Introduction to optical spectroscopy Chemistry 243.

Introduction to optical spectroscopy

Chemistry 243

Fundamentals of electromagnetic radiation

34Planck's constant 6.626 10 J s

=frequency in Hz

E h

h

8 mspeed of light 3.00 10 s=wavelength

c

c

(s-1)Eswavenumbercm

11

hE c

Electromagnetic spectrum

http://www.yorku.ca/eye/spectrum.gif

HighEnergy

LowEnergy

Terminology Spectroscopy is the study of the interaction of

light and matter NMR or X-Ray spectroscopy; spectroscopist

Spectrometry is the establishment of the pattern of interaction (as a function of energy) of light with particular forms of matter Mass spectrometry (MS); spectrometrist

Spectrophotometry is the quantitative study of the interaction of light with matter UV-Visible spectrophotometry (I’ve never heard anyone called a spectrophotometrist)

What chemical and/or material properties can we measure using spectral methods? Broad and powerful applications Elemental composition (often metals; CHNO) Identity of a pure substance (what is it?) Components of a mixture (purity?) Amount of a substance in a mixture (how much?) Bulk/major component, minor component,

trace component, ultra-trace component Surface composition Material property (stress/strain, polymer cross-

linking, change of state, temperature) Reaction rate, mechanism, products

What properties of incident or generated light can we measure?

Absorption Fluorescence (fast) & Phosphorescence (slow) Thermal Emission Chemiluminescence Scattering Refraction or Refractive Index Polarization, Phase Interference/Diffraction Coherence Chemistry consequent to the above

What atomic/molecular properties affect or are affected by light?

Rotation (typically refers to a molecule) Vibration (typically refers to a molecule) Electronic Excitation (atomic or molecular) Ionization (loss of electron to yield a cation) Combinations of the above:

Rotation-vibration (infrared/Raman) Rotational, vibrational, electron excitation (UV-Vis) Ionization with UV absorbance (strong excitation)

The properties you want to studyhelp to select a suitable wavelength

High Energy

Low Energy

Why wavenumber?

The energy difference between two wavenumbers is the same regardless of spectral region or λ

Wavelength is not proportional to energy; it has a reciprocal relation to energy, so:

The energy difference between two wavelengths (in nm or angstroms) varies as a function of spectral region.

11 cmunit

)(

hcE

hc

hc

hE

c

E1

Selecting the right optical method

Emission

Plasma, flame, or chemical

Focus

Sorting of Energy,

Space, and Time

Detection

Computer control enhances and optimizes the info extracted from each

instrument component.

Excitation Source

Chemiluminescence is emissioncaused by a chemical reaction.

Fluorescence is emissioncaused by excitation

Absorption

Light Source Focus Specimen Focus

Energy, Space, and

Time SortingDetection

Transmissionand/or

Reflection canalso occur

Nearly linear light path geometryfor multi-wavelength,

simultaneous light detection

Relaxation is non-radiative;sample warms up a bit via vibration and rotation

Ab

so

rba

nc

eWavelength (λ)

Fluorescence (fast) & Phosphorescence (slow)

Light Source

(Laser)

Focus

Focus Detection

Specimen

Energy, Space, and

Time Sorting

May include energy sorting

Typical geometry 90°, but angle variable

Em

issi

on

Po

wer

Radiative

Raman Scattering

Light Source

Laser

Focus

Focus Detection

Specimen

Energy, Space, and

Time Sorting

Typical geometry 90°, but angle variable

Same geometrical layout as fluorescence and phosphorescence,… But what happens is not the same as absorption or emission

Raman Scattering

Elastic scattering: Eex = Eout

Inelastic scattering: Ein < Eout and Ein > Eout Eexcitation

Eex +E-E

virtual statevirtual state

Emission

Flame, plasma, chemistry

Absorbance(UV/Vis or IR)Lamps, LEDs

Fluorescence/Phosphoresence

Lamps, LEDs, lasers

lasers

Raman scattering

Different classes of optical spectroscopy

Classes of light sources

Light sources:Common examples

Blackbody radiation Light emitting diode (LEDs) Arc lamp/hollow cathode lamp Lasers

Solid-state Gas/excimer Dye laser

Thermal excitation Combinations (laser to vaporize

sample leading to thermal emission)

