(UV) Tips and Tricks - Before You Create a Method

When using a UV Spectrometer, there are many issues which can affect your desired results and produce unwanted unusual spectra, absorbances or artefacts.  To prevent this from happening, it is a good idea to first of all understand the instrument being used in more detail, but moreover to use the instrument in a more basic systematic format first before fully employing all of its corrective powers and interpretations.

Very often, analytical chemists using a UV instrument will encounter artefacts such as unusual noise in one specific part of their sample spectrum, or a step in the spectrum that just will not go away. Effects such as these are very often related to the choice of solvent or cuvette used, or due to the way the instrument beam is travelling though the sample in comparison to that of the blank used.

The construction of a UV Spectrometer consists of a light source - usually a Deuterium Lamp and a Tungsten Iodine Lamp.  The emission profile of each lamp is projected forward to a monochromator system that separates the wavelength of interest and filters it forward to a beam geometry that either goes through a single pathway of a double pathway (sample and reference beam).  A double beam system uses ratio procedures to compare the sample in real time to a reference beam, correcting for changes in the Spectrometer stability or sample matrix chemistry (if the blank is in the reference beam).  Before analysis of the sample, a baseline procedure is used to account for the spectral absorbances of the blank solvent media and main spectrometer hardware.  It is important that the beam geometry travelling through the sample is the same shape as that obtained during the baseline procedure.

The Spectrometer uses a mirror to utilise the two emission spectral ranges from these individual lamps, stopping a graphical scan whilst doing so. If a more complex UV Spectrometer is used, then monochromators may be changed in the hardware procedure during a scan - again stopping the graphical trace as it does so.  It is at these change-over point that steps in the graphical trace can sometimes be observed.

This is caused where the comparison ratio of the sample to the baseline algorithm is distorted due to the fact that the beam shape coming from the sample to the detector may be a different shape to the blank baseline procedure.  Causes of distortion are usually from scatter within the sample, such as particulate, different refractive index, different temperature or dissolved solids chemistry and concentration effects. The beam distortion from the sample also dilutes the concentration of light impacting onto the detector and will artificially elevate and Absorbance reading (or reduce a %Transmittance reading) across the whole spectral range. Step changes are a frequent occurrence when analysing solid samples in transmission mode, or reflection - where the beam takes a turn of direction during analysis.

Another common problem is the presence of unwanted noise within the spectral scan of a sample.  This is normally caused by the choice of cuvette or solvent. The use of a UV Spectrometer employs the procedure of analysing a Blank solution in a cuvette to perform a background or baseline correction procedure before starting to analyse the sample.  This process takes the electronic signal from the detector for a specific or range of wavelengths and resets it to zero for that analysis methodology.  The important issue to recognise is that if your blank solvent media or cuvette absorb the light in that spectral region then there is very little light reaching the detector for the instrument to analyse and the residual sensitivity of the detector is severely reduced when you present your sample.  The outcome is that there will be a lot of noise instead of your expected absorbance peak/result in your sample spectrum.

Your choice of cuvette is important. The material of the cell must have no absorption at the measurement wavelength. Two materials often used for cells are glass and quartz. Polystyrene (PS) and polymethylmethacrylate (PMMA) are mainly used for disposable cells. The table below indicates the wavelengths for which the different types of cells can be used.

The spectral transmission profile for the above cuvette choices can be seen in the graphic below:

It can be seen that if the cuvette choices are used where their spectral transmission is starting to diminish, then the light intensity reaching the detector will be poor before your sample is analysed in the beam.  It means that you will have a sample spectrum that is very noisy in that region.

Your sample spectrum of your active peak is a chromophore dissolved in a choice of solvent.  The solvent is also acting in the spectrum, contributing to the result, even though the blank procedure and spectrometer hardware background correction procedure has effectively masked it from the analysis.  The table below indicates the spectral transmission ranges for some typical solvent choices available :

The above factors often cause unwanted artefacts in the sample spectrum.  If you are using the Photometric or Quantitation modes of analysis, then only the on-peak or on-point part of the sample spectrum is analysed to produce an individual Absorbance number calculation and no other associated underlying spectral information is provided.  It can therefore be difficult to troubleshoot those operational modes if there is a singular spectral interference or anomaly.

The best procedure to follow with molecular spectroscopy is to conduct an initial investigation of the sample matrix, solvent and container using the Spectrum Scanning mode.  The idea is to stabilise the Spectrometer as per the instruction manuals without any samples inside, and then gradually introduce the materials to the system each time producing a full spectrum scan or chemical and container component.  With the individual spectra overlaid on the screen it is then possible to visually understand how each material absorbs and contributes to the intended sample result.  You will be able to interpret any of the above mentioned artefacts and behaviours of substances in the beam and the reasons why they are generated.

Investigation Procedure

A typical procedure for doing this would be :

Take all cuvettes out of the spectrometer and ensure that the windows of beam entry into the sample compartment on the left and the detector windows on the right inside the sample compartment look clean and free of hairs and dust.  For example, hair can flicker in the way of the beam and cause partial distortion of the baseline and it is not easy to see.  Please don’t be tempted to rub, touch or clean the windows, as cleaning should be done by an engineer.

Next, make sure that the lid closes properly and no light can get in the sample compartment when closed.

