Basic operating principles of the FTS6000


The instrument is based around a Michelson interferometer, the theory and details of which can be found in most spectroscopy, optics, physical chemistry and standard physics text books for example:

J. M. Hollas, Modern Spectroscopy Wiley
E. Hect, Optics Addison-Wesley
P. W. Atkins, Physical Chemistry OUP
H. Benson, University Physics Wiley

Given here is a brief explanation of some of the specific features of the FTS6000 which may not be found in the references above, including:

Rapid Scan Operation

In rapid-scanning interferometry, the moving mirror is driven at a constant velocity by a linear motor under computer control to vary the optical path difference. The mirror physical speed is typically 0.16 to 0.64 cm/sec, producing optical velocities of twice that: 0.32 to 1.28 cm/sec. The mirror speed is monitored by an internal He-Ne laser, operating at 632.8 nanometers.

The He-Ne laser beam is also passed through the interferometer to its own detector, where it generates a reference signal which enables the spectrometer electronics to sample the interferogram at precise intervals. The laser detector, monitoring the He-Ne beam, observes a cosine wave of frequency 5 KHz for 0.16 cm/sec, and 20 KHz for 0.64 cm/sec mirror speeds. Thus, scan speeds are commonly referred to in terms of Hz, with 5 KHz usually employed with DTGS detectors, and 20 KHz for MCT detectors. The computer converts the interferogram into a single-beam spectrum by a Fourier transform.

The single-beam background spectrum is the spectrum of the source modified by the transmission characteristics of the beamsplitter and the response characteristics of the detector. Most grating Instruments (dispersive) are double beam spectrometers which measure the spectrum of the radiation that passes through the sample and automatically ratio it to the spectrum of the source. This results directly in a transmission spectrum.

FT-IR Instruments, however, first collect the spectrum of the source (single-beam background spectrum) and store it on disk. Then the sample is inserted in the instrument. The single-beam spectrum of the source, modified by the absorption due to the sample, is collected and ratioed against the single-beam background spectrum to obtain the desired transmission spectrum.

Conventionally, the infrared detector is sampled when the laser signal passes through zero; this position is called a zero-crossing. When the infrared signal is sampled at alternate zero crossings (undersampled), the spectral range over which data can be unambiguously acquired is affected. If data points are taken on every other zero crossing, the range covers 0 to 7900 cm-1. Sampling at every other zero crossing is called an undersampling ratio (UDR) of 2. In rapid-scanning, no signal-to-noise gain occurs in undersampling; a single scan requires the same amount of time regardless of how frequently the signal is sampled.

The rapid mirror movement modulates individual infrared frequencies. If the mirror moves at a speed v, the laser frequency is modulated at the indicated frequency, and the infrared radiation of frequency f is modulated at 2vf (f = 15800.8235 cm-1).

Mirror Speed (in cm/sec) Associated Laser Modulation Frequency (in KHz) IR Modulation Frequencies (in Hz) at 400 cm-1 IR Modulation Frequencies (in Hz) at 4000 cm-1
0.16 5 126 1260
0.64 20 504 5040

Infrared radiation produces an interferogram, which is decoded by the Fourier transform to generate a spectrum. The spectrometer keeps track of the position of the moving mirror at all times by monitoring the cosine wave generated by the laser detector, a process known as fringe counting. This enables interferograms to be co-added to increase the signal-to-noise ratio in the spectrum.

All the experimental conditions (source, sample, detector, etc.) should remain constant during a scan; otherwise artifacts will be present in the resulting spectrum. This has spurred the development of a high performance, step-scan interferometer, such as the 896, which is the core of the FTS 6000.

The dynamically-aligned air-bearing interferometer, and rapid scan turnaround algorithms, makes the FTS6000 capable of monitoring fast chemical events at better than 10 ms time resolution per scan. The figure below displays a series of scans showing the ignition of a butane cigarette lighter in the sample compartment of an FTS6000, taken 175 ms apart at 8 cm-1 spectral resolution, and are plotted on the same scale with the baselines offset for clarity.