Continuum spectra and blackbody radiation

6

6

6

2.898 10 K nm

2.898 10 K nm9.82 m

295 K

2.898 10 K nm9.35 m

310 K

peak

roomtemp

human

T

A solid is heated to incandescence It emits thermal blackbody radiation in a continuum

of wavelengths

Skoog, Fig. 6-22

High E = Low λ = High T

T

bblackbodypeak Wien’s

Law

b is Wein’s displacement constant

Continuum spectra and blackbody radiation

http://en.wikipedia.org/wiki/Image:Blackbody-lg.pnghttp://en.wikipedia.org/wiki/Black_body

T ≈ 1200° CT ≈ 1473 K

Continuum sources Common sources

Deuterium lamp (common Ultraviolet source) Ar, Xe, or Hg lamps (UV-vis)

Not always continuous; spectral structure possible

http://www1.union.edu/newmanj/lasers/Light%20Production/LampSpectra.gifhttp://creativelightingllc.info/450px-Deuterium_lamp_1.png

Light emitting diodes (LEDs) First practical visible region LED

invented by Nick Holonyak in 1962 (GE; UIUC since 1963) “Father of the light-emitting-diode”

http://en.wikipedia.org/wiki/Nick_Holonyakhttp://upload.wikimedia.org/wikipedia/commons/7/7c/PnJunction-LED-E.PNG

http://www.pti-nj.com/images/TimeMasterLED/LED-spectra_remade.gif

An LED is a semiconductorwhich emits electroluminescence

Light emitting diodes (LEDs) Cheap, low energy, long-lasting, small, fast Commonly used in display screens, stoplights,

circuit boards as state indicators Lots of colors Infrared LEDs used in remote controls

http://en.wikipedia.org/wiki/File:Verschiedene_LEDs.jpg

Line (emission) sources

Continuous wave Hollow cathode discharge lamp Microwave discharge Flames and argon plasmas

Pulsed Pulsed hollow cathode Spark discharge

All these are non-laser

A line source is a light sourcethat emits at a narrow wavelength

called an emission “line”

Lasers

Light Amplification byStimulated Emission

of Radiation

• Intense light source• Narrow bandwidth (small range λ < 0.01 nm)• Coherent light (in phase)

Lasers

Light Amplification byStimulated Emission

of Radiation

• Pumping• Spontaneous Emission• Stimulated Emission• Population Inversion

Laser design

Lasing medium is often: • a crystal, like ruby• a dye solution• a gas or plasma

A photoncascade!

Skoog, Fig. 7-4

Pumping

Generation of excited electronic states by thermal, optical, or chemical means.

Skoog, Fig. 7-5

Spontaneous emission or relaxation Random in time No directionality Monochromatic (same λ), but incoherent (not in phase) Solid vs. dashed line – 2 different photons

Skoog, Fig. 7-5

Stimulated emission

The excited state is struck by photons of precisely the same energy causing immediate relaxation

Emission is COHERENT Emitted photons travel in same direction Emitted photons are precisely in phase

Skoog, Fig. 7-5

Population inversion

When the population of excited state species is greater than ground state, an incoming photon will lead to more stimulated emission instead of absorption. Inverted population

Normal population distribution

Pexcited > Pground

Pexcited < Pground

Skoog, Fig. 7-6

3- and 4-state lasers

Population inversion easier in 4-state system

Skoog, Fig. 7-7

Things stackup here.

Populationinversion easily

achieved.

Populationrelatively low

down here

Laser design

Lasing medium is often: • a crystal, like ruby• a dye solution• a gas or plasma

A photoncascade!

Skoog, Fig. 7-4

Continuous wavelaser sources Nd3+:Yttrium aluminum garnet (YAG: Y3Al5O12)

Solid state 1064 nm, 532 nm, 355 nm, 266 nm

The GTE Sylvania Model 605, uses a Nd-YAG laser rod set in a "double elliptical“ reflector, is pumped by two 500-W incandescent lamps, and is limited to a low order mode by an aperture in the laser cavity.

