Journal Pre-proof

Biomachining properties of various metals by microorganisms

Fei Ma, Hui Huang, Changcai Cui

PII: S0924-0136(19)30485-6
DOI: https://doi.org/10.1016/j.jmatprotec.2019.116512
Reference: PROTEC 116512

To appear in: Journal of Materials Processing Tech.

Received Date: 7 March 2019

Revised Date: 15 November 2019
Accepted Date: 18 November 2019

Please cite this article as: Ma F, Huang H, Cui C, Biomachining properties of various metals
by microorganisms, Journal of Materials Processing Tech. (2019),
doi: https://doi.org/10.1016/j.jmatprotec.2019.116512

This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.

© 2019 Published by Elsevier.

 Biomachining properties of various metals by microorganisms
Fei Ma, Hui Huang*, Changcai Cui

1 Institute of Manufacturing Engineering, Huaqiao University, Xiamen, 361021, Fujian Province, P. R. China

* Corresponding Author / E-mail: [email protected], TEL: +86-0592-6162615 (Hui. Huang)

of
ro
-p
re
Abstract: The biomachining properties of five pure metals (Cu, Co, Fe, Sn, and W) and four sintered components
made from binary powder mixtures (Cu-Co, Cu-Fe, Cu-Sn, and Cu-W) were synthetically studied under the same
lP

conditions. The material removal rate (MRR) in the machining process was recorded. The surface roughness (Sa),
topography, and element distribution of the surface were compared before and after machining. The cross-sections
of machined surfaces from nine samples were also observed. The electrochemical corrosion resistance of the five
pure metals was evaluated by potentiodynamic polarization curves. The experimental results showed that the
na

MRR of all samples presented good linear behavior in the machining process with different slopes. Sa of all
samples increased from below 0.1 µm to 0.35.4 µm after machining. Material removal mechanisms involved in
the biomachining process were also explored. The MRR of pure metals was mainly affected by the
electrochemical corrosion resistance of the material. However, the generation of a reaction layer could reduce the
ur

MRR for the pure Co. The MRR of sintered components was affected by the reaction processes between the
binary powder mixtures in the sintering process. Three material removal mechanisms were proposed for the
sintered components corresponding to the different reaction processes.
Jo

Keywords: Biomachining, Acidithiobacillus ferrooxidans, Sintered components, Tafel extrapolation, Passivation

1 Introduction

Biomachining utilizes the biological energy of chemolithotrophic bacteria for metal removal as demonstrated
by Valdés et al. (2008) with a renewable removal effect similar to that of chemical machining (Díaz-Tena et al.,
2016). This process is an environmentally-friendly machining method with no damage layer left on the machined
surface. Bacteria, including Acidithiobacillus ferrooxidans (A. ferrooxidans) can act as a catalyst for continually
oxidizing Fe2+ to Fe3+ and obtaining energy; Fe3+ is constantly oxidizing and converting to Fe2+ resulting in a

1
circulatory system. Uno et al. (1996) first reported the possibility of biomachining by the bacteria.
Surface topography and surface roughness are important subjects in manufacturing. Maruda et al. (2018)
studied the influence of the factors on the surface topography in the turning of double-phase steel by measuring
the mean arithmetic surface height Sa, the height of the highest surface peak Sp, the depth of the lowest surface
valley Sv, and the surface topography map after machining. Wieczorowski et al. (2010) investigated the possibility
of predicting brake behavior by analyzing the surface topography and roughness using a white light interferometer
(NT1100, Veeco). Leach (2004) reviewed some issues in surface topography measurement.
Studies on biomachining have also been reported. Kumada et al. (2001) reported the biomachining rate of
mild steel (SS400) and copper varied with machining conditions. However, the SS400 always had a rougher
surface than that of the copper after machining because pitting corrosion and stepwise crystallographic dissolution
were found on the surface by scanning electron microscopy (SEM) inspection. Johnson et al. (2007) showed a
slow MRR of copper, but the surface roughness (Ra) increased to 1.72.4 µm after 24 h and 48 h of biomachining.
In another study, Ra was also increased after biomachining regardless of the fine or coarse initial surface
(Istiyanto et al., 2010). Hocheng et al. (2012) reported the feasibility of biomachining copper, nickel, aluminum,
zinc, and tin. The MRRs were quantified with great diversity. They also observed the copper surface texture after

of
machining by SEM and WAYCO (Veeco) machine—Ra increased to 2.29 µm (0.43 µm in control using medium
alone). Xenofontos et al. (2015) investigated the influence of process parameters on the copper removal rate and

ro
evaluated the copper removal mechanisms by analyzing the SEM images of copper surfaces and the RMS
roughness (Rq). Muhammad et al. (2018) optimized the MRR and surface roughness (Ra) of copper in the
biomachining process via an empirical model.
-p
Previous references show that most studies in biomachining focus only on pure metals. Compared with pure
metals, sintered components have higher strength (and stiffness)-to-density ratios, better wear (and fatigue)
resistances, and lower coefficients of thermal expansion. Sintered components are widely used industrially
re
including in bonds of abrasive tools, cutting tools, and brake linings. Research on the biomachining properties of
sintered components is important for expanding biomachining applications.
This paper reports on the biomachining characteristics and material removal mechanisms of five pure metals
lP

(Cu, Co, Fe, Sn, and W) and sintered components made from copper powder and other powder mixtures. The
steady-state MRRs of all samples were measured by the weight loss test (long-term test). The potentiodynamic
polarization curves (short-term test) of pure metals were measured to evaluate their electrochemical corrosion
resistance. The surface roughness of all samples was investigated. The sample surface topography and the
na

elemental composition of certain locations were examined together with an integrated SEM and energy dispersive
spectroscopy (EDS). The associated material removal mechanisms of the pure metals and sintered components
was discussed.
ur

2 Materials and methods

2.1 Culturing of Acidithiobacillus ferrooxidans

Jo

The microorganisms used in this study were Acidithiobacillus ferrooxidans (hereafter A. ferrooxidans) that
was cultured in pure form. An SEM photo of the bacteria is shown in Fig. 1. This culture was originally
withdrawn and isolated from acidic pit water containing A. ferrooxidans from an iron mine in China via 9K
medium.

