Chap 4_Atomic Spectroscopy

8/13/2019 Chap 4_Atomic Spectroscopy

http://slidepdf.com/reader/full/chap-4atomic-spectroscopy 1/67

Rezaul KarimEnvironmental Science and Technology

Jessore Science and Technology University

Instrumental Technique for Environmental AnalysisChapter 4 Atomic Spectroscopy



Chapter contentOverview of AS 1,2 Advantage and disadvantage of AAS 1 Theory of AS 2 Instrumentation 1 Atomization (flames, furnace and plasmas) 1 How temperature affects on AS 1 Background correction 1

Detection limits 1

Interference 1 Virtues of the ICM 1 Analytical Applications3

2



Reference1. Daniel C. Harris , 2010, Quantitative

Chemical Analysis , 8th edition, W. H. Freemanand Company , Madison Avenue New York, NY10010

2. S. Ahuja and N. Jespersen (Eds), 2006,Comprehensive Analytical Chemistry ,Volume 47, Elsevier B.V.

3. Robinson, 1995. Undergraduate instrumentalanalysis, Marcel Dekker, Inc. NY, USA.

3



AA spectrometry

the absorption of discrete wavelengths oflight by ground state, gas phase free atoms• .

Free atoms in the gas phase are formed from thesample by an “atomizer ” at high temperature • .

AAS was developed in the 1950s by Alan Walsh and rapidly became a widely used analytical tool

4



Advantages

5

1• an elemental analysis technique capable of

providing quantitative information on 70elements

2• practically independent of the chemical form

of the element in the sample• .e.g. A determination of cadmium in a water

3• used routinely to determine ppb and ppm

concentrations of most metal elements



6

its high sensitivity.

its ability to distinguish one element fromanother in a complex sample.

its ability to perform simultaneous multi-element analyse s.

the ease with which many samples can beautomatically analyzed.



Disadvantage

no

informationis obtained onthe chemicalform of theanalyte (no

“speciation”)

often only oneelement canbe

determined ata time

limited usefor

qualitativeanalysis.

7



• An AAspectrometer,

Model AA 280equipped witha graphitefurnace and

Zeemandevice.

• A rotatingturret holdsheight HCL



Types of atomic spectroscopy:

9

• absorption of sharp linesfrom hollow cathodelamp1

• emission from a thermallypopulated excited state2

• fluorescence followingabsorption of laser radiation3



Basis of analyticalmeasurement

10



Theory: Atomic absorptionspectroscopyBeer Lambert equation:

I (λ )= I o (λ )10 -K (λ )

b,where I o(λ ) is the radiant power of the incident radiation

of wavelength λ ,I(λ ) the radiant power of the transmitted radiation,K(λ ) the absorption coefficient of the ground state atom,b the path length.

This equation can be expressed in terms ofabsorbance :

A(λ ) = log (I (λ )/Io(λ ))= K(λ )b

11



Atomic fluorescencespectroscopy

AFS quantifies the discrete radiation emittedby excited state atoms that have beenexcited by radiation from a spectral source.

If a line source is used for excitation and ifthe atomic vapor is dilute, then the radiantpower of the atomic fluorescence signal

(I f ) can be related to the concentration ofground state atoms by the followingequation:

12



◦ where ῼf/4π and ῼ A/4π are the solid angles offluorescence and excitation respectively;

◦ L the length of the atom reservoir in theanalytical direction;

◦ F the atomic fluorescence quantum efficiency ;◦ IL the integrated radiant power for the incident

beam per unit area;◦

∂ a correction factor that accounts for the relativeline widths of the source and absorption profiles; and◦ Δ λ D the Doppler half width of the fluorescence

profile



Atomic emission spectroscopy

AES quantifies discrete radiation that is emitted by anexcited atom when it deactivates to the ground state.This energy of excitation is provided by thermal,chemical, or electrical means.The Boltzmann distribution law gives the concentrationsof atoms in the excited and ground states:

◦ N j/N 0 = (g j / g o )e -Ej/kT

◦ where Nj and No are the number densities of atoms inthe excited (jth state) and ground states,

◦ gj and go the statistical weights of these states,◦ Ej the energy difference between the jth and ground

states,◦ K the Boltzmann constant ; (1.381*10-23 J/K )and◦ T the temperature (K) of the atom reservoir.