Continuous wavelaser sources Helium-Neon (HeNe)

Gas, but emission comes from generated plasma (very excited state atoms)

632.8 nm, 612 nm, 603 nm, and 543.5 nm; 1.15 & 3.39 μm Emission lines all the way out to 100 μm

99.9% reflective

99% reflective

Continuous wavelaser sources Ar+

Gas laser, but emission comes from ions Uses lots of electrical power to generate ions

351.1 nm, 363.8 nm, 454.6 nm, 457.9 nm, 465.8 nm, 476.5 nm, 488.0 nm, 496.5 nm, 501.7 nm, 514.5 nm, 528.7 nm, 1092.3 n

Coherent Innova 90Up to 5 W of output!~100x my laser pointer

Other continuous wavelaser sources Cu vapor

520 nm HeCd

440 nm, 325 nm Dye lasers

Pulsed lasers sources Nd:YAG

Solid state Often nanosecond pulses 1064 nm, 532 nm, 355 nm

Ti:sapphire Solid state—often pumped by Nd:YAG Tunable output aroudn 800-1200 nm Produces femtosecond pulses

Nitrogen Gas 337 nm

Excimer lasers (gas mixtures; excited state is stable) Tunable dye lasers (λ is selective within limits)

Laser diodes

Used in CD and DVD players (not very strong)

Wavelengths now available from IR to near UV regions

Band gapenergy, Eg

Skoog, Figs. 7-8 & 7-9.

ResonantCavity emitsAt 975 nm

Tip going forward

Keep your variables straight v for velocity or for frequency

Microsoft equation editor gives:

I will use m for integer, textbook uses n Easy to get mixed up with refractive index, n

(m/s) vee

(1/s)nu

v

Properties of electromagnetic radiation

Transmission Refraction Reflection Scattering Optical Components Interference Diffraction

Properties of electromagnetic radiation

tAy

tAy

2sin

2

sin

y = magnitude of the electric field at time tA = ymax – also called the amplitude of yν = frequency in s -1 (cycles per second)φ = phase angle (an offset relative to a reference sine wave)ω = angular velocity in radians/sec (a handy definition)

Recall: π radians = 180 degrees

Constructive Interference - In Phase

-7

-6

-5

-4

-3

-2

-1

0

1

2

0 2 4 6 8 10 12

Time

Am

plit

ud

e

A

B

A+B

B is in phase with A

Interference – magnitudes add or subtract

Destructive Interference - Half Wave Shift

-7

-6

-5

-4

-3

-2

-1

0

1

2

0 2 4 6 8 10 12

Time

Am

plit

ud

e

A

B

A+B

B is 180 degrees (π radians) shifted from A

Interference – magnitudes add or subtract

Interference -- Quarter Wave Shift

-7

-6

-5

-4

-3

-2

-1

0

1

2

0 2 4 6 8 10 12

Time

Am

plit

ud

e

AB

A+B

B is 90 degrees (π / 2 radians) shifted from A

Interference – magnitudes add or subtract

Interference between waves of different frequency

Wave 1 + 2

21

Skoog, Fig. 6-5

Transmission through materials Compared to vacuum, the velocity of light is reduced

when propagating through materials that have polarizable electrons. Wavelength also decreases All electrons are polarizable to some extent

ii

cn

v

Skoog, Fig 6-2.

mediumc

constant

hE

cvacuum = 2.99792 x 108 m ● s-1

Index of Refraction

Refractive index is measure of how much light is slowed:

Refractive index is wavelength- and temperature-dependent for many materials:

1.46Quartz

1.49Toluene

1.43Hexadecane

1.58Glass(light flint)

1.33Water

1.00Vacuum (air)

n @ 589.3 nmMaterial Wavelength-dependence of nSiO2

http://www.rp-photonics.com/refractive_index.html

nwater vs Temperature

1.316

1.318

1.32

1.322

1.324

1.326

1.328

1.33

1.332

1.334

1.336

0 20 40 60 80 100

temperature (oC)

refr

ac

tiv

e in

de

x

lengthgiven wave aat velocity

1)(lengthgiven wave aat index refractive

i

i

vacuumi

v

v

cn

Refraction Snell’s law:

Oil immersion lenses for high magnification microscopy

2

1

1

2

2

1

sin

sin

v

v

n

n

Velocities, not frequencies

Medium 1Medium 2

Here, n2 > n1

Skoog, Fig. 6-10

For your information …

Book Error on page 141, equation 6-12:

