Hydrogen-powered polymer electrolyte membrane fuel cells (PEMFCs) show promising potential to power a wide range of mobile and stationary applications and to reduce greenhouse gas emissions significantly. In PEMFCs, the oxygen transport and the water transport are essential for a long lifetime and high-performance characteristics. The diffusion media (DM), located between the bipolar plate and the catalyst-coated membrane, is a crucial component of the fuel cell that significantly affects the cell-internal processes. Usually, the DM is a two-layer material system consisting of a microporous layer based on carbon black particles coated onto a porous gas diffusion layer (e.g., carbon paper). The properties of the microporous layer regarding the water transport at high current densities and, consequently, the fuel cell’s performance and lifetime can be improved by laser structuring. Within this work, different microporous layers with varying binder content and porosities were structured by locally ablating the material using ultrashort-pulsed laser radiation in the infrared wavelength range. The effect of varying process parameters was additionally investigated. Furthermore, the ablation efficiencies were calculated for increasing pulse repetition rates to qualify a process window for an industrial structuring process. The size of the micro-drillings and the heat-affected zone surrounding the hole were evaluated through topographic and microstructure analyses using a laser scanning microscope and a scanning electron microscope with energy-dispersive x-ray spectroscopy. The results showed a rather small influence of the porosity and composition of the microporous layer on the ablation behavior. In contrast, the laser structuring parameters influenced the micro-drilling geometry significantly.

The devastating and partly irreversible consequences of the climate crisis are becoming increasingly observable worldwide. Decarbonizing the industrial and transport sectors is one of the largest challenges in the upcoming years. A possibility to reduce CO2 emissions in shipping, air, and heavy-duty transportation is the usage of renewably produced hydrogen instead of fossil fuels. As hydrogen is one of the most promising energy carriers of the future, fuel cells, which can convert the chemically stored energy into electrical energy, show great potential to power a wide range of applications. Due to its high power density and moderate operating temperature, the polymer electrolyte membrane fuel cell (PEMFC) is currently the dominant fuel cell technology.1 However, a further reduction in material and manufacturing costs is necessary for the further commercialization of PEMFC.2 In addition to the considerably high costs, the management of the reaction product within the cell is one of the most significant challenges for further commercializing the PEMFC.3 In recent years, various research papers have investigated the modification of the diffusion media (DM) in PEMFCs and have achieved promising results of improved water management and cell performance.4 

In a PEMFC, hydrogen and oxygen react to water. For an efficient and controlled reaction, a multilayer structure is necessary. Hydrogen and oxygen are supplied through the flow channels of the bipolar plates (BPPs). The hydrogen (H2) and oxygen (O2) molecules move through the DM, which consists of a gas diffusion layer (GDL) and a microporous layer (MPL), to the catalyst layer (CL). At the anode side, the hydrogen evolution reaction (HER) and, at the cathode side, the oxygen reduction reaction take place. During the HER, liquid water is generated. A part of the water is needed to hydrate the membrane and enable the ion transport. In contrast, excess water must be removed from the fuel cell through the BPP to prevent the formation of water accumulations, which extend or completely block the diffusion paths for oxygen. Restricted oxygen transport leads to an increase in diffusion resistances and a drop in fuel cell voltage, resulting in a reduced performance.2 

A schematic illustration of the PEMFC structure, as well as the flow direction of the hydrogen, oxygen, and liquid water, is presented in Fig. 1.

FIG. 1.

Schematic illustration of a PEMFC with the supply of the oxidant (H2) as well as the reactant (O2) and the reaction product (H2O) drainage.

FIG. 1.

Schematic illustration of a PEMFC with the supply of the oxidant (H2) as well as the reactant (O2) and the reaction product (H2O) drainage.

Close modal

The DM is of great importance for the performance of the PEMFC as this layer fulfills multiple functions: First, the reaction gases from the flow channels of the BPP should be distributed uniformly to the CL. Second, the electron transport between the CLs and the BPP should be enabled. Third, the retention of some product water is necessary to ensure a sufficiently humidified PEM while simultaneously removing the excess water as well as the heat. Fourth, the DM should provide physical support to the membrane electrode assembly.5 In order to meet these requirements, the DM must provide high gas and water permeability, high electrical and thermal conductivity, as well as mechanical stability along with chemical resistance.6 The DM comprises a macroporous gas diffusion layer and a microporous layer. The GDL is typically a carbon fiber paper with a hydrophobic polytetrafluoroethylene (PTFE) impregnation, which has a layer thickness of about 200–400 μm.7 The MPL is composed of a mixture of carbon or graphite particles with a proportion of 60–80 wt. % and a hydrophobic polymer binder, such as PTFE with a proportion of 20–35 wt. %.4 The layer thickness of the MPL is between 10 and 100 μm,8 and the MPL has pore sizes of less than 500 nm6 

The MPL porosity significantly influences the mass transport in the PEMFC. Higher porosities and larger pore sizes increase the effective diffusivity and, thus, the reduction in oxygen transport resistance.9 However, MPL pore sizes in the region of GDL pore sizes (typically between 10 and 30 μm2) lead to flooding conditions and consequently to restrictions in mass transport at high current densities10 or a membrane drying at low current densities.11 The pore size distribution is another important property of the MPL and is affected by the manufacturing process and the applied materials.12 Simon et al. demonstrated that MPLs with a broad pore size distribution exhibit an improved mass transfer mechanism.9 Since small hydrophobic pores have a high capillary pressure, the liquid reaction product water (H2O) is preferentially transported through larger pores, while smaller pores remain free for O2 diffusion.9 For commercial MPLs, the porosity values and the material compositions are usually not available. Chen et al. investigated the porosity, as well as the porosity distribution and morphology of 15 MPLs from three different commercial manufacturers using x-ray tomographic microscopy.13 The results of the study revealed that the MPLs differ significantly in their porosity. Furthermore, the majority of the MPLs examined exhibited a heterogeneous porosity distribution.13 

