Humboldt State University ® Department of Chemistry

Richard A. Paselk

Chem 432	Biochemistry Laboratory	Fall 2002
Lecture Notes:: February 21	© R. Paselk 1999

Characterization of Macromolecules

Introduction to Spectrochemical Methods

Instrumentation for Optical Spectroscopy

Instrument Components: want to look at common sorts of components found in many instruments, these include the: source, sample holder, wavelength selector, detector, signal readout. [overhead 40]

Optical materials: obviously transparent materials are needed for the manufacture of various types of components. The most common are silicate glasses (pass approx. 320-2000 nm), fused silica or quartz (pass approx. 180-3000 nm) and salts such as NaCl (pass approx. 200-15,000 nm) or KBr (pass approx. 200-30,000 nm).

• Radiation Sources
  • Continuous- sources giving all wavelengths within a spectral region.
    • Deuterium (D₂) lamps- give UV radiation (about 180-350 nm)
    • Tungsten (W) lamps- use resistively heated W-filament as source (about 340-2200 nm); now more common to see quartz/halogen lamps-still tungsten lamp, but halogen allows higher operating temp. so higher UV output (240-2500 nm). Note that these lamps have a radiant intensity which varies as V⁴
    • Xenon - (250-600+ nm) used in fluorescence etc.
    • IR Sources (400-40,000 nm)
  • Line Sources- sources giving discontinuous, discreet line spectra
    • Hollow cathode lamps- cathode of specific metal provide metal element line spectra (can be more than one element in lamp.
    • Gas discharge & metal vapor lamps (neon lamps, sodium lamps, etc.)
  • Lasers- monochromatic (bandwidths of 0.01 nm or less), coherent, very narrow beams with very high intensities possible, some are tunable, can operate continuously or in pulses (down to femtoseconds)
• Wavelength Selectors: Generally need a more or less narrow band of wavelengths for spectroscopic studies. Frequently in fact would like completely monochromatic light of a single wavelength. However no available source achieves this ideal. Get a distribution instead with a Gaussian-like shape. Define the bandpass or bandwidth as the width of this distribution at half-height {Sketch}. Two major types: filters, and monochromators.
  • Filters- only allow a restricted selection of wavelengths to pass. Two types:
    • Absorption- less expensive, use dyes etc. to absorb certain portions of spectrum. Generally have a broad bandpass (30-250 nm). A special type of absorption filter is the cut-off filter: T drops from about 100% to 0% over a short wavelength range. (Sometimes referred to as long-pass filters since they allow wavelengths from the cut-off to about 4000 nm to pass.)
    • Interference-operate by internal reflections and constructive/destructive interference. Set up a cavity (metal film/CaF or MgF dielectric/metal film) which has dimensions which is a multiple of the desired wavelength, all other wavelengths will then be rejected by destructive interference. Good filters stack a series of identical cavities to get better rejection. Note that the filter will allow multiple orders of light to pass that is light with multiples of the same wavelength. Thus these filters often use colored glass absorption filters as a substrate to block the unwanted orders. Note also that they are angle dependent, since the angle of incidence will affect the pathlength through the cavity!
      • The bandpass of these filters is determined by: l_max= 2th/n; where t is the thickness of the dielectric layer, h is the refractive index, and n is the interference order. (Why does the refractive index appear here? Recall that refractive index is a measure of how much light is slowed down in a medium, which in turn affects wavelength: l_vacuo= l_mediumh. And the change must be in the wavelength and not frequency, since E = hn and energy is conserved.)
  • Monochromators-allow the selection of a narrow band of wavelength over a considerable range. There are two types: Prism and grating. All monochromators have the following components (the second though fourth items are often combined by having a concave, focusing, grating) [overhead]
    • an entrance slit to provide an optical image
    • a collimating lens or mirror (makes light beam parallel)
    • a dispersing element (grating or prism)
      • Prism: a 60° angle gives the best dispersion. Generally of glass, quartz, or fused silica. Note that prisms give higher throughput than gratings, but are more expensive, and give a non-linear dispersion (higher in the UV region), so harder to produce instruments with constant bandpass. Not very popular today, see mostly in older instruments.
      • Grating: disperse light via positive and negative interference. Basically the grating creates a whole series of light sources which are allowed to interfere. At any given angle only a single wavelength (and multiples of it) will be positively reinforced [overhead] Generally use a replica grating made by casting plastic onto the original and silvering.
        • Traditionally used a scribe to create lines on glass or metal to create the master grating. Now can also use laser interference patterns to photo-etch a holographic grating. Generally holographic gratings have better perfection with respect to line shape and dimensions giving lower stray radiation and fewer "ghosts."
        • Most common is the Echellette grating where reflections off the larger faces combine to give the spectrum. For UV commonly have 1200-1400 grooves/mm, IR 100.
        • Gratings can be made concave to give optical focusing, thus eliminating the need for collimating and focusing lenses in the monochromator.
    • a focusing element (lens or mirror) which focuses the image of the entrance slit on the exit slit
    • an exit slit
      • Slits-generally the same dimension (often the same slit in folded designs). Optimize between resolution and intensity.
    • Note the resolution of a monochromator will depend on (given a high quality dispersing element) the slit width and the pathlength from slit to slit.
  • Radiation Detectors-devices which indicate the presence of radiant energy by creating some sort of signal. Ideally want signal to be proportional to the radiant energy: s=signal=k P, where P=radiant power. In real instruments often haves=K P+k_d, where k₂=dark current. Two basic types of detectors:
    • Photoelectric (Quantum) detectors-respond to individual events (photons), either by releasing an electron, or by enhanced conduction. Limited by shot noise (junctions).
    • Thermal Detectors-respond to average power, widely used in IR work. Limited by thermal noise.
    • Photon detector types:
      • Photovoltaic cells-Sense, 380-780 nm, rugged, low cost, no external power required. However, the signal is difficult to amplify so low sensitivity, they exhibit fatigue, and have a slow response.
      • Phototubes- [overhead] Sense from 150-1000 nm. Work by photoelectric effect. Have a sensitive cathode, such as Na metal, which readily releases photo-electrons. Then have an anode at about +90 (relative) so electrons will be drawn to it, generating a current. Easy to amplify, quick response, high resistance. But frequently produce significant dark currents from thermal electrons and ⁴⁰K decay in glass envelope.
      • Photomultiplier tube (PMT)- [overhead] similar to the phototube, but amplify with an electron cascade using a dynode. Get amplification of 10⁶-10⁷ electrons/photon. Fast response, with cooling (liquid He) can get single photon detection (photon counting), but can be irreversibly damaged by hi-intensity (room level) light.
      • Photodiodes- [overhead] Si diodes sense 190-1100 nm. Higher sensitivity than phototubes, less than photomultipliers. Can be manufactured in an array to detect many signals simultaneously.
      • Charge- [overhead] transfer Device (CTD) detectors- Sense 170-800 nm. Performance approaches PMT's in sensitivity and dynamic range, but multichannel, usually manufactured in 2-D arrays. (basis of CCD astronomical detectors, video cameras etc.)
    • Thermal types: IR photons don't have enough energy to dislodge photoelectrons so detect by the heat they generate. Have a blackened surface which absorbs heat photons, raising temperature and resulting in an increase an electrical signal. Must measure temperature differences of a few milli °C.
      • Theromopile-tiny thermocouples, generate a voltage when heated.
      • Bolometer- a conductor which changes conductivity with temperature.
  • Signal Readout devices.

