Today’s compact spectrometers deliver SNR performance, optical resolution and acquisition speed comparable to more expensive and less flexible spectrometers. Here we explore the absorbance linearity of the Ocean HR2 spectrometer.
Ocean HR2 offers high resolution performance (<1.0 nm FWHM), rapid acquisition speed (1 µs integration time) and great signal to noise ratio (SNR) performance (380:1) in a single instrument. These characteristics make Ocean HR2 an excellent option for applications including protein sample concentration measurements, which we demonstrate here.
IMPORTANT: PLEASE REFRESH YOUR BROWSER IF YOU DO NOT SEE ANY IMAGES/SCREENSHOTS/SPECTRA ON THE PAGE.
Absorbance of Bovine Serum Albumin
To evaluate the HR2, we measured nearly 30 different concentrations of 503 mg bovine serum albumin (BSA) protein samples in distilled water. BSA is often used as a protein concentration standard in biochemical applications.
Our setup comprised an Ocean HR2 spectrometer (190-880 nm), a deuterium-tungsten halogen light source with attenuator, a pair of 400 µm optical fibers, and a quartz cuvette in a cuvette holder. OceanView operating software and OceanDirect, a device driver platform, completed the system.
To ensure best measurement results, we first warmed up the spectrometer and light source for 30 minutes. Light source output can change slightly until the source is in thermal equilibrium, affecting measurements.
Also, as part of the sampling process, we never removed the quartz cuvette from the cuvette holder, as doing so can introduce errors. Instead, we made our dilutions in the cuvette, using a disposable pipet to remove some of each sample, then mixing in the DI water, and repeating the process for each sample.
Our measurements focused on the BSA absorbance peak at 280 nm (Figure 1), which mapped to absorbance linearity of 0.999, up to 2.5 AU (Figure 2). This level of performance would be beneficial in quantifying other types of protein concentration, as well as for biotechnology applications including quality assurance of pharmaceutical formulations.
High Speed Averaging Mode to Improve SNR
High Speed Averaging Mode is a hardware-accelerated signal averaging function accessible via the OceanDirect device driver that improves SNR performance in Ocean Insight spectrometers. With better SNR comes higher quality spectra and more accurate results.
SNR is a function of several factors, some more easily managed than others. For example, signal averaging to improve SNR can be carried out in operating software on the host computer but can take longer to process than what may be acceptable for your application. HSAM overcomes that limitation by enabling many more spectral averages over a given time period, yielding a much higher SNR per unit time. This can be important for time-critical or real-time applications, where decisions must be made very quickly and with high accuracy.
The advantage of HSAM can be quantified easily. For example, the standard operating software can process 10,000 averages in about 2 seconds. But because HSAM performs the averages on the hardware, the OceanDirect device driver can process 10,000 averages in 200 ms – 10x faster.
Summary
From determining concentration levels of proteins to identifying contaminants in pharmaceutical materials, the Ocean HR2 spectrometer delivers research-grade spectrometer performance in an instrument that works equally well in the lab or on the line, as part of a standalone setup, or as a key subsystem or component within another device.
New HR2 spectrometers offer high resolution with rapid acquisition speed, high SNR performance, and excellent thermal stability. We explore the effectiveness of HR2 for absorbance and irradiance applications.
IMPORTANT: PLEASE REFRESH YOUR BROWSER IF YOU DO NOT SEE ANY IMAGES/SCREENSHOTS/SPECTRA ON THE PAGE.
With its proprietary CCD-array detector and low stray light optical bench design, the HR2 provides research-grade spectrometer performance for applications ranging from plasma monitoring to pharmaceuticals analysis.
As part of our evaluation of HR2, we focused on three areas: the remarkable absorbance linearity possible with the spectrometer; its high-resolution performance; and the dramatic gains in signal to noise ratio made possible using High Speed Averaging Mode (HSAM), a hardware-accelerated spectral averaging technique now available with HR2 and SR2spectrometers.
Note: Ocean Optics Yvette Mattley and Brianna Waggoner provided the measurements for this application note.
