Simultaneous culture and biomachining of copper in MAC medium: A comparison between Acidithiobacillus ferrooxidans and Sulfobacillus thermosulfidooxidans

Biomachining will not be considered as a full-scale manufacturing technology until a stable, controlled and continuous metal removal rate (MRR) is achieved. In this research work, a novel strategy that could promote its industrial implementation, namely simultaneous bacterial growth and machining of copper contained in oxygen-free copper (OFC) workpieces, was investigated. This proposal has the major advantage of being a single-stage process, thereby reducing total operating times and becoming more economical in comparison with conventional biomachining (downtime due to bacterial growth would disappear). The study was carried out using mesophilic (Acidithiobacillus ferrooxidans) and thermophilic (Sulfobacillus thermosulfidooxidans) extremophile bacteria in order to prevent the progressive decrease in the amount of metal removed per unit time. A constant MRR of 43 mg h –1 was achieved with A. ferrooxidans in the simultaneous process. Despite the accomplishment of a constant MRR, this value is lower than the maximum MRR obtained in conventional biomachining (109 mg h –1 ), probably due to the inability of ferric ions to come into contact with the metallic surface. With regard to the culture period in MAC medium, S. thermosulfidooxidans showed a slower growth rate (0.11 h –1 ) and lower ferrous ion oxidation level (0.12 g Fe 2+ L –1 h –1 ) than A. ferrooxidans (0.17 h –1 and 0.22 g Fe 2+ L –1 h –1 , respectively) under optimal pH (1.5) and Fe 2+ concentration (6 g L –1 ) conditions.


Just Accepted
"Just Accepted" manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides "Just Accepted" as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. "Just Accepted" manuscripts appear in full in PDF format accompanied by an HTML abstract. "Just Accepted" manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). "Just Accepted" is an optional service offered to authors. Therefore, the "Just Accepted" Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the "Just Accepted" Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these "Just Accepted" manuscripts.

INTRODUCTION
Copper has been identified as a key metal for industrial and social progress. Indeed, both copper and its many alloys (e.g., oxygen-free copper (OFC) pieces) have become strategic materials due to their economic and environmental benefits 1 . It is estimated that two-thirds of the 550 million tonnes of copper produced since 1900 are still in productive use 2 . Its excellent properties, such as mechanical strength, conductivity, corrosion resistance, machinability, ductility and recyclability, make copper ideal for use in a large number of products and a wide range of applications, including electronics, communication, electricity and transport. For instance, a typical medium-sized car may contain up to 22.5 kg of this metal, and new generation aircraft and trains between 2 and 4 tons 1 .
With regard to OFC microfabrication, physicochemical methods, such as milling, turning, drilling, laser, electrical discharge machining and chemical machining, have traditionally been applied. However, the high fabrication rates achieved have masked the drawbacks associated with these processes, namely high production costs, thermal and/or structural damage, and the application of harmful compounds for both employees' health and the environment. As a result, and given the environmental impact caused by the traditional manufacturing sector, concepts such as ecological and sustainable production are gaining more and more importance, and industries are increasingly being forced to retrain their processes (either by integrating new processes or transforming existing ones), thereby moving towards a greener manufacturing technology. According to Pusavec et al. 3 , any environmentally sustainable process is based on four basic principles: (1) process cost depletion (including energy consumption and manufacturing process costs), (2) generated waste reduction and waste-treatment upgrading, (3) occupational health progress, and (4) a decrease in environmental impact.
Numerous approaches based on traditional manufacturing processes that improve progress towards socalled "green manufacturing" have arisen recently. These new processes are more efficient and ecofriendly than existing ones. Although their application to date has been limited to a laboratory scale, biomachining, which is defined as a process based on microbiological activity to form microstructures on a metal workpiece by metal removal (or dissolution) using microorganisms, is one of the possible alternatives for OFC microfabrication purposes 4 . Biomachining is characterised by machining components with minimum heat or residual stress and without damaging the metallurgical properties of the workpieces. Microorganisms responsible for the metal processing are not only commercially available, but also can be cultured and be readily available at any moment. In addition, energy consumption is low, thus resulting in a subsequent cost saving in the manufacturing of high quality pieces 5,6 .
Two main mechanisms (direct and indirect pathways) have been postulated to explain metal bioetching 7 . Regarding the direct mechanism, bacterial adhesion on metal surface allows the enzyme activity to be responsible for metal oxidation and dissolution 5 . Nevertheless, according to the literature, this process is not the priority pathway for zero-valence copper (Cu 0 ) biooxidation 6,8 . Xenofontos et al. 8 observed that even if the effect of microorganisms on copper foil surface roughness was significant, the direct mechanism contributed only 5 % to the total biomachining process.
Copper biomachining via indirect mechanism typically involves two consecutive stages 9 . Initially, chemolithotrophic and acidophilic bacteria are grown using the energy generated by the oxidation of ferrous ions into ferric ions (Reaction 1), and in a second step, the biogenic ferric ion causes the dissolution of copper metal present in the bacterial medium (Reaction 2). (Reaction 2) Ideally, a closed cycle should be formed by joint interaction between the biological (Reaction 1) and chemical (Reaction 2) reactions. However, the relatively low biological metabolism in comparison with the fast chemical copper dissolution rate results in a depletion of the Fe 3+ concentration in the liquid medium (Figure 1a), thereby decreasing the metal removal rate (MRR) over time 10 . In previous studies 4 , the MRR was optimized by means of regeneration periods between successive stages of biomachining ( Figure 1b). Thus, during the chemical solution bioregeneration, the piece of metal was submerged in a second bioreactor in order to continue the leaching process at the highest possible MRR 11 .