2
 Fig. 1. SEM image of A. ferrooxidans.
The 9K medium used here is an aqueous dissolution of six kinds of inorganic salts including ferrous sulfate

of
as the energy source. The composition of the medium is shown in Table 1. The pH was adjusted to 1.8 via 1:1
(volumetric ratio) sulfuric acid and distilled water.

ro
Table 1 9K medium compositions.
Composition Qty. (g/L)
(NH4)2SO4 3
MgSO4·7H2O
KCl
K2HPO4
-p 0.5
0.1
0.5
re
Ca(NO3)2·4H2O 0.01
FeSO4·7H2O 24.83
A. ferrooxidans was cultivated in the liquid 9K medium. The process of cultivation is as follows:
lP

● First, the 80 mL sterilized 9K medium was prepared in a 250 mL Erlenmeyer flask.

● Second, 20 mL culture solution with the initial generation of A. ferrooxidans was transferred to the
prepared medium.
● Third, the flask with 100 mL inoculated medium was incubated in a thermostatic incubator shaker
na

(BSD-WX2280, Boxun) at 30ºC and shaken at 160 rpm until the ratio of Fe3+ concentration in solution
approached 95%. This produced the first-generation A. ferrooxidans.
● Fourth, 20 mL culture solution with first-generation A. ferrooxidans was taken out from the 100 mL
ur

inoculated medium and cultivated in 80 mL sterilized 9K medium similar to above.

● Finally, the same culture process was repeated four times to improve the bacterial activity and obtain a
stable bacterial density and Fe3+ concentration.
Jo

The details of measuring pH, Fe3+ concentration, and bacterial density have been described by Ma et al.
(2019) and Huang and Ma (2019). The variation of Fe3+ concentration and bacteria density for the fourth culture
process is shown in Fig. 2. The initial ratio of Fe3+ was approximately 19%: the iron was added by inoculation
prior to the conversion, and it took 21 h to complete the conversion of Fe2+ to Fe3+ for fourth-generation A.
ferrooxidans.

3
 of
Fig. 2. Fe3+ to total iron ratio and bacterial density with culture time.

2.2 Biomachining experimental condition

ro
The workpiece materials tested in this experiment include five pure metals (Cu, Co, Fe, Sn, and W) and four
sintered components made from binary powder mixtures (Cu-Co, Cu-Fe, Cu-Sn, and Cu-W). In this experiment,
-p
four categories of sintered components were sintered via the corresponding pure metal powders with a purity of
99.9%. The average particle diameter of the powders was less than 50 μm. The powders were pre-mixed in a
mixer for 4 h. The same weight percent of each pure metal powder in Cu-Co, Cu-Fe, and Cu-W components was
re
used to evaluate the impact of elements. Sn powder was generally used in fairly small quantities; therefore, the
weight percent of Sn and Cu in Cu-Sn component was selected to be 20 and 80, respectively.
The mixing ratios and the maximum sintering temperature of four sintered components are designed
lP

separately as shown in Table 2. The sintering temperature was gradually increased to the maximum values with a
stair type of sintering process. The maximum temperature 920ºC could create chemical bonds of Cu-Co and
Cu-Fe and only the physical bond of Cu-W component. Some intermetallic compound could be produced for
Cu-Sn components with a maximum temperature of 450ºC. The sintering pressure increased successively to 33
na

MPa (holding the maximum temperature and pressure for one minute).
Table 2 Sintering parameters.
Mixing ratio Maximum Temp.
Sintered component
ur

(weight percent) (ºC)

Cu-Co 1:1 920
Cu-Fe 1:1 920
Jo

Cu-W 1:1 920

Cu-Sn 4:1 450
The metal sheets (15 × 15 × 1 mm) used here were inlaid in polystyrene resin blocks (ϕ25 × 7 mm), which
ensured a constant machining surface area as shown in Fig. 3(a). The surfaces were polished with successively
decreasing grit to obtain a fine surface with Sa under 100 nm. The workpieces were successively rinsed with
distilled water and ethanol (analytical reagent) in an ultrasonic cleaner and then dried in the oven at 60ºC.

4
Fig. 3. (a) Dimensions of the exposed faces to be machined and (b) placement of the workpieces in biomachining.

of
ro
-p
re
lP

Fig. 4. Flow-diagram of the biomachining process.

