|
Notas
| Description of the project: Titanium is widely used as a material for bipolar plates (BP) PEM fuel cells, but they require conductive coatings that help with corrosion and have low interfacial contact resistance (ICR). Among many materials, polymer composites reinforced with graphene nanoparticles have the potential to become an effective coating, but the influence of different types of graphene-based materials on the performance remains has not been properly established. In this study, different nanoparticles (graphene monolayers, reduced graphene oxide and few-layer graphene) were used to process PTFE/graphene composite coatings for Ti, and their performance was evaluated. Most of the coating formulations improved the corrosion resistance of the Ti in one or two orders of magnitude, but only the combination of few-layer graphene and carbon black as additive resulted in an ICR of 12 mΩcm2, in range with the target values.
Methodology: For a complete description of the methodology, including raw materials, their characteristics and the specific equipment, refer to the related publication.
Processing of the materials:
Ti substrates were prepared via powder metallurgy from titanium hydride powders (GfE Metalle und Materialien GmbH). These powders were shaped into 16 mm disks using uniaxial pressing at 200 MPa and then sintered in Ar, with a sintering temperature of 1000 °C and a holding time of 2 h. To ensure the hydride decomposition, the heating ramp was set to have three regions with different rates: between RT and 450 °C, the heating rate was 5 °C/min; between 450 and 650 °C/min, 2 °C/min; and between 650 and 1000 °C, 5 °C/min. The cooling rate was 5 °C/min until RT.
Several formulations for the coatings were implemented, with different graphene-based materials used and different additions of carbon black. The polymer matrix for the coating was a commercial PTFE resin for one-coat application, with the use of a commercial organic solvent (as the medium to produce the graphene-based suspensions and as a thinner, with a dispersant agent. Three graphene-based materials were used: graphene nanoplatelets (GN), rGO produced by the Hummer’s method, and few layer graphene (FLG). Additionally, conductive carbon black (CB) was added to some formulations to improve the conductivity of the coatings.
The coating mixes were prepared following this procedure: first, a stock solution is prepared with the solvent, the graphene-based nanoparticles and the dispersant. This stock solution is then exfoliated in a laboratory mixer . Then, the PTFE resin and the solution are mixed in the appropriate ratio, obtaining the coating mix. Depending on the composition, CB is also added in this step. For the formulation with FLG and CB, thinner was added to obtain a suspension with the right rheology for processing. Resulting in the following formulations: PTFE+GN; PTFE+rGO+CB; PTFE+FLG; PTFE+FLG+CB
Before coating, the titanium substrates were ground with SiC paper of grit size #1000 to remove the oxide layer and subsequently cleaned in an ultrasonic bath with acetone for 10 min. The substrates were immersed in the mix for 3 minutes before removal at a controlled rate in range between 5 and 200 mm/min. Then, the coating was placed in a convection oven at 230 °C for 10 min to dry the solvent and cure the resin, obtaining a dense coating.
Characterization:
Raman spectra:
The Raman spectra were acquired using an AFM-Raman system (NT-MDT) with 532 nm wavelength laser. The data is stored in a dat file in tabulated columns, with the Raman shift in the first column and the intensity of the measurement in the second column.
Water contact angle:
The hydrophilic/hydrophobic behaviour of the coatings were measured using water contact angle (WCA) measurements, using the sessile drop method with a droplet of 15 µL. Several images were taken to perform the calculations.
Corrosion testing:
The corrosion behaviour of the coatings was characterized by combining Electrochemical Impedance Spectroscopy (EIS), linear polarization scans and chronoamperometric tests. All the tests were performed in a three electrode cell, with a Pt wire as counter electrode and a Ag/AgCl reference electrode. The tests were performed in 0.5 M H2SO4 at room temperature. The surface area in contact with the electrolyte was 1 cm2, which is required to calculate corrosion current densities.
A preliminary test, simulating the anode environment, was perfomed capturing the open circuit potential (OCP) for 1 hour before obtaining an EIS spectra, then a polarization by linear sweep voltammetry (LSV) was performed between -1 V and +1 V vs OCP. The results of these tests are identified as [material]_corrosion-anode-EIS.dat and [material]_corrosion-anode-LSV.dat. A test simulating the cathode environment was performed in a 0.5 M H2SO4 electrolyte at room temperature with air bubbling, using a chronoamperometric (CA) test at 0.6 V vs Ag/AgCl, which was preceded by an EIS measurement. After 12 h of CA testing, the OCP was measured and another EIS measurement was performed. Then, 0.6 V vs Ag/AgCl was applied again for another 12 h. Afterwards, the OCP and EIS were measured. The results of these tests are presented in files with the naming [material]-corrosion-cathode-[measurement]-[time].dat, where the measurement can either be CA or EIS, and the time represents the interval measured in CA or the time during the test where EIS was performed. As an example, [material]-corrosion-cathode-CA-0h-12h.dat stores the CA results performed from 0h to 12h, while [material]-corrosion-cathode-EIS-12h.dat stores the EIS data performed after 12h of CA. In all cases, EIS was acquired between 100 kHz and 10 mHz, applying a 10 mV amplitude.
EIS files: the data stored is in tabulated columns representing the following magnitudes:
freq/Hz
Re(Z)/Ohm
Im(Z)/Ohm
|Z|/Ohm Phase(Z)/deg
freq is the frequency evaluated in Hz. Z is the electrochemical impedance, measured in Ohms. Re(Z) is the real component of the impedance, Im(Z) is the imaginary component, |Z| is the modulus and Phase(Z) is the phase of the impedance in degrees.
LSV files: the data is stored in tabulated columns with two magnitudes: Ewe/V, which is the potential applied in volts, and < I >/mA, which is the corrosion current measured in mA.
CA files: the data from these tests is stored as four magnitudes in tabulated columns: time/s, which is the time since the beginning of the whole cathode simulation in seconds, control/V, which is the potential applied as control in V, which is similar to the third magnitude, Ewe/V, which is the potential applied to the working electrode, and I/mA, which is the corrosion current measured in mA.
ICR:
ICR was measured with a micro-mechanical testing device, Au-coated copper plates, a DC power source, and a precision multimeter, using the four point probe method to measure voltage and current intensity. The procedure was the following: a GDL (or carbon paper) was cut in a circle, whose area determines de contact area used in the ICR calculation. Then, the GDL was placed in between the copper plates, that are connected to the DC source, and as the force was increased stepwise, the voltage was recorded. In this setup, the current intensity is fixed. For each value of force, a voltage V_GDL is recorded. Afterwards, the specimen to measure was sandwiched between two GDLs with the same area, placed between the copper plates, and force and current are applied simultaneously. The current is fixed, the force increases stepwise to similar values as in the previous case, and for each step the voltage V_assembly is recorded. Finally, the ICR including the coating and the uncoated Ti surface is calculated for each force value using the formula (V2-V1)/I*A, where I is the intensity and A the contact area. The applied pressure is calculated dividing the force applied in each step by the contact area. The ICR of the uncoated Ti is 12 mOhm·cm2, so this quantity is substracted to the ICR_rec+Ti to obtain the ICR of the coating (ICR_coat). The other important magnitude, included in the file, is the compaction pressure, which is stored in P(N/cm2). |