Figure 1.
Evolution of Fe 2+ concentration (black line) and metal removal rate (grey line) for a conventional biomachining process (A), an alternative biomachining process including regeneration periods (B), and a simultaneous bacterial growth and biomachining process (C).
The work described herein explores the possibility of carrying out the bacterial growth and machining stages simultaneously in order to provide a continuous, stable and sustainable biomachining process. The total operating time would be reduced in comparison with conventional biomachining as the growth period (Reaction 1) would act as a producing stage that occurs alongside copper dissolution (Reaction 2). In addition, the energy source for bacterial growth (i.e., Fe 2+ ) would be readily available since the ferric ions generated would be rapidly reduced upon interaction with elemental copper (Cu 0 ) (Reaction 2). Such an alternative approach could provide an endless energy source for the microorganisms, thus 4 meaning that they would be immersed in a growth stage at a suitable activity level for machining purposes (Figure 1c). The most commonly used bacterium in biomachining is the autotrophic, acidophilic and mesophilic bacterium Acidithiobacillus ferrooxidans (A. ferrooxidans). Thermophilic bacteria typically employed in the bioleaching of sulfide ores and environmental detoxification of heavy metals have not yet been investigated as possible agents for increasing the biomachining rate. As such, experimental data regarding the growth and machining capacity of Sulfobacillus thermosulfidooxidans (S. thermosulfidooxidans) were also compared with those for A. ferrooxidans in this study.

Preparation of copper samples
OFC workpieces with a minimum purity of 99.99 % and a size of 10 × 15 × 2 mm were manufactured in the Department of Mechanical Engineering of the University of the Basque Country (UPV/EHU) using an 800-grit abrasive disk wheel (Petrographic cleaner LS1, REMET, Italy). Each workpiece had a hole (2 mm in diameter and 2 mm thick) to allow complete immersion in the active medium when performing the tests (SI Figure S1). Prior to immersion in MAC medium, the OFC blocks were rinsed with deionized water and ethanol (96 % v v -1 ) and then heated to remove surface moisture.

