129
1 Basic Principles of Atomic Absorption and Atomic Emission Spectroscopy

Basic Principles of Atomic Absorption and Atomic … 8...Basic Principles of Atomic Absorption and Atomic Emission Spectroscopy 2 Source Wavelength Selector Sample Detector Signal

  • Upload
    others

  • View
    22

  • Download
    1

Embed Size (px)

Citation preview

1

Basic Principles of

Atomic Absorption and

Atomic Emission

Spectroscopy

2

Source Wavelength Selector

Sample

Detector Signal Processor

Readout

P

Flame Atomic Emission Spectrometer

3

Flame Atomic Emission Spectrometer

4

Emission Techniques

Type Method of Atomization Radiation Source

Arc sample heated in an sample electric arc (4000-5000oC)

Spark sample excited in a sample high voltage spark

Flame sample solution sample aspirated into a flame (1700 – 3200 oC)

Argon sample heated in an sample plasma argon plasma (4000-6000oC)

5

Flame and Plasma Emission Spectroscopy are based upon those

particles that are electronically excited in the medium.

The Functions of Flame and Plasma

1. To convert the constituents of liquid sample into the vapor

state.

2. To decompose the constituents into atoms or simple molecules:

M+ + e- (from flame) -> M + hn

3. To electronically excite a fraction of the resulting atomic or

molecular species

M -> M*

Emission Spectroscopy

6

Measure the intensity of emitted radiation

Emission Spectroscopy

Ground State

Exc ited State

Emits Special

Electromagnetic Radiation

7

8

Argon Inductively Coupled Plasma as an

atomization source for emission spectroscopy

9

10

• Flame techniques are limited to alkali (Li, Na, K, Cs,

Rb) and some alkaline earth metals (Ca and Mg)

• More energetic sources are used for

– more elements specially transition elements

– simultaneous multielement analysis since spectra

od dozen of elements can be recorded

simultaneously

– Interelement interference is eliminated

– Liquids and solids

Why plasma source or other high

energetic sources?

11

Advantages and disadvantages of

emission spectrometry

Advantages

• rapid

• Multielement (limited for alkali and some alkaline earth metals

• ICP-AES has become the technique of choice for metals analysis.

Disadvantages

• initial cost of ICP instrumentation

• continuing cost of operation (Ar required)

12

Flame Atomic Absorption Spectrometer

Source Wavelength Selector

Sample

Detector Signal Processor

Readout

P Po

Chopper

13

Flame Atomic Absorption Spectrometer

T

14

Atomization Techniques Type Method of Atomization Radiation Source

Atomic sample solution aspirated HCL

(flame) into a flame

atomic sample solution evaporated HCL

(nonflame) & ignited (2000 -3000 oC)

(Electrothermal)

Hydride Vapor hydride generated HCL

generation

Cold vapor Cold vapor generated (Hg) HCL

15

Advantages and disadvantages of atomic

absorption

Advantages

• sensitive (GFAA)

• selective

Disadvantages

• intended for metallic/metalloid atomic species, not

nonmetals or intact molecular species

• lamps - one element at a time

• not easy for solids

• calibration curves nonlinear above A = 0.5

• more costly and less widely applicable than UV vis

• matrix effects - easy for some metals to be masked

16

17

Flame Fluorescence Spectrometer

Source

Wavelength Selector

Sample

Detector Signal Processor

Readout

P Po

90o

18

Fluorescence Techniques

Type Method of Atomization Radiation Source

atomic sample soln. aspirated sample

(flame) into a flame

atomic sample soln. evaporated sample

(nonflame) & ignited

x-ray none required sample

fluoresence

19

Processes that take place

in flame or plasma

T 1

20

Flames

Regions in Flame Temperature Profile

21

22

23

24

25

26

27

Flameless atomization

Electrothermal atomization

Graphite furnace atomization

28

29

30

31

Why do we use a temperature program in

flameless AA?

32

Why do we use an inert gas with

Flameless atomization?

33

Light source used for AA

• What light source do we use with AA?

• Would it be a continuous light source or a

line light source?

• A line light source is used for AA

34

T 2

35

36

How does the line source

provide a single wavelenght?

T3

37

38

39

AA spectrophotometer

40

41

42

43

44

• Flame Emission -> it measures the radiation emitted by the

excited atoms that is related to concentration.