In addition to the porosity, the PTFE content of the MPL has a crucial impact on mass transport in the cell. Due to the material properties of PTFE, an increase in the PTFE content in the DM leads to enhanced hydrophobic characteristics of the DM.4 While high hydrophobicity would lead to water flooding in the CL, too much hydrophilicity would cause flooding in the GDL.14 In both cases, the performance of the PEMFC is severely impaired by restricted oxygen transfer.14 Furthermore, excessive hydrophobic additives commonly result in reduced porosity, conductivity, and permeability. Related to the type of carbon and hydrophobic agent, an optimum ratio is available.15 In recent years, several studies have investigated the optimal value of the PTFE content to improve the water management. The results of these studies varied between 20 (Ref. 14) and 35 wt. %16 depending on the manufacturing process of the MPL and the operating conditions of the cell. Additionally, research focused on the impact of MPL surface cracks on the water management.4 Sasabe et al. showed that 10 μm wide cracks promote water removal by acting as pathways to the GDL.17 The smaller pores of the MPL are, thus, not blocked by product water and facilitate oxygen transport to the CL.17 Hou et al. also investigated the influence of 10 μm wide cracks in the MPL in a simulative study using a 3D Lattice Boltzmann Model.12 The simulation result showed that the cracks are the preferable water transport paths into the GDL because a smaller capillary force against the water transport has to be overcome in comparison to the small hydrophobic MPL pores. However, the formation of these cracks is a random process during the fabrication of the MPL without the possibility of controlling the distribution and size of the coating defects, limiting the potential of further improving the functionality of the water transport reproducibly.12 Promising approaches for a systematic introduction of the cracks are mechanical structuring11 or laser micro-drilling.18 Within the previous studies, an improvement of the power density of single fuel cells18 and also fuel cell stacks19 with structured DM by up to 20% was shown in comparison to reference fuel cells and fuel cell stacks without structured DM. So far, Alink et al. have indicated that laser structuring of the DM with nanosecond laser pulses removes the hydrophobic binder in the vicinity of the perforation.11 The presence of hydrophilic regions resulted in an aggregation of water near the micro-drillings, counteracting the favored water removal properties.11 It was found that laser perforation under an argon atmosphere reduced the PTFE removal around the holes, which resulted in better water transport and cell performance.11 Using ultrashort laser pulses with a duration of 8 ps for the structuring task showed advantageous drilling properties. A smaller heat-affected zone surrounding the laser perforated holes and simultaneously fewer hydrophobic binder evaporation was detected.20 

The publications mentioned above indicate that systematically introducing laser-structured holes in the DM is suitable for improving the water management and the performance characteristics of a fuel cell. However, as already described, the DM for fuel cells is continuously optimized in terms of composition, porosity, and material thickness. Additionally, for different fuel cell types, the DM composition can vary. Therefore, in this study, the influence of the diffusion media’s MPL composition and porosity on the ablation behavior was examined. Furthermore, focusing on the industrialization of the process, an efficient drilling process concerning the drilling time and the energy required per micro-drilling is needed, which is the focus of the second part of the paper.

The experiments were carried out with six different DMs. The properties of the DMs are summarized in Table I. The MPLs Li100-20, Li100-40, and Li400 were produced on a lab scale as described by Simon et al.21 

TABLE I.

Characteristics of the laser-structured MPLs.

DMs with an MPL produced on lab scale and commercially available GDL
NameLi100-20Li100-40Li400
GDL thickness 140 μ130 μ135 μ
MPL thickness 40 μ35–40 μ35–40 μ
Acetylene black 80 wt. % 60 wt. % 80 wt. % 
PTFE content of the MPL 20 wt. % 40 wt. % 20 wt. % 
MPL porosity 79% ± 1% 79% ± 1% 68% ± 1% 
Source 21  21  21  
Commercially available DMs 
Name Freudenberg H14CX653 Freudenberg H14C10 SGL Carbon BC24 
GDL thickness 145−158 μ120–125 μ200 μ
MPL thickness ca. 40 μca. 45 μ35–40 μ
PTFE content of the MPL — — 23 wt. % 
MPL porosity 66% ± 4% — — 
Source 13, 22  22  23  
DMs with an MPL produced on lab scale and commercially available GDL
NameLi100-20Li100-40Li400
GDL thickness 140 μ130 μ135 μ
MPL thickness 40 μ35–40 μ35–40 μ
Acetylene black 80 wt. % 60 wt. % 80 wt. % 
PTFE content of the MPL 20 wt. % 40 wt. % 20 wt. % 
MPL porosity 79% ± 1% 79% ± 1% 68% ± 1% 
Source 21  21  21  
Commercially available DMs 
Name Freudenberg H14CX653 Freudenberg H14C10 SGL Carbon BC24 
GDL thickness 145−158 μ120–125 μ200 μ
MPL thickness ca. 40 μca. 45 μ35–40 μ
PTFE content of the MPL — — 23 wt. % 
MPL porosity 66% ± 4% — — 
Source 13, 22  22  23  

The respective surface morphologies of the six MPLs used for laser structuring recorded with the scanning electron microscope (SEM) at a 4.500× magnification are shown in Fig. 2. A clear difference in the particle size and shape, the pore size and shape, and the surface structure can be seen.

FIG. 2.

SEM top views of the different MPL surfaces, which were laser structured within this study: (a) Li100-20, (b) Li100-40, (c) Li400, (d) Freudenberg H14CX653, (e) Freudenberg H14C10, and (f) SGL Carbon BC24.

FIG. 2.

SEM top views of the different MPL surfaces, which were laser structured within this study: (a) Li100-20, (b) Li100-40, (c) Li400, (d) Freudenberg H14CX653, (e) Freudenberg H14C10, and (f) SGL Carbon BC24.

Close modal

In Table II, the material constants of the two main MPL components, acetylene black and PTFE, used for Li100-20, Li100-40, and Li400 are shown.

TABLE II.

Material constants of the main MPL components.

