UV-Visible Molecular Absorption Spectrometry
Mainly Chapters 7 and 13, a little of Chapter 6 and 14 also.
Based on an increase in electronic energy (Eelectronic) from the absorption
of a photon in the UV or visible region, as previously discussed.
Absorption spectrophotometry is the single most common method for
quantitative analysis of molecules, but…
Beer’s Law has its limitations (Section 13B).
1. Already know about linear range, for any instrumental method. In
the derivation of Beer’s Law, it is assumed that solutes act
independently. If the solute conc. is too high, then that is not a
good assumption and the absorptivity changes.
2. In the derivation of Beer’s Law monochromatic incident radiation
is assumed to irradiate the sample. If a source is used that puts out
light of all wavelengths (a continuous source), then Beer’s Law
does not apply.
To illustrate this 2nd point, consider a beam of incident radiation on an
absorbing sample consisting of 2 wavelengths
3. Stray Light: Minimize by
closing the cover! Cannot eliminate.
Noise in spectrophotometric measurements
UV-Vis Molecular Absorption Instrumentation (Section 13D + parts
of Chapters 6 & 7)
First – sources.
A continuous source that emits all wavelengths of radiation in the
region is used.
Mostly – Blackbody radiators. When a conducting solid is heated, it
will emit electromagnetic
1. The total amount of light
energy increases with
2. The spectral intensity
shifts to higher energies
The W filament lamp is the most common source for visible (and
near-infrared) spectrometers. (~2900K)
The W/halogen lamp can operate at a higher temperature, and can
be used as a UV source (if housed in quartz which does not absorb
UV light, rather than glass which does). ~3500K. They are also
superior visible sources.
The continuous spectral output of a blackbody radiator can be
understood qualitatively from the MO diagram of a solid
conductor. Compare this to the Atomic orbital diagram of Na
A more common UV source is the D2 lamp which provides a
continual spectral output from 190-400 nm.
LED for smaller instruments like the OOI spectrometer we have often
been using. LED does not produce a continuous output, so how does
it put out a continuous spectrum of white light?
If the spectrometer’s source emits a continuous range of wavelengths
of light, and the goal is to obtain a spectrum of absorbance as a
function of wavelength, there must be a way to select a wavelength, or
range of wavelengths, for sample irradiation.
Second – Wavelength selectors. Section 7C
Ideally – output from a wavelength selector is a single wavelength.
Reality – output from a wavelength selector is a band of wavelengths.
The narrower the band, the better the wavelength selector, the greater
the spectral resolution attainable.
Wavelength selectors can be
either filters or
monochromators. We will
exclusively. In any case the
effective bandwidth is one
important defining factor of a
performance, defining its
Monochromators can be
based on the use of prisms,
which work on the principal
of refraction, or gratings,
which work on the principal
You can see that a traditional monochromator consists of more
than a grating or a prism, but these are the parts of a wavelength
selector most important to selecting a band of wavelengths from a
continuous source. The next figure shows why diffraction gratings are
superior to prisms for most applications.
Now a few specifics about grating monochromators.
Function to disperse different wavelengths of light at different angles.
Angular dispersion: dr/dλ
Remember from earlier:
This would be a transmission grating,
where it was shown that the following
conditions result in constructive
nλ = d sinθ
where d is the distance between the holes
in the grating
Monochromators invariably use a reflection grating, where closely
spaced graves are cut out of a mirrored face.
In this case constructive
interference occurs when
nλ = d(sin i + sin r)
i is the angle of incidence, r is
the angle of reflection. d is
defined at left, and n is an
integer value (diffraction order)
Differentiation of the above
equation at a constant angle of
While the angular dispersion is important, the linear dispersion D is
more relevant, since it refers to the variation in wavelength along the
focal plane (AB in figure of grating monochromator)
Most important is the inverse of the linear dispersion, the reciprocal
linear dispersion D-1
This to a large extent determines the spectrometer’s resolution. The
ability to distinguish absorbances at different wavelengths close to
A grating monochromator’s resolving power depends largely on
d – the space between grating blazes
f – the monochromator’s focal length
n – the diffraction order
Say a conventional grating has 1200 blazes/mm and a focal length of
0.5 m. (Both of these numbers are pretty standard).
Find D-1 for n = 1
Find D-1 for n = 2
Find D-1 for a grating with 600 blazes/mm
Will a large D-1 or a small D-1 provide better resolution?
Why is spectral resolution important?
The higher the resolution the lower the S/N
The higher the resolution the greater the information content
Here is an illustration of vapor phase UV spectra of benzene at 2
Resolution will be revisited soon. First, a discussion of the last major
instrument component of UV-Visible molecular absorbance
Portions of Section 7E, detectors.
In the UV-Vis region we discuss 3 types, the first 2 are very similar in
principle but have very different performance characteristics.
1. Vacuum Phototube
2. Photomultiplier tube
Both of these are “photoemissive” devices. Shine light on them, and
they emit electrons.
3. Multichannel photon transducers. Shine light on these, and they
conduct electricity. Advantage of smaller size and can construct
instruments with a different design as a result.
The photoemissive detectors work on the principle of Einstein’s
photoelectric effect (Section 6C).
Here is a vacuum phototube.
The cathode is coated
with a low ionization
117 = K-Cs-Sb, S11 = Cs3Sb
Vacuum phototube: 1 photon > ionization energy of photocathode =>
1 photoelectron emitted.
Photomultiplier tube: 1 photon > ionization energy of photocathode
=> ~106 photoelectrons emitted.
PMT – much greater sensitivity. Much lower light levels give
In a PMT there is still a photocathode, but between that and the anode
are a series of dynodes.
