metal layers – Jackfish SEC

What does “Conditioning” mean?

The empirical observation is that prolonged potential cycling of a Au thin film electrode for ATR-SEIRAS results in higher vibrational signals. The exact mechanism of this process is still unclear, but is often referred to as electrochemical annealing, electropolishing, or electrochemical conditioning of the film. Au electrodes are known to undergo surface reconstruction at oxidizing potentials. It is suggested that by repeatedly cycling the potential into Au oxidation, the mobility of surface Au atoms is increased, resulting in the slow growth of Au islands. Since the enhancement effect is known to depend on the morphology of the metal film, the conditioning process is thought to improve the morphology to provide better enhancement. There is likely also a cleaning effect at play – by cycling the potential, contaminants may be forced off the surface.

Conditioning in Practice

A variety of aqueous electrolytes have been used for film conditioning, including 0.1 M sulfuric acid, 0.1 M perchloric acid, 0.1 NaF, and 0.1 M KClO4. Although some authors have reported using prolonged (many hours) cycling in the double layer region in aqueous sulfuric acid, we find that careful excursions into Au oxidation are beneficial.

Disclaimer

Before describing a typical conditioning procedure and best practices from our lab, we will note that Au layers are notoriously unstable and great care should be taken when working with them. Au oxidation and gas evolution both cause large stresses in the Au thin film which can cause it to buckle and delaminate from the internal reflection element. A Au film which fails in this way is not recoverable. Therefore, while following this procedure, we recommend occasional visual checks of the film to verify its integrity. As you grow more comfortable with the process, cues from ATR-SEIRAS spectra collected during the process will help determine whether the layer is still usable.

In the process below, we’ll often suggest that you collect a spectrum. We refer to a “single beam” spectrum; that is, the power spectrum that results from averaging many interferograms and computing the Fourier transform. This should not be confused with an absorbance or transmission curve, which are calculated from the ratio of two such spectra, e.g. A = -log(I/I0). Collecting individual spectra is useful because absorbance or transmission curves can be calculated using spectra collected at the beginning of the process. If your data collection software is not already set up to easily collect single beam spectra and perform calculations with them, we suggest you set it up to do so before attempting the conditioning process for the first time.

With the Au layer at the open circuit potential (OCP), collect a reference spectrum. (You can use this to compare changes in the SEIRAS response as the conditioning process progresses.)

Starting from OCP, cycle the potential in a window ±200 mV around the OCP. Complete three cycles in this window.

Increase the potential bounds by 100 mV, and complete three cycles in this new window.

Repeat step 3, under the following constraints:

Do not enter hydrogen evolution! Set the cathodic potential bound to be just positive of hydrogen evolution.
When the anodic potential bound reaches Au oxidation, proceed to step 5

In small increments (typically 50 mV), increase the anodic potential bound into Au oxidation. Complete three cycles, and, if necessary, increase the upper bound by another 50 mV.

As a rule of thumb, we very rarely push the window past the peak of Au oxidation (roughly 100 – 200 mV beyond the onset), as the stress induced by oxidation can delaminate the layer. The goal is to strike a delicate balance between “just enough” gold oxidation and “too little”. We find this is often dependent on layer prepration method and electrolyte composition and encourage end-users to experiment.
We recommend collecting spectra to compare against the spectrum which was collected in step 1. Overall, a more active Au layer should show increased absorbance in liquid water bands, indicating improved enhancement.

Acetate Adsorption as a Probe System

Completing the conditioning process can be challenging without clear evidence of increasing SEIRAS activity. We’ve found that using 50 – 100 mM aqueous acetate buffer (pH 3.6 – 5.6) is an excellent tool for conditioning Au layers, particularly for novice users. Acetate adsorbs on Au at anodic potentials (around +600 mV vs. Ag/AgCl) but is replaced by at lower potentials (0 mV vs Ag/AgCl), and the asymmetric carboxylate stretch at 1400 cm-1 is a great diagnostic for the changes in the SEIRAS enhancement.

You can find out more about this system in the application note here.

ATR-SEIRAS exploits the near-field interaction between incident infrared radiation and a thin metal film electrode, often a layer of gold. Working in the Kretschmann configuration, the surface of the metal layer can be probed under a variety of electrochemical conditions.