Material constantAcetylene black (Ref. 24)PTFE (Ref. 25)
Fusion point 3652–3697 °C 327 °C 
Boiling, sublimation temperature 4200 °C >400 °C 
Material constantAcetylene black (Ref. 24)PTFE (Ref. 25)
Fusion point 3652–3697 °C 327 °C 
Boiling, sublimation temperature 4200 °C >400 °C 

For the experiments, two different laser beam sources were used. The first system was a pulsed ytterbium fiber laser (IPG YLPP-25-3-50-R, IPG Photonics Corporation, USA) that emits laser pulses with a duration of 2 ps. The beam deflection and focusing were performed by a 2D scan head (SUPERSCAN IV-15, RAYLASE GmbH, Germany) equipped with an F-Theta lens (JENar 160-1030…1080-110, JENOPTIK Optical Systems GmbH, Germany).

The second laser system used was a pulsed fiber laser (Dart Picosecond Laser, Novanta Europe GmbH, Germany), which provides laser pulses with a duration of 8 ps. For beam guidance and focusing, a 2D scan head (Racoon 21, Novanta Europe GmbH, Germany) with an F-Theta lens (JENar 160-1030…1080-110, JENOPTIK Optical Systems GmbH, Germany) was utilized. The schematic representation of the experimental setup consisting of the laser beam source, the scanning optics, and the diffusion media placed on an aluminum plate is shown in Fig. 3.

FIG. 3.

Schematic drawing of the experimental setup.

FIG. 3.

Schematic drawing of the experimental setup.

Close modal

Further information regarding both laser systems is shown in Table III.

TABLE III.

Characteristics of the laser beam sources and of the optical setup.

Laser system12
Producer IPG Novanta 
Product name YLPP-25-3-50-R Dart picosecond Laser 
Operation mode Pulsed Pulsed 
Wavelength λ 1030 nm 1064 nm 
Max. laser power P 50 W at 2.75 MHz 56 W at 15 MHz 
Pulse duration τ 2 ps 8 ps 
Pulse repetition rate (amplifier) fr 50–2750 kHz 100–15 000 kHz 
Pulse energy Ep 25.7 μJ at 50 kHz 341 μJ at 100 kHz 
18 μJ at 2.75 MHz 3.7 μJ at 15 MHz 
Spot diameter df (calculated) ≈33 μ≈26 μ
Beam quality factor M2 1.2 1.2 
Focal length 163.3 mm 160 mm 
Laser system12
Producer IPG Novanta 
Product name YLPP-25-3-50-R Dart picosecond Laser 
Operation mode Pulsed Pulsed 
Wavelength λ 1030 nm 1064 nm 
Max. laser power P 50 W at 2.75 MHz 56 W at 15 MHz 
Pulse duration τ 2 ps 8 ps 
Pulse repetition rate (amplifier) fr 50–2750 kHz 100–15 000 kHz 
Pulse energy Ep 25.7 μJ at 50 kHz 341 μJ at 100 kHz 
18 μJ at 2.75 MHz 3.7 μJ at 15 MHz 
Spot diameter df (calculated) ≈33 μ≈26 μ
Beam quality factor M2 1.2 1.2 
Focal length 163.3 mm 160 mm 

High-resolution surface images and element maps were obtained using an SEM and integrated energy-dispersive x-ray spectroscopy (EDX) (JSM-IT200, Jeol Ltd., Japan). The topography of the micro-drillings was analyzed with a 3D laser scanning confocal microscope (LSM) (VK-X 1000, Keyence Corporation, Japan). All LSM measurements were performed at a 480-fold magnification. The lowest point of the micro-drillings compared to the average MPL surface was measured to determine the ablation depth D.

The pulse peak fluence Φ0 is defined as follows:26 
Φ 0 = 2 E p r 0 2 π ,
(1)
where Ep represents the pulse energy and r0 represents the radius of the laser beam at the focus spot.
In order to calculate the depth ablation efficiencies η D, the following formula was used:27 
η D = D N pulses 1 E p ,
(2)
where D is the ablated depth and Npulses is the number of pulses applied per drilling.

In order to systematically analyze the influence of the material composition and process parameters on the hole geometry, for each varied process parameter, a 3-by-3 matrix of micro-drillings was created on the respective MPL with a distance of 200 μm between each hole. For the nine micro-drillings, the mean value was calculated.

With the aim to characterize how the material composition and the porosity, as well as shielding gas, influence the ablation behavior, the MPLs presented in Table II and Fig. 2 were laser structured. For the structuring process, laser system 1 was used and 100 pulses with a peak fluence of 3.59 J/cm2 at a pulse repetition rate of 50 kHz were applied for one drilling. The number of pulses, the pulse energy, and the resulting pulse peak fluence were chosen, as previous studies showed a complete removal of the MPL layer at this parameter set.20 The pulse repetition rate was set to 50 kHz. Effects of pulse-to-pulse interaction, such as heat accumulation or shielding by ablation products, had to be minimized during these experiments. Furthermore, laser system 2 was used to show the influence of using a slightly different laser beam source as well as optical setup on the ablation behavior and to investigate the cross-system validity of the results obtained within the first experiments. In empirical studies, the influence of pulse peak fluences in the range of 0.6–6 J/cm2 and of different pulse repetition rates, between 50 and 1883 kHz, on the ablation depth and width was shown. In order to further investigate the effect of the pulse repetition rate on the ablation behavior in experiments, a constant pulse energy (EP = 25.70 μJ) and a constant number of pulses (Npulses = 37) were applied to one micro-drilling while varying the pulse repetition rate. A more detailed listing of the experimental designs with the fixed and the variable factors can be found in Table IV in the  Appendix.

TABLE IV.

Overview of the experimental designs constructed in order to achieve the results.