The spectra show the ignition sequence of a butane, from top to bottom: the baseline spectrum before lighting, pre-ignition detection of butane (weak peak at 2,950cm-1), early ignition (weak butane peak at 2,950cm-1 and CO2 peak at 2,350cm-1), and complete ignition (strong CO2peak and water vapor).

Dynamic Alignment

The FTS 6000 incorporates a novel, patented method for dynamic alignment control and step scan operation. The He-Ne laser beam is expanded before passing through the interferometer (collinear with the infrared path, shown offset here for clarity). After being modulated by the interferometer the laser beam falls on three detectors, the electrical outputs of which are cosine waves. The alignment control electronics alter the position of the three piezoelectric actuators located on the interferometer fixed mirror to lock the phases of these three signals together. This process keeps the interferometer accurately aligned in real time, as the interferometer is scanning and also while it is idling between scans. This is known as dynamic alignment. Dynamic alignment effectively isolates the spectrometer from the ambient environment, such as external vibrations or long term temperature changes. And lastly, the dynamic alignment mechanism provides the basis for step scan operation of the FTS6000 spectrometer.


Step Scan Operation

Step scan operation is achieved on the interferometer with a combination of motion of the moving mirror and the fixed mirror.

The moving mirror is translated very slowly, while the fixed mirror is drawn back by the piezoelectric actuators, keeping the optical retardation constant. After a period of time, determined by the step scan rate, the fixed mirror is rapidly moved forward by the piezos, effecting a step in the optical retardation. The process then repeats.

Throughout this process dynamic alignment control is active, so alignment is never lost. Small errors in the motion and position of the mirrors are also compensated. The scan control passes information to both the fixed and moving mirrors. The moving mirror is large, and its drive mechanism (the coil and magnet) responds relatively slowly, while the fixed mirror mass is small and the piezos can respond rapidly. Errors in the mirror positions and velocities are corrected in two ways: low frequency corrections are applied to the moving mirror and high frequency corrections to the fixed mirror. The piezos allow the interferometer to "settle" rapidly (in less than 100 ms) after a step. If no piezo velocity and position corrections were done, the error correction bandwidth would be low and poor performance would result, with the system being very sensitive to ambient conditions (vibrations, acoustic noise, etc.) and having a long settling time (approximately 10 ms). With the piezos, the interferometer can be operated in step scan mode under normal laboratory conditions, placed on a normal bench. The servo control is AC-coupled and is immune from drift in the He-Ne laser output, the laser detectors, and the laser signal amplifiers. The long- and short-term position error in step scan mode is approximately 0.001 of a He-Ne laser fringe (RMS), approximately 0.6 nm. This value arises from the design and AC-coupling of the scan servo control.

The interferometer can be stepped as rapidly as 800 Hz (steps per second) or as slowly as one step every 250 seconds (0.004 Hz) with intermediate speeds differing by factors of two. The settling time is fast compared to all the step rates, giving a very high measurement time duty cycle. Both fringe counting and dynamic alignment are maintained during the forward step scan and the retrace; thus step scans are easily co-added, just as rapid scans are.

Why step scan? Rapid scan spectroscopy limits the analysis of dynamic events to the time it takes to scan the interferometer. Even with the rapid scan speeds available on the FTS 6000, the best time resolution that can be achieved is ~10 ms/scan. Many dynamic processes that researchers want to measure occur on time scales much faster than this, time scales which were typically unavailable to the researcher for FTIR analysis prior to the introduction of the step scan spectrometer. Step scan permits the separation of the dynamics of scanning the interferometer from the dynamics of the chemical event to be measured. Time domain studies for time resolved spectroscopy (TRS can be used to monitor transient chemical species to the nanosecond time scale.

Time Resolved Spectroscopy

Time Resolved Spectroscopy (TRS) is used to study the dynamics of repetitive events (events that occur on a time scale much faster than the rapid scan capabilities of the spectrometer). The FTS 6000 has two TRS options available: with the internal digitizer on the spectrometer electronics board a time resolution to 5 microseconds can be achieved; or to use a 12-bit digitizer in the spectrometer workstation for time resolution to 200 ns. From ferroelectric liquid crystals to biological reactions the TRS modules can be used to understand fast changing phenomena.

