one h before profile measurements. The pH microsensor was connected with an Ag/AgCl reference milli-electrode (MI-401, Microelectrodes Inc., Bedford, NH) and the sensor tip was positioned by controlling a 3D manipulator (UNISENSE A/S, Denmark) under a microscope (World Precision Instruments, Sarasota, FL). The microprofiles were then measured from the bulk to 100 μm above the metal surface. A used microelectrode was used as a guide microelectrode, which protrudes 5–10 μm ahead of a working microelectrode. This approach made the profiling possible in the proximity of the very surface without breaking the microelectrode tip. After positioning a pH microsensor, 10–30s was spent stabilizing the pH readings initially. Then, the pH (as mV versus Ag/AgCl) was measured by moving the tip of the sensor (3s), waiting before the signal recording (3s), and recording the signal (3s). 1D profiles were measured in triplicate (or duplicate) to ensure that the measured pH microprofile is representative and reproducible. Under flow conditions (the flow was controlled at 2 ml min⁻¹) the flow effect on the pH measurements was very minimal and negligible. Pre- and post-calibration were conducted for every pH profile measurement. The metal coupon surface was initially identified with a used microelectrode. The electrode signals were monitored and recorded using a multimeter (UNISENSE A/S, Denmark), a data acquisition system and a software program (SensorTrace Pro 3.0, UNISENSE A/S, Denmark). Microelectrode profile measurements using a control coupon (e.g. polycarbonate slides) were conducted and validated previously [32, 33].

Figure 2.

pH 2D contour mapping experimental setup. (a) microprofiling setup, (b) flow cell system for microelectrode profile measurements.

For the 2D pH map, pH was measured at 100 μm above the metal surface with a certain grid. For the 80 d aged brass coupon, a total of 25 points (5×5) were measured over the coupon surface of 6400 μm (W) × 7200 μm (L) with a distance interval of 1600 μm (W) and 1800 μm (L). For the 200 d aged brass coupon 2D pH map, 25 points (5×5) were measured over the coupon surface of 4000 μm (W) × 7200 (L) with a distance interval of 1000 μm (W) and 1800 μm (L). The triplicated data was recorded in each point for 2D mapping. The triplicates for the pH measurements indicate three profile measurements at one point before moving to the next point. 1D pH microprofiles were measured from the bulk (2000 μm above the metal surface) with a 50 μm interval, and at least two random locations were selected to investigate the heterogeneity of measured pH profiles (figure 3). Each measurement was replicated at least once. The total duration for obtaining a 2D pH contour map of a single brass coupon was 1.5–2.5h. For the galvanic solder coupon measurement, 30 points (3×10) were selected over the coupon surface of 4000 μm (W) × 18000 (L) with an interval 2000 μm (W) and 2000 μm (L) under flow conditions, and 65 points (5×13) were selected over the coupon surface of 4800 μm (W) × 24000 (L) with an interval of 1200 μm (W) and 2000 μm (L) under stagnant conditions. Triplicated data was also recorded in each point for 2D mapping. Three different locations (brass|solder|brass) were selected for 1D pH microprofile measurement (figure 4). Each measurement was replicated at least once. The total duration for obtaining a 2D pH contour map of a galvanic coupon was 3–4 h. After surface pH scanning with the pH microelectrode, 2D maps were constructed using a programming language (Python 2.7.10, Python software). Free chlorine concentrations in the bulk solution in a flow cell were measured before and after the experiments by a colorimetric test kit (Hach–8021) and a DR 5000 spectrophotometer (Hach Co.). The weight loss of the galvanic solder coupon was measured on a weekly basis. The pH microsensor was pre- and post-calibrated during the profiles to make sure that the pH changes are not from the electrode signal drift. The coupon images were taken using a built-in CCD camera (World Precision Instruments, Sarasota, FL) during pH profile measurements simultaneously.

Figure 3.