Experimental design 1 (one factor at a time)
FactorsBaseline setLevel 1Unit
Material porosity 80 66 
Material binder content 20 40 wt. % 
Atmosphere Argon Air — 
Pulse duration/pulse peak fluence 2/3.60 8/5.80 ps/J/cm2 
Constant parameters 
Pulse energy 15.42  μ
Pulse repetition rate 50  kHz 
Number of pulses 100  — 
Measured variables of the micro-drilling 
Ablation depth Ablation width Volume of the edge elevation surrounding the drilling Binder distribution surrounding the drilling 
Experimental design 1 (one factor at a time)
FactorsBaseline setLevel 1Unit
Material porosity 80 66 
Material binder content 20 40 wt. % 
Atmosphere Argon Air — 
Pulse duration/pulse peak fluence 2/3.60 8/5.80 ps/J/cm2 
Constant parameters 
Pulse energy 15.42  μ
Pulse repetition rate 50  kHz 
Number of pulses 100  — 
Measured variables of the micro-drilling 
Ablation depth Ablation width Volume of the edge elevation surrounding the drilling Binder distribution surrounding the drilling 
Experimental design 2 (full factorial)
FactorsLevel 1Level 2Level 3Level 4Level 5Level 6Level 7Level 8Level 9Level 10Unit
Pulse energy/pulse peak fluence 2.57/0.60 5.14/1.20 7.71/1.80 10.28/2.40 12.85/3.00 15.42/3.60 17.99/4.21 20.56/4.81 23.13/5.41 25.70/6.01 μJ/J/cm2 
Pulse repetition rate/number of pulses 50/10 100/20 201/40 501/100 1003/200 1838/368 — — — — kHz/— 
Constant parameters 
Drilling time 200          μ
Pulse duration          ps 
Atmosphere Argon          — 
Binder content 20          wt. % 
Porosity 80          
Measured variables of the micro-drilling 
Ablation depth           Ablation width 
Experimental design 2 (full factorial)
FactorsLevel 1Level 2Level 3Level 4Level 5Level 6Level 7Level 8Level 9Level 10Unit
Pulse energy/pulse peak fluence 2.57/0.60 5.14/1.20 7.71/1.80 10.28/2.40 12.85/3.00 15.42/3.60 17.99/4.21 20.56/4.81 23.13/5.41 25.70/6.01 μJ/J/cm2 
Pulse repetition rate/number of pulses 50/10 100/20 201/40 501/100 1003/200 1838/368 — — — — kHz/— 
Constant parameters 
Drilling time 200          μ
Pulse duration          ps 
Atmosphere Argon          — 
Binder content 20          wt. % 
Porosity 80          
Measured variables of the micro-drilling 
Ablation depth           Ablation width 
Experimental design 3 (full factorial)
FactorsLevel 1Level 2Level 3Level 4Level 5Level 6Unit
Pulse repetition rate 50 100 201 501 1003 1838 kHz 
Constant parameters 
Pulse energy 25.70      μ
Pulse peak fluence 6.01      J/cm2 
Number of pulses 37      — 
Pulse duration      Ps 
Atmosphere Argon      — 
Binder content 20      wt. % 
Porosity 80      
Measured variables of the micro-drilling 
Ablation depth     Ablation width   
Experimental design 3 (full factorial)
FactorsLevel 1Level 2Level 3Level 4Level 5Level 6Unit
Pulse repetition rate 50 100 201 501 1003 1838 kHz 
Constant parameters 
Pulse energy 25.70      μ
Pulse peak fluence 6.01      J/cm2 
Number of pulses 37      — 
Pulse duration      Ps 
Atmosphere Argon      — 
Binder content 20      wt. % 
Porosity 80      
Measured variables of the micro-drilling 
Ablation depth     Ablation width   

In Fig. 4, the depth and width values of the micro-drillings for the six MPLs are shown. The mean of the ablation depth and width was similar for the MPLs produced on a lab scale and only a few variations between these characteristics were found. The standard deviation for both the MPL with a lower porosity (Li400) and the MPL with higher PTFE content (Li100-40) increased, indicating a more unstable ablation process. The depth and width values of the Freudenberg H14C10 and the SGL BC25 were located close to the lab scale produced MPLs. In contrast, the micro-drillings introduced to the Freudenberg H14CX653 had a significantly smaller ablation depth and width. As it is obvious from Fig. 2, the particle size and geometry differ from the other MPLs. Based on the SEM image, it can be assumed that the H14CX653 material consists of graphite particles instead of carbon black particles. A different MPL material would be an explanation for a different drilling geometry. Additionally, the presented data indicate that a varying PTFE content between 20% and 40% and the porosity change between 79% and 68% does not cause a significant change in the way the material is removed. The high standard deviations are explainable by the presumably occurring ablation behavior. Due to the material composition of the MPL, an ablation behavior similar to the laser structuring of graphite anodes for lithium-ion batteries is assumed. Lithium-ion battery anodes mainly consist of graphite particles that are held together by a polymeric binder (often polyvinylidene fluoride, PVDF). Schmieder investigated the material removal using high-speed video recording.28 As the dominant ablation process, the ejection of complete graphite particles instead of the graphite evaporation was detected. This behavior/ablation mechanism was explained by the significantly smaller evaporation temperature (Tevap = 660 °C) of the polymeric binder (PVDF) surrounding the particles compared to the evaporation temperature of graphite particles (Tevap = 4098 °C). The considerably lower temperature value of PVDF results in vapor generating a force that ejects the graphite particle out of its compound.28 It can be assumed that a similar ablation behavior occurs during MPL structuring as the evaporation temperatures of the main compound are in the same order of magnitude (see Table II). Further comparing the laser structuring of battery anodes to the MPL structuring, a notable influence of the binder content on the ablation behavior, which was shown by Habedank et al.,29 is not visible within the experiments. Within the study of Habedank et al., deeper micro-drillings at a smaller binder content (2 wt. %) were explained by a reduced cohesion between the graphite particles facilitating the ablation process in contrast to electrodes with higher binder contents.29 In comparison to the 2–8 wt. % binder content within the battery anodes, the binder contents of the MPL were between 20 and 40 wt. %. Therefore, a significant change in the cohesion between the carbon black particles was not expected.

FIG. 4.

Ablation depth and width with corresponding standard deviations of laser-structured MPLs with a varying binder content and porosity; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ and Φ0 = 3.60 J/cm2. The bars in the figures indicate the standard deviation.

FIG. 4.

Ablation depth and width with corresponding standard deviations of laser-structured MPLs with a varying binder content and porosity; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ and Φ0 = 3.60 J/cm2. The bars in the figures indicate the standard deviation.