But how are metal layers for ATR-SEIRAS prepared? This post will describe the two main fabrication methods, physical vapour deposition and chemical (electroless) deposition. It will also describe a newly-developed method for preparing mechanically stable hybrid gold films using electrodeposition.

A general observation in ATR-SEIRAS is that metal films have very poor mechanical stability. Because of the poor adhesion of gold on silicon oxide surfaces, film delamination is the dominant failure mode. Metal films prepared by vapour or electroless should be handled very carefully. Prolonged potential excursions into gas evolution (highly anodic or cathodic potentials) will often result in high film stresses which contribute to failure.

Physical Vapour Deposition

Physical Vapour Deposition (PVD) methods generally use magnetron sputtering or thermal evaporation systems. Metal thin films in the 10 to 20 nm thickness range are deposited at low rates (typically around 0.01 nm/s) on the basal plane of an internal reflection element. The low deposition rate and thickness provide a rough surface with an island structure which is known to contribute favourably to signal enhancement.

From a practical standpoint, PVD methods are relatively quick and usually highly repeatable with a well-maintained system. Because the films are prepared in vacuum, it often takes several hours to pump the substrates down to base pressure. Deposition usually takes under an hour. Most PVD chambers can accommodate multiple substrates, allowing several metal layers to be prepared in parallel. This is particularly evident when using IRUBIS GmbH ATR wafers, which are inexpensive, small, and can be used with our J1W and J2W cells.

Although metal deposition with PVD methods is relatively quick and easy, the layers can be less mechanically stable than those prepared with chemical (electroless) methods. In our hands, enhancement factors are often slightly lower as well.

Chemical (Electroless) Deposition

Electroless deposition using gold on a silicon substrate was first described by Osawa in 2002, and involves a series of mechanical and chemical steps. Compared to PVD, it is highly challenging and potentially hazardous, but results in robust films with good enhancement factors. The basal plane of the internal reflection element is first mechanically polished with diamond suspensions (decreasing to sub-micron particle sizes) then treated with HF to remove the native oxide layer and hydrogen terminate the surface. Then, a plating solution containing a gold precursor (often gold tetrachloroaurate) is introduced and allowed to react for one to two minutes before the reaction is quenched by rinsing with water.

We find that in the hands of a skilled practitioner, this method generates relatively robust films with strong enhancement. However, even an experienced user will only have about a 50% success rate. Furthermore, the highly corrosive solutions involved require extreme care in use and should be treated with appropriate caution. Deposition in this manner can only be performed one crystal at a time. In total, including preparation and cleanup, electroless deposition usually takes around two hours to complete.

Another option that has been reported in the literature is electroless deposition on ZnSe internal reflection elements. PIKE Technologies offers ZnSe face-angled crystals which are compatible with our JxF-series cells should you wish to try this.

Because of the difficulty and chemical hazards involved, new users should strongly consider avoiding electroless deposition if they have the resources to do so.

Hybrid Metal / Conductive Metal Oxide Films

JackfishSEC workers have recently demonstrated the preparation and use of hybrid films for ATR-SEIRAS. In Andvaag et al., a 50 nm film of indium zinc oxide (IZO) was deposited by magnetron sputtering and then gold nanoparticles were electrodeposited in the presence of a shape-directing ligand (4-methoxypyridine). Such films have remarkably high enhancement and are robust enough to withstand repeated potential cycles into hydrogen evolution. The paper linked above illustrates the stability of the hybrid film by reductively desorbing a thiol monolayer three times from the same film with no meaningful change in the spectrum quality. This is essentially impossible on a conventional gold film.

Hybrid films, which can also use indium tin oxide (ITO), require access to a PVD chamber to prepare the conductive metal oxide (CMO) layer. Electrodeposition can be completed in situ in a spectroelectrochemical cell so that the enhancement can be monitored during the process. Although multiple CMO layers can be deposited in parallel, electrodeposition takes place in series, although it takes only around an hour to complete.

“Activating” A Metal Layer

Most of the literature describes a potential cycling routine to “activate” the metal layer. The exact processes involved are still unclear, but the empirical result is that the signal enhancement increases during the process. This will be covered in detail in a future post.

Tag: metal layers

Conditioning Au layers for ATR-SEIRAS