Microorganisms and culture media
A. ferrooxidans DSM-14882 was kindly provided by the Department of Chemical Engineering and Food Technology of the University of Cádiz (UCA) and S. thermosulfidooxidans was obtained from Copahue geothermal ponds (Neuquen, Argentina).
The mineral medium (MAC 12 ) composition (g L -1 ) was: As the pH was not adjusted during the culture period, this stage was concluded when complete oxidation of Fe 2+ to Fe 3+ was achieved (less than 1 % of the initial Fe 2+ concentration), which indicated that bacterial growth was satisfactory, or jarosite precipitation due to progressive basification was observed 4 .
To analyse the effect of pH and Fe 2+ concentration, two pH values (1.5 and 2.0) and three Fe 2+ concentrations (3, 6 and 9 g Fe 2+ L -1 ) were tested. pH tests below 1.5 were not performed since a marked reduction in A. ferrooxidans viability has already been observed by Gomez and Cantero 13 when working at a pH of 1.25. With regard to S. thermosulfidooxidans, prior research indicated that ferrous ion oxidation rates were similar between pH 1.4-2.0, with an optimum pH of 1.7 14 . Deionized water was added during the experiments to compensate water losses due to evaporation. Blank tests following the ). Each experiment was performed in duplicate and the mean results were used.

Conventional biomachining process
Biomachining tests were carried out by suspending pre-weighed OFC copper blocks (Sartorius Practum 124-1S analytical Balance, Spain) in 250 mL conical flasks containing the aforementioned culture media (A. ferrooxidans or S. thermosulfidooxidans, both cultured up to stationary phase 15 ) under the optimal conditions based upon the results obtained in "Parameters affecting the culture period" section (pH of 1.5 and 6 g Fe 2+ L -1 ) ( Table 1). The influence of workpiece dimensions 16 was avoided by selecting analogous pieces in terms of both weight and total surface area. A pH threshold value of 1.8 was maintained in all experiments by addition of sulfuric acid (25 % v v -1 ) 4 .

Biomachining during the culture period
Tests to analyse the effect of conducting bacterial culture (A. ferrooxidans or S. thermosulfidooxidans) and biomachining simultaneously were performed in four different solutions, as shown in Table 1. In this case, copper bioleaching was considered to be complete either when dissolutions presented jarosite formation (in the samples with no pH control) or the MRR ceased to increase over time. Irrespective of the fact that copper OFC samples were placed within the flasks together with the uncultured inoculum (SI Figure S1), the same conditions as those mentioned in the previous section were applied in these experiments. The metal samples tested in the latter two sections were removed from the solution at user-defined intervals, rinsed with deionized water and ethanol (96 % v v -1 ), dried, weighed, and then reintroduced into the liquid samples. Each experiment was performed in duplicate and the mean results were used. The copper removal rate (MRR) achieved was calculated as follows 9 :

Other analyses
The total cell number was calculated using a counting chamber (Improved Neubauer counting chamber, Zuzi, Spain) in conjunction with a phase-contrast microscope (Labophot, Nikon, USA). Additionally, cell densities were compared using the 4',6'-diamidino-2-phenylindole (DAPI) stain technique 18 . Supporting data, namely pH (GLP 21+ pH-meter equipped with a sensION+ 5014T glass combination pH electrode, Crison, Spain) and redox potential (Orion 9778BNWP Sure-Flow® electrode with epoxy body (combination of a platinum redox and a silver/silver chloride reference electrode in one body (Ag/AgCl, 4M KCl)) connected to a Thermo-Orion 920A+ meter, ThermoFisher Scientific, Germany), were measured to provide additional information regarding the stage of the process. All potentials in this paper are given with respect to the reference electrode (+220 mV vs. Ag/AgCl).