Close modal

The LSM pictures of the laser-structured MPL surfaces (Li100-40, Li100-20, and Li400) are shown in Fig. 5. Compared to the depth and width values, where no real difference between the three MPLs was observed, the LSM pictures showed differences in the borehole opening. While at the surfaces of the Li100-40 [compare Fig. 5(a)] and Li400 [compare Fig. 5(c)], no particles and also almost no edge elevation surrounding the holes were detected, both phenomena occurred at the Li100-20 surface [see Fig. 5(b)]. In order to quantify the difference, the edge elevation volume surrounding the drilling at a radius of 50 μm was measured and plotted [according to Fig. 5(d)]. More than twice the volume of accumulated material could be measured in the observed area.

FIG. 5.

Influence of the binder content on the ablation behavior: (a) Li100-40, (b) Li100-20, (c) Li400 structured with the laser system 1 in an argon atmosphere; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ and Φ0 = 3.60 J/cm2; (d) comparison of the edge elevation volume surrounding the micro-drilling of the three MPL.

FIG. 5.

Influence of the binder content on the ablation behavior: (a) Li100-40, (b) Li100-20, (c) Li400 structured with the laser system 1 in an argon atmosphere; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ and Φ0 = 3.60 J/cm2; (d) comparison of the edge elevation volume surrounding the micro-drilling of the three MPL.

Close modal

In literature, the laser structuring of the DM in a shielding gas atmosphere showed beneficial results as the heat-affected zone was smaller with reduced PTFE evaporation in the area surrounding the micro-drilling.11 

In order to evaluate the effect of the atmosphere on the ablation behavior using ultrashort-pulsed laser radiation, an analysis of the micro-drillings created within the argon atmosphere was compared to micro-drillings produced in the ambient air atmosphere. Figure 6 shows the values of the depth and width measurements from the Li100-20 DM structured using laser system 1 and 100 laser pulses with a peak fluence of 3.59 J/cm2 at a repetition rate of 50 kHz. The measured depth and width were almost the same, indicating a negligible influence of the shielding gas on the ablation behavior.

FIG. 6.

Ablation depth and width with corresponding standard deviations of MPLs, laser structured in shielding gas atmosphere and ambient air atmosphere; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ and Φ0 = 3.60 J/cm2. The bars in the figures indicate the standard deviation.

FIG. 6.

Ablation depth and width with corresponding standard deviations of MPLs, laser structured in shielding gas atmosphere and ambient air atmosphere; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ and Φ0 = 3.60 J/cm2. The bars in the figures indicate the standard deviation.

Close modal

Besides the measured values, the LSM pictures of the micro-drillings did not show any difference in the micro-drillings’ geometry on the surface area [according to Figs. 7(a) and 7(b)]. In order to examine the heat-affected zones, EDX analyses were carried out, showing the distribution of the fluorine binder at the surface of a laser-structured MPL in an argon atmosphere [compare Fig. 7(c)] and air atmosphere [see Fig. 7(d)]. The binder distribution in the vicinity of the holes was nearly similar and the binder-free areas had almost the same size. A negligible influence of the atmosphere is explained by the pulse duration used for laser structuring. As it was already described by Geiger et al., the introduced heat and the resulting heat-affected zone were decreased by using laser systems emitting pulses in the low picosecond range.20 

FIG. 7.

Comparison of the micro-drillings produced with the laser system 1 and the Li100-20 material; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ and Φ0 = 3.60 J/cm2: (a) argon environment, (b) air environment, (c) fluorine distribution of the MPL surface shown in subfigure (a), and (d) fluorine distribution of the MPL surface shown in subfigure (b).

FIG. 7.

Comparison of the micro-drillings produced with the laser system 1 and the Li100-20 material; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ and Φ0 = 3.60 J/cm2: (a) argon environment, (b) air environment, (c) fluorine distribution of the MPL surface shown in subfigure (a), and (d) fluorine distribution of the MPL surface shown in subfigure (b).

Close modal

From a series production point of view, the results shown in this section are beneficial. Ensuring a shielding gas atmosphere in a production environment is expensive due to the high costs of investing in infrastructure (e.g., gloveboxes or microenvironments) and operating in an inert gas atmosphere.30 It is therefore avoided by companies as far as possible.

The laser system 1 (τ = 2 ps) was compared to the laser system 2 (τ = 8 ps). Even though the laser system 2 had a smaller focal diameter due to the different optical setups, smaller drillings were achieved by applying laser system 1. In contrast, deeper holes were created with laser system 2. From Figs. 8 and 9, the differences in the borehole geometry are observable.

FIG. 8.

Ablation depth and width with corresponding standard deviations of MPLs, laser structured with laser system 1 and laser system 2; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ, Φ0 = 3.60 J/cm2 (laser system 1) and Φ0 = 5.80 J/cm2. The bars in the figures indicate the standard deviation.

FIG. 8.

Ablation depth and width with corresponding standard deviations of MPLs, laser structured with laser system 1 and laser system 2; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ, Φ0 = 3.60 J/cm2 (laser system 1) and Φ0 = 5.80 J/cm2. The bars in the figures indicate the standard deviation.

Close modal
FIG. 9.

Li100-20 material with laser structured micro-drillings; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ; (a) laser system 1 with Φ0 = 3.60 J/cm2; (b) laser system 2 with Φ0 = 5.80 J/cm2; (c) comparison of the edge elevation volume surrounding the micro-drilling of the MPLs shown in subfigures (a) and (b).

FIG. 9.

Li100-20 material with laser structured micro-drillings; a constant number of 100 pulses per drilling was applied at fr = 50 kHz with EP = 15.42 μJ; (a) laser system 1 with Φ0 = 3.60 J/cm2; (b) laser system 2 with Φ0 = 5.80 J/cm2; (c) comparison of the edge elevation volume surrounding the micro-drilling of the MPLs shown in subfigures (a) and (b).