Spectroscopic Instrument Designs

Typical Instrument Designs: [overhead]

• Photometers: select wavelength with filters.
• Spectrophotometers: prism or grating based instruments; can select narrow bandpass beam of radiation over a broad wavelength range. Discuss three examples, noting the components, and the important parts comprising these components (e.g. slits, focusing elements and dispersing element in a monochromator), in each:
  • Single-beam (Spectronic 20) [overhead 42]
  • Double-beam (Cary 14 & Perkin -Elmer) [overhead]
  • Multichannel (HP 8452A) [overhead]

Molecular Spectroscopy

UV-Vis Absorption Spectroscopy: First want to look at the species which absorb radiation and the phenomena involved. Let's go back for a moment and review the variations seen in absorption spectra with environment. [overhead]. This figure appears to show the fine structure for a single electronic transition. Spectra have greatest resolution (peaks are sharpest and narrowest) when the molecules behave independently (as independent resonators) so that there are limited numbers of rotational and vibrational states available to them. Thus get pretty good resolution in gas phase. In a non-polar solution phase the rotation of the molecules is obstructed, and the vibrational lines are greatly broadened by jostling in the solution. Finally, in polar solutions, the interactions with rotating dipoles of the solvent spread out the vibrational levels by creating many new ones and we see a very much broadened spectra. These types of broadening are pretty general, and will be seen for most species.

 

Let's look then at some particular types of species and their expected spectral characteristics.

