Tarnishing of a Cu-Al-Zn-Sn Alloy Compared to Commercially Pure Copper: Implications Toward Antimicrobial Function

Methicillin-resistant staphylococcus aureus (MRSA) is a highly contagious bacterium and is spread mainly by hand-to-surface contact. In 2005, 94,650 patients in the United States contracted MRSA, and 18,650 of these patients died. MRSA is especially prevalent in hospitals, where patients are infected with MRSA by touching surfaces that will, in turn, be touched by other patients whose immune systems may be compromised.

This research is related to the goal of minimizing the spread of antibiotic-resistant diseases in hospitals enabled by corrosive release of Cu+ and Cu2+ from copper alloys in solutions such as human perspiration. At the same time it is desirable to maintain color stability on hospital surfaces by minimizing tarnishing in the form of corrosion product formation. This creates a clearly contradictory situation. Unfortunately, the copper alloys with the best antimicrobial efficacy are often those that tarnish more readily. Conversely, the most corrosion resistant copper alloys in human perspiration often do not exhibit very good antimicrobial efficacy especially after long periods of oxide passivation after abrasion. Nordic Gold (89% Cu, 5% Zn, 5% Al, 1% Sn) was reported to kill MRSA when freshly cleaned. However, Nordic Gold passivated and copper release was limited when exposed to ambient lab conditions for 7 days.2 However, kill rate studies have not been conducted in a solution that mimics hospital conditions (i.e. salt transfer from hand perspiration due to frequent skin contact).

The goal of this thesis was to understand tarnishing, copper release and color stability of a 89% Cu- 5% Al-5% Zn-1 % Sn solid solution alloy (Nordic Gold) during corrosion in various “high touch” hospital environments such as full immersion in synthetic perspiration solution or full immersion in concentrated synthetic perspiration solution simulating the equilibrium conditions at low %RH, as well as during cyclic wetting and drying of synthetic perspiration solution. This alloy was compared to commercially pure Cu, C11000. Additionally, the role of thin oxides such as those formed by prolonged prior air oxidation was examined during subsequent full immersion testing. This was accomplished by tracking the “fate of copper” during oxidation from the elemental state in solid solution through the corrosion process by analysis of the total copper oxidation, the oxide layer formed, as well as solution analysis for dissolved copper. Electrochemical, gravimetric, surface science and solution spectrometry tools were utilized.

In full immersion exposure to synthetic perspiration solution, Nordic Gold exhibited decreased instantaneous corrosion rates, a lesser degree of tarnishing, and a thinner corrosion product layer while maintaining comparable copper ion release rates to C11000. The only corrosion products detected were CuO and Cu2O on both Nordic Gold and C11000. Additionally, the presence of thin air formed oxides during lab air passivation had minimal effect on subsequent corrosion of C11000, but more noticeable effects on Nordic Gold. Air oxidized Nordic Gold exhibited decreased instantaneous corrosion rates compared to the freshly ground alloy surface, a thinner corrosion layer, but comparable copper release to the freshly ground sample. This behavior is hypothesized to be caused by the complex effect of Nordic Gold’s alloying elements. Sn, Zn, and Al help form a more compact oxide, but still enable the release of similar quantities of Cu cations. This indicates that there is insufficient Zn present to limit Cu release by suppression of the OCP. The Cu release enhancing effect of Sn seen by Goidanich is also operative in Nordic Gold.

Concentrated synthetic perspirations solution was determined to result in a stronger thermodynamic driving force to oxidize copper alloys without direct oxide formation. Following full immersion exposure to concentrated synthetic perspiration, both alloys exhibited increasing instantaneous corrosion rates over time. Nordic Gold was observed to have a thinner compact oxide layer form after exposure to concentrated synthetic perspiration, and released less total metal ions into solution compared to C11000. Prior air oxidation resulted in an increase in Cu release from the alloy. A greater variety of corrosion products, including precipitated Cu2(OH)3Cl, were detected compared to freshly ground alloys where only CuO and Cu2O were detected.

Corrosion behavior as a result of synthetic perspiration deposition and cyclic wetting and drying was determined to be strongly dependent on the number of cycles samples were exposed to. Nordic Gold and C11000 were observed to have no statistical difference in total oxidation charge determined by mass loss. Evidence of CuO, Cu2O, Cu2(OH)3Cl and Cu2(OH)2CO3 were observed on both alloys. CuO and Cu2O were theorized to form by direct oxidation while Cu2(OH)3Cl and Cu2(OH)2CO3 products were formed by chemical supersaturation of the droplet, resulting in precipitation by homogeneous chemical reaction. Nordic Gold had slightly lower mass gain over 12 dry-wet cycles, suggesting less precipitated corrosion products as governed by these homogeneous chemical reactions. However, Nordic Gold resulted in thinner electrically connected compact oxide layers. These thinner corrosion films enable more copper release from the bulk alloy despite comparable corrosion rate and a lower Cu composition in the bulk alloy. Moreover, since the overall corrosion rate was similar, ion release was comparable or slightly greater on Nordic Gold. Sn, Al or Zn directly or indirectly enhance Cu release. Based on dissolved Cu concentrations measured, it is suspected that both alloys would provide sufficient Cu2+ necessary to kill E. coli (HCB1) bacteria in time periods less than 8 hours.

The most significant factor in corrosion behavior was found to be the environmental condition (i.e. full immersion, dry-wet cycles, synthetic perspiration, concentrated synthetic perspiration) given that both alloys possess greater than 89% Cu. While all scenarios facilitate copper ion release at concentrations above those necessary to kill bacteria, concentration build-up in drying-wetting testing with synthetic perspiration solution resulted in the greatest copper ion release rate, followed by full immersion testing with concentrated perspiration solution simulating the concentration just before drying or low %RH. Lastly, full immersion testing with normal concentration synthetic perspiration solution was the least aggressive. Additionally, corrosion layer thickness and Cu ion release varied by solution and alloy. Cyclic wetting and drying in synthetic perspiration yielded Cu based precipitates due to limited droplet volume subject to solubility laws, followed by full immersion with synthetic perspiration solution and lastly full immersion with concentrated perspiration. All solutions generated CuO and Cu2O corrosion products, with the formation of Cu2(OH)3Cl especially aided by the high Cl- content in drying-wetting testing. Additionally, copper carbonate products formed during cyclic drying-wetting in synthetic perspiration, likely due to extended exposure to lab air, thin electrolyte films and the governing solubility laws. It is hypothesized that oxide formation occurs by direct oxidation to form a compact CuO and Cu2O film, and by homogeneous chemical precipitation due to supersaturation of the droplet volume with metal cations.
The work in this thesis contributes to the assessment of copper alloys for hospital applications by assessing corrosion behavior as opposed to empirical bacteria kill testing. As such, it evaluates the corrosion process by which a copper alloy is antimicrobial. Results in this thesis indicate that alloy composition and environment are the primary factors controlling corrosion behavior. A complex effect is evident in Nordic Gold alloy with Sn, Zn, and Al. Future work is suggested continue to define the role of corrosion environment on the fate of copper. For instance, antimicrobial efficacy is typically tested in nutrient broths instead of perspiration. Additionally, this work established the methodology and foundation to bridge empirical antimicrobial testing through bacteria kill rates with corrosion studies determining ion release directly from copper alloys.