Published: 2020-04-18 | Categories: [»] Engineering, [»] Opticsand[»] Chemistry.

Lots of things have been moving in one year since I launched the initial [»] breadboard version of the OpenRAMAN spectrometer and today I am really glad to announce the official first prototype of OpenRAMAN: OpenRAMAN Starter Edition. The spectrometer with the standard cuvette is shown in Figure 1 and the full assembly instructions are available at the [∞] companion website.

Figure 1 – Photography of the OpenRAMAN Starter Edition with the standard cuvette

From the experience I gained with the [»] breadboard version, I decided to offer the spectrometer as a modular platform available in different versions called here editions. The first version, or starter edition, will be ideal for teaching the fundamentals of Raman spectroscopy to students and guide them through the alignment and/or operation of a spectrometer. It is easy to assemble, operates with a class 3R 4.5 mW laser and costs about 2500€ with the standard cuvette. Since this version is mostly designed as an educational version, it has mid-performances and achieves a 35 cm-1. For researchs needs, the Performance Edition will come later this year with all the performances that is currently available with the [»] breadboard version , so a 12 cm-1 resolution and a 10× more powerful laser to achieve high SNR at low exposure time. You can however technically repeat some of the experiments with the starter editions, such as the [»] methanol in ethanol experiment, the tracking of [»] ethyl acetate production or the [»] addition of bisulfite to aldehydes!

Figure 2 shows typical spectra obtained with both the starter and the performance editions to help you make a choice which edition is best for you.

Figure 2 – Spectra of iso-propanol taken with both the starter edition and the performance edition of the spectrometer

I am still struggling a little bit with the performance edition at the moment which is the reason it is not available yet. The spectrometer is functional but the price is still too high due to the expensive laser and TEC drivers used. These drivers are clearly overkill for the application and I am planning to replace them with custom electronic but this takes time and money and I must admit that I lack both of these two resources at the moment. There is also the question of the safety since this version operates with a 40 mW laser. As this one is likely to be used for day-to-day analysis, it must operate safely in a lab environment which means that I also have to include a housing around the spectrometer and safety switch to turn the laser off when the cuvette is opened.

So two editions: one for teaching and one for research. But it is not all! I designed the system to be as modular as possible which means that it will feature all of the appreciated stuff of an open system! The system does not require you to use a fixed cuvette design and you will be able to swap any holder type you would like, including your own holders. At the moment I focused on liquids and the only cuvette available is the one to accept the thin-walled 10 mm vials (called here the standard cuvette). But I am also planning to develop other types of holder such as one to study flows or one to study powder samples! And since one good news never comes alone, the starter edition is built on the same backbone structure as the performance edition and it will be possible to upgrade it later! So why waiting? You can already start building your setup and make it all nice and shiny after you have mastered the fundamentals ;-)

Ok enough talking. The question is now: what is in the box?

The optical design is given in Figure 3 and it basically is the design of the [»] breadboard version with the imaging lens upgrade plus some technical improvements embedded on a custom base plate to ease up alignment.

Figure 3 – Optical design for the OpenRAMAN Starter and Performance editions

One of the first improvement that I gained from the former version was that there was a lack of freedom in the steering of the laser beam into the cuvette. The dichroic beamsplitter was fixed at roughly 45° and only the incoming beam angle could be modified. It was therefore extremely difficult to get a beam entering the cuvette perfectly straight and my last attempts with the breadboard was to include an expensive translation stage to get the job done. Here, I therefore swapped the fixed dichroic beamsplitter by one mounted on a kinematic base (Thorlabs KM100). These bases are cheap enough and provide an extremely accurate laser steering at low cost. Also, I developed an alignment method to get the laser perfectly straight in regards to the optical axis. The method is described on the [∞] companion website so please have a look at there too.

As a minor improvement, I also decided to replace the edge filter by the premium version of the same filter. It is a bit more expensive but it offers better performances overall so I think it worth the extra expense. You don’t have to apply this fix if you are a bit tight on money (extra 50€).

Last improvement was to include a flat transparent window. This is not a joke and it actually does something! When the beam with the Raman signal goes back into the system in its collimated state, it is laterally shifted to the side by the refraction in the dichroic beam splitter and the edge filter. Orienting the edge filter the other way with a negative angle relative to the dichroic beamsplitter compensate this effect but not does not completely cancel it. The consequence is that the collimated beam will hit the condensing lens a bit on the side which will create a cone of light that is not centred relative to the optical axis of the spectrometer. When working on liquid, this has little effect because the source is more or less Lambertian and thus create a cone of light that completely fills the numerical aperture of the system. However, when working on larger particles the effect might become visible and some light/resolution will be lost due to this offset position. The compensation window corrects for this effect by shifting the beam back on the optical axis before it hits the condensing lens. This is represented in Figure 4. It is a small effect that you can potentially neglect when working on liquids but I recommend keeping it in place unless you are really tight on money (the window with the support costs about 100€).

Figure 4 – Effect of the compensation window

That is all for the optical design since the rest was already described in the post about the [»] breadboard version. I strongly urge you to read it if you haven’t yet because it describes the limitation of the system in terms of resolution and all other small (but important!) effects of the laser, the grating and so on.

If you are wondering what type of spectra you can expect with the setup, just take a look at Figure 5 to Figure 8. It is all kind of common solvents that you can easily get in a lab. All these spectra are raw measurement without any type of baseline subtraction. Also please note that the wavenumber axis is only partially calibrated for these measurements.

Figure 5 – Spectrum of acetone, 1 min exposure
Figure 6 – Spectrum of ethanol, 1 min exposure
Figure 7 – Spectrum of methanol, 1 min exposure
Figure 8 – Spectrum of ethyl acetate, 1 min exposure

Finally, you will also find on the [∞] companion website the drawings and step files of all the elements of the system. All the parts can be printed in 3D at manufacturers like Materialise at relatively fair prices but I also designed some parts such that they can readily be made by machining aluminum in the case you have access to a mechanical workshop. I’m myself a big fan of crafting stuff so I know some of you will enjoy machining the base plate themselves :)

That is all for today folks! I would really like to thank James who has been supporting since several months already through Patreon. And if you like the OpenRAMAN project or other stuff I produce, please, please, consider [∞] donating too as this will really make a difference! Do not hesitate to share this post with your friends/colleagues!

[⇈] Top of Page

You may also like:

[»] Launching OpenRaman DIY Spectrometer !

[»] Raman Cuvette Upgrade

[»] Raman Imaging Lens Upgrade

[»] Complete 400-800 nm Spectrometer Design from A to Z

[»] Confocal Spectroscopy