Close modal

The micro-drillings created with pulses by laser system 2 are wider and a pronounced edge elevation surrounds the drillings, which was again measured from the volume of the accumulated material [see Fig. 9(c)]. Due to the different optical setups, the pulse peak fluence differed, which is assumed to be the reason for the increased ablation depth reached with laser system 2. At the same time as Φ0 was increased by a factor of 1.56, the ablation width increased by a factor of 1.61 if the drillings produced with the different laser systems were compared. This leads to the assumption that there is a correlation between the drilling width and the pulse peak fluence, which was further investigated and presented, in addition to the influence of the PRR, in Sec. IV D.

In order to pave the way toward an industrial usage, a high-speed laser structuring process is needed. For an increase in the processing speed, laser beam sources with significantly higher output power and high pulse repetition rates, using either beam splitting (with, e.g., diffractive optical elements) or high-speed scanners (e.g., polygon scanners), are the most promising path. It is, therefore, necessary to better understand the influence of the laser process parameters such as the pulse peak fluence Φ0 and the pulse repetition rate fr on the drilling geometry. For Fig. 9(a), the influence of Φ0 and fr on the ablation depth was determined. At an fr of 1883 kHz, the MPL was ablated already at a Φ0 of 0.6 J/cm2. In order to also ablate the carbon fibers of the GDL, a Φ0 > 2.4 J/cm2 was necessary. The effect of a change within the ablation behavior at the material junction was also seen at the other PRR. For instance, at an fr of 1003 kHz, the intersection was reached at a Φ0 of 2.4 J/cm2 and ablation of the GDL was observed at Φ0 > 4.8 J/cm2. The analysis of the ablation width [compare Fig. 10(b)] showed a steady increase with the pulse peak fluence. The junction between the MPL and GDL was not identifiable in the data as the width also increased for Φ0, where the depth stayed almost constant. A possible explanation for these observations is that the introduced pulses were heating the bottom and the walls of the micro-drilling, leading to further melting and evaporating of the polymer. In Fig. 11, the LSM pictures of micro-drillings introduced at an fr of 501 kHz and different Φ0 are shown. With increasing Φ0, the width increased while the depth stayed nearly the same. Additionally, the solidified melt was observed, especially at higher Φ0 at the edge of the micro-drillings.

FIG. 10.

(a) Ablation depth, (b) ablation width, and (c) aspect ratio as functions of the pulse peak fluence at different pulse repetition rates with the Li-100 MPL and the laser beam source 1 with a drilling time of 200 μs per hole.

FIG. 10.

(a) Ablation depth, (b) ablation width, and (c) aspect ratio as functions of the pulse peak fluence at different pulse repetition rates with the Li-100 MPL and the laser beam source 1 with a drilling time of 200 μs per hole.

Close modal
FIG. 11.

Comparison of the micro-drillings introduced with laser system 1, an fr of 501 kHz and a drilling time of 200 μs: (a) Φ0 =2.4 J/cm2, (b) Φ0 =3.0 J/cm2, (c) Φ0 = 3.6 J/cm2, (d) Φ0 = 4.2 J/cm2, (e) Φ0 = 4.8 J/cm2, and (f) Φ0 = 5.4 J/cm2.

FIG. 11.

Comparison of the micro-drillings introduced with laser system 1, an fr of 501 kHz and a drilling time of 200 μs: (a) Φ0 =2.4 J/cm2, (b) Φ0 =3.0 J/cm2, (c) Φ0 = 3.6 J/cm2, (d) Φ0 = 4.2 J/cm2, (e) Φ0 = 4.8 J/cm2, and (f) Φ0 = 5.4 J/cm2.

Close modal

So far, Haußmann et al. and Wang et al. found that micro-drillings with a diameter between 60 and 100 μm were most suitable for the increase in the limiting current density.31,32 Based on these findings, the results of the empirical study show that it is possible to use high PRRs (e.g., 1.8 MHz) at a pulse peak fluence between 1.8 and 4.2 J/cm2 to introduce the micro-drillings into the MPL.

In order to further investigate the influence of fr. on the micro-drilling’s geometries, studies with a constant pulse energy (Ep = 25.70 μJ) and a constant number of pulses (Npulses = 37) applied to the drilling were performed. The results are shown in Fig. 12. For fr > 500 kHz, the influence of fr on the ablation depth and ablation width was significantly higher, resulting in an asymptotic approach to a limit value. This phenomenon is, with high probability, caused by shielding effects, which were also observed within other studies.27,29,33 The shielding effects arise when pulses are attenuated by the particles and the vapor generated by the previous pulses before reaching the surface. The particles and the vapor are partly absorbing the energy of the pulses, resulting in a reduced pulse energy available for the ablation process.

FIG. 12.

Ablation depth and ablation width at different pulse repetition rates; a constant number of 37 pulses per drilling was applied resulting in an energy per hole of 0.95 mJ with laser beam source 1. The bars in the figures indicate the standard deviation.

FIG. 12.

Ablation depth and ablation width at different pulse repetition rates; a constant number of 37 pulses per drilling was applied resulting in an energy per hole of 0.95 mJ with laser beam source 1. The bars in the figures indicate the standard deviation.

Close modal

In Fig. 13, the depth ablation efficiency for different fr is shown. For all fr, a maximum efficiency was reached at 0.6 J/cm2, where the efficiency is almost twice as high compared to the other pulse peak fluences. As it was shown in previous studies, plasma shielding effects occur with increasing pulse peak fluences and pulse repetition rates due to the growth of the size and the amount of generated plasma, particles, and plume.34 

FIG. 13.

Depth ablation efficiency as a function of the pulse peak fluence at different pulse repetition rates with the Li-100 MPL and the laser beam source 1.

FIG. 13.

Depth ablation efficiency as a function of the pulse peak fluence at different pulse repetition rates with the Li-100 MPL and the laser beam source 1.

