Dear Doan -Quan Mai , I completely agree Liam Kelleher , SERS is a wide and diverrse field as the reviews Liam shared will show you.

Let me add just a couple of comments: silver and gold are the more common metals used in SERS, with copper being also a plasmonic material but its application is a lot less popular in bibliography, both as colloidal solution or as a solid structured substrate.

Silica is popular for solid structured substrates but many other materials can be used too, for instance, polymers work pretty well in my own experience, because they allow an easy fabrication process well by the methods described by Liam or by methods like NanoImprint Lithography (NIL). In this way you get the nanostructure, then you have to deposit the metal.

Some other metals and conductive materials can do the job too: like graphene or carbon nanotubes, alone or just coating a gold or silver substrate. I my own experience aluminium (Al) works quite well, comparable to Au or Ag.

SERS enhancement is caused by a combination of two contributions:

Electromagnetic and chemical. Often the electromagnetic is the main contribution by localizing the electric field in some points of the SERS sensor and if it happens that a molecule of the analyte is close to that point it will feel the strong field and the Raman signal will be enhanced.

The chemical contribution comes from the formation of chemical bonds between the molecule and the surface of the metal.

This chemical contribution can also be the main part when your substrate is made of a dielectic or semiconducting material.

Hope it helps.

Hi Doan -Quan Mai , I have a feeling this is a question that will get you some very broad answers. Can you add some specifics about what sort of SERS you want to do? From my own understanding, you can roughly split SERS up to colloidal based and substrate-based approaches. For colloidal solutions there are different methods and geometries utilised, but ultimately these are deposited on foil or silica and then measured.

For substrates these can be made in a range of ways depending on the geometry you want to achieve. A surface can have a pattern imprinted and then a material like gold spluttered on, this could be chemically etched or imprinted using ion-beam processes. I think silica again is used a lot, some materials can be used to improve adhesion, but those I can't think of right now. Some other things to consider are wavelength to be used and whether you want to use labelling or not. Hope that helps a bit! Best of luck.

Here are two recent reviews on SERS that may help.

https://bit.ly/3nTqimV

https://bit.ly/2HwGY4d

Dear Doan -Quan Mai , I completely agree Liam Kelleher , SERS is a wide and diverrse field as the reviews Liam shared will show you.

Let me add just a couple of comments: silver and gold are the more common metals used in SERS, with copper being also a plasmonic material but its application is a lot less popular in bibliography, both as colloidal solution or as a solid structured substrate.

Silica is popular for solid structured substrates but many other materials can be used too, for instance, polymers work pretty well in my own experience, because they allow an easy fabrication process well by the methods described by Liam or by methods like NanoImprint Lithography (NIL). In this way you get the nanostructure, then you have to deposit the metal.

Some other metals and conductive materials can do the job too: like graphene or carbon nanotubes, alone or just coating a gold or silver substrate. I my own experience aluminium (Al) works quite well, comparable to Au or Ag.

SERS enhancement is caused by a combination of two contributions:

Electromagnetic and chemical. Often the electromagnetic is the main contribution by localizing the electric field in some points of the SERS sensor and if it happens that a molecule of the analyte is close to that point it will feel the strong field and the Raman signal will be enhanced.

The chemical contribution comes from the formation of chemical bonds between the molecule and the surface of the metal.

This chemical contribution can also be the main part when your substrate is made of a dielectic or semiconducting material.

Hope it helps.

Hi Liam Kelleher ,

Following your opinion, we will have two kinds of substrate that colloidal-based and substrate-based. I care more about colloidal-based. Is it a kind of flexible substrate? Can you share more about this substrate' kind?

Hi Manuel Gómez ,

Yes, your experiences are very helpful to me. As far as I know, the conductive or insulator property of the substrate is vital in SERS. Furthermore, Electromagnetic mechanism (EM) and Chemical mechanism (CM) are two mechanisms which most recognized. Can you propose some polymer that you feel proper to prepare a suitable substrate?