If at each dynode 5 e- are emitted for each electron that strikes it, for
the 9 diode arrangement above the PMT current gain = 59
Dynodes are irreversibly damaged by high intensity light, which is
why we are paranoid when using the fluorescence spectrometer.
These 2 photoemissive detectors are fine, but they are large. The
usefulness of a very small detector will be shown after a brief
discussion of the basic principles of how one works.
Small photoconductive detectors are based on semiconductors.
Since each detector is about 25μm, a small instrument can fit many
detectors. This allows for a different type of instrument.
Types of Instruments – Section 13D-2
Single Channel instruments use a single large detector like a vacuum
phototube or a photomultiplier.
These single channel instruments can be single beam
These single channel instruments can also be double beam
The small semiconductor detectors afford the capability of building a
like the Ocean Optics single beam Array Spectrometer.
Now as promised, spectral resolution revisited.
First – single channel instruments (Section 7-C3), then multichannel
Spectral resolution is determined by the monochromator.
Thus far only the grating has been
discussed, and its D-1. To understand
resolution we must also think about
entrance and exit slits.
For a single channel instrument almost invariably:
The size of the entrance slit width = size of exit slit width
Different wavelengths are brought to the exit slit by rotating the
Consider the following scenario if we have monochromatic source input
into the monochromator, say from a Na vapor lamp.
1. Monochromator illuminated with line source λo = 589 nm
2. Entrance slit width = exit slit width = 1 mm
3. Monochromator D-1 = 20 nm/mm
D-1 = dλ/dy = range of wavelengths spread over the distance dy along
the exit slit focal plane.
1. When monochromator set to 589 nm, entrance slit image fills exit
slit – maximum signal intensity.
2. When monochromator set to 579 or 599 nm, entrance slit image
half fills exit slit – half of maximum signal intensity.
3. When monochromator set to 569 or 609 nm, entrance slit image
misses exit slit – no signal intensity.
This is the origin of the triangular slit function shown above.
Little sense in illuminating a monochromator with monochromatic
Consider the more realistic scenario
1. Monochromator illuminated with polychromatic light
2. Entrance slit width = Exit slit width = 1 mm
3. Monochromator D-1 = 20 nm/mm
Now – For every λ present there is a triangular distribution of
energies exiting the monochromator.
When the monochromator is set to 589 nm – what range of wavelengths
are passed through the exit slit?
Setting the monochromator to 589 nm:
100% of source power at 589 nm passes through
50% of source power at 579, 599 nm passes through
The sample is illuminated with a polychromatic band of light of various
wavelengths each with varying intensity!
To define the width of the wavelength band, we go back to the effective
bandwidth, FWHH, now also called the spectral slit width (S)
S = W x D-1 where W = physical slit width
Spectral slit width or effective bandwidth
defines spectrometer resolution
affects the applicability of Beer’s Law
affects the ability to acquire detailed spectral information
Earlier the deviations from Beer’s law due to polychromatic radiation
were addressed. If the monochromator’s effective bandwidth < 10% of
the width of the FWHH of the absorption band, then Beer’s law is
obeyed. Since absorption bands in UV-Vis absorption in liquids are so
broad, this is usually not a problem.
Spectral resolution with a polychromator (i.e. a multichannel instrument)
must be looked at somewhat differently.
Here the entrance slit width is fixed such that the image from the
continuous source illuminates the entire detector array.
Commonly, 1024 detectors are lined up in a 1-D array. [If the detectors
are 0.025 mm apart, this many detectors fits along a line of 25.6 mm
(about 1 inch)!]
Since there is no exit slit there is no slit function as with a single channel
instrument. If there are 1024 detectors over a given spectral range Δλ,
say 800 nm (200 nm 1000 nm) then the spectral resolution is given
by the range of wavelengths that a single detector is sensing.
Res. = 800 nm/1024 detectors = 0.78 nm/detector
The range of wavelengths distributed in a given distance along the focal
plane is still dependent on D-1.
Advantages of multichannel instruments: No moving parts, rapid
spectral acquisition (signal averaging), enhanced source throughput.
Disadvantage of multichannel instruments: spectral resolution is not
Finally a little on applications of UV-visible absorption spectrometry
The absorption of UV/visible light generally results from excitation of
bonding electrons. In organic compounds, useful transitions are n π*
and π π*. Compounds with useful transitions are said to contain
chromophores. Transitions below 200 nm are not useful (vacuum UV).
You would think that the wavelengths of absorption bands could be
correlated with the types of bonds and functional groups in a compound.
In theory that is correct.
In practical terms, UV-Vis molecular absorption spectrometry is almost
totally useless for qualitative analyses. By far the most important
application for this type of spectroscopy is for quantitative analysis.
Chapter 6 – Wave & particle properties of EM radiation.
Diffraction. Interaction of radiation with matter, absorption,
emission. Line spectra, continuous spectra, atomic/molecular
absorption and emission. Quantitative aspects of absorption and
emission measurements. Problems/Questions 1e,f,g,h,l, 2, 3, 7,
8, 9a, 13, 14, 15
Chapter 7 – General designs of optical instruments, UV-Vis
continuous sources, grating monochromators and performance
characteristics and resolution, UV-Vis detectors (radiation
transducers), optical Fourier transform spectroscopy.
Problems/Questions 1, 3, 4, 5, 8, 19a, 20,
Chapter 13 – Transmittance, absorbance, Beer’s Law and
limitations/deviations, slit widths and effects on spectra,
instrumentation (sources, types of instruments).
Problems/Questions 1, 2, 5, 7, 8, 9, 13b,g,h, 17, 18, 22, 23