Electrochemically-Assisted Low Power Density Laser Writing on Stainless Steel via Enrichment of Chromium Oxides

Abstract

Laser color marking produces nearly permanent, environmentally friendly, vibrant colors on surfaces. However, previous work has used high-power-density pulsed lasers to induce the physicochemical reactions for marking. Here, laser color marking on stainless steel 304 (SS304) is performed with a less expensive continuous wave (CW) laser and a power density five orders of magnitude below that previously reported by combining an electrochemical cell with a fluorescence microscope. Using a combination of optical microscopy, x-ray photoelectron spectroscopy, and bulk electrochemistry, it is demonstrated that the laser-induced luminescence and colors are due to enrichment (32 ± 9% increase) of Cr₂O₃ in the SS304 passive film. It is shown that the enrichment proceeds by a different chemical mechanism than the oxygen pyrolysis that occurs in typical laser color marking. The technique provides a new pathway for laser color marking of metals in industrial settings with applications as diverse as solar absorbers or corrosion prevention.

1 Introduction

Laser writing involves using laser energy to create vibrant, visible marks at the surface of materials, including metals.^[1-7] Laser-induced color changes on metals have been attributed to three physical mechanisms: 1) periodic surface restructuring, 2) nanostructures, or 3) oxide thin films. Most previous methods used pulsed lasers in the 10 000 kW cm⁻² range to deliver thermal energy in an ultrafast timescale to change the surface.^{[3, 8-10]} While there is much experimental characterization of the laser parameters required to produce different colors, there is little experimental work investigating the nanoscale redox chemistry taking place within the laser area.^[11-15] Additionally, the accepted chemical mechanism leading to oxide thin films has rarely been put to the test.^[16]

Here, we investigate the oxidation mechanism occurring in a laser-exposed area at the surface of stainless steel 304 (SS304). First, we describe a new method of continuous wave (CW) laser color marking with irradiance at levels four – five orders of magnitude lower than peak power densities in pulsed lasers. Then, we determine the electrochemically-assisted chemical mechanism of oxide formation with a combination of in situ and ex situ analysis. Finally, we propose a chemical mechanism to explain the oxidation taking place in the laser-exposed area, highlighting the differences from accepted laser writing reaction mechanisms. This study opens the door to new laser writing techniques and mechanistic studies involving the combination of electrochemistry and optical microscopy.

We combined an inverted fluorescence microscope with a three-electrode electrochemical cell in a windowed petri dish (Figure 1). Double-sided tape holds an epoxy-embedded metal sample within the working distance of an objective lens in epifluorescence, and a piezoelectric sample stage controls the relative motion of the laser and the metal. By applying a polarization in the pitting region of the SS304 (Figure S1, Supporting Information), we modified the passive film structure of the SS304, which lowers the energy barrier for the laser-induced oxidation reactions as explored in Section 2.

Figure 1

Laser writing set up and electrochemical cell diagram. a) An Olympus IX-83 fluorescence microscope with piezostage is used to optically image the b) sample assembled in a No. 1.5 glass-bottomed petri dish. c) The side-view shows the three-electrodes, solution, and 25 µm thick double-sided tape that allows for both electrochemical control of the sample and CW laser exposure to the SS304 sample to achieve laser writing. Art by Vivian Wattle, reuse copyright CC BY-NC-ND 4.0.

2 Results

Laser writing is observed in real time on a SS304 surface with a 561 nm CW diode laser only after electrochemically modifying the surface. Prior to polarization, no modification to the SS304 occurred upon irradiation for 1 min with the laser (Figure 2a–c). Next, the SS304 was potentiostatically polarized for 90 min above the pitting potential (Figure S1, Supporting Information) at +0.19 V versus Ag/AgCl. A current drop of ≈1 mA was observed over 90 min and small corrosion pits could be seen on the SS304 surface (Figure 2f; Figure S2, Supporting Information). However, the surface remained unmodified by the exposure to the white light (Figure 2d,f). After 90 min, the polarization was cut off and the sample was exposed to the 561 nm laser operating at a power density of 1.4 kW cm⁻². Within 30 s, bright photoluminescent spots appeared (Figure 2h) that grew in density and intensity with time (Video S1, Supporting Information). Based on the high content of chromium in the laser-exposed areas (see Section 2.2, Figure 5) and previous literature showing visible absorption bands from Cr₂O₃ nanoparticles,^[17-20] we ascertain that these photoluminescent spots are likely deposits of Cr₂O₃. In addition to the photoluminescence appearing, the morphology of the stainless steel surface appeared different when visualized by white light reflectivity images (Figure 2g vs i). The modification was localized only to the circular area of the laser beam (Video S1, Supporting Information), confirming that laser writing occurred.