Close modal

Within this study, the laser structuring of different MPLs for PEMFC was investigated. A slight influence of the binder and the porosity on the ablation characteristic could be seen. Additionally, no significant impact of shielding gas could be detected. In contrast, the used materials, notably the carbon black or graphite particles on the material side, the process parameters, and the laser beam source have a higher influence on the resulting drilling geometry. Furthermore, the effect of the material change at the MPL-GDL junction was clearly visible in the investigated parameter space. At a pulse peak fluence of 0.6 J/cm2, the highest depth ablation efficiency was reached for all fr. The influence of the pulse repetition rate was shown with a decrease in the ablation efficiency at increasing fr, which indicates that pulse-to-pulse interactions have to be considered. For an efficient process, a small fr and a small Φ0 are beneficial. On the contrary, for the scaling of the process, increasing fr to the MHz region can be used to accelerate the processing speed, as more material is removed at the same time than for smaller values of fr.

In further studies, the structuring of the GDL will be investigated in more detail. As it has been shown in this publication, the material change has a significant influence on the structuring process, and a change of the process parameters at the junction could be beneficial. Therefore, an inline determination of the material change, or even an inline depth measurement, will be advantageous for a process improvement. Additionally, the influence of the micro-drillings on the properties of the DM (good gas and water permeability, high electrical as well as thermal conductivity, and mechanical along with chemical resistance) shall be determined.

The authors gratefully acknowledge the diffusion media materials provided by the Chair of Technical Electrochemistry of the Technical University of Munich. Additionally, the authors thank the Novanta Europe GmbH for providing the Dart Picosecond laser system as well as the IPG Photonics Corporation for the YLPP-25-3-50-R system.

The authors have no conflicts to disclose.

Christian Geiger: Conceptualization (lead); Data curation (equal); Methodology (lead); Visualization (equal); Writing – original draft (lead). Sophie Grabmann: Conceptualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Tony Weiss: Methodology (equal); Visualization (supporting); Writing – original draft (supporting). Alena Gruendl: Data curation (supporting); Investigation (supporting); Writing – original draft (equal). Michael Zaeh: Funding acquisition (equal); Project administration (equal); Supervision (equal); Writing – review & editing (equal).

In Table IV, the experimental designs with the laser source and material parameters which were held constant and which were varied within the empirical studies can be seen.