14



Instrumentation

15



Light Source

Two radiation sources:◦ the Hollow Cathode Lamp (HCL) and◦ the Electrodeless Discharge Lamp (EDL).

Both types of lamps are operated to provide asmuch intensity as possible while avoiding line-broadening problems caused by the collisionprocesses.

Monochromators generally cannot isolate linesnarrower than 10 -3 to 10 -2 nm.

16



Hollow Cathode Lamp (HCL)To produce narrow lines of the correct frequency, we use a hollow-cathode lamp containing a vapor of the same element as thatbeing analyzed.The hollow-cathode lamp is filled with Ne or Ar at a pressure of130 – 700 Pa.

The cathode is made of the element whose emission lines we want .When 500 V is applied between the anode and the cathode, gas is ionizedand positive ions are accelerated toward the cathode.After ionization occurs, the lamp is maintained at a constant current of 2 – 30mA by a lower voltage.

17



18

The HCL process, where Ar+

is a positively charged argon ion,M0 is a sputtered ground state metal atom,M* is an excited state metal atom, andλ is emitted radiation at a wavelength characteristic for the sputtered metal



19

The HCL emits narrow, intense lines from theelement that forms the cathode.Applying a high voltage across the anode and cathodecreates this emission spectrum .Atoms of the filler gas become ionized at the anodeand are attracted and accelerated toward the cathode.The fast-moving ions strike the surface of the cathode

and physically dislodge some of the surface metal atoms(a process called “ sputtering ”).The displaced atoms are excited by collision withelectrons and emit the characteristic atomicemission spectrum of the metal used to make the

cathode.The emitted atomic lines are extremely narrow .Unlike continuum radiation, the narrow emission linesfrom the HCL can be absorbed almost completely byunexcited atoms .



20

Relative bandwidths of hollow-cathode emission, atomic absorption, and amonochromator. Linewidths are measured at half the signal height.

The linewidth from the hollow cathode is relatively narrow because gastemperature in the lamp is lower than flame temperature and pressure in thelamp is lower than pressure in a flame.



Instrumentation

21



Atomization: Flames,Furnaces, and PlasmasIn atomic spectroscopy, analyte is atomizedin◦ a flame,◦ an electrically heated furnace, (graphite) or◦ a plasma (inductively coupled plasma)

The path length of the flame is typically 10

cm .A detector measures the amount of lightthat passes through the flame.

22



Atomization Process

23

In a flame atomizer, a solution of the sample is nebulized by a flow ofgaseous oxidant, mixed with a gaseous fuel, and carried into a flame whereatomization occurs.



24

The processes occurring in a flame atomizer. M + is a metal cation; A2 is theassociated anion. Mo and Ao are the ground state free atoms of therespective elements



25

A complex set of interconnected processes thenoccur in the flame.◦ The first is desalvation , in which the solvent evaporates

to produce a finely divided solid molecular aerosol.◦ The aerosol is then volatilized to form gaseous

molecules.◦ Dissociation of most of these molecules produces an

atomic gas.◦ Some of the atoms in the gas ionize to form cations

and electrons.Other molecules and atoms are produced in theflame as a result of interactions of the fuel withthe oxidant and with the various species in thesample.A fraction of the molecules, atoms, and ions arealso excited by the heat of the flame to yield atomic ,ionic, and molecular emission spectra.



FlameMost flame spectrometers use a premix burner, in which fuel,oxidant, and sample are mixed before introduction into theflameIn atomic absorption, a liquid sample is aspirated (sucked)into a flame at 2000 – 3000 K.Sample solution is drawn into the pneumatic nebulizer by the

rapid flow of oxidant (usually air) past the tip of the samplecapillary.Liquid evaporates and the remaining solid is atomized (brokeninto atoms) in the flame. The spray is directed against a glass bead, upon which thedroplets break into smaller particles .The formation of small droplets is termed nebulization.A fine suspension of liquid (or solid) particles in a gas is called anaerosol.