2 medium invelocity

1 medium invelocity

sin

sin

2

1

1

2

2

1 v

v

n

n

This is correct: Snell’s Law of Refraction

Reflection Amount of loss at a reflection increases with refractive

index mismatch. For right angle light entrance into a medium:

Reflective loss is angle-dependent Fresnel equations (which we will skip) Most important case is: total internal reflection

22 1

20 2 1

r n nI

I n n

1

intensityincident

intensity reflected

Total internal reflection Light incident upon a material

of lesser refractive index is bent away from the normal so that the exit angle is greater than then incident angle. At the critical incident angle,

the exit angle is 90° - beam does not exit

Angles larger than the critical incident angle lead to total internal refection (TIR)

exitcriticalentry

exitcriticalentry

exitentry

nn

nn

nn

nn

sin

)90sin(sin

sinsin

sinsin

21

2211

Medium1

Medium2

nentry

nexit

1

2

Modified from Skoog

nentry > nexit

θ2 > θ1

When this is true, θ1 = critical entry angle for TIR

Total internal reflection

TIR fluorescence microscopy:

http://hyperphysics.phy-astr.gsu.edu/Hbase/phyopt/totint.htmlhttp://www.olympusmicro.com/primer/techniques/fluorescence/tirf/tirfintro.html

exitcriticalentry nn sin

When this is true,θ1 = critical entry angle for TIR.

If θ1 > θcritical result is TIR.

Evanescent wave samples a very narrow slice of the sample very near

to the dielectric interfaceTypically ~200 nm

Total internal reflection

TIR fluorescence microscopy:

exitcriticalentry nn sin

When this is true,θ1 = critical entry angle for TIR.

If θ1 > θcritical result is TIR.

http://hyperphysics.phy-astr.gsu.edu/Hbase/phyopt/totint.htmlhttp://www.olympusmicro.com/primer/techniques/fluorescence/tirf/tirfintro.html

Good for studying adhered cells; low background

Fiber optics Extruded strands of glass

or plastic that guide light via total internal reflection. Core has higher refractive

index than cladding. Flexible Material choice allows

transmission in UV, visible, or IR

Skoog, Fig 7-39.

Follows all the rules ofSnell’s Law

Scattering

Raman scattering Inelastic scattering

offset from by frequency of molecular vibrations Rayleigh scattering

Molecules or aggregates smaller than l Intensity ~ 1/l4

Mie scattering Particles large (or comparable) to l Used for particle sizing

Essential optical elements

Lenses Mirrors Prisms Filters Gratings

Basic optical components Mirrors

Reflection

Concave mirror is converging

Convex mirror is diverging

Prisms Refraction

Snell’s Law

1 1 2 2sin sinn n

Filters Absorption filters

Cheap, visible region; colored glass

Cutoff filters – long-passshort-pass

Interference filters2

cos

2

cosinterger

air

air

dm

n

dn

mm

Skoog, Fig. 7-12

θ is usually zeroso, cos θ = 1.

Also, m is usually 1

d = thickness of dielectric layern = refractive index of dielectric medium

m = integerʹ = wavelength in the dielectric material

Interference filters Almost

monochromatic

Skoog, Fig. 7-13

Bandwidth of a filter is width at half-height

(aka full-width @ half-max)

Diffraction of coherent radiation: Interference at work

Consequence of interference

Skoog, Figs. 6-7, 6-8

integer

sin

m

dm constructive

destructive

constructive

constructive

destructive

d = distance from slit B to C

Distance x to y is one λ

(m is the order of interference)

m is: • 0 for E• 1 for D

m used here, text uses n

Diffraction of coherent radiation: Interference at work

Consequence of interference

integer

sin

m

dm

(m is the order of interference)

OD

DEBC

dm

sin

m used here, text uses n

You can now determine thewavelength of light

based on things thatare easy to measure!

Skoog, Fig. 6-8

Monochromators

Used to spatially separate different wavelengths of light: prisms, gratings

Czerny-Turner grating monochromatorBunsen prism monochromator

Skoog, Fig. 7-18

Gratings and monochromatorsReflection + diffraction: echellette-type grating

sin

sin

sin sin

m CB BD

CB d i

BD d r

m d i r

Skoog, Fig. 7-21

The condition forconstructive interference.