• Atomic Absorption -> it measures the radiation absorbed by the

unexcited atoms that are determined.

•Atomic absorption depends only upon the number of unexcited

atoms, the absorption intensity is not directly affected by the

temperature of the flame.

•The flame emission intensity in contrast, being dependent upon

the number of excited atoms, is greatly influenced by

temperature variations.

Relationship Between Atomic Absorption and

Flame Emission Spectroscopy

45

Measured signal and analytical concentration

1. Atomic Emission

Signal = Intensity of emission = KNf = K’Na =K’’C

Nf = number of free atoms in flame

Na = number of absorbing atoms in flame

C = concentration of analyte in the sample

K, K` and K’’ depend upon:

• Rate of aspiration (nebulizer)

• Efficiency of aspiration (evaporation efficiency)

– Flow rate of solution

– Solution concentration

– Flow rate of unburnt gas into flame

• Efficiency of atomization (effect of chemical environment). This depends upon

– Droplet size

– Sample flow rate

– Refractory oxide formation

– Ratio of fuel/oxygen in flame

• Temperature effect (choice of flame temperature)

46

2. Atomic absorption

Signal = I absorbed = Absorbance = A = k l C

• For the measurement to be reliable k must

be constant; k should not change when a

change in matrix or flame type takes place.

• K depends upon same factors as those

for the atomic emission spectroscopy

47

Background and

Background Correction

48

• Definition

• Sources of Background

• Background correction:

– Blank correction method

– Two-line correction method

– Continuous source correction method

– Zeeman effect correction method

49

What is a background?

• It is the signal observed when the element sought is

absent

• The light at a specific wavelength (Analyte wavelength) is

attenuated by the effect of flame components or matrix

components in the sample

• Thus measured absorbance and analyte concentrations are

too high

• If problem from flame, blank aspiration will correct for it

• It is more serious at short wavelengths (< 430 nm) and with

graphite furnace

50

Sources of Background in atomic

absorption

1. Absorption by flame itself (Serious at

below 220 nm; e.g., As, Se, Zn )

• Water blank can compensate for this flame

absorption

• In flame absorption, background

interference is insignificant at > 230 nm

51

2. Absorption by concomitant molecular species originating from the matrix like NaX or solvents containing X (halogen) like CCl4.

• Halides absorb at < 300 nm

3. Scattering of radiation from the particulate material in the flame

• Particulate material: Unevaporated droplets; unevaporated refractory salt particles

• Scattering is more serious at short • 2 and 3 are most common with electrothermal atomizer

52

Specific Applications Require

background correction

1. Graphite furnace

2. Flame determination of low concentrations of an element in the presence of high concentrations of dissolved salts

3. Flame analysis where sample matrix may show molecular

absorption at of the resonance line

4. Flame determination of an element at where flame absorption is high

53

Background correction methods

1. Using a blank

• Measure the absorbance of the metal

resonance line by both flame & blank

(flame system)

• Measure the absorbance of the metal

resonance line by sample and flame system

(flame + blank)

• A is the difference

54

The main two methods used for background correction:

1. Continuous source correction method

2. Zeeman effect correction method

• These two methods are based on the fact that

• Atomic bands are very narrow

• Background bands are molecular in nature thus

they are broad bands

55

2. The continuous-source correction

method

• This method is an available option with most instruments

• Deuterium lamp is used in conjunction with the HCL lamp

• The two lamps are observed by detector alternatively in time

• Background usually absorb radiation from D2 lamp and HCL

• Absorption of the analyte from D2 lamp is negligible

• The lamps may be pulsed at different frequencies, thus the signal processing electronics can distinguish and process separate absorption signals

• Thus

Acorrected = A HCL - AD2

56

The continuous-source correction method

Chopper

57

58

Background correction by continuous source

correction method

HCL HCL

HCL

59

60

Disadvantages of the continuous source

correction method

• Correction may degrade detection limit. The current of HCL is reduced to match the D2 lamp

• The correction may be in error if the background s (sample and standards) are not the same

• Alignment of the two sources to pass through the same area of atomizer is tedious

• Beam splitter reduces the intensity of radiation which may affect the limit of detection

• Additional light source and electronics are needed

61

Zeeman effect correction method

62

Zeeman effect correction method

63

• Under the magnetic field, atomic spectral line (emitted or absorbed) splits into three or more polarized components

• Two components will be displaced at equal wavelength intervals higher and lower than the original line.