The biomachining process is shown in Fig. 4. Each workpiece was independently placed on the bottom of the
250 mL flask with 100 mL A. ferrooxidans culture solution. The machined surface was facing upward (Fig. 3(b))
na

during the biomachining processing. During the machining time, the flasks were placed in the incubator shaker at
30ºC and shaken at 160 rpm. A new bacterial solution with fourth-generation A. ferrooxidans was cultured for 12
h under the same condition as above. This was used according to preliminary results. At this time, the Fe3+
ur

concentration increased to approximately 3 g/L and the cell density increased to approximately 3.32 × 107 cells/ml
according to Fig. 2.
The workpieces were taken out from the solution and then rinsed and dried as described above after each
Jo

hour; the weight loss was measured with an electronic balance (BS214-D, Sartorius). The process was repeated
eight times, and each process took approximately 30 min. After measurement, the workpieces were then immersed
in the solution again. The polystyrene resin was corrosion resistant and not hygroscopic; thus, the weight loss
values indicated material removal from the workpiece. The weight loss per unit area was calculated for nine
workpieces. The MRR was calculated from the equation below according to the literature (Díaz-Tena et al. 2017):

amount of metal removal (mg)

MRR  (1)
time (h)  area (cm2 )

After eight hours of machining, the machined surface and cross-section topography were observed with SEM
(ProX, Phenom). The elemental distributions on the machined surfaces of sintered components were measured by
the EDS (ProX, Phenom). The machined surface roughness (Sa) was examined by a white-light interferometer

5
(NV7300, Zygo). This is the arithmetic mean of the absolute of the ordinate values within a definition area
according to the ISO 25178 standard. Here, the determination scope of the white-light interferometer is 0.70 ×
0.53 mm. The Sa values of the workpieces were averaged from nine measurements on the fixed position for each
sample.

2.3 Electrochemical polarization curves for pure metals in the culture solution

The polarization experiments of five pure metals were carried out with a standard three-electrode cell
assembly. A platinum sheet (25 × 25 × 0.1 mm) served as an auxiliary electrode. The silver/silver chloride
electrode (SSCE) was the reference electrode. 1 cm2 exposed area of polished pure metal was the working
electrode. The working electrode was inlaid in polystyrene resin blocks (ϕ25 × 7 mm) for the insulation of the
solution as shown in Fig. 5. The electrodes were immersed in 100 mL A. ferrooxidans culture solution cultured for
12 h same as that in Fig. 4 and connected to an electrochemical analyzer (660E, CHI) coupled to a computer with
supporting software. The polarization curves were obtained from the supporting software. Tafel extrapolation is a
frequently used law in electrochemistry, by which the important polarization parameters of corrosion resistance

of
and corrosion rate can be evaluated from the polarization curve, further characterizing the electrochemical kinetics
of the corroding metal. Therefore, the Tafel parameters for the five pure metals were calculated by the software
from the polarization curves.

ro
-p
re
lP

Fig. 5. Dimensions of the working electrodes.

na

3 Results
ur

3.1 MRR experiment results

The amount of removal per unit area of the nine workpieces is shown in Fig. 6. Under the experimental
Jo

conditions used here, all workpieces can be machined by A. ferrooxidans culture solution except pure W.
Meanwhile, all the weight losses of the workpieces present a linear behavior over time. This linear behavior is
consistent with the results reported by Zhang and Li (1998) and Díaz-Tena et al. (2014). The scattered points of
the nine workpieces are nicely fitted to linear curves by the least-squares method. The MRR of each workpiece
was implemented directly by the corresponding slope of the curve as described in the equation. The MRR of each
sample is clearly different. The MRR of Co is similar to the case of Fe. The Cu-W sample achieved the highest
MRR of 15.36 mg/h·cm2 for all the samples, although the MRR of pure W is almost zero. Pure Sn had the lowest
MRR of 2.35 mg/h·cm2. The MRR values are slightly higher than the results reported by Hocheng et al. (2012),
showing that pure Cu and pure Sn were 5.5 mg/h·cm2 and 1.04 mg/h·cm2, respectively. It is interesting that MRRs
were found in the initial phase (t=0). This is probable because of the different initial surface state of workpiece
and the experiment errors for the first hour. The other confusing result is the negative MRR for pure W—this may

6
be due to the continuous passivation and ion deposition on the W sample during the experiment. This should be
studied in more detail in the future.

Fig. 6. Extent of removal for the nine workpieces due to biomachining.

of
3.2 Sa change of machined surfaces

ro
The Sa values of the nine workpieces were all under 0.1 µm before machining (Fig. 7a). However, the values
have obviously increased after eight hours of machining except for pure W as shown in Fig. 7b. The Sa value of
Sn machined surface increased to 4.2 µm. This is the highest for the pure metals. The Sa value of other pure
-p
metals also increased to more than 1 µm. The Sa value of Cu-W component machined surface increased to 5.4 µm:
This is the highest in the sintered components. However, Sa of Cu-Co component and Cu-Fe component increased
to 0.3 µm and 0.6 µm, respectively.
re
lP
na
ur
Jo

7
 of
ro
-p
re
lP

Fig. 7. Surface roughness Sa of the various workpieces (a) before machining and (b) after machining.

3.3 Topography and EDS analysis of machined surfaces

na

The SEM photos of the initial surfaces of workpieces are shown in Fig. 8. The surfaces were comparatively
smooth. Scratches and indentations from polishing are visible on the surfaces. The pure phases are shown on the
five pure metals (see Fig. 8ae). The contrast of the two different phases on sintered components is evident (see
ur

Fig. 8fi).
Jo

8
 of
ro
-p
re
lP

Fig. 8. SEM micrographs of the polished surfaces before machining: (a) pure Cu, (b) pure Co, (c) pure Fe, (d)
pure Sn, (e) pure W, (f) Cu-Co component, (g) Cu-Fe component, (h) Cu-Sn component, and (i) Cu-W
na

component.