A. ferrooxidans
An optimal culture medium formulation and suitable growth conditions are essential in any biological process in which bacteria are involved in order to attain their maximal growth rate and replication. For A. ferrooxidans culture in MAC medium, samples with an initial pH of 2.0 presented a slightly lower growth rate in comparison with samples with an initial pH of 1.5. A difference of approximately 10 % in the bacterial concentration increase was observed when samples with a Fe 2+ concentration of 3 g L -1 but different pH values (1.5 or 2.0) were compared ( Figure 2). For the sake of clarity, samples with an initial pH of 2.0 and the highest Fe 2+ concentrations (6 g L -1 and 9 g L -1 ) were not included in Figure 2. In these latter two cases, jarosite precipitation (SI Figure S2) was evident when the pH rose above 2.3, which was in accordance with Nazari et al 19 . As expected, a higher bacterial density was achieved in those samples with a higher iron concentration, which was visually corroborated by means of DAPI analysis (SI Figure  S3).
Biological oxidation of Fe 2+ to Fe 3+ in samples grown at a pH of 1.5 resulted in an average increase in the pH value of 0.5 as a result of the consumption of hydrogen ions. Proton depletion was similar irrespective of the Fe 2+ content (Figure 2). The time required by the inoculum to accomplish Fe 2+ oxidation and achieve the maximum redox potential (590-630 mV vs. Ag/AgCl) essentially depended on the amount of ferrous ion in the media (SI Figure S4). In all cases, the culture period was completed within 38 h. Blank tests (abiotic tests developed to observe ferrous ion oxidation associated with dissolved oxygen) showed a negligible Fe 2+ depletion in samples at pH 1.5 or pH 2.0. For the same culture period (24 h), minor increases (6-17 mV (↑ 2-5 %)) in the redox potential associated with the conversion of Fe 2+ into Fe 3+ were measured, which confirmed the applicability of the redox potential as a rapid measurement for gaining information about the status of the biomachining process 4 .  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58 59 60 Taking into account that the variables with the highest influence on A. ferroxidans metabolism are ferrous and ferric ion concentration, pH, temperature, dissolved oxygen concentration, carbon dioxide concentration and inert solids 13 , the effect of pH and Fe 2+ concentration on the Fe 2+ oxidation rate (ν Fe 2+), specific bacteria growth rate (µ) and specific substrate consumption (q s ) were addressed in this section in order to establish the optimal conditions for A. ferroxidans culture in MAC medium ( Table 2). The µ and ν Fe 2+rates were calculated during the bacterial exponential growth phase, as defined by Blanch and Clark 20 . q s was obtained as the ratio between bacterial growth rate and cell yield, which is defined as the amount of cells produced per unit amount of substrate consumed 20 .
Previous studies have demonstrated that the rate of Fe   Samples with an initial pH of 2.0 and the highest Fe 2+ concentrations (6 g L -1 and 9 g L -1 ) were not included due to jarosite formation, a physical barrier which prevents effective diffusion of the reagents and products.  The ferrous ion oxidation rate was higher in those samples with higher ferrous ion concentration, probably due to the fact that increasing amounts of Fe 2+ resulted in greater densities of cells that are able to transform more Fe 2+ ions into Fe 3+ 9 . The linearity between initial ferrous ion concentration and its oxidation rate found in this study is not fully supported in the literature. Thus, Danis et al. 31 observed that increasing Fe 2+ concentrations up to 2.6 g L -1 gave increased Fe 3+ levels, but higher ones led to lower Fe 3+ concentrations. Karamanev and Nikolov 32 showed that the A. ferrooxidans oxidation rate was almost unaffected by ferrous ion concentration from 4 g L -1 upwards.
The stagnation or even decrease of the ferrous ion oxidation rate is related to the inhibition suffered by A. ferrooxidans 31,32 . This fact was evident in our study, since both the specific bacterial growth rate and specific substrate consumption decreased at the highest ferrous ion concentration (9 g Fe 2+ L -1 ).
Similarly, Barron  ) were slightly higher than those found by other authors, although it should be noted that different strains and nutrient media were employed in those reports [23][24][25][26][27] .
Consequently, the optimal conditions were found to be pH 1.5 and a ferrous ion concentration of 6 g L -1 , which ensured the highest bacterial growth rate, with moderate acid consumption and limited possibility of jarosite formation.