1.
Y.
Wang
,
D. F.
Ruiz Diaz
,
K. S.
Chen
,
Z.
Wang
, and
X. C.
Adroher
, “
Materials, technological status, and fundamentals of PEM fuel cells—A review
,”
Mater. Today
32
,
178
–203 (
2020
).
2.
H.
Li
,
Y.
Tang
,
Z.
Wang
,
Z.
Shi
,
S.
Wu
,
D.
Song
,
J.
Zhang
,
K.
Fatih
,
J.
Zhang
,
H.
Wang
,
Z.
Liu
,
R.
Abouatallah
, and
A.
Mazza
, “
A review of water flooding issues in the proton exchange membrane fuel cell
,”
J. Power Sources
178
,
103
–117 (
2008
).
3.
Z.
Lu
,
J.
Waldecker
,
M.
Tam
, and
M.
Cimenti
, “
Influence of MPL structure modification on fuel cell oxygen transport resistance
,”
ECS Trans.
69
,
1341
–1353 (
2015
).
4.
Y.
Yang
,
X.
Zhou
,
B.
Li
, and
C.
Zhang
, “
Recent progress of the gas diffusion layer in proton exchange membrane fuel cells: Material and structure designs of microporous layer
,”
Int. J. Hydrogen Energy
46
,
4259
–4282 (
2021
).
5.
J.
Zhang
,
H.
Zhang
,
J.
Wu
, and
J.
Zhang
,
PEM Fuel Cell Testing and Diagnosis
(
Elsevier
,
New York
,
2013
), p.
44
.
6.
M. F.
Mathias
,
J.
Roth
,
J.
Fleming
, and
W.
Lehnert
, “
Diffusion media materials and characterisation
,” in
Fundamentals, Technology and Applications
(
Wiley Interscience
,
Hoboken, NJ
,
2010
).
7.
R. R.
Rashapov
,
J.
Unno
, and
J. T.
Gostick
, “
Characterization of PEMFC gas diffusion layer porosity
,”
J. Electrochem. Soc.
162
,
F603
F612
(
2015
).
8.
P.
Antonacci
,
S.
Chevalier
,
J.
Lee
,
N.
Ge
,
J.
Hinebaugh
,
R.
Yip
,
Y.
Tabuchi
,
T.
Kotaka
, and
A.
Bazylak
, “
Balancing mass transport resistance and membrane resistance when tailoring microporous layer thickness for polymer electrolyte membrane fuel cells operating at high current densities
,”
Electrochim. Acta
188
,
888
–897 (
2016
).
9.
C.
Simon
,
J.
Endres
,
B.
Nefzger-Loders
,
F.
Wilhelm
, and
H. A.
Gasteiger
, “
Interaction of pore size and hydrophobicity/hydrophilicity for improved oxygen and water transport through microporous layers
,”
J. Electrochem. Soc.
166
,
F1022
F1035
(
2019
).
10.
J. M.
Morgan
and
R.
Datta
, “
Understanding the gas diffusion layer in proton exchange membrane fuel cells. I. How its structural characteristics affect diffusion and performance
,”
J. Power Sources
251
,
269
–278 (
2014
).
11.
R.
Alink
,
J.
Haußmann
,
H.
Markötter
,
M.
Schwager
,
I.
Manke
, and
D.
Gerteisen
, “
The influence of porous transport layer modifications on the water management in polymer electrolyte membrane fuel cells
,”
J. Power Sources
233
,
358
–368 (
2013
).
12.
Y.
Hou
,
X.
Li
,
Q.
Du
,
K.
Jiao
, and
N.
Zamel
, “
Pore-scale investigation of the effect of micro-porous layer on water transport in proton exchange membrane fuel cell
,”
J. Electrochem. Soc.
167
,
144504
(
2020
).
13.
Y.-C.
Chen
,
C.
Karageorgiou
,
J.
Eller
,
T. J.
Schmidt
, and
F. N.
Büchi
, “
Determination of the porosity and its heterogeneity of fuel cell microporous layers by X-ray tomographic microscopy
,”
J. Power Sources
539
,
231612
(
2022
).
14.
S.
Park
,
J.-W.
Lee
, and
B. N.
Popov
, “
Effect of PTFE content in microporous layer on water management in PEM fuel cells
,”
J. Power Sources
177
,
457
–463 (
2008
).
15.
G.
Athanasaki
,
A.
Jayakumar
, and
A. M.
Kannan
, “
Gas diffusion layers for PEM fuel cells: Materials, properties and manufacturing—A review
,”
Int. J. Hydrogen Energy
48
,
2294
–2313 (
2023
).
16.
G.
Velayutham
, “
Effect of micro-layer PTFE on the performance of PEM fuel cell electrodes
,”
Int. J. Hydrogen Energy
36
,
14845
(
2011
).
17.
T.
Sasabe
,
P.
Deevanhxay
,
S.
Tsushima
, and
S.
Hirai
, “
Soft X-ray visualization of the liquid water transport within the cracks of micro porous layer in PEMFC
,”
Electrochem. Commun.
13
,
638
–641 (
2011
).
18.
D.
Gerteisen
,
T.
Heilmann
, and
C.
Ziegler
, “
Enhancing liquid water transport by laser perforation of a GDL in a PEM fuel cell
,”
J. Power Sources
177
,
348
–354 (
2008
).
19.
D.
Gerteisen
and
C.
Sadeler
, “
Stability and performance improvement of a polymer electrolyte membrane fuel cell stack by laser perforation of gas diffusion layers
,”
J. Power Sources
195
,
5252
–5257 (
2010
).
20.
C.
Geiger
,
J.
Kriegler
,
T.
Weiss
,
A.
Berger
, and
M. F.
Zaeh
, “
Micro-perforation of the diffusion media for polymer electrolyte membrane fuel cells using short and ultrashort laser pulses
,”
Procedia CIRP
111
,
796
–799 (
2022
).
21.
C.
Simon
,
D.
Kartouzian
,
D.
Müller
,
F.
Wilhelm
, and
H. A.
Gasteiger
, “
Impact of microporous layer pore properties on liquid water transport in PEM fuel cells: carbon black type and perforation
,”
J. Electrochem. Soc.
164
,
F1697
F1711
(
2017
).
22.
Freudenberg Performance Materials SE & Co, KG, see https://fuelcellcomponents.freudenberg-pm.com/-/media/Files/fuelcellcomponents,-d-,freudenbergpm,-d-,com/FPM_technical-data-sheet_electrolysis_ENG_2022-02-11.pdf for “FREUDENBERG GAS DIFFUSION LAYERS – FUEL CELL. Technical Data” (accessed 14 February 2023).
23.
R.
Schweiss
,
S.
Hofmeister
,
C.
Meiser
,
D.
Dan
,
A.
Baumann
,
T.
Kuster
,
N.
Haak
, and
S.
Bacher
, see https://www.sglcarbon.com/pdf/SIGRACET-Whitepaper.pdf for “Powering up fuel cells” (accessed 14 February 2023).
24.
ThermoFisher Scientific, see https://www.alfa.com/en/msds/?language=EE&subformat=CLP1&sku=39724 for “Safety Data Sheet. Carbon Black, Acetylene, 50% Compressed” (accessed 23 February 2023).
25.
Bohlender GmbH, see https://www.bola.de/msdb/Englisch/MSDS_PTFE_e.pdf for “Safety Data Sheet PTFE” (accessed 23 February 2023).
26.
B.
Neuenschwander
,
B.
Jaeggi
,
M.
Schmid
,
V.
Rouffiange
,
P.-E.
Martin
, and
G.
Hennig
, “
Optimization of the volume ablation rate for metals at different laser pulse-durations from ps to fs
,”
Proc. SPIE
8243
,
824307
(
2012
).
27.
J.
Kriegler
,
E.
Jaimez-Farnham
,
L.
Hille
,
E.
Dashjav
, and
M. F.
Zaeh
, “
Pulsed laser ablation of a ceramic electrolyte for all-solid-state batteries
,”
Procedia CIRP
111
,
800
–805 (
2022
).
28.
B.
Schmieder
, “
Analytical model of the laser ablation mechanism of lithium-ion battery coatings
,”
Proc. SPIE
9351
,
93511C
(
2015
).
29.
J. B.
Habedank
,
J.
Endres
,
P.
Schmitz
,
M. F.
Zaeh
, and
H. P.
Huber
, “
Femtosecond laser structuring of graphite anodes for improved lithium-ion batteries: Ablation characteristics and process design
,”
J. Laser Appl.
30
,
032205
(
2018
).
30.
F.
Duffner
,
N.
Kronemeyer
,
J.
Tübke
,
J.
Leker
,
M.
Winter
, and
R.
Schmuch
, “
Post-lithium-ion battery cell production and its compatibility with lithium-ion cell production infrastructure
,”
Nat. Energy
6
,
123
–134 (
2021
).
31.
J.
Haußmann
,
H.
Markötter
,
R.
Alink
,
A.
Bauder
,
K.
Dittmann
,
I.
Manke
, and
J.
Scholta
, “
Synchrotron radiography and tomography of water transport in perforated gas diffusion media
,”
J. Power Sources
239
,
611
–622 (
2013
).
32.
X.
Wang
,
S.
Chen
,
Z.
Fan
,
W.
Li
,
S.
Wang
,
X.
Li
,
Y.
Zhao
,
T.
Zhu
, and
X.
Xie
, “
Laser-perforated gas diffusion layer for promoting liquid water transport in a proton exchange membrane fuel cell
,”
Int. J. Hydrogen Energy
42
,
29995
(
2017
).
33.
J.
König
,
S.
Nolte
, and
A.
Tünnermann
, “
Plasma evolution during metal ablation with ultrashort laser pulses
,”
Opt. Express
13
,
10597
(
2005
).
34.
L.
Hribar
,
P.
Gregorčič
,
M.
Senegačnik
, and
M.
Jezeršek
, “
The influence of the processing parameters on the laser-ablation of stainless steel and brass during the engraving by nanosecond fiber laser
,”
Nanomaterials
12
,
232
–241 (
2022
).