The surfaces of the SEM photos of the nine workpieces after eight hours machining are shown in Fig. 9. The
ur

surfaces became rough after machining except for pure W. The surface topographies of workpieces are noticeably
different according to the SEM photos. Interestingly, the internal texture of pure Cu, Fe, and Sn appeared on the
surfaces after machining (Figs. 9a, c, and d). The individual copper grains with different crystallographic
Jo

orientations were observed on the machined surface—especially for pure Cu. A similar result was seen for pure
copper by Xenofontos et al. (2015) after biomachining. For the pure Co, a layer of fine-sheet-like structure
appeared on the machined surface (Fig. 9b). The micro-topography of pure W does not show a discrete change.
The topography of the sintered component surfaces appears more diverse after machining. Fig. 9f and g show the
microstructure of Cu-Co component and Cu-Fe component changing from component to porous. Fig. 9h presents
the Cu-Sn component surface topography where some big pits formed: The original surface clearly retains
scratches from polishing. Fig. 9i clearly presents two types of particles that fill loosely on the Cu-W component
surface; one type of particle shares the same grain texture with pure copper.

9
 of
ro
-p
re
lP

Fig. 9. SEM micrographs of surfaces after eight-hour machining and EDS analysis on sintered component
surfaces: (a) pure Cu, (b) pure Co, (c) pure Fe, (d) pure Sn, (e) pure W, (f) Cu-Co component, (g) Cu-Fe
na

component, (h) Cu-Sn component, and (i) Cu-W component.

Next, the four sintered component surfaces were studied with EDS before and after eight hours of machining.
ur

The area of analysis for each sample after machining is shown in the red areas in Fig. 9fi. The elemental
distribution of each analysis is shown in Table 3. It was found that the elemental distribution is clearly different
before and after machining. For the Cu-Co and Cu-Fe component, the amount of Cu increases for the machined
Jo

surface. However, the content of Co and Fe decreased sharply. The amount of Sn element was dramatically
improved for the machined surface versus the initial surface. The elemental distribution of the Cu-W component
indicated that the particle filling on the machined surface was Cu and W element.

Table 3 Element distribution proportion changes of the four sintered components.

Before machining After machining
Sintered component
Element and proportion Range Element and proportion
Cu-Co Cu: 50.0% Co: 50.0% A Cu: 93.3% Co: 6.7%
Cu-Fe Cu: 43.8% Fe: 56.2% B Cu: 95.1% Fe: 4.9%
Cu-Sn Cu: 81.0% Sn: 19.0% C Cu: 48.8% Sn: 51.2%

10
 D Cu: 51.8% Sn: 48.2%
Cu-W Cu: 45.5% W: 54.5% E Cu: 100% W: 0%
F Cu: 21.3% W: 78.7%

3.4 Cross-section observations of samples

The cross-section morphologies of the nine samples after eight hours of machining are shown in Fig. 10. The
metallographic structures of pure Cu, Fe, Sn, and W were homogeneous along the entire cross section with a
clearly machined surface profile (Fig. 10a, c, d, and e). However, a new reaction layer was generated with a
relatively uniform depth of 68.6 µm on the pure Co machined surface; here, some strips of single phase extended
into the sample from the deep layer (see Fig. 10b). A porous single-phase layer was formed beneath the Cu-Co
component and Cu-Fe component machined surfaces with depths of 177 µm and 310 µm, respectively (see Fig.
10f and g). Furthermore, the demarcation line between the porous single-phase layer and component was much
clearer and maintained an approximately straight line. Some large pits with different sizes were irregularly formed
beneath the machined surface of the Cu-Sn component sample (see Fig. 10h). The metallographic structure of the

of
two different phases on the Cu-W component cross section was homogeneous along the whole cross section (see
Fig. 10i).

ro
-p
re
lP
na
ur
Jo

Fig. 10 SEM micrographs of section morphologies of samples after eight hours of machining: (a) pure Cu, (b)
pure Co, (c) pure Fe, (d) pure Sn, (e) pure W, (f) Cu-Co component, (g) Cu-Fe component, (h) Cu-Sn component,
and (i) Cu-W component.

11
3.5 Potentiodynamic polarization curves

The potentiodynamic polarization curves for the five pure metals (Cu, Co, Fe, Sn, and W) in bacterial
solution are shown in Fig. 11. The curve of pure W shows a striking difference. The current density decreases
rapidly from the potential value of approximately 0.6 V and remains relatively constant, along with the potential
value increasing from 0.7 V to the end of the test. By graphing the intersection point of the two straight lines of
the two branches, the corrosion potential (Ecorr) and the current density (Icorr) for the five samples were obtained.
Ecorr and Icorr for the five samples are presented in Table 4. Ecorr of pure W is the highest, and Icor of W is the lowest.
This suggests that that the pure W in bacterial solution has clear signs of passivation on the anodic branch. Icor of
Co is the highest for the five pure metals.

of
ro
-p
re
Fig. 11. Potentiodynamic polarization curves of five pure metals in bacterial solution after 12 h of culture.