About the Ocean HR2 Spectrometer
HR2 high resolution spectrometers are compact and robust, with integration times as fast as 1 µs and thermal wavelength drift of just 0.06 pixels/° C, helping to ensure reliable spectral performance as temperatures change. Ocean HR2 models cover various wavelength ranges within ~190-1150 nm, with a choice of slit width sizes to help users manage throughput and optical resolution.
The Ocean HR2 spectrometer is a high-resolution instrument compatible with Ocean Optics light sources, accessories and software, allowing users to optimize setups for different applications. Also, each HR2 spectrometer comes with OceanDirect, a powerful, cross-platform Software Developers Kit with an Application Programming Interface.
HR2 for Absorbance: Potassium Dichromate Standards
To test the absorbance linearity of the HR2, we measured potassium dichromate absorbance standards and bovine serum albumin (BSA) protein samples, each at varying concentration levels. BSA is often used as a protein concentration standard in biochemical applications.
Our setup for the absorbance standards comprised an Ocean HR2-UV-Vis spectrometer (190-880 nm), DH-2000 deuterium light source with attenuator, a pair of 400 µm optical fibers, a quartz cuvette and the SQ-1 Square One cuvette holder. OceanView software and the OceanDirect device driver platform completed the system.
The low stray light performance of the HR2 is reflected in its UV absorbance spectra (Figure 1), especially when plotted versus concentration, which shows absorbance linearity of a remarkable 0.9997 (Figure 2).
While the results for potassium dichromate standards are impressive – with that standard curve, you could predict almost any unknown sample within that wavelength range — we measured the standards only to 1.5 AU. That’s where the BSA protein samples came in.
HR2 for Absorbance: Bovine Serum Albumin
Using an HR2 spectrometer setup comparable to what we had for the potassium dichromate experiments, we measured nearly 30 different concentrations of 503 mg BSA in distilled water.
To ensure best measurement results, we observed several practices you should consider for your own experiments. First, prior to the measurements, the spectrometer and light source were warmed up for 30 minutes. Light source output can change slightly until the source is in thermal equilibrium, affecting measurements.
Also, as part of the sampling process, we never removed the quartz cuvette from the cuvette holder, as doing so can introduce errors. Instead, we made our dilutions in the cuvette, using a disposable pipet to remove some of each sample, then mixing in the DI water, and repeating the process for each sample. When possible, it’s also advisable to avoid moving the optical fibers in the setup.
Our absorbance linearity results with BSA were even more impressive than with potassium dichromate. We focused on the absorbance peak at 279 nm (Figure 3), then mapped the absorbance linearity at 0.999, all the way up to 2.5 AU (Figure 4). This level of performance makes HR2 an excellent option for quantifying other types of protein concentration, as well as for biotechnology applications including analyzing blood composition and performing quality assurance on pharmaceutical formulations.
Figure 4. Absorbance linearity measured over nearly 30 BSA samples is a remarkable 0.999.
HR2 for High Resolution: Identifying Spectral Lines
Depending on the optical bench configuration, the HR2 spectrometer has optical resolution that’s typically less than 1.0 nm (FWHM). For example, in measuring a mercury-argon gas-emission source, we observed numerous sharp, well defined peaks across the UV-Visible wavelength range (Figure 5).
We observed similar optical resolution performance when measuring solar irradiance, where an HR2 with a 25 µm slit detected spectral emission lines with optical resolution <1.2 nm (FWHM) across its spectral range (Figure 6). Our sampling setup comprised a cosine corrector attached to a 600 µm Vis-NIR optical fiber that was placed into a reflection probe holder and positioned at 90° into the sky. As you might expect, the absolute spectral intensity measured at each wavelength varied throughout the cloudless day.
HR2 and High Speed Averaging Mode
High Speed Averaging Mode is a hardware-accelerated signal averaging function accessible via OceanDirect that improves signal to noise ratio (SNR) performance in SR2 and HR2 spectrometers. With better SNR comes higher quality spectra and more accurate results.
SNR is a function of several factors, some more easily managed than others. For example, signal averaging to improve SNR can be carried out in operating software on the host computer but can take longer to process than what may be acceptable for your application. HSAM hardware-accelerated signal averaging overcomes those limitations by enabling many more spectral averages over a given time period, yielding a much higher SNR per unit time. This can be important for time-critical or real-time applications, where decisions must be made very quickly and with high accuracy.