Optical microscopic images of brass coupons and galvanic coupon in a microprofiling flow cell. (a) 80 days aged brass coupon. (b) 200 days aged brass coupon. (c) 14 days aged brass-lead galvanic connection coupon.

Figure 4.

2D pH contour map of 80 days aged brass coupon. (a) 2D map grid with scale, (b) with flow (2 ml/min). (c) under stagnation. The pH in the bulk solution was 9.0.

2.3. Metal surface characterization

Scanning electron microscopy (SEM) observation with energy dispersive x-ray spectroscopy (EDS) was conducted with a Zeiss ULTRA-55 FEG SEM to characterize the corrosion by-products which were found on the metal surface during the profiles.

3. Results and Discussion

3.1. 2D pH swing on aged brass coupons

After 80 and 200 d of flow cell operation, the colors of both brass coupon surfaces turned from their original bright yellow to dark brown, reflecting active corrosion and the development of corrosion by-products on the surface (figures 3(a) and (b)), as expected and reported in other brass corrosion studies [11, 13, 25, 34]. The 200 d brass coupon showed more homogenous dark brown corrosion by-products formation across the surface than the 80 d coupon.

2D pH surface profiles above the brass coupons showed that the water flow influenced the pH across the coupon surface (figures 4 and 5). During flow conditions, the pH near the surface was relatively uniform across the coupons after 80 and 200 d (table 2, figures 4(b) and 5(b)) and averaged pH 9.0 which was the same as the bulk pH. However, after 1.5–2.5h of stagnation, the pH ranged from 9.0–9.5 and averaged 9.4 ±0.08 across the brass coupon surface (figure 4(c)). At 200d, the pH ranged from 9.1- and averaged 9.2±0.08 across the brass coupon surface (figure 5(c)). Water stagnation generally increased the pH at the brass surface as compared to the bulk water. In addition, the pH across the surface was much more variable during stagnation. During the stagnation test, the free chlorine residual was decreased to 3.8 and 3.1 mg Cl₂/L for 80 and 200 d brass coupons, respectively. This indicates that the pH may continue to change as the stagnation time increases (i.e. free chlorine residual decreases). Water stagnation resulted in a wider range of local pH differences across the surface compared to flow conditions, indicating that even in a small area, brass coupons can be polarized and thus corrosion proceeds non-uniformly [25].

Figure 5.

2D pH contour map of 200 days aged brass coupon. (a) 2D map grid with scale, (b) with flow (2 ml/min). (c) under stagnation. The pH in the bulk solution was 9.0.

Table 2.

Summary of the surface pH changes on brass coupons and a galvanic solder coupon.

Metal Coupons	Flow condition		Stagnation
	Ave. pH	ΔpH	Ave. pH	ΔpH
Aged brass (80 days)	9.0 ± 0.03 max.: 9.1 min.: 9.0	0.1	9.4 ± 0.08 max.: 9.5 min.: 9.0	0.5
Aged brass (200 days)	9.0 ± 0.08 max.: 9.1 min.: 8.8	0.3	9.2 ± 0.08 max.: 9.4 min.: 9.1	0.3
Brass-lead galvanic connection (14 days)	9.1 ± 0.12 max.: 9.4 min.: 8.9	0.4	8.9 ± 0.51 max.: 10.0 min.: 7.9	2.1

^*Initial bulk pH was 9.0.

1D pH profiles (figure 6) further demonstrated that the pH across the brass surface under flow conditions was relatively constant. The pH was also nearly the same as the bulk water pH (at 200 d) and slightly higher to pH 9.1 (80 d). Water flow influenced surface pH near the brass surface within a diffusion boundary layer of 600–800 μm thickness. The pH increased from 9.5 to as high as 9.5 (at 80 d) and 9.3 (after 200 d) during the stagnation. The surface pH of the 80 d brass coupons under flowing and stagnant conditions was slightly higher than the 200 d brass coupons, suggesting that newer surfaces were more reactive in chlorinated water. For the 200d aged brass coupon, dark brown deposits covered the entire surface, which may have contributed to the lower pH surface change during stagnation (figure 6) as compared to the 80 d aged brass coupon. In contrast, corrosion by-products on the surface of the 80 d brass coupon were non-homogenous and poorly established, which may have resulted in relatively higher cathodic reactions [8] along with higher pH across the surface under stagnation (figure 4(c)). An interesting observation for the 200 d aged coupon is that, as shown in figure 3(b) which was taken after the experiment, white compounds were formed at the end of the brass coupon following a flow direction. On the other hand, with 80 d aged brass coupon, there were no similar findings. It seems that the formation of zinc meringue corrosion by-products [11] may be initiated after the completion of the surface oxidation (i.e. 200 d aged coupon with entire color change to dark brown).

