Spectroscopy

 
Back to main PHYS 317/318 page. Brief theory of the absorption spectra of gases

The study of atomic spectra is one of the oldest research topics in "modern" physics. Theoretical descriptions of atomic spectra may be found in a number of books dedicated to the topic, and introductory discussions are available in the beginning textbooks of quantum mechanics. The excellent book by Kuhn is at an intermediate level. Melissinos has an introductory section on atomic spectra. Descriptions of optical instruments may be found in Sawyer and in Jenkins and White.  A brief summary of the absorption spectra of gases is also included in these web pages.

Many of the features of atomic spectra can be explained using relatively simple quantum mechanics.  You should become familiar with the notation and terminology used to desceibe atomic energy levels.
 

A.    Procedure:

The apparatus to be used is shown schematically in Fig. 1.  The central component is a Jarrell-Ash quarter-meter spectrometer (or "monochrometer"), described in Fig. 2.  this instrument has fixed input and output slits, and the spectrum is scanned by rotating the grating.  Thanks to Mr. Jung '96, this is now done by a computer-controlled stepper motor.  The light which passes through the exit slit is collected by a photomultiplier and the output of the photomultiplier is measured by a picoammeter and then digitized in the computer (data aquisition and display added by Frank Hess in Summer '96.)  Operation of the computer program SPEC.BAS is described below



Figure 1: Schematic Diagram



Figure 2: Monochrometer



The dial on the monochromator is reasonably accurate, but improved accuracy should be obtained by using the mercury spectrum as a calibration.  Measure the (dial) wavelengths of the peaks appearing in the mercury spectrum (by the procedure below) and identify as many as possible with known lines (e.g. table in the Handbook of Chemistry and Physics.)  Use lines in second order diffraction, as well as first order.  Note that some lines are unresolved multiplets; choose what should be the strongest line.  Tabulate the differences (= correction.)  Fit a straight line or other simple curve, which can be used to apply the correction to other measured spectra.

Turn on the high voltage power supply for the photomultiplier, wait at least one minute, then turn on the high voltage switch.  Meanwhile turn on the computer, stepper motor power supply, picoammeter, and fileter.  All this before you start the scan, below.

The 486 computer contains a stepper motor controller card and an analog-to-digital converter card.  A program to control both cards for this experiment has been written in Quick Basic.  From the DOS prompt, type SPEC.  Click on Run in the menu bar, then Start.  The program then asks you for the settings of the spectrometer:

  1. High or Low blaze.  This tells the computer which of two gratings is in place.  Normally, use HighNOTE: Do not press Enter here.
  2. Current wavelength.  Type in the number visible on the dial inside the window associated with the blaze you have chosen.  Warning!  It is important to give the (approximately) correct setting for the blaze the computer thinks you are using.  Otherwise the computer will try to drive the grating beyond one of its stops, and may mess up the calibration, if not the drive mechanism.
You now get the operating menu (most of which you never need.)  With Scan highlighted, press enter.  Enter the low, then the high, end of the wavelength range to be scanned.  (By this point, the stepper motor power supply should have been turned on.)  The motor will run the spectrometer to the low end on the range (if the sound of the motor should change, due to grinding against a stop, turn off its power supply.)  Then the computer will ask for the scan rate.  Try 14 for a fast scan.  the spectrum will be displayed as it is recorded.

You can determine wavelengths on the screen, using the two cursors and the zoom feature.  You may (?) be able to screen print to get a hard copy.  You can Quit and then save the spectrum to a permanent file, which you can later import into a spreadsheet or graphics program or examine in the editor.  Note that a datum is recorded every 1/8 nm so the file can be quite long.  the spectrum can be restored to the screen using View last spectrum in the menu.

You can install another source and Scan without any further steps.
 