26





Graphite FurnacesAn electrically heated graphite furnace is moresensitive than a flame and requires lesssample.From 1 to 100 L of sample are injected into thefurnace through the hole at the center.

Light from a hollow-cathode lamp travels through windows at each end of the graphite tube .To prevent oxidation of the graphite, Ar gas ispassed over the furnace and the maximumrecommended temperature is 2550°C for notmore than 7s.In flame spectroscopy, the residence time of analyte inthe optical path is < 1s as it rises through theflame.

28



29

A graphite furnace confinesthe atomized sample inthe optical path forseveral seconds, therebyaffording higher sensitivity.1 – 2 mL is the minimumvolume of solution

necessary for flame analysis,as little as 1 L is adequatefor a furnace.Precision is rarely betterthan 5 – 10% with manualsample injection , butautomated injectionimproves reproducibility to1%.

A 38-mm-long, electricallyheated graphite furnace foratomic spectroscopy.



30

(a) Premix burner.(b) Endview of flame. The slot in the burner head is about 0.5 mm wide.(c) Distribution of dropletsizes produced by a particular nebulizer.



Matrix Modifiers for FurnacesEverything in a sample otherthan analyte is called thematrix, decomposes andvaporizes during the charringstep.A matrix modifier is asubstance added to thesample to reduce the loss ofanalyte during charring.E.g. The matrix modifierammonium nitrate can be added

to seawater to increase thevolatility of the matrix NaCl.

31

a. A graphite furnace heating profile used to analyze Mn inseawater.

b. When 0.5 M NaCl solution is subjected to this profile, signalsare observed at the analytical wavelength of Mn



Inductively Coupled Plasmas

ICP is twice as hot as acombustion flame.The high temperature,stability, and relatively

inert Ar environmenteliminate much of theinterference encounteredwith flames.The plasma instrumentcosts more to purchaseand operate.

32



33

The cross-sectional view of aninductively coupled plasmaburner shows two turns of a27- or 41-MHz induction coil

wrapped around the upperopening of the quartz apparatus.High-purity Ar gas is fedthrough the plasma gas inlet.After a spark from a Tesla coilionizes Ar, free electrons areaccelerated by the radio-frequency field .Electrons collide with atoms andtransfer energy to the entire gas,maintaining a temperature of6000 to 10000 K.The quartz torch is protectedfrom overheating by Ar coolantgas.



34

The concentration of analyte needed for adequatesignal is reduced by an order of magnitude with anultrasonic nebulizer , in which sample solution is

directed onto a piezoelectric crystal oscillating at 1 MHz.The vibrating crystal creates afine aerosol that iscarried by an Ar stream through a heated tubewhere solvent evaporates.In the next refrigerated zone, solvent condensesand is removed.Then the stream enters a desolvator containing amicroporous polytetrafluoroethylenemembrane in a chamber maintained at 16 °C.Remaining solvent vapor diffuses through themembrane and is swept away by flowing Ar.



Sensitivity with an inductively coupledplasma is further enhanced by a factor of3 to 10 by observing emission alongthe length of the plasma (axial view)instead of across the diameter of theplasma.Additional sensitivity is obtained by

detecting ions with a massspectrometer instead of by opticalemission.

35



How Temperature AffectsAtomic Spectroscopy?Temperature determines◦ the degree to which a sample breaks down

into atoms and◦ the extent to which a given atom is found in its

ground, excited, or ionized states.Each of these effects influences the strength ofthe signal we observe.Affects:◦

The Boltzmann Distribution◦ Effect of Temperature on Excited-StatePopulation

◦ The Effect of Temperature on Absorption andEmission

36



The Boltzmann Distribution

It describes the relativepopulations of differentstates at thermalequilibrium.