Figure 2

Colored laser writing is achieved with a combination of applied potential and continuous wave laser. a–c) Micrographs of an unpolarized SS304 before (white light reflectivity), during (fluorescence), and after (reflectivity) exposure to 561 nm laser light. d,f) Micrographs of SS304 surface before and after polarization (parameters in (e), as fluorescence images cannot be obtained without laser exposure). g–i) Micrographs of SS304 surface before, during, and after exposure to 561 nm laser light immediately after polarization shown in (e). Scale bars, 10 µm.

To demonstrate this method's potential for laser writing, we processed several samples with first exposing the sample to 561 nm laser excitation with our high resolution, 100× magnification microscope (Figure 3a, see Section 4.2) and then imaging on a separate white light microscope under lower magnification with a color camera (Figure 3b,e, see Section 4.5). Single spots were imaged under a 20× objective so the entire single area exposed to the laser could be observed (Figure 3b–c).

Figure 3

Laser-written marks remain post-sample processing, and the SS304 sample may be moved to draw patterns. a,b) Diagram showing the method to obtain color images of single spots. c) Color image at 20× magnification of a spot exposed to 561 nm laser for 1 min after polarizing the sample at +0.5 V for 40 min. d,e) Diagrams showing method to obtain color image of a zig-zag pattern. f) Color image at 5× magnification of a pattern created with this process.

User designed patterns may be created by moving the sample stage and thus the SS304 sample relative to the laser (Figure 3d). The sample surface was polarized for 30 min further into the pitting region at V = +0.5 V to decrease time before writing could be performed. While continuing to polarize the sample, the surface was exposed to 561 nm laser and the sample was moved relative to objective lens in 125 µm steps (slightly greater than the 100 µm laser diameter) with a 30 s laser exposure time for each step. Ex situ visualization was performed with 5× magnification to see the entire design in our field of view (Figure 3e). The obtained images showed a pattern (Figure 3f) in our intended zig-zag shape that remained even after cleaning with ethanol.

Laser writing can be achieved with different voltages, polarization times, laser exposure times, laser powers, and wavelengths. We provide color images of eight different parameter sets showing the colors we achieve, namely white, light brown, dark brown, black, and blue in addition to mixes of these colors (Figure S3, Supporting Information). In addition to our experiments at 561 nm, we've observed two different wavelengths (488 nm and 633 nm) cause similar changes to the SS304 surface (Figure S4, Supporting Information). A full exploration of these variables on the effect of final color will be the subject of a future study. Finally, we performed optical profilometry on spots created at different polarization times (Figure S5, Supporting Information) but fixed laser parameters (30 mW power, 5 min laser exposure). We observed a positive correlation between the area surface roughness (S_a, see Section 4.4)^[21] of the laser areas and the polarization time. These experiments also indicate that the oxide layers are thickening by tens to hundreds of nanometers as the S_a value increases from 0.01 ± 0.01 µm on a control sample of unpolarized stainless steel to 0.13 ± 0.11 µm on a sample polarized for 15 min at V = +0.5 V and exposed to the laser for 5 min. Ten to hundreds of nanometers thick layers matches previous reports for traditional laser-induced oxidation.^[15] The oxide layers are also nonuniform as evidenced by the high S_a values.

2.1 Laser Power Dependence Studies

Combining the electrochemical cell with a high-resolution microscope allows real-time observation of the SS304 surface which can uncover chemical kinetics of the laser-induced process. We measured the rate of surface modification by monitoring the change in luminescence intensity over time (see Section 4.3) with variable laser power, the time polarizing the sample (t_Vapp) and waiting time after ending the polarization before laser exposure (t_wait) (Figure 4). For all t_Vapp and t_wait, greater laser power resulted in faster rates of formation of the luminescent spots (Table S4, Supporting Information). Additionally, t_Vapp had a strong effect on the rate (Figure 4a,d) with longer polarization times leading to faster rates. Indeed, we could not observe any luminescent spots or surface coloring at polarization times shorter than 1 h at +0.19 V polarization (Figure S6, Supporting Information). The t_wait also affected the final observed rate. The fastest rates were observed immediately after polarization was stopped or was still being applied, whereas t_wait of 30 or 60 min resulted in a drastic decrease of an order of magnitude in the rate (Figure 4a–c). Generally, longer polarization time and higher laser power lead to a faster rate of formation. We note that relative intensity changes for Figure 4b,c do not always follow this trend in the first <10 s of laser exposure. This is attributable to autofluorescent contaminants within the sample photobleaching at the low intensities in these panels.

Figure 4

The rate of color marking increases with higher laser power and shorter wait time after applying polarization as quantified by luminescence intensity. Relative intensity change over time at three different laser powers under four different electrochemical polarization conditions (t_Vapp = applied polarization, t_wait = waiting time after stopping polarization before taking a video).