Figure 6.

Measured pH microprofiles. (a) pH microprofiles of 80 days aged brass coupon with and without flow. (b) pH microprofiles of 200 days aged brass coupon with and without flow. The pH in the bulk solution was 9.0. Two locations per one coupon (Fig. 3) were measured and an average value with error bar was presented for each pH profile.

3.2. 2D pH swing on a brass-lead soldered joint coupon

Previous studies have assessed lead corrosion and leaching from solders by analyzing bulk solutions [3, 35, 36] from which conceptual description of the anodic and cathodic conditions have been derived rather than direct surface measurements. In this experiment, direct pH measurements were performed at the surface of the brass-lead solder joint coupon after two weeks of conditioning under flow. During this period, white corrosion by-products developed on and near the joint area. Surface characterization using SEM/EDS of the material was attempted, but failed due to the interference of other salt precipitates. A large amount of sodium that remained after the sample treatment (i.e. centrifugation and dry) interfered with detecting the lead and tin. Based on the observations of others [7, 37], and material appearance, location, and geochemical considerations, the white deposits are likely to be one or more lead and/or tin corrosion by-products. The test water was gently flushed over the coupon surface for 5 min before profile measurements were initiated to avoid potential interference from the particulate deposits [1, 38]. The 2D pH map (figure 7(b)) clearly showed a pH increase across the leaded solder joint under flow conditions. The pH at both brass ends remained the same as the bulk pH. The pH increased, moving inward from the brass ends toward the leaded solder joint, reaching a maximum value of 9.4 at the downstream left end of the lead connection. A weight loss of 4.2mg (0.11%) was measured after exposing the coupon to the bulk water at a flow of 2 ml min⁻¹ continuously for 7 d, indicating the possible zinc and/or copper ions release. However, when water flow stopped and stagnant water contacted the brass and lead solder (simulated soldered joint), the 2D pH surface showed a different phase as opposed to the pH shift under flow conditions. Under stagnation, a corrosion cell was formed between metals in solder (Pb and Sn) and the brass, resulting in a relatively corrosive microenvironment at the solder surface which may contribute to rapid lead release [6, 39]. The different surface pH dynamics between flow and stagnation were quantitatively determined for a galvanic solder coupon (table 2). After 3–4h of stagnation, the lead surface became anodic with a pH drop to 7.9, while the pH on brass increased up to 10.0. The pH difference between cathodic area (brass) and anodic area (lead) was 2.1, and the average pH across the galvanic connection was 8.9±0.51.

Figure 7.

2D pH contour map of 14 days aged brass/lead solder joint coupon. (a) 2D map grid with scale for flow condition, (b) 2D map with flow (2 ml/min), (c) 2D map grid with scale for stagnant condition, and (d) 2D map under stagnation. The pH in the bulk solution was 9.0.