S. thermosulfidooxidans
Bioleaching of copper using mesophilic iron-oxidising acidophilic bacteria, such as A. ferrooxidans, which develop relatively high redox potentials during leaching processes by maintaining high ferric to ferrous ion ratios, has been extensively studied. However, the use of moderately thermophilic bacteria could be beneficial to boost bioleaching processes since higher temperatures improve both bacterial leaching kinetics and metal dissolution 33 . Cruz et al. 33 observed that greater and faster nickel extraction was achieved from nickel sulfide ores when using S. thermosulfidooxidans in comparison with A. ferrooxidans, mainly due to the temperature effect and the rapid increase in solution redox potential. In addition, S. thermosulfidooxidans has been reported to be tolerant to copper ions during ferrous ion biooxidation, and its cell-growth inhibition has been established at 22 g L -1 14 . In this study, only the sample supplemented with yeast extract was able to complete Fe 2+ oxidation.
Thus, S. thermosulfidooxidans acted as a relatively poor ferrous ion oxidiser when grown in pure culture ( Figure 3). This is in accordance with Pina et al. 29 , who highlighted the benefits of mixotrophic conditions at the expense of autotrophic ones (Table 2). S. thermosulfidooxidans showed slower growth kinetics (µ = 0.11 h -1 ) and lower ferrous ion oxidation rates (ν Fe 2+ = 0.12 g Fe 2+ L -1 h -1 ) than A. ferrooxidans under the same conditions. In light of the results obtained, and setting aside the benefits owing to the higher temperatures applied (which also involves the handicap of an additional energy consumption), no potential benefit for the process can be expected in this case when using S. thermosulfidooxidans instead of A. ferrooxidans.

Biomachining experiments
Proteobacteria of the genus Acidithiobacillus, and more specifically A. ferrooxidans or A. thiooxidans, are the most widely studied bacteria in copper biomachining 8,16 , although reports involving bacteria of the genus Staphylococcus sp. and a strain of Aspergillus niger fungi have been published recently 34,35 . However, to the best of our knowledge, little information regarding the ability of moderately thermophilic bacteria in this field is available. In this section, the performance of mesophilic (A. ferrooxidans) and thermophilic bacteria (S. thermosulfidooxidans) in MAC medium during copper biomachining process is compared (Figure 4). A progressive decrease in the amount of copper removed per unit time was observed for both bacteria after the maximum value had been reached during the first hour, with a maximum MRR of 109 and 117 mg h -1 being reached for A. ferrooxidans and S. thermosulfidooxidans, respectively. These values are in accordance with those obtained by Jadhav et al. 36 , who used A. ferrooxidans 13823 culture supernatant in 9 K medium and also observed a sustained decrease after the first hour (25-30 % and 40-45 % decrease after 2 and 3 h, respectively).  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58   . Evolution of redox potential (columns) and metal removal rate (circles) for A. ferrooxidans (white) and S. thermosulfidooxidans (grey) in MAC medium with an initial pH of 1.5 and Fe 2+ concentration of 6 g L -1 (with yeast extract supplementation for the thermophilic bacteria). Error bars represent one standard deviation of the duplicates.
As stated above, the decrease in metal leaching rate with Fe 3+ concentration has prevented biomachining from being implemented industrially. To address this problem, previous studies published by us 4 proposed the use of a multi-module carousel in which each module contains a ferric ion solution and a microbial consortium that is able to oxidize Fe 2+ ions. The workpiece remains submerged in each solution as long as the MRR is a maximum. During that period, acidophilic bacteria (i.e., A. ferroxidans) regenerate the ferric ion concentration in the rest of the modules. The novelty of this study lies in the achievement of a stable, controlled and continuous biomaching process by means of a unique dissolution tank in which bacterial culture and biomachining take place at the same time ("Biomachining during culture period" section). In this case, continuous bacterial growth at its maximum degrading activity would be accomplished since the higher kinetics of the chemical reaction would ensure the availability of the substrate (Fe 2+ ions) (Figure 1c). The key is therefore to overcome the difference between chemical and microbiological reaction rates to achieve a good balance and the desired MRR stability.