• Organic compounds: [overhead] Absorption between 180 and 780 nm generally results from bonding electron transitions (p Æ p*, sigma bonds are too strong to show transitions in this range, while nÆ s*{non-bonding transitions} occur in the range 150-250 nm, but with a low value of epsilon, so the transitions are not favorable), or transitions for electrons localized on non-carbon atoms (O, S, N, X). Note that conjugation of double bonds generally moves the spectral line to longer wavelengths (lower energies, that is there are more energy levels more closely spaced), and increases epsilon (the transition is more likely).
• Inorganic compounds:
  • Transition metal ions and complexes of Periods 4 and 5 often absorb due to transitions of d -electrons (dsp hybrids for complexes). The energy differences, and thus wavelengths corresponding to these transitions, depend on the liganding species, the oxidation state of the ion, and its placement on the Periodic table. [overhead]
  • Lanthanide and actinide inner-transition series elements exhibit electron transitions due to f -electrons. Since these are inner-shell electrons, they are relatively shielded from environmental effects and thus show fairly sharp spectra more reminiscent of gas-phase spectra. [overhead]

UV-Vis Absorption Spectroscopy

• Charge-transfer complexes: These consist of an electron-donor group bonded to an electron-acceptor group. Absorption of radiation results in the electron being transferred to an acceptor-associated orbital. The process thus resembles an internal redox reaction. These transitions are often very favorable, leading to high epsilon values (>10,000) and thus high sensitivity, so folks like them as analytical reagents. In most complexes involving a metal ion, the ion is the electron-acceptor. [overhead]

Qualitative Absorption Spectroscopy: UV-Vis spectra are of limited value for qualitative identification because of the limited amount of information in these spectra. For qualitative spectroscopy want a narrow bandpass to get sharp peaks and good resolution. Note how peaks not only narrow, but also gain in magnitude as the bandpass narrows. Generally the bandpass is limited by the increasing noise vs. signal strength as the slit narrows. Solvents must generally be transparent throughout UV-Vis region. Note also the shift to longer wavelengths (a few nm) and higher absorbance that occurs in going from polar to non-polar solvents.

 

Quantitative Absorption Spectroscopy: Extremely wide usage. Some characteristics:

• Wide applicability: many species absorb light, or can react with reagents to make light absorbing species. Perhaps 90% of clinical analysis involve UV-Vis absorption.
• High sensitivity: Typical detection limits range from 10^-4-10^-5M. And can get to 10^-7M.
• Good selectivity: Can often select analyte by looking at particular wavelengths which are specific to analyte.
• Good accuracy: relative errors of 1-5%. Can get down to tenths of % with special procedures.
• Convenience: Methods are usually easy and readily automated.

As we have seen in lab, usually do a series of standards and create a standard curve for finding unknown concentrations. This has the advantages of assuring us that Beer's Law is in fact being followed (no curvature in plot of A vs. c), and the best fit line, whether found by a least-squares fit or just eyeballed, gives an averaged value for standards and is thus better than any single value. Note , however, that for a procedure in which Beer's Law is known to be followed (in your laboratory), can just use Beer's Law to calculate unknown on basis of a single standard or epsilon, if known.

 

Standard-Addition Method: A particular problem which can occur in real analysis is matrix effects. That is substances in the unknown can interfere with the absorbance of the unknown by interference with complex formation etc. Can compensate if you know what is there, but usually of course you don't. One way to deal with this situation is to include the unknown and its matrix in al of the standards. That is the standard curve is made by adding known volumes of a standard to a series of unknown samples all of the same volume, and then all samples are made up to the same volume in volumetric flasks. The concentration of the unknown will then be related to the intercept of the resulting best-fit line. Thus the concentration of the unknown, c_x may be found from:

where b is the y-intercept, c_s is the concentration of the standard, m is the slope for the line, and V_x is the volume of the unknown. Of course m and b are readily found via a least-squares analysis.

 

Analysis of Mixtures: Can do simultaneous analysis of mixtures by looking at the absorbances of standards for each unknown substance and for the unknown mixture at a series of wavelengths (one absorbance wavelength for each unknown substance). Can then find the concentration for each unknown by solving a series of simultaneous equations. This analysis can be readily accomplished for many two component mixtures, but becomes difficult for more complex mixtures. However, computerized systems can do much more complex mixtures (I am familiar with up to at least 8) by "over determining" the system, that is get more than the theoretical wavelengths--look at complete spectra.

---

Laboratory

C432 Home

Lab Lecture Notes

Last modified 25 April 2002