Table 4 The corrosion potential (Ecorr) and current density (Icorr) of five pure metals.
lP

Metal Ecorr (V) Icorr (mA/cm2)

Cu 0.055 0.9089
Co -0.347 1.180
na

Fe -0.507 0.7538
Sn -0.429 0.5457
W 0.429 0.2161
ur

4 Discussion

4.1 Machining properties of the five pure metals

Jo

The order of MRR values of the five pure metals was Cu > Co ≈ Fe > Sn > W (Fig. 6). The SEM micrograph
of the machined surface (Fig. 9ad) and the cross-section morphology (Fig. 10ad) of the four pure metals show
that the internal texture of elemental Cu, Fe, and Sn appeared on the surfaces, and a thick reaction layer was seen
on Co. At the same time, Sa of Cu, Co, Fe, and Sn increased significantly due to the distinct changes in the surface
topography after eight hours of machining. The MRR value of W is almost zero. The topography and Sa value of
W also had minor changes indicating that W was not machined by the bacterial solution.
During biomachining, the metal was actually removed by corrosion in the bacterial solution, which is in
essence a chemical reaction. The weight loss test for measuring the MRR values is a commonly used method for
evaluating the corrosion rate for metal samples. It is more suitable for the steady-state test and was classified into
long-term tests according to Shi et al. (2010). The potentiodynamic polarization curves reported the corrosion rate

12
generalized into a short-term test (Shi et al. 2010). Some detailed information on passivation and the corrosion
rate can be obtained.
The corrosion rate is proportional to Icorr, as stated by Shi et al. (2010) and Tavakkolizadeh and
Saadatmanesh (2001). In the case of pure Cu, Co, Fe, and Sn, the passivation did not appear on the polarization
curves, and thus the corrosion was sustainable. According to the Icorr value of each sample, the corrosion rates of
the four metals can be ranked as Co > Cu > Fe > Sn, which is similar to the order of the MRR value. However, the
Co is distinct. Co had a thick reaction layer deposited on the sample surface, but there was no sign of passivation
on the anodic branch (Fig. 11). Therefore, the reaction layer should be loose and conductive in the solution.
According to Badawy et al. (2000), Co is always covered by a native passive film that consists mainly of CoO or
Co(OH)2 during the corrosion at different pH values of aqueous solutions. The structure of the nanosheets in the
Co(OH)2 films observed by Nguyen et al. (2017) was very similar to that of machined surface of Co in this work.
Therefore, the reaction layer on the pure Co would probably be composed primarily of Co(OH)2. The thick
reaction layer, however, would hinder the diffusion of ions during the long-term test in a bacterial solution. This
finding may explain the distinction of Co MRR. Similar research on Co film was also carried out by Reikowski et
al. (2016). Regarding the Cu, Fe, and Sn, they also produce films in aqueous solutions according to Pieraggi et al.

of
(2005), but the thickness is far lower than that of Co in the long-term test. Thus, the influences on the Cu, Fe, and
Sn are relatively weak.

ro
The W had a lower Icorr value than the other metals in the bacterial solution (Table 4). The potentiodynamic
polarization curve of the W also showed a strong sign of continuous passivation along with the potential increase
on the anodic segment (Fig. 11). Maurice and Marcus (2018) stated that the redox reaction of metal in a solution
-p
will be seriously affected by the passive film, which is an oxide of the metal substrate and self-forming at the
metal surface in an aqueous solution. Continuous passivation probably indicates a condition in which the pure W
was permanently protected from further corrosion by a component film of corrosion product on its surface.
re
Similar papers have reported that the W film was produced during the chemical removal process as well: Coetsier
et al. (2011) investigated the W film dissolution rate for chemical mechanical polishing; Lim et al. (2013) found a
WO3 layer formed on W in chemical mechanical planarization by Fe(NO3)3 and H2O2; and Seo et al. (2006) stated
lP

that WO3 prevents further oxidation by a slurry in chemical mechanical polishing.

4.2 Machining processes and properties of four sintered components

The MRR values of Cu-Co component and Cu-Fe component were 5.71 mg/h·cm2 and 6.34 mg/h·cm2,
na

respectively. In contrast, the MRR value of Cu-Sn was relatively low (3.89 mg/h·cm2), and the MRR value of
Cu-W was relatively high (15.36 mg/h·cm2). The machined surfaces of the Cu-Co component and the Cu-Fe
component presented a similar surface topography of a smooth porous surface (Fig. 9f and g). Sa maintained
ur

relatively low values. The topographies of the Cu-Sn component and the Cu-W component were quite different
because there were big pits and loose particles on the machined surfaces, respectively (Fig. 9h and i). Sa
maintained a relatively high value. A similar case could be found in the cross-section morphologies (Fig. 10fi).
Jo

The EDS results on the Cu-Co component and Cu-Fe component surfaces (Fig. 9f and g and Table 3) show
that the porous layers mainly consist of Cu element. The Co powder and Fe powder were removed faster than the
Cu powder in the bacterial solution. Combined with the result of the potentiodynamic polarization curves, the data
show that the Co and Fe powders with lower corrosion potential than Cu powder will be removed first due to the
electrochemical effect. Pores were then obtained on the machined surface in the biomachining process.
The MRRs of the Cu-Co component, Cu-Fe component, and the corresponding pure metals are ranked as Cu >
Co > Cu-Co, Cu > Cu-Fe > Fe (Fig. 6). The reaction layer on the Co powder also influences the MRR rankings for
the Cu-Co component. When the outermost Co powders were removed, the corrosion rate of the inner Co powders
will be severely decreased as a result of the hindering of the diffusion of ions through the pores by the reaction
layer. However, for the Cu-Fe component, the diffusion resistance of ions through the pores was much smaller
than that for the Cu-Co component. Therefore, the MRR of the Cu-Fe component (6.34 mg/h·cm2) is slightly