We tested the advantage HSAM offers by measuring the relative output of an 880 nm LED using 600 µm optical fibers and attenuator. This is a useful area within the spectral range to evaluate, as in some silicon CCD-array detector setups, the spectral response at NIR wavelengths may drop off as grating dispersion and detector quantum efficiency drops off.
In one sense, what HSAM accomplishes is to “pull” signal out of the noise. This is demonstrated in Figure 7, where we compared results for a single average scan, which is quite noisy; and then for 1,000 scans (less noise); and finally, for 10,000 scans (minimal noise).
The big advantage with HSAM running in OceanDirect is that you can perform more spectral averages, over less time, than with OceanView software. For example, OceanView can process 10,000 averages in about 2 seconds. But because HSAM performs the averages directly on the hardware, OceanDirect can process 10,000 averages in 200 ms – 10x faster. This can enhance SNR for existing applications dramatically and open up new possibilities.
Summary
From detecting narrow emission lines in plasmas and gases to determining concentration levels of proteins, the HR2 spectrometer delivers research-grade spectrometer performance in an instrument that works well in the lab or on the line, as part of its own setup or as a key subsystem or component within another device.
The small size yet big performance of the Ocean ST microspectrometer makes it an ideal option for applications including medical diagnostics. In this example, we use Ocean ST to measure absorbance of whole blood.
High-performance absorbance linearity in spectrometers has been limited to complex and expensive instruments. Now, simpler and more accessible spectrometers like Ocean ST are achieving comparable measurement results.
The new High Speed Averaging Mode improves signal to noise ratio (SNR) in Ocean SR2 and other new spectrometers using hardware-accelerated signal averaging. With better SNR comes higher quality spectra and more accurate results.
We describe spectral data acquisition parameters in OceanView operating software, from integration time to scans to average. Use OceanView acquisition and processing features to optimize experiment settings.
Note: Please refresh your browser if you do not see any screenshots or icons in this material.
Q:I’m getting started in OceanView and don’t understand how best to use the data acquisition parameters that are available. What do I need to know?
A: Understanding the key data acquisition parameters available in OceanView will help ensure best results for your spectral measurements. This is important even if you’re using one of OceanView’s wizards, which guide you through the steps for specific measurements (absorbance, color and so on) but still require you to input most data acquisition parameters.
Additional resources offering insight on this FAQ are available at our Glossary and via the links at the bottom of this page.
Integration Time
This is the first parameter that you will set. Integration time is the period of time over which the spectrometer collects photons and correlates with the intensity of the signal that is captured. We find that the best measurements are made when the signal intensity is between 80% and 90% of its full range. When setting integration time, be careful not to allow any pixels to become saturated, as these pixels will not provide useful data.
Quick tip: In OceanView 2.0.8 and higher versions, there’s a convenient feature that automatically selects the integration time appropriate for your setup. Activating this feature is as simple as selecting the Automatic option in the Acquisition Group Window of OceanView.
Scans to Average
This time-based averaging function specifies the number of discrete spectral acquisitions that the device driver accumulates before OceanView receives a spectrum. The higher the value, the better the signal-to-noise ratio (SNR). The SNR will improve by the square root of the number of scans averaged.
Quick tip: To convey the “true” SNR of its spectrometers, Ocean Insight reports SNR specifications without signal averaging applied.
Boxcar Width
Boxcar smoothing is a technique that averages a group of adjacent detector elements across spectral data. For example, a boxcar width value of 5 averages each data point with 5 points to its left and 5 points to its right.
The greater the boxcar width value, the smoother the data and the higher the SNR. If the value entered is too high, a loss in spectral resolution will result and peaks will become flattened. The SNR will improve by the square root of the number of pixels averaged.
Quick tip: There’s a more detailed explanation of boxcar smoothing and its trade-offs in our Glossary.