The m = 1 line is most intense.

The surface contains “grooves” or “blazes”.

Take a look atExample 7-1,

Page 184.

Monochromators

Used to spatially separate different wavelengths of light: prisms, gratings

Czerny-Turner grating monochromator

Skoog, Fig. 7-18

Useful metrics for monochromators Dispersion (page 185); high dispersion is good

Integration of at constant i gives the angular dispersion:

Linear dispersion, D, is the variation of λ along the focal plane position, y:

Reciprocal linear dispersion, D-1:

sin sinm d i r

cos

dr m

d d r

focal length

dy f drD

d df

1 cos for small r

d d r dD

dy mf mf

r = angle of reflectiond = distance between blazes

More useful, results inD-1 in nm per mm

or similar

A measure ofthe ability to

separate wavelengths

Useful metrics for monochromators—continued Resolving power (R; unitless)

Limit of monochromator’s ability to distinguish between adjacent wavelengths.

Light gathering power (f-number, F; unitless) Collection efficiency—improve for maximizing S/N Efficiency scales as the inverse square of F

focal length of collimating mirror or lens

diameter of collimating optic

F f d

f

d

dilluminate blazes grating ofNumber

(unitless)

N

mNR

2

f

dE

Complications with monochromators

Overlap of orders m = 1, l= 600 nm and m = 2, l = 300 nm spatially overlap

You can get ’s mixed up if light source contains many wavelengths

Additional wavelength selection often needed Filter, prism, detector λ selection device, digital

analysis after data collection, background subtraction

Might need to use a different light source if your wavelength of interest is not “clean”

Slit width and spectral resolution of a spectrometer Tradeoff exists between sensitivity and resolution

High intensity = high sensitivity (low noise) Two basic concepts:

If you make the entrance slit width too big, you let in a lot of light (that’s good – high intensity), but it can be multi-wavelength; a large section of light dispersed in l is let in Good light intensity, poor spectral resolution

If you make the entrance slit width too small, you let in less light (less intensity), but its l range is smaller Poor light intensity, good spectral resolution

Entrance slit (creates image) and exit slit (output filter) Usually the same width

Optimal slit width based upon grating dispersion

Slit width

Skoog, Fig. 7-24

For just passing l2

Skoog, Fig. 7-25

Ptotal

If spectral bandwidth is Dl/2, good spectral

resolution

Both entrance and exit slits

narrowed from top to bottom

Slit width

w is slit width

Skoog, Fig. 7-26

Slit width

Watch the effect of adjusting the slit width and the resultant spectral bandwidth on the following data sets of benzene vapor.

w is slit width

Optical Photodetectors

A. Photomultiplier tube (PMT)

B. CdS photoconductivity

C. GaAs photovoltaic cell

D. CdSe photoconductivity cell

E. Se/SeO photovoltaic cell

F. Si photodiode

G. PbS photoconducitivity

H. Thermocouple

I. Golay cell

These generally make currentor voltage when light hits them.

More sensitive

Less sensitive

Ideal photodetector(photon transducer)

High sensitivity High S/N Fast response time Signal directly

proportional to # of photons detected

Zero dark current The blank is zero

counted PhotonsN

S

kPS Or, equivalently,

High sensitivity High S/N Fast response time Signal directly

proportional to # of photons detected

Zero dark current The blank is zero

Ideal photodetector (photon transducer)

High sensitivity High S/N Fast response time Signal directly

proportional to # of photons

Zero dark current

S kP

darkS kP k Reality

Intrudes

Here’s what really happens:

Signal isFunction of λ

Constant darkcurrent term (non-zero)

Three mainphotodetector types

Photon transducers (directly “count” photons) Photomultiplier tubes (PMTs)

Charge transfer devices Charge injection devices (CID) Charge coupled devices (CCD)

Thermal transducers Photons strike the transducer

Temp increases Temp increase increases conductivity

Current or voltage are measured

Vacuum phototube

Cathode is coated with photo-emissive material

Emitted electrons are collected anode.