If equilibrium exists, therelative population (N*/N0 )of any two states isBoltzmann distribution:N*/N 0= (g* / g o )e - E/kT where

◦ T, temperature (K)◦ K, Boltzmann’s constant (1.38*

10 -23 J/K)◦ the degeneracies g0 and g*.

37

Two energy levels with different

degeneracies.

The number of states at each energy iscalled the degeneracy , denoted as g



Effect of Temperature onExcited-State PopulationThe lowest excited state of a sodium atom lies 3.371x10 -19

J/atom above the ground state.The degeneracy of the excited state is 2, the ground state is 1.The fraction of Na in the excited state in an acetylene-air flame

at 2600 K is,◦ N*/N0 = (2/1)e - (3.371*10-19 J)/[(1.381*10-23 J/K)(2 600 K)]

= 1.67*10 -4

That is, less than 0.02% of the atoms are in the excited state.

If the temperature were 2610 K,◦ the fraction of atoms in the excited state would be N*/N0 =

1.74*10-4 .

The fraction of atoms in the excited state is still less than 0.02%,but that fraction has increased by 100(1.74 - 1.67)/1.67 = 4%.

38



The Effect of Temperature onAbsorption and Emission

We see that more than 99.98% of the sodium atomsare in their ground state at 2600 K.Varying the temperature by 10 K hardly affects the

ground-state population and would not noticeably

affect the signal in atomic absorption.How would emission intensity be affected by a 10 Krise in temperature?Emission intensity is proportional to the population

of the excited state.Because the excited state population changes by 4% whenthe temperature rises10 K, emission intensity rises by4%.

39



Atomic LinewidthsBeer’s law requires that the linewidth of the radiation source should

be substantially narrower than the linewidth of the sample.Otherwise, the measured absorbance not proportional to thesample concentration.Atomic absorption lines are very sharp, with an intrinsic width ofonly ~10 4 nm.Linewidth is governed by the Heisenberg uncertainty principle,which says that the shorter the lifetime of the excited state,the more uncertain is its energy :

∂E∂t ≈ h/4 π ◦ where ∂E is the uncertainty in theenergy difference between ground and

excited states,◦

∂ t is the lifetime of the excited state before it decays to the ground state

the uncertainty in the energy difference betweentwo states multiplied by the lifetime of theexcited state is at least h/4 .

40



41

If ∂t decreases, then ∂E increases.The lifetime of an excitedstate of an isolated gaseous atom is ~10 -9 s. Therefore, theuncertainty in its energy is

∂E≤ h/4 π ∂t = 6.6*10 34 J .s/ 4 π (10 -9 s) ≈ 10 -25 J

Suppose that the energy difference ( Δ E) between the groundand the excited state of an atom corresponds to visible lightwith a wavelength of λ = 500 nm.

◦ This energy difference is Δ E= hc/ λ = 4.0*10 -19 J.

The relative uncertainty in the energy difference is∂E/ Δ E ≈ (10 -25J ) / (4.0*10 -19 J) ≈ 2*10 -7 .

The relative uncertainty in wavelength ( δ λ /λ ) is thesame as the relative uncertainty in energy:◦ δ λ /λ = ∂E/ E ≥ 2*10 -7 ¬ δ λ = 2*10 -7 * 500 nm = 10 -4

nmTh e inherent linewidth of an atomic absorption or emissionsignal is ~ 10 -4 nm because of the short lifetime of the excitedstate.



Broadening lines:

Two mechanisms broaden the lines to 10 -3 to10-2 nm in atomic spectroscopy. One is theDoppler effect

pressure broadening

42



Doppler effect

An atom moving toward theradiation sourceexperiences moreoscillations of theelectromagnetic wave in agiven time period than onemoving away from thesource.That is, an atom movingtoward the source “sees”higher frequency light thanthat encountered by onemoving away.

43

The Doppler effect. A moleculemoving (a) toward the radiationsource “feels” the electromagneticfield oscillate more often than onemoving (b) away from the source.