Electric Dark Correction
Many spectrometers have a small number of dark pixels. A dark pixel is an electrically active CCD detector pixel that has been physically coated to prevent any light from entering. Due to leakage, a small amount of electrons escape from the detector well even if there is absolutely no light.
To eliminate any error due to leakage, electric dark correction computes an average value of these dark pixels over 15 consecutive scans, and then subtracts this average value from all pixels in the spectrum. Although the algorithm specifies that 15 consecutive scans be averaged, the correction factor is applied as soon as the first spectrum of data becomes available. Then additional scans are averaged into the correction factor as these spectra become available. A running average of the most recent 15 scans is maintained.
The amount of electron leakage is proportional to the duration of the integration time. So, whenever the integration time is changed, the previous correction factor is discarded and we then begin to compute a new correction factor based entirely on the new integration time.
Non-linearity Correction
All Ocean Insight spectrometers are calibrated at the factory to maximize accuracy. One of the calibrations performed is to correct for detector non-linearity. This calibration consists of eight numbers used as the coefficients of a 7th order polynomial, which adjusts for the phenomenon that CCD detectors do not respond to stimuli photons uniformly as more electrons drain from the detector well. In other words, the efficiency of CCD detectors may be 30% when the well is half-full but may be only 20% when the well is completely drained of electrons.
By “efficiency” we mean the probability that an incoming photon will drain an electron from the CCD well; 100% efficiency means every incoming photon will drain one electron, while 50% efficiency means an incoming photon has only a 50% chance of causing an electron to drain. Non-linearity calibration is made by averaging together all pixels of the CCD array. Thus, we are assuming that all pixels respond about the same.
Quick tip: Detector linearity is different than absorbance linearity, which is associated with Beer’s Law.
Ocean Optics optical fiber assemblies, probes and accessories allow users to transmit and collect light in our spectrometer setups. From off-the-shelf patch cords and custom fibers to specially engineered OEM assemblies, your fiber options are as varied as your applications. Here are some tips to ensure reliable, long-lasting fiber and probe performance.
Rule #1: Choose Wisely Modular spectral systems are only as good as the sum of their individual parts. The same care you put into choosing a spectrometer should go into choosing a light source, the sampling optics, and the fibers or a probe. Are you measuring absorbance or reflectance? Are you measuring wavelengths below 270 nm, where UV exposure can solarize certain fibers? Where will the fiber be placed in your setup? Is the sample environment chemically harsh? Determining those criteria will help us guide you to the optimum combination of components – including fibers – that meets your needs and tolerates sample conditions.
Rule #2: Handle Fiber Connectors and Terminations Carefully Without proper care, SMA 905 and other fiber connectors can be scratched or damaged and affect measurements. On occasion, customers have even inadvertently separated the connector or ferrule from its fiber or probe assembly by pulling on the end too forcefully.
Since the ends of fibers receive the most wear and tear, manufacturers design terminations with extra strain relief and boot collar protection. But just be mindful when removing end caps to use one hand to hold the fiber by the connector and the other to pull off the end cap. Ocean Optics XSR extreme solarization-resistant fibers go one step further by having an end cap that screws on to the end of the fiber — no pulling necessary.
Rule #3: Mind the Bend Radius Although optical fibers and probes are used to move light around your spectrometer setup, there is a limit to how much bending those assemblies can tolerate. The bend radius of a fiber denotes how sharply the fiber can bend before damage occurs in the fiber. This damage can range from fiber attenuation to fiber breakage, which leads to even more dramatic light loss.
That’s why it’s good practice to inspect fibers periodically to ensure that light transmission is occurring. Broken fibers stop transmitting light.
Ocean Optics reports both long-term bend radius (LTBR) and short-term bend radius (STBR). LTBR is the minimum bend radius recommended for storage conditions. STBR is the minimum radius recommended during fiber use.
