Spectra detection. Detector selection.

Spectrum detection is the estimation of light energy as a function of a wavelength with use of a detection system. Typically, the detection system includes a photoelectric detector that converts the electromagnetic energy into electrical signals, hardware and software for data processing and visualization. A graph of light energy versus wavelength is normally called a spectrum.

There are two types of detection systems: integral system (single-channel) and imaging system (multi-element/array detector).

Integral detection systems use detectors converting the integral flux of light into electrical signal without wavelength distribution of light energy. This type of detectors includes single-element detectors based on photomultiplier tubes (PMTs) and solid-state photodetectors.

The systems which use linear or array multi-element sensors are called imaging detector systems. Such detectors are utilized when in addition to measurement of integral light energy the light energy each wavelength should be measured.

Photometric, Raman/fluorescence measuring system based on monochromator-spectrograph MSDD1000.

For spectrum detection with the use of integral detection system, the detector is installed directly behind the exit slit of a spectral device. The width of exit slit determines the width of a spectral band of light which passes through a monochromator. Width of a spectral band is calculated as a product of the slit width value and the reciprocal linear dispersion of a monochromator.

Spectra detection with the use of integral detection systems is carried out point-by-point. The monochromator is tuned step by step according to wavelengths (spectrum scanning) at the preset step of scanning with simultaneous detection of electrical signal on a detector in each point of a spectrum . Scanning on a spectrum is performed by rotation of a grating around its axis. Here the wavelength interval in the focal plane and thus the spectral band of the light that passes through exit slit of a monochromator are changing, while the detector is detecting the integral light energy at the selected spectral bandwidth.

The process of spectra scanning with the use of a monochromator and the detection process should be synchronized in order to obtain the correct spectrum graph. Two types of such synchronization are possible: synchronization in time and step-by-step synchronization. With time synchronization, monochromator’s scan rate should be constant across the scan range (it is set in nm/sec). When using a detector or a signal recorder as a detection system, the scanning rate should be correlated with the recorder’s tape rate. In this case, the spectrum image is visualized with the help of the recorder. Some other ways of time synchronization are possible, for example, with the sync pulses that are generated by a monochromator in certain wavelength intervals in the process of scanning.

However, the systems “scanning monochromator- detection system” with time synchronization are not convenient for practical applications. Scanning monochromators, produced by SOL instruments, make use of more advanced principle of synchronization – step-by-step synchronization of a spectrum scanning and spectrum detection. An embedded analog-to-digital converter (ADC) is one of the main part of such detection system, which is offered as an option with some spectral devices. The ADC board receives electrical signals from photodetectors (photodiodes, PMT, etc.) to process them using a 16bits ADC. The ADC board is built into the spectral instrument. Both a spectral device and ADC board are controlled through PC. Such ADC board allows to build a measuring system on basis of a spectral device. That is to say, the spectrum can be scanned along a specified spectral range at a set scan step with simultaneous detection of a photodetector signal. The obtained spectrum is displayed on monitor as a graph. Our specially developed spectroscopic software SpectraSP is used to control the system operation via a PC.

А spectrum detection is performed point-to-point. The algorithm of a spectrum scanning is as follows: the spectral scan range, the scan step and the average measurement time of a detector signal in each spectrum point should be set. Once the scanning process is initiated, the monochromator starts tuning to the initial wavelength, then stops and the signal with the set measurement time is measured. After that, the monochromator tunes to the next wavelength, which differs from the previous one by one scan step, stops again and the signal at this wavelength is measured, and so forth. The entire scanning range is covered with such point-to-point measurements. Thus, the scan process consists of two sequential cycles: setting of monochromator to a needed wavelength (spectrum point) and measuring the photodetector signal at this wavelength. The spectrum is visualized simultaneously with the scan process.

During a measurement cycle the analoge-to-digital converter executes more than one readout in each spectrum point. The readouts are averaged that allows to improve the measurement accuracy. The time period between ADC readouts is 82 µsec approximately. The minimal measurement time in each point of a spectrum is 1 µsec. Thus, even with the shortest measurement time, the ADC executes 12 readouts in each point of spectrum. The longer time of measurements, the greater number of ADC readouts (linear dependence), the closer the averaged value of a detector signal to an actual value, i.e. the higher accuracy of measurement . It is possible to change the measurement accuracy by changing time of measurement. In order to obtain the desired accuracy of measurement, the set measurement time for detection of low signals has to be longer that that for the detection of large signals. The absolute value of signals, detected at different measurement time, will not change but accuracy of measurement will be increased. Taking into account this specific feature of measurement algorithm, the time of measurement in each point of a spectrum is named as “averaging time”.

With such a measurement algorithm, the scan rate (speed) is not specified. The scan rate is generally set in nm/sec and is defined as a ration of the scan range (in nm) to the scan time (in sec) of the set range. The scan time for the same spectral range depends on the number of spectral points and the averaging time in each point, because a scan process consists of two sequential cycles, a cycle of monochromator tuning to a specified spectrum point and a cycle of measurement at this point. The minimum averaging time in each point of a spectrum is 1 msec, the maximum one is 5000 msec.