13
higher than the Cu-Co component (5.71 mg/h·cm2). The porous layer depth of the Cu-Fe component (310 µm) is
much deeper than the Cu-Co component (177 µm) after eight hours of machining (Fig. 10f and g).
For the Cu-Sn component, the Sn elemental proportion on the residual surface reached up to 51.2% after
eight hours of machining. This is significantly higher than the original ratio of 19% (Fig. 9h and Table 3).
Combined with the Cu and Sn phase graph shown by Fürtauer et al. (2013), a part of the Cu powder and Sn
powder should be generated to intermetallic compounds with a high Sn proportion under the sintering temperature
(450ºC). The intermetallic compound with excellent corrosion resistance capability can withstand the corrosion of
a bacterial solution according to Walsh and Low (2016). Sürme et al. (2009) found that the corrosion resistance of
Cu-Sn alloy in the acid medium increased with proportion of Sn; in their study, the alloys with Sn contents of 54.5%
and 63.5% present the best corrosion resistance. That is the main reason why the original polishing scratches were
still retained after eight hours of machining (Fig. 9h). Therefore, the MRR of the Cu-Sn component kept a
relatively low value.
The Cu-W component made it very difficult for the Cu powder and W powder to form a chemical connection
at a temperature of 920ºC. With corrosion of the Cu powder in the solution, the physical connection of the surface
powders would become looser, and the powders would be easily desquamated. Finally, a surface with loose Cu

of
and W particles was formed after eight hours of machining (Fig. 9i). High MRR of the Cu-W component was
achieved due to the desquamation of the metal powders during processing. The desquamation of the copper

ro
powders during the processing may result in a rapid depletion of Fe3+. Due to the limited conversion rate of Fe3+
by microorganisms, the Fe3+ concentration could be decreased resulting in a slightly lower material removal rate
near the end of machining. Thus, the linear curve of the Cu-W component in the first hours becomes more like an
-p
asymptote or logarithmic curve during the final hours of machining (Fig. 6).

4.3 Biomachining mechanism

re
It is a general consensus that biomachining includes the two main processes as shown in the literature
reviewed by Díaz-Tena et al. (2017): The conversion of Fe2+ to Fe3+ by bacteria and the removal of metal by Fe3+,
in which the Fe3+ is reduced to Fe2+ by metal simultaneously. Therefore, the two processes can form an endless
lP

chain. The detailed process of converting Fe2+ to Fe3+ by bacteria is described as follows. First, A. ferrooxidans
chose Fe2+ that comes in the culture medium, and then Fe2+ causes iron to lose one electron through an internal
catalytic process to produce Fe3+ as described by Díaz-Tena et al. (2014). During this process, the bacteria can
obtain energy from the catalytic oxidation reaction for metabolic activity and reproduction. The detailed process
na

of removing metals by Fe3+ is discussed below. The Fe3+ has already been widely used as an etchant in chemical
machining due to the capacity of removing many kinds of metals. When these metals are immersed in a solution
with Fe3+, they are first oxidized by Fe3+ to a high valence and then dissolved into the solution. At the same time,
ur

the Fe3+ will be reduced to Fe2+ by metal material. In short, the Fe3+, as a product of bacteria, is the executor for
metal removal; and the Fe2+, as the emission of metal removal, is the source of living bacteria. Therefore, a
circulatory system exists in biomachining. As the redox mediator in biomachining, the Fe2+ and Fe3+ are always
Jo

converted throughout the process.

In this study, there are two types of workpiece: pure metals and sintered components. The biomachining
mechanism for the pure metals has already been demonstrated by many scholars: Cu by Díaz-Tena et al. (2014)
and Istiyanto et al. (2012, 2010), and Cu and Fe by Zhang and Li (1998). See Fig. 12. Pure W cannot be machined
by microorganisms as mentioned above. However, for the pure Co, a loose and very thick reaction layer is
generated on the machined surface resulting in a serious hindering of diffusion of ions. The illustration of the
machining process for pure Co is shown in Fig. 13.

14
 Fig. 12 The biomachining mechanism for pure metals.

of
ro
-p
Fig. 13 The biomachining process for pure Co.
re
The machining mechanisms of the sintered components are attributed to metal removal by Fe3+. The typical
sintered component is made of two different powders bonded together by chemical connections such as the Cu-Co
component and the Cu-Fe component. The powder material with a lower corrosion potential (higher activity) was
lP

removed first acting as a sacrificial anode. After a certain time of machining, the outermost powder with a higher
removal speed will be removed first. The inner powder is then removed, and thus the surface pores of component
have been extended and expanded. As this occurs, numerous micro-pores will be formed among the top layer of
powders. A schematic of the biomachining mechanism for the typical sintered components is shown in Fig. 14
na

(where the corrosion potential of M1 is lower than M2).

ur
Jo

Fig. 14 The biomachining mechanism for typical sintered components.

In some cases, two different powders result in an intermetallic compound with completely different chemical
properties. Some of the intermetallic compounds can resist the corrosion of a bacterial solution such as tin bronze.
The Cu-Sn component in this study was of this type. After a certain machining time, the metal powder was easily
removed, but the intermetallic compound with the original surface remained. A schematic of the biomachining
mechanism for the sintered component with an intermetallic compound is shown in Fig. 15.