44

In the laboratory frame of reference, the atommoving toward the source absorbs lowerfrequency light than that absorbed by the onemoving away.The linewidth, , δλ , due to the Doppler effect, is Doppler linewidth: δ λ ≈ λ (7 *10 -7 ) √(T/M ) ◦ T is temperature (K) and◦ M is the mass of the atom in atomic mass units.For an emissionline near λ =300 nm from Fe ( M =56 atomic mass units) at 2500 K, the Dopplerlinewidth is ,◦ 300 nm (7 *10 -7 ) √(2500/56)◦ =0.0014 nm which is an order of magnitude greater than

the natural linewidth.





Multi-element Detection withthe Inductively Coupled Plasma

An inductively coupled plasma emissionspectrometer does not require any lamps and canmeasure as many as 70 elements simultaneously.One photomultiplier detector is required at thecorrect position for each element.Dispersed radiation lands on a charge injectiondevice (CID) detector , which is related to the chargecoupled device (CCD)Capabilities of CID detector:◦ pixels are individually addressed◦ rapidly filling pixel can be read, re-zeroed, and read again◦ filled pixel does not bloom into neighbors

46



47



Types of interference

spectral: unwanted signalsoverlappinganalyte signalchemical: chemical reactionsdecreasing the

concentration of analyte atomsionization: ionization of analyte atomsdecreasing the concentration of

neutral atoms

48



Spectral interference

Spectral interference refers to theoverlap of analyte signal with signals dueto other elements or molecules in the sampleor with signals due to the flame or furnace.Interference from the flame can be subtractedby using D 2 or Zeeman backgroundcorrection .

The best means of dealing with overlapbetween lines of different elements in thesample is to choose another wavelength foranalysis.

49



A Cd line at 228.802nm causes spectralinterference with theAs line at 228.812nm in mostspectrometers. Withsufficiently high

resolution, peaks areseparated and thereis no interference.

50



Chemical interference

Chemical interference is caused by anycomponent of the sample that decreases theextent of atomization of analyte.

◦ For example, SO 42- and PO 4

3- hinder the atomization of Ca 2+,perhaps by forming nonvolatile salts.

◦

Releasing agents are chemicals added to a sample todecrease chemical interference.◦ EDTA and 8-hydroxyquinoline protect Ca 2+ from interference

by SO4-2 and PO 4

3-.◦ La3+ is a releasing agent, apparently because it preferentially

reacts with PO 43 and frees the Ca 2+.

◦ A fuel-rich flame reduces certain oxidized analyte species thatwould otherwise hinder atomization.

Higher flame temperatures eliminate many kinds ofchemical interference.

51



Ionization interference

Ionization interference can be a problem in theanalysis of alkali metals at relatively low temperatureand in the analyses of other elements at higher temperature.For any element, we can write a gas-phase ionizationreaction:

◦ M( g) ↔ M+(g) + e -

(g) ; K=? Because alkali metals have low ionization potentials, they aremost extensively ionized.At 2 450 K and a pressure of 0.1 Pa, sodium is 5% ionized.With its lower ionization potential, potassium is 33% ionized.Ions have energy levels different from those of neutral atoms,

so the desired signal is decreased.If there is a strong signal from the ion, you could use the ionsignal rather than the atomic signal .

52



Background Correction

Atomic spectroscopy mustprovide backgroundcorrection to distinguishanalyte signal fromabsorption, emission, andoptical scattering of thesample matrix, the ame,

plasma, or red-hot graphitefurnace.

53

the spectrum of asample analyzed in agraphite furnace.

Sharp atomic signals with



54

Sharp atomic signals witha maximumabsorbance near 1.0are superimposed onabroad background withan absorbance of 0.3.If we did not subtractthe backgroundabsorbance, signicant errors would result.Background correction iscritical for graphitefurnaces , which tend tocontain residual smokefrom charring .

d l f



Adjacent pixels of CIDdisplay

Figure 20-20 shows how background is subtracted in an emissionspectrum collected with a charge injection device detector.The gure shows 15 pixels from one row of the detectorcentered on an analytical peak.Pixels 7 and 8 were selected to represent the peak .