Biomachining during culture
A copper bioleaching process in which bacterial growth and workpiece biomachining take place at the same time has been envisaged in this section. Jarosite accumulation was visually observed in those samples with no pH control after 38 h (pH final 2.3) and 86 h (pH final 2.5) for A. ferrooxidans and S. thermosulfidooxidans, respectively, thereby reducing iron availability in the solution (precipitation above 60 % of total iron) and concluding the process prematurely.

12
In those experiments with pH control, S. thermosulfidooxidans showed a seven-times lower MRR in comparison with A. ferrooxidans after operation for 1 day (1.72 vs 12.6 mg h -1 ). In addition, growth of S.
thermosulfidooxidans was limited in comparison with its performance during standard culture (SI Figure  S5), which might be related to its low toleration to copper cations, thereby significantly reducing previous inhibition limits reported in the literature for different strains of the same species 14 . Similarly, Watking et al. 37 previously observed that metal tolerance (Fe, Cu, Zn, Ni and Co) varied significantly between strains of the same specie. Regarding A. ferrooxidans cell density, bacteria seemed to grow faster when biomachining was performed together with the growth step (SI Figure S6). In addition, the bacteria developed in the latter experiments reached a smaller size, which could be related to the increase in ferrous-oxidizing activity 38 .
The biomachining test using A. ferrooxidans showed an increase in MRR (~43 mg h -1 ) for the first 50 hours ( Figure 5), subsequently remaining constant over the following 30 h. Although a stable copper removal rate over time was achieved, the maximum MRR reached was less than half the value attained during conventional biomachining (109 mg h -1 ) ( Figure 4). Moreover, the copper removal rate remained constant even though the amount of Fe 3+ available also increased throughout the experiment. Jarosite formation was discarded as a cause of this stagnation since the total iron content in the medium remained approximately constant throughout the test (~6 g L -1 ). One possible explanation might be related to biofouling as a result of the adhesion of microorganisms and subsequent biofilm formation on the workpiece surface 39 . The resulting passivating layer would prevent ferric ion migration, thus inhibiting the biomachining process 40 .

SUMMARY
Despite the potential benefits of biomachining (e.g., environmentally friendly nature) as an alternative to conventional chemical and physical manufacturing processes, its industrial application has not yet been completed. However, more in-depth studies of the chemical and biological aspects of this process, as well as the influence of the most significant process parameters, could promote its industrial implementation. In this case, the use of extremophile mesophilic (A. ferrooxidans) and thermophilic (S. thermosulfidooxidans) bacteria to remove copper from OFC workpieces has been investigated. Unlike the obligate chemolithoautotroph A. ferrooxidans, S. thermosulfidooxidans only exhibited measurable growth rates in mixotrophic conditions, therefore its nutrient medium had to be supplemented with yeast extract. Despite this, A. ferrooxidans showed a higher specific bacterial growth rate and specific substrate consumption under the optimal conditions (initial pH and Fe 2+ concentration of 1.5 and 6 g L -1 and 30 °C (for A. ferrooxidans) or 45 °C (for S. thermosulfidooxidans)) in MAC medium. Jarosite formation and precipitation was observed when the pH exceeded the value of 2.2 for samples with a total iron content ≥6 g L -1 .
As expected, when the A. ferrooxidans growth (cultivation period) and machining stages were performed simultaneously, a stable copper removal rate in the range 40-45 mg h -1 was obtained. This method avoided Fe 3+ depletion in the medium, which is the key factor governing the machining rate. In contrast, the maximum MRR achieved was less than half of the value attained during conventional biomachining with pH adjustment (109 mg h -1 ). A linear Fe 3+ concentration increase in the medium (51 mg h -1 , R 2 = 0.951)) from the hour 20 onwards suggested a decrease in cation diffusion, probably caused by the attachment of microorganisms to the metallic surface, thus acting as a weak diffusion barrier.