The process is diagrammed schematically in the left figure above. The spectrometer steps to optical retardation point Xn, where after a programmed settling time, the spectrometer triggers an external excitation event (typically a laser flash) to perturb the sample and initiate digitization to record the transient event. The sample relaxes back to its initial state, and the process is repeated throughout each step of the interferogram. The right figure below shows how the data is sorted. At each retardation point Xn, a complete time trace of the excitation event is recorded as a series of columns in this two-dimensional representation. After data collection is complete, the data is sorted in rows to yield time-resolved interferograms.

Attenuated Total Reflectance

In ATR (Attenuated Total Reflectance) the beam of the spectrometer is directed in to a prism at an angle which exceeds the critical angle. As the beam is directed in to the crystal at an angle which exceeds the critical angle, internal reflections take place. When a sample is placed in optical contact with the prism at the point at which an internal reflection occurs, the sample absorbs IR energy at wavelengths equivalent to those which would be noted in a transmission experiment.

It has been proposed that the internal reflection generates an evanescent wave which extends beyond the surface of the crystal into a sample held in contact with the surface. A property of this wave, which makes ATR such a useful technique, is the intensity of the wave decays exponentially with the distance from the surface. This distance which is often less than 2um makes ATR generally insensitive to sample thickness. The technique is readily applied to the analysis of strongly absorbing samples

Solid samples must be mechanically pressed in to contact with the crystal to achieve optical contact. The intensity of collected data is a function of optical contact, and the best data results from samples in intimate optical contact with the ATR crystal; that is, when the contact efficiency approaches 100%. When liquids are analysed by ATR, intimate optical contact is achieved readily. When solids and powders are analysed using internal reflection methods, the spectral intensity will be largely governed by optical contact.

The factors which effect spectral intensity include:

  • The number of reflections. With 100% contact efficiency, a multiple reflection system will give higher absorbance values than a single reflection system. Theoretically, the ratio will be that of the number of reflections.
  • The face angle of the crystal and the angle of incidence of the radiation. In general, this means the higher the angle, the less intense the spectrum, and vice versa. (Note it is still essential to exceed the critical angle and hence maintain internal reflection.)
  • The refractive index of the sample. The higher the refractive index, the less interaction exists between the sample and crystal and the less the absorbance. The lower the refractive index, the higher the absorbance.

The wavelength of the radiation. The depth to which the radiation penetrates the sample is proportional to the wavelength; that is as the wavelength increases, the penetration also increases. However, as most FT-IR users prefer to think in terms of wavenumbers, then the depth of penetration increases with decreasing wavenumbers.

Choosing an ATR crystal

To help select the material best for your application the table below contains some of the physical properties of different ATR materials. The penetration depths quoted are for a sample with a refractive index 1.4 at 1000cm-1.

material Germanium KRS-5 Silicon Zinc Selenide Diamond
Useful range:
maximum/cm-1 5500 20000 8300 20000 4500
minimum/cm-1 830 400 70 650 33
Refractive Index
at 1000cm-1
4 2.37 3.4 2.4 2.4
Penetration Depths:
at 45o/microns
0.65 1.73 0.81 1.66 1.66
at 60o/microns
0.5 1.06 0.61 1.04 1.04
    1500-360   2500-1667

As well as the material performance, compatibility is also essential; below is a (by no means exclusive) list of reactants which are incompatible with each of the materials, and also what they must be cleaned with.

  Incompatible with Clean with
Germanium Hot H2SO4 and aqua regia Alcohol. Acetone, water
KRS-5 complexing agents/water MEK
Silicon HF + HNO3 Alcohol. Acetone, water
Zinc Selenide Acids Strong alkalis Alcohol. Acetone, water
Diamond K2Cr2O7, conc H2SO4 Alcohol. Acetone, water