2.2 XPS Determination of Elemental Composition

Oxidation of chromium takes place within the laser-exposed area as determined by x-ray photoelectron spectroscopy. Previously blue, brown, and black colors have been assigned to chromium and iron oxide layers on stainless steels.^[22-26] To determine if similar oxide layers were responsible for the visible marks in our system, we performed XPS at different points of the electrochemically-assisted low power density laser writing process (Table S5, Supporting Information; Figure 5). We measured a control sample, sample S1, which was unpolarized and not exposed, but otherwise treated the same. We also measured sample S2, which had been polarized at V = +0.5 V for 45 min but had not been exposed to the laser. Finally, we measured the XPS spectra from two spots that had been polarized at V = +0.5 V for 45 min but exposed to the laser for different lengths of time: one colored white, S3, and one black, S4 (Figure S3e,g, Supporting Information).

Figure 5

Electrochemically-assisted laser writing increases chromium oxide content in SS304. XPS characterization of elemental and oxidation state changes to the passivating layer. a) Survey spectrum of S4 area showing peaks identified for iron, chromium, and oxygen. b) The ratio of atomic percent from the survey spectra for chromium (green) and oxygen (purple) in areas S1-S5. c). High resolution spectrum of Cr2p peak in area S4 showing a three Gaussian model fit to the data. d) The relative percent of each oxidation state of chromium in areas S1-S5.

We measured survey spectra (Figure 5a; Figure S7, Supporting Information) for each of these conditions and calculated the overall atomic percentage of detected elements. We note that all surveys show some carbon and other elements, likely due to contamination^[27] in sample processing (see Section 4.5). Yet, consistent for all areas were the major components of oxygen, iron, and chromium, which are expected in the SS304 passive film. We calculated the ratio of chromium atomic percent to chromium and iron atomic percent (Figure 5b, green bars) to be 0.27±0.02. After polarizing the sample, the ratio was reduced significantly to 0.19±0.02, indicating chromium content in the passive layer formed after 45 min at +0.5 V versus Ag/AgCl had decreased. After exposing the surface to laser light for 1 and 15 min, respectively, the ratio increased to 0.42 ± 0.03 and 0.38 ± 0.04, a 37% increase of chromium:iron ratio from the initial passive layer. The atomic percent of oxygen considering the chromium, iron, and oxygen peaks was also quantified in Figure 5b (purple bars). The oxygen percent in the passive layer increased slightly from 79.3% to 83.7% after polarization. This oxygen concentration increase suggests the potential increases in the passive layer thickness, as further described in Section 2.4. Laser exposure for 1 and 15 min significantly increased the atomic percent of detected oxygen to 87.3% and 90.7% respectively. Errors for the atomic percentages were ±0.1%.

High resolution spectra of chromium reveal changes in the relative amounts of chromium species at each of the conditions measured. These high-resolution spectra were fit to a multiple Gaussian model (Figure 5c; Figure S8, Supporting Information). For the chromium 2p_3/2 peak, three peaks were fit corresponding to Cr⁰ (metal, 574 eV), Cr₂O₃ (oxide, 576.5 eV), and Cr(OH)₃ (hydroxide, 577.5 eV) as established for stainless steel passive films in previous studies.^[28-33] For the native passive layer, the area percentages were 11.8% Cr⁰, 48.8% Cr₂O₃, and 39.4% Cr(OH)₃ with error ±0.1%. After polarization alone, the Cr⁰ decreased to 2.6%, Cr₂O₃ stayed consistent at 48.5%, the Cr(OH)₃ content increased to 48.9%. Upon 1 min laser exposure, Cr⁰ increased slightly to 4.7%, Cr(OH)₃ was enriched even further to 55.4%, and Cr₂O₃ fell to 39.9%. In contrast, a 15 min laser exposure time practically flipped the ratio of chromium oxide to hydroxide with 55.8% Cr₂O₃ to 41.3% Cr(OH)₃.

2.3 Oxygen Deprived Experiments

Our optical and elemental observations support that the luminescence and black and blue colors observed are Cr₂O₃ enrichment at the surface of SS304. Cr₂O₃ was identified to be one of the major coloring compounds and is likely the primary photoluminescent species that was observed under 561 nm excitation (Figure 2h; Video S1, Supporting Information). We have additionally observed that Cr₂O₃ powder synthesized in solution luminesces under 561 nm excitation compared to no luminescence observed for Fe₂O₃ particles (Figure S9, Supporting Information).^[34] The photophysics underlying the Cr₂O₃ luminescence is beyond the scope of this study, but is an intriguing area of future investigation. The increase in chromium/iron ratio and specifically Cr₂O₃ increase (32 ± 9% increase) as measured by the high resolution XPS after 15 min of laser exposure when the highest luminescence intensity and the most intense colors were observed strongly indicates that Cr₂O₃ is the coloring component in this laser writing process.