Dear Doan -Quan Mai , polymers are convenient to make substrates with nanostructures, because most of they can be easily molded with such tiny patterns using diverse techniques of fabrication. These techniques can be as complex and expensive as with anyother material, but you also can use cheap ones like NIL where you only need a mold with the reverse pattern of the nanostructure you want to replicate, a polymer that can be conformed by heat or by exposure to UV light. The polymers you can use are those you have at hand and you know how to manipulate. They could be photoresists, silicones, common polymers (PE, PS, PP, PET, PMMA, PU, PTFE.... ) In my work you can see what polymers and methods I use as main options: Ormocers (a composite), PFPE (PerFluoroPolyEthers), Silgard 184 (a silicone or PDMS), Photoresists, etc. The polymer you choose will depend on the fabrication methods you have access to.

Some of these polymers I use are not commercial, but made by synthesis, while other are readily available on the market. I don´t use thermal impression (melting of the polymer, but it works pretty well if you have access to a printing press with heating function. Silicones like Silgard 184 from Dow comes as 2 components you mix before pouring on your mold, and the thing cures by heating or just leaving it at room temp for 24 or 48 h. Photoresists, mainly negative tone, get hard

(polymerized or crosslinked ) by exposure at UV light normally, so you would need a UV light with appropiate wavelength and intensity to harden your material.

The main problem with NIL is the need of a mold, which could be spensive because it would be fabricated by conventional lithographic techniques. Once you have it, you can use and reuse it many times- properly used and mantained!!.

When you have your nanostructured polymeric replica of the mold, a metal coat must be applied to give it the plasmonic properties.

My advice is, get familiar with traditional SERS substrates and metals before experimenting with new or difficult ones. In my experience, dielectric substrates require a proper design and election of materials and geometries to get enhancement factors that are often lower than those you can achieve with conventional plasmonic materials like Au or Ag.

In my work I use the polymer as a mere support because it is easier to mold than silicon, glass or metals for instance. In this way I can obtain a nice micro or nanostructure which is coated with the metal of interest to provide the plasmonic properties to the structure. The metal coat thickness depends on the metal, but the general idea is fully coating the polymeric structure. If you left holes or the metal layer is too thin it could work but when you hit the sample and substrate with the laser, your Raman spectrum will show a large background due to the polymer´s Raman signal. One houndred nm or so will be enough, but it is a matter of trial and assay.

Metal deposition can be done by diverse methods, like sputtering, physical vapour deposition (PVC), e-beam deposition, electrodeposition...

All this information is clearly stated in the reviews Liam Kelleher suggested, so have a look to them and any reference cited by the reviews' authors.

Hope it helps.

Hi Doan -Quan Mai , I think Manuel Gómez has answered your question very well here for the solid substrate side of things.

My statement for grouping of colloidal and a solid substrate is mainly due to the process to create them and their use. As the name colloidal suggests this is just a suspension of nanoparticles, very typically Au or Ag, these can have stability issues that cause aggregation. They can be purchased or synthesised. They can also be functionalised with different material to capture specific molecules. Upon mixing the colloidal NPs with the target solution or interest there can be an incubation time, after this, the solution is deposited on a slide (Al or Si) for measurement, this itself is why studies try to address reproducibility in their work.

This is about the extend of what I can provide you with information-wise, my thoughts for you moving forward might be to read those reviews and also to search for more specific studies that relate to the experiment you want to establish. Good luck!

Dear Manuel Gómez ,

Thanks for your very detailed answer. It will help me a lot. Thanks again!

Quan-Doan Mai

Dear Liam Kelleher ,

Thanks for your experience and advice. It will help me a lot!

Quan-Doan Mai

You are welcome Doan -Quan Mai . Good luck with your projects and my best wishes for the New Year.

Manuel

Hi.

Take a look at DOI: 10.1021/acsami.7b13965. It may be useful.

This team worked broadly SERS, both on modeling and experimental.