Pixels 1and 2 represent the baseline at the left and pixels 14and 15 represent the baseline at the right .The mean baseline is the average of pixels 1, 2, 14, and 15.The mean peak amplitude is the average of pixels 7 and 8.The corrected peak height is the mean peak amplitudeminus the mean baseline amplitude.

55



Beam chopping

For atomic absorption, beam chopping orelectrical modulation of the hollow-cathode lamp(pulsing it on and off) can distinguish the signal ofthe ame from the atomic line at the samewavelength.Figure 20-21 shows light from the lamp beingperiodically blocked by a rotating chopper.Signal reaching the detector while the beam isblocked must be from ame emission.

Signal reaching the detector when the beam is notblocked is from the lamp and the ame.The difference between the two signals is thedesired analytical signal.

56



57

D 2 lamp background



D 2 lamp backgroundcorrection

Deuterium lamp background correction , broademission from a D 2 lamp is passed through the flame inalternation with that from the hollow cathode.The mono-chromator bandwidth is so wide that anegligible fraction of D 2 radiation is absorbed by the

analyte atomic absorption line.Light from the hollow-cathode lamp is absorbed byanalyte and absorbed and scattered by background.Light from the D 2 lamp is absorbed and scattered onlyby background.The difference between absorbance measured with thehollow-cathode lamp and absorbance measured withthe D 2 lamp is the absorbance of analyte.

58

Z ff ( d



Zeeman effect (pronouncedZAY-mon) An excellent, but expensive,

background correctiontechnique When a magnetic field isappliedparallel to the light path

through a furnace, theabsorption (or emission) lineof analyte atoms is split intothree components.

Two are shifted to slightlylower and higher wavelengths(Figure 20-22), and onecomponent is unshifted.

59

Zeeman effect on Cofluorescence in a graphite furnacewith excitation at 301 nm anddetection at 341 nm. The magneticfield strength for the lowerspectrum is 1.2 tesla



The unshifted component does not have the correctelectromagnetic polarization to absorb light traveling parallelto the magnetic field and is therefore “invisible.” To use the Zeeman effect for background correction, astrong magnetic field is pulsed on and off.Sample and background are observed when the field is off.Background alone is observed when the field is on .The difference is the corrected signal.The advantage of Zeeman background correction is that itoperates at the analytical wavelength.In contrast, D

2 background correction is made over a broad

band.

60





62

Measurement of peak-topeak noise level and

signal level.The signal is measuredfrom its base at the

midpoint of the noisealong the slightlyslanted baseline .This sample exhibits asignal-to-noise ratio of2.4.

The detection limit for furnaces is typically two orders of



63

yp ymagnitude lower than that observed with a flame and thatfor the inductively coupled plasma are intermediate betweenthe flame and the furnace.

Comparison detection limits for flame, furnace, and ICM

C i f t i



Comparison of atomicanalysis methods

64



Analytical applications of AAS

AAS is used for the determination of all metal andmetalloid elements .Nonmetals cannot be determined directlybecause their most sensitive resonance lines arelocated in the vacuum UV region of the spectrum.It is possible to determine some nonmetalsindirectly by taking advantage of the insolubility ofsome compounds.◦ For example, chloride ion can be precipitated as

insoluble silver chloride by adding a known excess ofsilver ion in solution (as silver nitrate).

◦ The silver ion remaining in solution can be determined byAAS and the chloride ion concentration calculated fromthe change in the silver ion concentration.

65



Qualitative Analysis

The radiation source used in AAS is an HCL or anEDL, and a different lamp is needed for eachelement to be determined .Because it is essentially a single-elementtechnique , AAS is not well suited for qualitativeanalysis of unknowns.To look for more than one element requires asignificant amount of sample and is a time-consuming process .

For a sample of unknown composition, multielementtechniques such as XRF, ICP-MS and other atomicemission techniques are much more useful andefficient .

66



Quantitative Analysis

Quantitative measurement is one of theultimate objectives of analyticalchemistry .

AAS is an excellent quantitativemethod .It is deceptively easy to use, particularly

when flame atomizers are utilized.

Chap 4_Atomic Spectroscopy

Documents