• The original line is polarized in plane parallel to the magnetic field and the other lines are polarized perpendicular to the magnetic field

• The original line is absorbed by both background and analyte. However, the other lines are absorbed by background only

• The signal corresponding to the analyte is the difference

64

Advantages of Zeeman method

• Only a single lamp is required

• Problems of alignment of two beams is eliminated

• It provides more accurate correction for background than other methods because the background absorbance is measured at or very near the wavelength that the analyte absorbance is measured

65

Disadvantages of Zeeman method

• The implementation is rather complex and expensive

• The original and the split bands may overlap causing an increase in the background signal

• This causes a curvature in the calibration curve

• The instrument is bulky because of the magnet

• Spectral lines for some elements may undergo more complex splitting. Thus sensitivity will be reduced (It gave good results for 44 elements)

• It is more difficult to engineer with flames

66

Source self reversal

background correction method

• It is based on self-reversal of the HCL spectral lines

at high current

• At high current, many free atoms are generated

causing an increase in the number of unexcited

atoms that absorb the center of the emitted line

causing the self reversal

• Specially good for correcting for high molecular

absorption for that by phosphate on Se and As

• A total is measure

67

• Hollow cathode lamps normally operate at currents of 3-15 mA.

• If the applied power is raised to several hundred mA, they exhibit a phenomenon called self-reversal.

• This giant pulse of current changes the nature of the analyte absorption line so it will only measure the background absorbance.

• Atotal is measured during the low current period. Abackground is measured during the high current period

• The low current (6-20 mA) is pulsed at 100-500 mA

Corrected signal = Signal before the pulse

–Signal After the pulse

68

Advantages of Smith-Hieftje technique

• Only a single source is used

• It offers all the advantages of other methods at low cost

• It does not lose radiation like in Zeeman but this mehtod with the Zeeman one are less sensitive than the D2 one

• Drawbacks – less sensitivity

– lamp life decreased

69

Background in atomic emission

• Sources

1. Emission from flame

2. Emission from the sample matrix

70

Electrothermal Atomizer

(Graphite Furnace)

• Main components

• Atomization steps

• Interferences and matrix modification

71

Graphite Furnace Atomic Absorption

Spectrometry, GFAAS

A technique to minimize dilution during atomization of the analyte prior to its determination with atomic absorption spectrometry

A technique with more interferences than the more reliable flame atomization

A technique with high sensitivity and very good detectability but not so good throughput and precision

72

Advantages of GFAS

• Increase in the residence time thus more sensitivity will be obtained

• Possibility of analyzing samples of various matrices

• Analyzing small size samples even on the microliter scale

• Avoiding the formation of refractory oxides that cause serious interference sourcse

73

• Minimize the surface adsorption

74

Why do we use an inert gas with GFAAS?

75

Atomization Process

76

• The steps of a GF atomization

• 5-100µl sample

– drying

– ashing

– atomization

– burnout

– cool

Steps in graphite furnace atomization

77

Steps in GF atomization

• The sample introduction

– The sample is injected, usually

between 10-30µl but sometimes the

range could be extended to 100µl.

– The solution is preferably made up of

dilute nitric acid as matrix

– sometimes solid samples are

introduced

78

• The drying step

– to volatilize the sample solvent

– use some degrees above the solvents

boiling point (for water use 110°C) and

about the same time in seconds as the

injected volume in µl

– check that the solvent only volatilizes

and not boil, otherwise bad

reproducibility may occur

79

• Ashing of the sample

– Maybe the most important step

– Remove sample matrix without losing the

analyte

– Highly dependent on both the analyte and

sample matrix

– Conditions for every new sample type

should be optimized

– Inorganic matrix more difficult, because

higher temperatures are needed

80

• The atomization step

– This step is to vaporize and atomize the

analyte

– the vaporization pressure of the analyte

dominate the supply of analyte atoms

inside the tube

– The convection of expanding gas and

diffusion dominate the transport of

analytes out of the tube

– rapid heating is necessary for good

sensitivity

81

• The burnout step

– it is used for taking away parts of the sample not completely volatilized during atomization

– temperatures in the range 2700-3000°C is most often used

– if vaporization of the sample is achieved, completely, the sensitivity will slowly decrease and the imprecision increase