15
 Fig. 15 Biomachining mechanism for sintered components with an intermetallic compound.
There are other conditions that are difficult for the two different powders to form a chemical connection, e. g.,
the Cu-W component. Surface powders became looser and continued desquamating after the physical connections
were destroyed due to corrosion. A schematic of the biomachining mechanism for sintered components under a
physical connection is shown in Fig. 16 (here M1 is encrusted in M2).

of
ro
-p
re
Fig. 16 The biomachining mechanism for sintered components under a physical connection.
lP

5 Conclusions

1. The weight loss of the nine workpieces presents linear behavior along the test time. The MRR of the five pure
metals was determined by their corrosion kinetic parameters in the culture solution. For the pure Co, the MRR
na

was reduced by a thick reaction layer in the weight loss test.

2. The MRR of the sintered components can be divided into three categories: (a) close to both of the two
component metals for the chemically bonded components, (b) lower than the average of the two component
ur

metals for the component with an intermetallic compound, or (c) higher than both of the two component
metals for the physically bonded component.
3. The surface qualities of the nine metals turned worse after machining. Their Sa increased from under 0.1 µm
Jo

to 0.35.4 µm. Sa of the sintered components after machining was obviously affected by the material removal
mode.
4. The machining mechanisms for pure metals and sintered component were supplemented using a model
diagram. The powder bonding state of sintered components has a significant influence on the machining
mechanisms.

Author Contributions Section

Mr. Fei Ma contributed to implementation of the experiments, analysis of the data and preparing the
manuscript.

16
Prof. Hui Huang guided the experiments and discussed the study as well as modified the drafts.

Prof. Changcai Cui helped to analyze measuring data.

Funding

The authors would like to acknowledge the support of the National Natural Science Foundation of China
(NSFC) (Grant Nos. U1805251 and 51375179) and Fujian Science and Technology Project（2017H6014）and from
Changjiang Scholars and Innovative Research Team in University (IRT_17R41).

Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal relationships that
could have appeared to influence the work reported in this paper.

of
☐The authors declare the following financial interests/personal relationships which may be considered as

ro
potential competing interests:

Acknowledgment
-p
The authors would like to acknowledge the East China University of Technology for supplying the culture of
A. ferrooxidans.
re
References
lP

Badawy, W.A., Al-Kharafi, F.M., Al-Ajmi, J.R., 2000. Electrochemical behaviour of cobalt in aqueous solutions of
different pH. J. Appl. Electrochem. 30, 693–704. https://doi.org/10.1023/A:1003893122201
Coetsier, C.M., Testa, F., Carretier, E., Ennahali, M., Laborie, B., Mouton-Arnaud, C., Fluchere, O., Moulin, P., 2011.
Static dissolution rate of tungsten film versus chemical adjustments of a reused slurry for chemical mechanical
na

polishing. Appl. Surf. Sci. 257, 6163–6170. https://doi.org/10.1016/j.apsusc.2011.02.023

Díaz-Tena, E., Barona, A., Gallastegui, G., Rodríguez, A., López de Lacalle, L.N., Elías, A., 2017. Biomachining:
metal etching via microorganisms. Crit. Rev. Biotechnol. 37, 323–332.
ur

https://doi.org/10.3109/07388551.2016.1144046
Díaz-Tena, E., Gallastegui, G., Hipperdinger, M., Donati, E.R., Ramírez, M., Rodríguez, A., López de Lacalle, L.N.,
Elías, A., 2016. New advances in copper biomachining by iron-oxidizing bacteria. Corros. Sci. 112, 385–392.
Jo

https://doi.org/10.1016/j.corsci.2016.08.001
Díaz-Tena, E., Rodríguez-Ezquerro, A., López De Lacalle Marcaide, L.N., Gurtubay Bustinduy, L., Elías Sáenz, A.,
2014. A sustainable process for material removal on pure copper by use of extremophile bacteria. J. Clean. Prod.
84, 752–760. https://doi.org/10.1016/j.jclepro.2014.01.061
Fürtauer, S., Li, D., Cupid, D., Flandorfer, H., 2013. The Cu-Sn phase diagram, Part I: New experimental results.
Intermetallics 34, 142–147. https://doi.org/10.1016/j.intermet.2012.10.004
Hocheng, H., Chang, J.H., Jadhav, U.U., 2012. Micromachining of various metals by using Acidithiobacillus
ferrooxidans 13820 culture supernatant experiments. J. Clean. Prod. 20, 180–185.
https://doi.org/10.1016/j.jclepro.2011.07.019
Huang, H., Ma, F., 2019. Parameter Optimization for Improvement in Biomachining Performance. Trans. Nanjing Univ.
Aeronaut. Astronaut. 36, 376–386.

17
Istiyanto, J., Ko, T.J., Yoon, I.C., 2010. A study on copper micromachining using microorganisms. Int. J. Precis. Eng.
Manuf. 11, 659–664. https://doi.org/10.1007/s12541-010-0077-1
Istiyanto, J., Saragih, A.S., Ko, T.J., 2012. Metal based micro-feature fabrication using biomachining process.
Microelectron. Eng. 98, 561–565. https://doi.org/10.1016/j.mee.2012.07.002
Johnson, D., Warner, R., Shih, A.J., 2007. Surface Roughness and Material Removal Rate in Machining Using
Microorganisms. J. Manuf. Sci. Eng. 129, 223. https://doi.org/10.1115/1.2401629
Kumada, M., Kawakado, T., Kobuchi, S., Uno, Y., Maeda, S., Miyuki, H., 2001. Investigations of fine biomachining of
metals by using microbially influenced corrosion - Differences between steel and copper in metal biomachining
by using Thiobacillus Ferrooxidans. Zair. to Kankyo/ Corros. Eng. 50, 411–417.
https://doi.org/10.3323/jcorr1991.50.411
Leach, R., 2004. Some issues of traceability in the field of surface topography measurement. Wear 257, 1246–1249.
https://doi.org/10.1016/j.wear.2004.06.004
Lim, J.H., Park, J.H., Park, J.G., 2013. Effect of iron(III) nitrate concentration on tungsten chemical-mechanical-
planarization performance. Appl. Surf. Sci. 282, 512–517. https://doi.org/10.1016/j.apsusc.2013.06.003
Ma, F., Huang, H., Xu, X., 2019. Material Removal Mechanisms of Cu–Co Metal-Powder Composite by