Oxygen-deprived experiments indicate that the chemical mechanism of Cr₂O₃ formation is different from the accepted oxygen pyrolysis described in previous laser writing studies. The field has described the chemical mechanism for the formation of oxides as pyrolysis of diatomic dissolved oxygen with essentially five steps attributable to the local heating produced by the laser and adsorbed oxygen dissociating into anions that react with metal cations in the film.^[3] However, the power density of our CW lasers is significantly lower than in the previous ultrafast reports to such a level that it seems unlikely oxygen dissociation would be the physicochemical mechanism. To test whether pyrolysis of dissolved oxygen was occurring within the laser area, we performed electrochemically-assisted laser writing in an oxygen-deprived condition with nitrogen flow in the sample cell (Figure 6a–f). We observed similar marking and luminescence induced by the laser in the same amount of time with or without nitrogen flow, indicating the chemical mechanism does not require dissolved oxygen.

Figure 6

Chromium oxide is enriched without dissolved oxygen by dehydration and oxidation reaction of chromium hydroxide. a–d) Micrographs of SS304 surface a,b) before, c,d) during, and e,f) after exposure to 561 nm laser for 5 min with a,b,e,f) white light reflectivity and c,d) fluorescence microscopy. Scale bars are 15 µm. g) Proposed reaction mechanism for how Cr₂O₃ layers are made without dissolved diatomic oxygen.

2.4 Proposed Chemical Mechanism

We propose that electrochemically-assisted CW laser writing without a dependence on dissolved oxygen proceeds by a chemical mechanism which can be explained with assistance from the point defect model (PDM) for passive film growth. Essentially, the passive film is treated as a bi-layer structure with point defects (e.g., cation vacancies and oxygen vacancies) in an inner barrier layer and a non-defective outer barrier layer under the PDM. The PDM has been particularly successful at predicting the properties of the passive oxide layer in aggressive corroding environments like high concentration chloride and the strongly polarized environment in this study.^[35-38] MacDonald introduced the PDM and others have suggested that illumination of stainless steel surface can influence the passive film properties by interpreting the passive film growth with this model.^{[27, 39-43]} Because of this strong history of understanding the influence of radiation on the passive film composition, we use this model to interpret our findings. We now introduce a three-step chemical mechanism based upon the XPS experimental data that explains how the formation of a Cr₂O₃-enriched passive film occurs without dissolved oxygen as seen in the data in Figure 6a–f.

First, the applied electrochemical potential decreases the chromium content in the passive film and lowers the passive film bandgap for subsequent laser-induced reactions. The chromium content in the passive film decreased with the anodic potential, consistent with previous reports,^{[30, 44, 45]} due to dissolution of highly oxidized chromium species. The passive film also thickened as evidenced by the decrease in Cr⁰ and Fe⁰ (Figure 5d; Figures S8 and S10, Supporting Information) and oxygen content increase (Figure 4b) in the integrated intensity of elemental composition in the ≈10 nm thickness probed by XPS. Additionally, there were more iron oxides in this layer (Figure S10, Supporting Information). This more iron oxide-rich passive film will have a lower energy bandgap than the native passive layer (formed in liquid at open circuit potential) which explains why the laser can influence the composition during or directly after polarization but not before.^[46] Additionally, the high oxidizing power environment at anodic potential will be favorable for oxidation reactions.

Second, the exposure of the metal to the CW laser raises vacancy diffusion within the metal as described by the PDM, growing the inner and outer passive film layers and enriching chromium content. The PDM has been used to explain the influence of super bandgap illumination on passive film composition for stainless steels and other metals.^{[27, 42, 43, 47-50]} In particular, the formation of electron-hole pairs by illumination lowers the electric field strength within the passive film raising diffusivity of defects, such as oxygen vacancies. According the Nernst-Planck transport Equation (1), oxygen vacancy diffusion coefficient is inversely proportional to electric field.

$$$$\begin{equation}{{D}_0} = \frac{{{{i}_{ss}}RT}}{{4eF{{\omega }_2}{{\varepsilon }_L}}}\end{equation}$$$$(1)

where D₀ is the oxygen vacancy diffusion coefficient, i_ss is the steady state current, R is the gas constant, T is the temperature, e is the electron charge, F is the Faraday constant, ω₂ is experimentally determined with Mott-Schottky analysis, and ε_L is the electric field strength. Therefore, the oxygen vacancy diffusion coefficient will be raised under the laser illumination which is favorable for passive film formation. In the chloride environment, iron (II) and iron (III) will preferentially dissolve because iron chloride is more soluble than chromium chloride in aqueous environments leading to the immediate increase in overall chromium content we observe via XPS in Figure 4b sample S3.^[51-53]

Theoretical work has shown the composition of oxide and hydroxide components could be influenced by the electric field strength^[54] and our results qualitatively match previous observations of chromium hydroxide enrichment at lower illumination fluence and more chromium oxide at higher fluence.^[39] While exact theoretical explanations for the variations in the chromium forms (Cr(OH)₃ versus Cr₂O₃) observed are beyond this work's scope, it is a promising area of future theoretical and experimental spectroscopic work.