B.    Characterization of Common Atomic Spectral Sources.

  1. Using a mercury source, determine the corrections to be applied to the monochromator wavelength readings.
  2. Record spectra for H2, He, Na, and Rb.
  3. Identify the major lines in the 200 - 800 nm range.  You may want to do a slow scan over a limited range of particular interest.  Note that shorter wavelength lines may also appear in second order.
  4. Identify in addition to the strong lines, other medium strength lines which are seen (change picoammeter scale and re-scan, if necessary).
  5. Measure the relative heights of the various lines.  Applying, if possible, corrections for the phototube efficiency and the transmission of the spectrometer, make up a table listing the lines you have identified with their relative strength.  (Normalize to the strongest line).
  6. Make a Grotrian diagram for each source observed, identifying the transitions leading to each spectral line observed.  Relate to theory, in the cases of the simplest spectra (see below.)
 

C.    Other Spectra.

  1. Obtain Light-Emitting-Diodes of three or more colors.  Record their spectra and also measure the voltage drop across the diode at its operating current. Try to relate spectral peaks and the voltage drop to the energy gap of the semiconducting materials.  See, for instance, Kittel, Introduction to Solid State Physics.
  2. Study (an approximation to) blackbody spectra by recording the spectrum of a high-intensity lamp. Repeat (at the same gain settings) using a Variac to reduce the lamp voltage, and hence temperature.  This will require the corrections described in B.5 above.  There may be small irreproducible glitches due to the lamp.
  3. Use the white-light source to study absorption spectra of filters, such as the Hg light filters used in the Zeeman experiment.  Can you see absorption lines in a sodium vapor cell?
 

Useful Information

  1. The photomultiplier is an EMI 9855. A voltage of -600 should be adequate for part A. Never expose the photomultiplier to room light with the high voltage turned on or off.
  2. NO-NO'S:
  3. WARNING!  The mercury vapor lamp has a fused silica bulb which transmits dangerous UV radiation.  Do not expose to the eye unless ordinary glass (which cuts off beyond about 310 nm) intervenes.

References:

  1. F.A. Jenkins and H.E. White, "Fundamentals of Optics", McGraw-Hill.
  2. H.E. White, "Introduction to Atomic Spectra", McGraw-Hill.
  3. H. Kuhn, "Atomic Spectra", Academic Press.
  4. R.A. Sawyer, "Experimental Spectroscopy", Dover.
  5. A.C. Melissinos, "Experiments in Modern Physics", Academic Press.
  6. R.B. Leighton, "Principles of Modern Physics," McGraw-Hill.
 

Data on Atomic Spectra:

[All on Phys. Lib. Reference Shelf] 
 
  1. W.F. Meggers et al., Tables of Spectral Line Intensities, NBS monogram 145, Part I [QC453. M4 pt. 1]
  2. Wiese et al., Atomic Transition Probabilities, NSRO-NBS 22, vol. I [QC 783. W5 (1966)]
  3. Striganov and Sventilskii, Tables of Spectral Lines of Neutral and Ionized Atoms (Plenum Press, NY 1968)
 

Spectroscopy: Suggestion on Analysis of Monochromator Data

  1. Hydrogen. The exact solution is an elementary quantum mechanics problem and you should be able to make detailed comparison with theory.
  2. Helium. The low excited states involve one electron remaining in the ls state and one excited to a higher state. Thus the energy levels have a certain similarity to hydrogen levels, but there are more. For one thing, the two electrons may have their spins either antiparallel or parallel, giving rise to singlet and triplet levels.
  3. Sodium. Can you resolve the doublets? What is the resolution of our instruments? (Why are doublets double?) From your knowledge of chemistry and physics, what impurity line(s) would you most probably expect to see in this source, and is it present?
  4. Rubidium.  to a first approximation, the series are the same as for hydrogen, but with a correction (called the "quantum defect") to allow for the overlap of the valence electron wavefunction with the core.
 Estimate the center of a peak at half height.  If the top of the peak is cut off, you may have to measure at an unknown height.  this makes little difference if the peak is symmetrical.  Peaks may be asymmetrical because they are actually two or more overlapping peaks (handle on a case-by-case basis); because you are sweeping too fast (i.e., the sampling rate exceeds the low-pass cutoff frequency); or because of the properties of the spectrometer (the image of the entrance slit on the exit slit is intrinsicly curved).

Back to main PHYS 317/318 page. Brief theory of the absorption spectra of gases