A minimal scan step is equal to that of a wavelength tuning step and depends on the grating parameters (number of lines per mm). For example, the minimum scan step of monochromator-spectrograph MS3501 is 0.01 nm (when using a diffraction grating with 1200 lines/mm).

As SOL instruments scanning monochromators utilizes two output ports to hold two detectors, the ADC board also has two inputs to get signals from two detectors. Each detector can be coupled with one of the ADC input ports.

Switching over the output ports of scanning monochromator is performed with the computer controlled output mirror. The output port which gets light from the output mirror is called active one. The detector mounted on the active port will be active too. If two detectors are in use, only one of them is active at the moment. The ADC board can receive a signal only from one of the detectors coupled with its active input port. The active input port of ADC can be selected (depending on the output mirror position) either automatically or manually.

Photomultiplier tubes (PMT) and photodiodes are integral detectors which are mostly used in the detection systems.

When using an integral detection system, full light that passes through the exit slit should reach the active area of photodetector. This is very important. The light diverges from the slit of monochromator at a definite spatial angle and it is not a parallel beam. Therefore, it would be best to put the detector directly behind the exit slit, where there is the smallest aperture of a beam. Unfortunately it can not be always realized in practice. Some of light detectors (e.g. PTA-928 or a photodiode with active area of 10х10 mm) have a large size of active area. The detectors based on such sensors can be mounted directly on exit slit of monochromator.

A special conjunction unit should be used to couple detectors with active area less than 5 mm. A toroidal mirror inside this unit transfers an image of the exit slit to the photosensitive area of detector. In this case, full light from the output of monochromator will fall to the active area of the detector without any losses.

We can calculate and produce a conjunction unit with a detector on request.

PMT is a detector which uses an external photoelectrical effect. It is a vacuum device with an internal amplification of photocurrent resulting to a secondary emission. PMT contains a photocathode, and a large quantity of secondary emitters (dynodes) with accelerating electrical potential between them. Electrons emitted by the photocathode under the action of light are accelerated by the electrical field and go to the first dynode. And, generated secondary electrons are also being accelerated and go to the next dynode, etc. The shape of a photocathode and dynodes, their relative positions, magnitude of accelerating fields are estimated so that all electrons emitted by a dynode (photocathode) should come to the next one. So, the low photocurrent, produced by electrons which emits by the photocathode under the action of light, is amplified by the dynode system to generate an output current signal on the collector (anode). The amplification coefficient is determined by the number of dynodes.

We offer two models of detector system on basis of PMT to be used with our scanning monochromators: PTA-928 (based on Hamamatsu R928 photomultiplier) and a high sensitive detector PTM-7844 (based on Hamamatsu sensor module H7844) having one internal hermoelectrical Peltier element and a forced air cooling Both detectors are designed for detection of light in the spectral range of 185 to 900 nm. Detectors are mounted directly on the exit slit of a spectral device.

Photodiode is a photoelectric detector with an internal photoeffect. The photoconductivity effect, i.e. amplification of electrical conductivity in semiconductors at phonon absorption, is used in such detectors.

A great variety of photodiodes is available depending on the semiconductor material. They differ in spectral range and in sensitivity. Photodiodes based on Si, Ge and InGaAs are most frequently used. These photodiodes are designed for operation in the spectral range from the UV to near IR (0,2 to 2,3 µm). InAs and InSb photodiodes or PbS, PbSe and HgCdTe photoresistors can be used for longer wavelengths (1,5 to 5,5 µm). Pyroelectric detectors or thermocouples are used in the far IR spectral range (2 до 40 µm). More information about the types of photodiodes we offer: here.

One of the principal disadvantage of a photodiode is its low sensitivity compared to that of a PMT. PMT, a photodetector with the internal amplification of photocurrent, offers sensitivity several orders larger than any photodiode. However, the high voltage (up to 2500 V) is required for PMT operation, what sometimes restricts its application. With photodiodes low supply voltage is required what allows design compact detectors.

The main advantage of photodiodes is their broad operational spectral range. The spectral range of PMT is determined by the photocathode material, and generally confirmed to the UV an VIS range (200 to 650 nm), and sometimes to the near infrared spectral range (up to 900 нм). Infrared PMTs (up to 1,3 µm) are expensive and their sensitivity is not high. Another advantage of a PMT is its fast response and low output capacitance what allows to design fast detectors for measuring of fast-rated processes.

Transmission / Reflection measuring system based on MS3504i.

Conjugation unit (mirror) with small size active pixel area photodetector.

With the use of a imaging detection system, the detector of light in mounted in the focal plane of a spectral device. The exit slit is not used in this case, while a diffraction grating is installed in a position that enables formation of the preset wavelength interval in the focal plane.

Linear or matrix multi-element photodetectors are used in imaging spectra detection systems.