Third, the laser assists in Cr(OH)₃ dehydration to form the final Cr₂O₃ that is primarily responsible for the black or blue color changes and luminescence in the laser writing. In high oxidizing power conditions Cr(OH)₃ in the passive film can convert to Cr₂O₃ via:^[55]

$$$$\begin{equation}{\mathrm{Cr}}{{\left( {{\mathrm{OH}}} \right)}_3} + {\mathrm{C}}{{{\mathrm{r}}}^0} \to {\mathrm{C}}{{{\mathrm{r}}}_2}{{{\mathrm{O}}}_3} + 3{{{\mathrm{H}}}^ + } + 3{{{\mathrm{e}}}^ - }\end{equation}$$$$(2)

At longer laser exposure times and higher fluence (more energy imparted, Figure 4) the laser is assisting this reaction by dehydrating the hydroxyl groups (Figure 6e, green squiggle), thus building up a pure chromium oxide layer (Figure 6e, blue rectangle). This reaction may start as small micro inclusions of pure Cr₂O₃, which could be the photoluminescent spots we observe. Eventually, these microinclusions become dense enough that the film appears as a homogeneous dark blue or black color and oxide layer. Notably, this reaction does not require any oxygen source outside of the hydroxyl groups already present in the passive film, which explains why our method occurs in an oxygen-deprived environment in contrast to previously reported laser-induced oxidation on metal surfaces. Indeed, no step in the proposed reaction mechanism requires additional oxygen: the combination of the polarization and laser modifies the existing passive film. This reaction mechanism is graphically summarized in Figure 6e.

3 Conclusion

Overall, this research study has resulted in four main discoveries. First, laser writing may be performed on SS304 at a power density of 0.1–1 kW cm⁻² with visible wavelength CW lasers assisted by an anodic polarization of the metal. Second, luminescent spots are formed on the metal surface, and the rate of formation is linearly dependent on the applied laser power in the power density range of 0.1–1 kW cm⁻². Third, blue/black colors and luminescent spots were identified as Cr₂O₃. Finally, the formation of the Cr₂O₃ passive film does not require dissolved or atmospheric diatomic oxygen in contrast to previously reported for laser-induced oxidation on metal surfaces. The mechanism may be explained by the PDM, whereby changes in the flux of defects in the passive film structure lead to energetically favorable conditions for the laser energy to convert Cr(OH)₃ into Cr₂O₃.

The cost, energy, and safety requirements for writing on metal is reduced by orders of magnitude compared to ultrafast approaches. Currently, class 4 lasers that cost $10 000–$100 000 (e.g., Nd:YAG, Ti:sapph lasers) with 100 kHz–1 MHz repetition rates producing ns-fs pulses with peak irradiances at ≈10 000 kW cm⁻² are required. In contrast, our class 3 CW laser costs ≈$1000 and achieves a ≈1 kW cm⁻² irradiance. The additional cost of the potentiostat (≈$5000), could be reduced with a less expensive (≈$300) DC power supply. Hypothetically, a simple battery and laser diode could achieve direct laser writing for only dollars based on the voltage and laser conditions tested, showing the possibility for application of electrochemically-assisted laser writing beyond fundamental research.

While this method reduces cost and increases safety for laser color marking applications, there are several potential limitations to electrochemical-assisted low power density laser writing. The spatial resolution achievable for designs depends on the laser spot size. We only use a 100× objective yielding a 125 µm diameter beam in this study meaning the smallest writable feature would need to be 125 µm or greater. Lower magnification objectives could be used if a larger colored area is needed, but this will raise the power input requirements for a given color since the rate is power dependent. Here, we have tested the formation rate at three powers in the range of 0.1–1 kW cm⁻². In addition to the spatial limitations, the colors produced by our method span only brown and blue hues and show variability even within a single laser spot. It is possible that other colors could be produced on different grades of stainless steel as the chromium content in the alloys can vary. We have additionally shown that our method could work on a different grade of stainless steel with promising early results (Figure S11, Supporting Information). Additionally, electrochemically assisted low power density laser writing could be extended to non-ferrous alloys provided the oxide layer on those samples show strong colors, but the chemistry of such reactions would need to be determined. A last limitation is that the samples are polarized beyond the pitting potential meaning that we degrade the metal after polarization times greater than several hours. In fact, we have observed crevice corrosion from the joint between the epoxy and the SS304 cube to consume the surface of the sample after several hours. Therefore, we anticipate this method could be more suited to applications where localized small areas need to be laser marked without color specificity compared to traditional, high-power laser color marking.

