What is microspectroscopy?
What information is contained in the scattering spectra / what can I learn?
A number of different molecular, structural and mechanical modes in a sample can couple to photons (light) resulting in a scattering or absorption peak of the measured spectrum. These often correspond to quite distinct processes, and thus contain complimentary information. In certain cases, they can also be probed using time-domain methods if the processes are sufficiently slow or detection sufficiently fast. Here we however focus only on “wavelength” domain measurements, as these are what are currently offered at the Advanced Microscopy facility.
In life-science research the most common spectrally selective detection modality involves the use of different fluorescence-labels with spectrally distinct emission spectra. The emission spectra from fluorophores (or similarly the auto-fluorescent emission of certain molecules such as chlorophyll or NADH) is the result of so-called resonant electronic transitions in the respective molecules, and though often the emitted light is very bright and can be subject to variations in certain cases as a result of local environmental conditions (pH, temperature, etc.), it is also typically quite broad and washed out and contains only limited information beyond the abundance of the fluorescence species which can be correlated to its total intensity. In contrast, the dynamics of the fluorescence or polarization selective measurements may yield more interesting information on the local environment (see TRFM). In microspectroscopy very often the fluorescence signal is undesirable as it can mask that of the more structured and weaker label-free scattering processes.
The most popular label-free optical spectroscopy scattering technique used in the life sciences is “Raman Scattering Spectroscopy”. This involves the scattering from vibrational modes of different molecular bonds, and results in scattering peaks separated anywhere between one to several hundred nanometers in wavelength (one wavenumber to several thousand wavenumbers in cm-1) from the probing laser wavelength. The structure of the Raman spectra is generally quite complex, often requiring a multi-variate analysis as opposed to a more straight forward biochemical decomposition, due to the many different bonds and molecules present in the probed volume.
A particular region of that is very rich in information and usually of the most interest is the so-called “Raman fingerprinting” spectral range, which corresponds to spectral wavenumber shifts of ~1500-3000cm-1 from the probing laser frequency. This can yield information on the lipid and protein content in the probed volume.
Recently of increasing interest (especially in the pharmaceutical industry) is the so-called THz- or low-frequency Raman spectra which yields information on the lower energy collective vibrations within a molecule and occurs at spectral wavenumber shifts of ~1-200cm-1 from the probing laser frequency. This is however more challenging to measure due to its vicinity to the probing laser, requiring a spectrally very narrow probing laser and special (notch) filters.
Spontaneous Brillouin scattering is the result of scattering from inherent low energy thermal vibrational modes in a sample (so-called acoustic phonons). In a liquid or gel it is manifested by a single scattering peak due to the coupling of the light to so-called longitudinal acoustic phonons, which is spectrally separated by less than 1cm-1 wavenumber from the probing laser frequency. The position of this peak is directly proportional to the acoustic velocity of this mode, which in turn is proportional to the square root of the longitudinal storage modulus – which is a measure of how compressible (“stiff”) the sample is. The inherent width of the peak on the other hand is proportional to the lifetime of these acoustic phonons and proportional to the longitudinal loss modulus. Together the two comprise the complex longitudinal modulus and give information on the viscoelastic properties of the probed region. The longitudinal modulus itself is related to the Youngs (or Tensile) modulus via the Poisson ratio (which is a relative measure of how much a material would extend perpendicular to the direction of a given mechanical perturbation). The energy of the acoustic phonons probed in Brillouin spectroscopy in a typical biological sample is on the order of tens of meV, corresponding to a frequency on the order of tens of GHz. Thus, the derived Longitudinal Modulus is also a measure of the corresponding high-frequency (GHz) response of the sample and correspond to distinct mechanical relaxation mechanisms to those probed at lower frequencies. The elastic modulus obtained from a Brillouin scattering measurement thus differs from that obtain from an Atomic Force Microscopy (AFM) measurement in several ways and should be seen as providing complimentary as opposed to comparable information. The relevance of the Brillouin scattering derived modulus for biological processes is subject to active debate, discussions of which can be found here [T1] .
While the size of the modes in Raman spectroscopy are smaller than the molecules, the characteristic wavelength of the phonons probed in Brillouin scattering are on the order of 100-200nm – and should be thought of as a collective phenomenon – with the sample behaving as an effective media over this scale - as opposed to a single-molecule property.
Due to the very small shift of the Brillouin signal relative to the probing laser frequency, it is quite tricky to measure, since there exist no edge- or notch-filters sharp enough to spectrally separate the relatively strong elastic scattering at the probing wavelength of the laser and the Brillouin scattering peak. A distinct spectrometer meter with the a very high spectral resolution and finesse is required.