A linear photosensor (detector) consists of a large number (up to several thousands) of photosensors arranged in a line. These photosensitive elements of the linear detector are often called as pixels (from the English word “pixel”, i.e. “the smallest imaging element”). The linear detector is mounted on the focal plane of a spectral device (spectrograph) so that the incident light (spectrum) spreads out along the line of pixels. Each pixel of a linear detector detects the light energy with a spectral band whose width is determined by the pixel width and the instrument dispersion. The signals from all photosensitive pixels of a linear detector are recorded simultaneously. If the measurement result of a linear detector signals is visualized as a plotted curve, the ordinate axes is giving the pixel signal intensity and abscissa axes is giving the pixel order number, then the obtained graph shows the wavelength dependence on the emitted light energy, i. e. the spectrum.

For example, the averaged linear dispersion of MS3501 monochromator-spectrograph with a 1200 lines/mm grating is 2.37 nm/mm. For spectrum detection with the use of HS 103H-2048/64 digital camera whose pixel width is 14 µm, the spectral bandwidth detected with one pixel will be 2,37 [nm/mm] x 0,014 [mm] = 0,033 [nm]. Considering that the number of pixels in this detector is equal to 2048, the detector wavelength interval will be equal to 2048 [nm] x 0,033 [nm] = 68 [nm].

For full spectrum detection, a spectrum needs to be recorded by sections. This is achieved by a successive rotation of a grating in a spectrograph with a definite step size and then detecting a spectrum at each step. Usually, the spectrum section, which is most informative for a user, is being recorded. For this purpose the gratings that allow to expand the wavelength interval, which is simultaneously detected by a linear detector, are to be used. For example, the averaged reciprocal linear dispersion of MS3501 spectrograph with a 150 lines/mm grating is 17 nm/mm. Thus, the range of wavelength is 487 nm with use of a digital camera HS 103H-2048/64 (17 х 0,014 х 2048 = 487).

The required and narrower wavelength interval can be selected from a full spectrum taken with such grating (sometimes it is called panoramic spectrum) and will be further detected in details with the use of a grating having a larger density of lines. In order to realize this detection method, spectrographs with changeable gratings are to be used. Spectrographs which use an automated turret with four gratings are the most suitable for such purpose. Gratings are changeable automatically with the PC commands.

Linear detectors are widely used with compact spectrographs that have fixed gratings. Unlike scanning spectrographs, there are no movable parts in compact spectrographs and the grating is installed at a certain fixed angle. Spectrum of the required spectral interval is formed in the focal plane depending on angle grating rotation and other parameters of a spectrograph. Spectrograph with the embedded registration system is called a spectrometer. SOL instruments offers several models of compact spectrometers of SL40 Duo series and S150 Duo series.

Alongside with the linear image sensors, area image sensors are used in the imaging detection systems in which pixels are arranged as an array with many lines called “rows”. The area image detectors can be used as linear detectors through the hardware summation of the required quantity of rows or by selecting the required quantity of rows on the array.

Detectors with area image sensors are mostly extensively applied in multi-track spectroscopy where imaging spectrograph is used for a spectral device. If the light from several light sources strikes the input of a spectrograph, the spectra from each light source will be generated in the focal plane. The spectra from several light sources can be recorded simultaneously by locating the area detector so that the spectra spread along the array rows, and using for visualization only the rows accessible for output light.

If ordering a spectrograph with a multi-element detector (linear or array) as a detection system, the latter comes wavelength calibrated. Control both of a spectrograph and a detector is effected by a computer through the specially developed spectroscopic software SpectraSP. The software calculates the multielement detector wavelength range with high accuracy according to the parameters of a spectrograph and a grating parameters for the output wavelength set the spectrograph.

When choosing detection system, the operational wavelength range of a detector and detection time should be considered.

The spectrum detection time of imaging detection systems, when full spectral wavelength interval is recorded all at once, is generally larger than that of integral detection systems with point-to-point measurements. For detection of relatively low level signal with the use of an image detection system, the exposure time should be increased. At the same time, if an integral detection system uses a high sensitive detector (for example, PMT) the similar low intensity light can be detected with high speed. In this case, the spectrum detection time for both detection systems can be comparable, especially when a spectrum is to be detected in the narrow wavelength interval.

Imaging detection systems have evident advantages over the integral systems at detecting spectra of fast processes, such as pulse laser radiation, for example.

The detectable wavelength range in most cases depends on spectral sensitivity of a detector, since spectral devices can generally provide rather wide output wavelength range.

Linear and area detectors based on Si structures provide spectral sensitivity in the range of 200 to 1100 nm (UV, VIS and near IR spectral ranges). These are most commonly used linear detectors because of their relatively low cost. Linear near-infrared detectors with spectral sensitivity of 900 to ~2500 nm are rather expensive. Linear detectors for others spectral ranges are either very expensive or are not produced at all. Therefore, one of the restrictions for use of linear detectors is their rather limited spectral sensitivity.

Integral detectors offer a considerably broader range of the spectral sensitivity. Depending on the type such detectors can detect light in the wavelength range of 190 nm to 40 µm.

If required, our specialists can help you to choose a registration system best suited for your task.

Raman spectrometer based on monochromator-spectrograph MS3504i.

Raman spectrometer with laser twizer based on monochromator-spectrograph MS7504i.