Future work could pursue monitoring the laser writing process in real time, along with scanning a full range of variables beyond the conditions reported here to empirically determine possible colors from a materials science perspective, in addition to the redox mechanism. The seconds-long time scale of electrochemically-assisted CW laser writing allows for observation and insight into the mechanism of the laser-induced changes (Video S1, Supporting Information). The discovery that this oxidation reaction can be performed without dissolved oxygen in solution points out the need for greater understanding of the chemistry occurring within a laser area. Alternative color palettes could be developed for industrial laser color marking with the discovery and physical understanding of reactions and rates under different illumination conditions. Additionally, this work has implications for electrochemistry and corrosion research. Oxide layers formed under illumination are more resistant to pitting corrosion, but the underlying reasoning for this resistance is debated.^{[27, 43, 50]} Furthermore, the redox chemistry at a metal surface has implications in hydrogen evolution.^{[56, 57]} Fluorescence microscopy is a powerful tool to understand corrosion in real time and will be the topic of future work.^[58-61]

4 Experimental Section

Materials and Reagents

SS304 (0.0–2.0 Mn, 0.0 –1.0 Si, 0.0 –0.11 N, 0.0 –0.5 P, 0.0 –0.03 S, 8.0 –15.0 Ni, 17.5 –19.5 Cr, Fe Balance wt%) bars with a 1 cm × 1 cm size were purchased from McMaster Carr. SS304 bars were cut into cubes having a size of 1 cm × 1 cm × 0.5 cm and embedded in epoxy (Buelhler EpoThin 2 resin and hardener) with a hook up wire attached via copper tape (Electron Microscopy Sciences 77 802). The epoxy embedded SS304 was polished on the exposed sample surfaces by 100, 400, 600, 800, and 1200 grits SiC sandpaper (Allied High Tech Products Inc.). The samples were cleaned by ethanol ultrasonication and dried by nitrogen just prior to experiment. For moving and storing samples when not actively performing data collection, samples were placed in a portable desiccator vacuum chamber.

The epoxy-embedded SS304 was stuck to a petri dish with a No. 1.5 glass coverslip bottom (Mat Tek P50G1) via two pieces of double-sided tape with a thickness of 25 µm leaving an exposed SS304 area of 1 cm × 0.5 cm (Figure 1). The petri dish was cleaned by immersion in 5% v/v Contrad 70 soap (Decon Labs) for 1 h, then rinsed ten times with ultrapure water, and dried via nitrogen prior to experiment. The petri dish was filled with 5 mL of 20 mm HEPES buffer at pH = 7.3 (diluted from 1 m stock Thermo J16924) and 3.5 wt% NaCl (Fisher S271).

Chromium (III) oxide (Millipore Sigma 203 068) and iron (III) Oxide (Aldrich 544 884) were drop cast onto clean 1.5 coverslips cleaned by TL1, then O₂ plasma cleaning for imaging.

Fluorescence Microscopy and Electrochemical-Assisted Laser Writing Process

The home-built microscopy setup for imaging the SS304 working electrode surface comprised of a microscope body from Olympus (IX83). For luminescence images, the SS304 surface was illuminated by a 100 mW, 561 nm laser module from CNI lasers. The laser beam was expanded to a diameter of 7.5 mm and then eventually to 22.5 mm before it passed through a lens (focal length = 400 mm) that focused the beam to the back aperture of the objective lens. The dichroic mirror that directs the laser beam toward objective lens was obtained from Chroma (ZT561rdc). A 100×/1.49NA oil immersion objective lens from Olympus (UAPON100XOTIRF) was used to excite and image the sample using power densities specified in the text. The same objective lens was used to capture the fluorescence signal from the sample. Another emission filter (ZET561, Chroma) was used before the detector to remove any stray reflected excitation light. A Photometrics Prime 95B 22 mm scientific CMOS camera was used to capture the images using an exposure time of 30 ms at a rate of 10 Hz. Camera and microscope stages were controlled using the Micromanager 2.0 open source microscopy software. The same microscope (Olympus IX83) was fitted with a white halogen light source (Thorlabs MWWHL4) that illuminates the sample plane and the white light reflectivity image of the sample is captured by the same objective lens (100X/1.49 oil immersion) and camera (Photometrics Prime 95B 22 mm sCMOS).