1D pH profiles (figure 8) measured at three different locations showed that under flow conditions, the pH profile on the brass coupon was similar to the bulk water pH, while the pH on the lead joint increased from the bulk pH of 8.8–9.3 (ΔpH: 0.5). However, under stagnation, the pH on lead joint showed an opposite aspect. During the stagnation, the bulk pH dropped to 8.6, however, the pH on the brass increased up to 9.9, while the pH at the galvanic connection location dropped further to 8.0 near the lead surface (figure 8(b)). This observation, along with 2D pH surface, demonstrated that: (1) under water stagnation, the galvanic connection significantly promoted a cathodic reaction at the brass, contributing to pH increase, and (2) lead ions released at the joint could potentially contribute to the formation of soluble complexes or insoluble precipitates that contain OH⁻, Cl⁻ and SO₄^2- [3]. The chlorine residual was 2.4 mg Cl₂/L after the experiment. DeSantis et al [8] showed deep corrosion of brass or copper piping immediately adjacent to soldered joint, and summarized brass/copper materials behaving anodically when coupled with lead, contrary to conventional wisdom of commonly referenced galvanic series tables and standard electro-potential series.

Figure 8.

Measured representative pH microprofiles. (a) pH microprofiles of 14 days aged brass/lead solder joint coupon under flow condition (2 ml/min) and (b) pH microprofiles of 14 days aged brass/lead solder joint coupon under stagnation. The pH in the bulk solution was 9.0.

The main work in this article was to construct, by direct measurement, the 2D pH map to demonstrate large pH variation across a galvanic couple. The direct pH information on the metal surface in this study supported the theories which have been assumed for many years. For example, from two theories in a previous study [6], one of the hypotheses was a galvanic corrosion of lead can release lead ion (Pb²⁺) into the water environment leading to a local pH decrease at the lead surface. The 2D pH map in this study showed the direct evidence (i.e. local pH drops) supporting this theory in a microscale environment other than buffer solution measurement. The second theory was the ‘deposition corrosion of leadʼ [39] which can lead to a local pH rise at the lead surface. This was also measured and proven using microelectrode measurements. In this study, it was obvious that the 2D pH surface under stagnation was completely opposite to the 2D pH surface under flow conditions, indicating that the water flow condition may be one of the most significant parameters controlling anodic-cathodic relationships of coupled metals in addition to water chemistry. It appeared that under flowing conditions, micro-galvanic deposition corrosion [6, 39, 40], or the formation of hydrocerussite (Pb₃(CO₃)₂(OH)₂) or Pb (IV) oxides [2, 41, 42], would be the dominant reaction (i.e. deposition corrosion of lead), whereas under stagnation, galvanic corrosion releasing lead ion [6] would be the main cause. The observations reflect an opposite pH trend with different flow conditions and galvanic interactions between lead solder-brass joints under stagnant water conditions. Under continuous flow conditions, the surface pH was relatively uniform and close to the bulk water pH. However, under stagnation, the surface pH varied by as much as 2.1 pH units. In the case of brass-solder connections. pH scanning images under flow and stagnation showed opposite pH changes over the solder surface, which could be the possible different mechanisms of the lead corrosion process under flow and stagnant conditions. The pH was notably lower at the solder surface while the brass surface pH was much higher. The galvanic connection between the brass and lead/tin solder appeared to ‘protectʼ the brass, but the corrosion of the solder may be accelerated under stagnation. The differences in the pH on the surface likely reflected anodic and cathodic regions across the metals. As practical implications from the study, the pH in the bulk solution and at the metal surface may be different. In particular, when it comes to the brass-lead galvanic joint under stagnation, the bulk pH monitoring as a corrosion assessment would not provide accurate information regarding the magnitude of corrosion at the interface of galvanic joints.

4. Conclusion

This work demonstrated a new experimental method to directly measure pH in the proximity of the metal surface, which has never been successfully conducted previously. This measurement provides direct evidence of localized metal corrosion mechanisms in a micro-environment, confirming what others have previously speculated or indicated by alternative ‘macroʼ methods. Future work will seek to collect more spatiotemporal microprofile data with regard to longer-term corrosion processes under various water conditions, along with other important chemical profiles, such as free chlorine (or monochloramine), oxygen, redox potential, lead, zinc, and phosphate as a corrosion inhibitor. The extension of the localized water chemistry measurements using microelectrodes will lead to a better understanding of the corrosion mechanisms, and will validate existing theories.