Bend Radius for Visible-NIR, UV-Visible and Solarization-resistant and Extreme SR Fibers
Fiber Core Size
Fiber Types
LTBR
STBR
50 ± 5 μm
VIS-NIR, UV-VIS
4 cm
2 cm
100 ± 3 μm
VIS-NIR, UV-VIS
4 cm
2 cm
113 ± 6 μm (115 μm nominal)
XSR
4 cm
2 cm
200 ± 4 μm
VIS-NIR, UV-VIS, SR
8 cm
4 cm
230 ± 12 μm
XSR
4 cm
2 cm
300 ± 6 μm
SR
12 cm
6 cm
400 ± 8 μm
VIS-NIR, UV-VIS, SR
16 cm
8 cm
455 ± 22 μm
XSR
8 cm
4 cm
500 ± 10 µm
VIS-NIR, UV-VIS
20 cm
10 cm
600 ± 10 μm
VIS-NIR, UV-VIS, SR
24 cm
12 cm
600 ± 30 μm
XSR
24 cm
12 cm
1000 ± 3 µm
VIS-NIR
30 cm
15 cm
1000 ± 20 µm
UV-VIS
30 cm
15 cm
Rule #4: Avoid Excessive Heat Avoid exceeding the temperature thresholds for the fiber materials: For standard fibers, the temperature threshold for the silica fiber is 300 °C, while the epoxy and PVDF zip tube are rated to 100 °C. For premium-grade fibers, the entire assembly is rated to 220 °C. Jacketing options including stainless BX offer better protection, but it’s always best to consult your Ocean Optics representative for assistance with applications in challenging environments.
As one university professor recently shared with us, some Ocean Optics optical fibers in his freshman chemistry lab had “survived” 20 years in the hands of beginning chemists. These fibers might have lasted longer, but a few students got those fibers too close to the flame of the Bunsen burner they were measuring, melting the fiber boot collar and PVDF zip tube.
Chemical resistance is another criteria that may be important for your application. Avoid immersing the fiber in materials that can damage quartz, nickel, steel, aluminum, or the epoxy. In harsh sample environments, choosing durable jacketing materials including silicone monocoil or stainless steel BX is your best bet. Custom sleeves and ferrules are another option.
Rule #5: Remember the Little Stuff Although this might not always be practical, it’s helpful to replace the end caps on optical fiber connectors when the fibers aren’t in use. This helps to prevent scratches and avoid contamination from dust and fingerprints. Also, we suggest cleaning the fiber ends periodically with lens paper and distilled water, alcohol, or acetone. Avoid scratching the surface.
Our FAQs section has simple, clear-cut answers to questions about Ocean Optics products, measurements and software. In the spotlight: two keys to reference and dark background measurements for absorbance applications.
Q:How do I take the best reference and dark background spectrum for an absorbance measurement?
A: To take the best reference measurement, first allow the light source to come to thermal equilibrium (this can take up to 30 minutes in some cases).
Ensure that all surfaces of the cuvette or sampling optics are clean and clear of fingerprints, dust, and dirt. Fill the cuvette with the exact solvent or buffer solution to be used for the sample and check for bubbles. It is especially important to check for bubbles if you’re using a transmission dip probe or flow cell, as those devices are prone to this source of error, particularly at the shorter pathlengths.
In OceanView software, optimize integration time so that maximum signal is at ~80% of full scale. Use the highest number of averages tolerable and keep boxcar width to approximately the same value as the pixel resolution of the spectrometer, otherwise the setting can affect the spectral resolution.
Once the reference has been acquired, enter transmission mode in software and review the resulting spectrum. If a good quality reference spectrum has been stored, the transmission of the reference solution should be at 100%, with some noise around this value. Wavelength regions with more noise indicate the wavelengths at which accuracy will be the least (i.e., most affected by noise and other factors) for the sample measurement (typically the shortest and longest wavelengths).
When taking a dark measurement, it’s best to block the light at the light source if possible. Turning the light source off and on again will throw the light source out of thermal equilibrium and require a new reference measurement.
Alternatively, many cuvette holders have a filter slot where the light can be blocked with an opaque object. Just be sure to use a piece of metal or other object that is guaranteed to be 100% opaque. Paper, even cardboard, can be deceptively transmitting; it takes only a very low level of light to affect a measurement.
Aquaculture and fish farming require careful monitoring of conditions to ensure healthy, sustainable production. Techniques including spectroscopy and oxygen sensing are useful analytical tools in aquaculture settings.