# of electrons is directly proportional to # of photons.

Current is easy to amplify. Usually have small dark

current. Operate at ~ 90V bias Not so portable

Skoog, Fig. 7-29.

Photomultiplier tube (PMT) # of electrons is amplified

by photoelectric effect upon acceleration towards dynodes Each dynode biased ~ 90V

more positive than previous dynode (or cathode)

Voltage drop accelerates electrons to dynode cascade

Amplification: 106-107 electrons per incident photon; electron cascade

http://www.nt.ntnu.no/users/floban/KJ%20%203055/PMT.jpg

Photomultiplier tube (PMT) Advantages:

Very sensitive in UV-Vis region, single photon sensitivity

Cooled PMT has very low background (kdark approaches zero)

Linear response Fast response

Disadvantages Easily damaged by intense (ambient)

light Noise is power dependent Single channel: can’t use for imaging

Photovoltaic cell Light strikes a

semiconductor (Se) and generates electrons and holes

Magnitude of current is proportional to # of photons

Requires no external power supply!

Disadvantages: hard to amplify signal and fatigue (wears out)

Useful for portable analyses, field work, outdoor setting

Skoog, Fig. 7-28

Photodiodes(Silicon 190-1100 nm, InGaAs 900 – 1600 nm)

Reverse-biased p-n junction Conductance goes to near

zero Photons create electron hole

pairs that migrate to opposite contacts and generate current

Battery powered Portable applications Are not as prone to some

electronic noise sources 60 Hz line noise

Skoog, Fig. 7-32

Multichannel transducers

Allow simultaneous interrogation of multiple wavelengths

Imaging Photodiode arrays (1-D array) Charge-transfer devices (2-D array)

Charge-injection devices Charge-coupled device (CCD) CMOS

Photodiode arrays Each diode has defined

spatial address Advantages

Multichannel (used for imaging)

More robust than PMT Disadvantages

Not as sensitive as PMT Slower response time

Common in cheaper UV-Vis instruments Often perfectly adequate

Skoog, Fig. 7-33

Charge transfer devices Converts light into charge Negative-biasing leads to

increased capture of holes under pixel electrodes Potential well

Photon ejects electron and the device collects and stores charges 105-106 charges per pixel

Configured as CID or CCD

Skoog, Fig. 7-35

Schematic is for CID

Charge transfer devices (continued) Charge-injection device (CID)-measures

accumulated voltage change (nondestructive read; persistent after read) Measurements can be made while integrating

Charge-coupled device (CCD)-moves accumulated charges to amplifier and readout (destructive read; gone after read) Very high sensitivity; 104-105 pixels High resolution spectral imaging

Complementary metal oxide semiconductor (CMOS) Webcam technology: CHEAP! OK sensitivity, large pixel density

CCD (charge coupled device) Pixels read one at a time by sequential

transfer of accumulated charge

From: “CCD vs. CMOS: Facts and Fiction” by Dave Litwiller, in Photonics Spectra, January 2001

CMOS detectors

Digital camera and webcam technology Each pixel can be read individually

From: “CCD vs. CMOS: Facts and Fiction” by Dave Litwiller, in Photonics Spectra, January 2001 Image from Wikipedia

CCD CMOS Essentially serial Each pixel read one at a

time by common external circuitry Voltage conversion and

buffering Outputs an analog signal Historically gave higher-

resolution images Relatively expensive

High power consumption Up to 100x more than

CMOS

Essentially parallel Each pixel has its own red

out circuitry “on-chip” Allows amplification and

noise correction More susceptible to noise

Outputs a digital signal Reduced area for light

absorption Relatively inexpensive

Highly commercialized fab Runs on less power

Requires less “off-chip” circuitry

Both approaches exist today

Photoconductivity transducers

Semiconductors whose resistance decreases when they absorb light

Absorption promotes electron to conduction band.

Useful in near IR( = 0.75 to 3 m) Cooling allows extension to longer

wavelengths by reducing thermal noise

Thermal transducers

Solution for IR region (low energy photons) Thermocouples

Light absorbed heats the junction (two pieces of dissimilar metal) which leads to a change in voltage relative to a reference thermocouple.

Bolometer (thermistor) Material changes resistance as a function of temp

Pyroelectric devices Temperature-dependent capacitor Change in temperature leads to change in circuit current