A constant potential was applied to SS304 sample by a CH Instruments potentiostat (CHI630) via a three-electrode system including SS304 sample as a working electrode, MI-402 flexible dip-type Ag/AgCl 3 m KCl reference electrode (Microelectrodes, Inc) as a potential reference, and a platinum wire as the counter electrode. The potentiostatic polarization was applied for varying times, which produced different colors (Figure S3, Supporting Information). Three different power density levels were used and are noted in the text. The exposure of the sample to the laser was controlled by flipping a mirror that switched between the laser and the white light halogen source mentioned above. Laser exposure time was thus measured from the time the mirror was swapped out, exposing the sample to the laser, to the time the mirror was swapped back in, returning to a white light imaging. These laser exposure times are noted in the text and figure captions or shown in the data in Figure 4. For experiments in oxygen deprived conditions pure nitrogen gas (Airgas UN1066) was bubbled through the liquid in the petri dish for 5 min before exposing the surface to the laser while electrochemical polarization was still being applied. After laser writing and imaging, the tested sample was rinsed with deionized water, dried with nitrogen, and stored in a desiccator for later ex situ analysis.

Image Analysis for Measuring Intensity Changes

Image stacks collected under laser illumination were analyzed in FIJI for intensity changes using the following procedure. The stack was loaded in and cropped to a square region inscribed around the Gaussian beam spots maximum intensity as the center with the side length being equal to the 1/e size of the beam. A threshold was applied to this image to distinguish signal pixels. For three of four samples analyzed, the signal in the first frame of the video was sparse so the threshold was set to the mean pixel intensity plus three sigma. For one sample, a large amount of signal was already present, so the threshold was set using the moments threshold,^[62] which is suited to cases where the distinction between signal and background pixels is small (Figure S12, Supporting Information). For each frame at time (t) in the stack the intensity (I) was calculated as the sum of the i pixels above the threshold and the relative intensity (RI) was calculated as

$$$$\begin{equation}RI = \frac{{\sum_i {{I}_0} - \sum_i {{I}_t}}}{{\sum_i {{I}_0}}}\end{equation}$$$$(3)

The curve of relative intensity versus time was linearly fit in Origin 2022.

Optical Profilometry for Surface Roughness Characterization

Surface roughness of the samples were quantified through optical profilometry. Images were acquired on a Zygo NewView 7300 Scanning White Light Interferometer (SWLI) Optical Profilometer. Images were taken using a 20× objective on a 2× zoom tube setting for an effective 40× magnification of the surface. An area of 0.18 mm × 0.13 mm was imaged for each condition. Data was saved in a xyz format and analyzed using MATAB 2024a. A plane was fitted to the raw data and subtracted from it, to ensure any tilt between the sample plane and imaging plane was removed, to give a surface profile. The surface profile was visualized in 2D and a line profile was selected through the center of the laser spot (where applicable) to extract height deviations across the laser spot. A S_a roughness parameter was calculated for a manually selected smaller region of interest (55 µm × 41 µm) inside the laser spot (where applicable) using the following expression

$$$$\begin{equation}{{S}_a} = \overline {\left| {{{h}_{{x},{y}}}} - {{h}_{{x},{y}}} \right|} \end{equation}$$$$(4)

where, h_x,y is the height at the x and yth pixel in the image. The uncertainty in the measurement was taken as the standard deviation of the absolute value of the difference between each height data point and the average height point.

Post-Process Color Imaging and X-Ray Photoelectron Spectroscopy (XPS) Measurements

After laser exposure (Section 4.2) was complete, the SS304 sample was removed from the buffer solution and cleaned by ultrasonication in ethanol. SS304 samples were cut from the epoxy using an Allied High Tech saw without cutting into the laser-exposed surface. The samples were imaged with a Zeiss Imager A1m microscope with a Axio MR3 color camera under a EC Plan-Neofluar 20x or 10x objective. While imaging, a micro razor blade was used to make correlative marks to help find the laser-exposed areas in the XPS.

The samples were mounted on a standard 1-inch sample mount (provided by Physical Electronics) using a double-sided tape, ensuring the tape was only applied to the non-measured side. PHI 5000 Versaprobe XPS with monochromatized AlK_α source (x-ray size of 100 µm) was used for measuring the as-received sample surfaces. The electron gun was used for charge compensation. Chamber pressure during the measurements remained below 10⁻⁷ Pa. Survey scans were carried out with a pass energy of 93.8 eV, energy step of 0.4 eV, and time/step of 25 ms. High resolution scans were carried out with a lower pass energy of 23.5 eV, energy step of 0.1 eV, and time/step of 50 ms. Atomic percentages of the elements of interest were calculated using the Multipak software (version 9.0). Atomic percentages were calculated by dividing the peak area of a specific element peak by its relative sensitivity factor. Then, that value was divided by the sum of such values calculated for all element peaks detected in the survey spectra. The high resolution peaks of chromium (Cr 2p_3/2) and iron (Fe 2p_3/2) were separated into contributions from each oxidation state based on well-established data analysis techniques to fit a multiple Gaussian model to these peaks.^{[29-31, 63-65]} Well-characterized standards in the literature for SS304 allowed each Gaussian component to be assigned to a specific species and from the area percent of each Gaussian component to determine the relative amounts of each oxidation state.^{[28, 29, 31-33, 66]}