of
Microorganisms. Int. J. Precis. Eng. Manuf. Technol. https://doi.org/10.1007/s40684-019-00110-8
Maruda, R.W., Krolczyk, G.M., Wojciechowski, S., Zak, K., Habrat, W., Nieslony, P., 2018. Effects of extreme

ro
pressure and anti-wear additives on surface topography and tool wear during MQCL turning of AISI 1045 steel. J.
Mech. Sci. Technol. 32, 1585–1591. https://doi.org/10.1007/s12206-018-0313-7
Maurice, V., Marcus, P., 2018. Current developments of nanoscale insight into corrosion protection by passive oxide
-p
films. Curr. Opin. Solid State Mater. Sci. 22, 156–167. https://doi.org/10.1016/j.cossms.2018.05.004
Muhammad, I., Khatoon, T., Ullah, S.M.S., Ko, T.J., 2018. Development of Empirical Model for Biomachining to
Improve Machinability and Surface Roughness of Polycrystalline Copper. Int. J. Precis. Eng. Manuf. - Green
re
Technol. 5, 201–209. https://doi.org/10.1007/s40684-018-0020-y
Nguyen, T., Boudard, M., Carmezim, M.J., Montemor, M.F., 2017. Layered Ni(OH) 2 -Co(OH) 2 films prepared by
electrodeposition as charge storage electrodes for hybrid supercapacitors. Sci. Rep. 7.
lP

https://doi.org/10.1038/srep39980
Pieraggi, B., MacDougall, B., Rapp, R.A., 2005. The role of the metal/oxide interface in the growth of passive films in
aqueous environments. Corros. Sci. 47, 247–256. https://doi.org/10.1016/j.corsci.2004.03.004
Reikowski, F., Maroun, F., Di, N., Allongue, P., Ruge, M., Stettner, J., Magnussen, O.M., 2016. In situ surface X-ray
na

diffraction study of ultrathin epitaxial Co films on Au(111) in alkaline solution. Electrochim. Acta 197, 273–281.
https://doi.org/10.1016/j.electacta.2016.01.052
Seo, Y.J., Kim, N.H., Lee, W.S., 2006. Electrochemical corrosion effects and chemical mechanical polishing
ur

characteristics of tungsten film using mixed oxidizers. Microelectron. Eng. 83, 428–433.
https://doi.org/10.1016/j.mee.2005.11.006
Shi, Z., Liu, M., Atrens, A., 2010. Measurement of the corrosion rate of magnesium alloys using Tafel extrapolation.
Jo

Corros. Sci. 52, 579–588. https://doi.org/10.1016/j.corsci.2009.10.016

Sürme, Y., Gürten, A.A., Bayol, E., Ersoy, E., 2009. Systematic corrosion investigation of various Cu-Sn alloys
electrodeposited on mild steel in acidic solution: Dependence of alloy composition. J. Alloys Compd. 485, 98–
103. https://doi.org/10.1016/j.jallcom.2009.05.150
Tavakkolizadeh, M., Saadatmanesh, H., 2002. Galvanic Corrosion of Carbon and Steel in Aggressive Environments. J.
Compos. Constr. 5, 200–210. https://doi.org/10.1061/(asce)1090-0268(2001)5:3(200)
Uno, Y., Kaneeda, T., Yokomizo, S., 2011. Fundamental Study on Biomachining. Machining of Metals by Thiobacillus
ferrooxidans. Trans. Japan Soc. Mech. Eng. Ser. C 59, 3199–3204. https://doi.org/10.1299/kikaic.59.3199
Valdés, J., Pedroso, I., Quatrini, R., Dodson, R.J., Tettelin, H., Blake, R., Eisen, J.A., Holmes, D.S., 2008.
Acidithiobacillus ferrooxidans metabolism: From genome sequence to industrial applications. BMC Genomics 9,
1–24. https://doi.org/10.1186/1471-2164-9-597

18
Walsh, F.C., Low, C.T.J., 2016. A review of developments in the electrodeposition of tin-copper alloys. Surf. Coatings
Technol. 304, 246–262. https://doi.org/10.1016/j.surfcoat.2016.06.065
Wieczorowski, M., Mrozek, R., Andrałojć, P., 2010. The use of surface asperities analysis to investigate wear of bodies
in contact on example of brake elements. Metrol. Meas. Syst. 17, 271–278.
https://doi.org/10.2478/v10178-010-0023-7
Xenofontos, E., Feidiou, A., Constantinou, M., Constantinides, G., Vyrides, I., 2015. Copper biomachining mechanisms
using the newly isolated Acidithiobacillus ferrooxidans B1. Corros. Sci. 100, 642–650.
https://doi.org/10.1016/j.corsci.2015.08.041
Zhang, D., Li, Y., 1998. Possibility of biological micromachining used for metal removal. Sci. China, Ser. C Life Sci.
41, 151–156. https://doi.org/10.1007/BF02882720

of
ro
-p
re
lP
na
ur
Jo

19