Switch on the UV-1800/1900i/UV instrument and allow to warm up for 30 minutes – this is the maximum time to ensure the highest resolution performance of the spectrometer with lamps at full stability (double beam configuration corrects for lamp energy fluctuations anyway, but this 30 minutes gives the best performance).

The spectrometer should pass its start-up initialization and notify you that the hardware is ok.

With the instrument on, and stable as above, connect the UV1800/1900i/UV via USB to the PC, select PC control on the UV instrument console, and connect to the LabSolutions UV-Vis or UVProbe software.

Set up a full wavelength scan method in the spectral module of the Software, 190nm to 800nm.

Keep the sample compartment empty and lid closed.

Baseline the spectrometer for the 190nm to 800nm scan.

Open the lid and place your clean empty cuvette in the front sample holder and front beam only.

Perform a scan of this empty cuvette.

Fill the same cuvette in the front position beam with your blank solvent and perform a spectral scan.

Remove the blank solvent from the front sample cuvette and replace that liquid with the desired sample chromophore dissolved in the same solvent in that same cuvette, (you should pre-rinse the cuvette with the sample) and replace that cuvette in the front sample beam and close the lid. Perform a spectral scan of this sample.

When you overlay the scans, you will be able to pinpoint the following aspects:

• Does the cuvette absorb in the far UV ? Is the cuvette clean ?
• Does the solvent have an absorbance profile, especially in the UV ?
• How does the chromophore absorb on top of the cuvette and solvent underlying absorbance profile?

Points to note are:

• Quartz cuvettes have the best transparency across the entire wavelength range
• Dirt absorbs in the UV below 200nm
• Choice of solvent is critical and getting the same solvent from a different supplier can sometime cause confusion with the result if the wrong purity is selected (and higher absorbance profile in the UV).
• If the cuvette and solvent absorb in the UV, then after a Baseline routine is performed, the remaining residual detector sensitivity is extrapolated along with the noise.  So the sample spectrum will exhibit noise in that UV region - but it is because the container and solvent have already absorbed a lot of lamp energy.
• Always be wary of the lower UV spectral range because the choice of solvent and cuvette are critical for this region.
• Never us plastic cuvettes below 300nm – some analysts use plastic low volume cuvettes for protein and DNA measurements and get noisy, or blocky spectra because the plastic and solvent are mopping up the beam and leaving virtually nothing for the detector to see.  However, biological samples have certain toxicity and proteins are sticky – so a disposable cuvette is a logical choice.

The beam geometry of the instrument beam profile is 0.8mm wide and 9mm high for the UV-1800/1900I systems and different for the higher range systems according to the slit width selected (see diagram below).

If you are using reduced volume cuvettes, then the beam has a high probability that it is going to be clipped.  If the clipped part of the beam travels though the quartz walls, then a lot of baseline noise will be evident. Reduced aperture cuvettes should always be made of black quartz walls, so that they are self-masking.  The black-walled cuvettes are not used, then an additional aperture mask can be used to reduce the beam access to the cuvette.

Reduced aperture cuvettes will impart more noise to the analysis, so more signal-averaging of the detector may be required by scanning slower.

If the sample itself is not truly transparent, such as having particulates, then it will scatter the beam. Scatter is the enemy of spectroscopy because the shape of the beam on the detector window will have changed in comparison to the baseline assessment – and you will get a step-change elevation which cannot be corrected unless you employ an integrating sphere accessory.

Analysis Procedure

Next is an explanation of the preferred analysis sequence :

Choose matched cuvettes made of quartz, so that they are the exact same materials and wall thickness construction.

Fill both cuvettes with the chosen blank solvent that does not have any significant absorbance profile.

Place both cuvettes in the stable warmed-up spectrometer, one in the front and one in the rear beam.

Perform a baseline scan for the multiple wavelength method scan.

Leave the rear cuvette in the reference beam.

Remove the front beam sample cuvette and replace the liquid inside with the chromophore dissolved in the same blank solvent. Replace that same cuvette now into the front sample beam and close the compartment lid.

Perform your spectral scan.

Points to note about that procedure:

• The true double beam spectrometer uses the rear beam to perform a ratio assessment to normalise any variance of the source energy.
• Performing a baseline with the same two cuvettes and blank contents in both beams normalizes the instrument absorbance to account for the absorbance of the container materials, the blank solvent and anything else in the beam such as gases and moisture.
• Leaving the blank solvent and cuvette in the rear beam during sample measurement will automatically correct for any difference in chemistry of the solvent with time and temperature.
• Using the rear beam and using the normalization baseline procedure with both cuvettes in the beam is the best optical procedure because the dual detectors experience the same illumination level at the start of the analysis procedure.
• If no blank cuvette is in the rear beam, then that detector is always having full light shone onto it.  But the sample detector is having a lot less light exposure. Ratioing the beams will be using disproportionately different energy levels and more errors will propagate to your sample results.
• If there is a blank cuvette in the rear beam, then it is not experiencing the full lamp energy and will have a similar energy exposure to the sample, so ratioing the beams will employ less disproportionate error.

Hopefully, these tips and tricks will provide a good resource of information to mitigate against spectroscopic errors when analysing samples on the UV-Vis-NIR Spectrometers.