Bulk Electrochemistry Cyclic Polarization

The constant potential used in electrochemically-assisted writing was selected to be above the SS304 pitting potential (E_pit) to ensure the modification of metal surface. To determine the pitting potential of SS304 in the testing condition, cyclic potentiodynamic polarization testing was performed on SS304 in bulk solution of 3.5 wt% NaCl with 20 mm HEPES buffer at pH = 7.3, in accordance with ASTM G5 and ASTM G61 standards.^{[67, 68]} The polished sample, with a final surface roughness of 1200 grit and a testing area of 1 cm², was tested using a CHI630 potentiostat. Figure S1a (Supporting Information) illustrates the cyclic potentiodynamic polarization setup, which includes a three-electrode system: an SS304 working electrode, a standard-sized Ag/AgCl reference electrode (Fisherbrand Accumet glass body Ag/AgCl reference electrode, 4 m KCl), and a Pt wire counter electrode. The open circuit potential (OCP) was measured for at least 1 h until it stabilized, after which the potentiodynamic scan began at −0.5 V versus OCP, ramped up to 1.2 V versus OCP, and then back to −0.5 V versus OCP. The scan rate was set to 0.005 V s⁻¹, with a sample interval of 0.001 V and auto sensitivity (A/V). The measured results were analyzed by CHI630 software, and the value of E_pit was determined at the inflection point.^[69-71]

Since different reference electrodes were used in the various tests, the measured E_pit versus the standard-sized Ag/AgCl 4 m KCl reference electrode in cyclic potentiodynamic polarization was converted to a value relative to the MI-402 flexible dip-type Ag/AgCl 3 m KCl reference electrode used in electrochemical-assisted laser writing as follows:

$$$$\begin{equation}{{{\mathrm{E}}}_{{\mathrm{pit}}}}{\mathrm{versus}}\,{\mathrm{MI}} - 402\,{\mathrm{Ag}}/{\mathrm{AgCl}} = {{{\mathrm{E}}}_{{\mathrm{pit}}}}{\mathrm{versus}}\,{\mathrm{standard}}\,{\mathrm{size}}\,{\mathrm{Ag}}/{\mathrm{AgCl}} + \Delta {\mathrm{V}}\end{equation}$$$$(5)

where ΔV represents the potential difference between the two reference electrodes. This value is measured as OCP between reference electrodes immersed in the testing solution for at least 30 min until the reading stabilizes. Unless otherwise specified, all potentials mentioned in this study are referenced to the MI-402 flexible dip-type Ag/AgCl 3 m KCl electrode.

Statistical Analysis

For the slopes and 95% confidence intervals constructed in Figure 4, least squares fitting was performed in Origin, and the confidence intervals were calculated from the associated t-value for 95% confidence given the degrees of freedom of the measurements of intensity multiplied by the standard error of the estimated slope. The error bars in Figure 5b were calculated by propagating the instrument on atomic percentage (0.1%) through the calculation for the ratio of chromium to total iron and chromium.^[72] Mean and standard deviation definitions for optical profilometry measurements are given in Section 4.4.

Acknowledgements

The National Science Foundation Award #2142821 made through the CHE CSD program supported this research, along with support for V.V. by an Allen Distinguished Investigator Award, a Paul G. Allen Frontiers Group advised grant of the Paul G. Allen Family Foundation. The authors thank Case School of Engineering's Swagelok Center for Surface Analysis of Materials at Case Western Reserve University for use of their instruments. They also thank Professors Clemens Burda, Christy Landes, and Stephan Link for helpful discussions on the research and chemical mechanism presented. They further thank Dr. Gregory Nielson of Nielson Scientific LLC and Ed Caner for helpful discussions surrounding the industrial applications for laser writing. They also thank Professor Xuan Gao's lab at Case Western Reserve University for the use of their metrology microscope. The authors thank Vivian Wattle, who created the illustration of the electrochemical cell in Figure 1. They also thank Manish Kumar and SciDraw for the drawing of the microscope objective used in Figure 3 (10.5281/zenodo.4914800) under a CC BY 4.0 license. They also thank the MORE Center at Case Western Reserve University for use of the optical profilometer.

Conflict of Interest

The authors declare no conflict of interest.

Open Research

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.