• Cooling step

– its purpose is to cool down the furnace before next sample is added

82

Steps in GF atomization

temperature progamme

0

500

1000

1500

2000

2500

3000

0 20 40 60 80

time (sec)

Tem

p (

°C)

Drying

Ashing

Atomization

83

Interferences with graphite furnace

and its control

• In the early days this technique was

accepted as a highly interference

technique

• Later, new instrumental and analytical

procedures were developed which

avoided the conditions found to be the

source of interference problems

84

Interferences in GFAAS

• The graphite furnace atomization technique is much

more affected by interferences than the flame which

depends on higher concentrations of non-analytes

–physical

• change in surface properties of the graphite tube

influences the atomization rate

–Spectral

• molecular absorption by molecules or atoms other

than the analyte atoms

• light scattering

–non spectral or chemical

• molecular binding between analyte and matrix

• formation of less volatile compound

• reactions with the graphite tube

85

Interference removal

Spectral interferences

• Emission interference

• Non-analyte absorption and scattering

86

1. Emission interference

• Radiation emitted by the hot graphite tube or

platform that covers a wide range of

waqvelengths (250-800 nm)

• Following elements will be obscured:

– Zn 213.9 nm; might be ok

– Cr 357.9 nm

– Ca 422.7 nm

– Ba 553.7 nm mostly affected

• This effect is eliminated by removing the

graphite tube wall or platform from the view

of the detector. This is achieved by ine of the

following arrangements:

87

• Setting a narrow monochromator slit

• Furnace alignment

• Cleaning graphite tube and windows to

prevent light scattering

• Atomization temperature should not be

higher than required

88

2. Non-analyte absorption and scattering

• It is considered as a background absorption

• It is the most severe spectral interference problem

• It occurs as a non specific attenuation of light at the analyte wavelength due to matrix components in the sample

• It leads to having a broad band covering 10-100’s nm due to molecular absorption caused by “undissociated sample matrix” components in the light path at atomization

Remedy?

• Matrix modification

• Furnace control procedures

• Background correction as with flame techniques

89

Matrix modification

• Matrix should be more volatile than the anayte so as

to be removed before atomization

1. Formation of easily volatile matrix components, e.g.,

Using NH4NO3 or ammonium acetate to modify NaCl

matrix

NaCl + NH4NO3 NaNO3 + NH4Cl

nonvolatile matrix More volatile compounds

Matrix modifier

• The more volatile components can be driven off at

lower pyrolysis temperature

90

2. Converting the analyte element into a less

volatile (if it is too volatile) component using

a modifier e.g., Ni (NO3)2 for Se

determination.

• Se is a highly volatile component

• With the modifier Se form nickel selenide.

Thus, Se can be heated up to 900 oC without

loss. Consequently the matrix would be

removed without affecting Se.

91

3. Varying the sample volume

• Can be done if concentration allows

• Reduce the sample size by injecting smaller

samples, thus the mass of background producing

matrix components will be reduced.

4. Spectral background correction

• A continuous background correction is not very

efficient specially for complex matrices. Matrix

modification is a must.

• Zeeman background correction is very efficient for

complex matrices

92

Chemical Interference

• It occurs when other components of the sample

matrix inhibit the formation of free analyte atoms

Mo + Xo MX

Free analyte atoms Different matrix component

93

Interference Removal

1. Ashing with the optimal temperature

–matrix modification (mentioned above)

• Delay of anayte release to furnace allows time for a

constant furnace temperature

• thermal stabilization of analyte (to hold the analyte

on the graphite surface to a higher temperature)

and vaporization of interference during the ashing

step

– Thus, volatility of interfering matrix compound

is increased

• On the other hand volatility of interfering

compound is decreased and selective vaporization

during the atomization step

takes place

94

2. Use of pyrolytically coated Tubes and Platforms

•Ordinary graphite has a porous surface thus it forms

with the analyte a carbide that is nonvolatie

•The tubes are more or less always fabricated from

pyrolytic carbon or covered with a pyrolytic carbon layer.

It’s a more inert form of the graphite. It is more dense

surface and less absorption

•Tubes could sometimes be coated with tungsten or

tantalum for prolonged lifetime or other special

applications

•The platform technique is used to vaporize the analyte

into a more uniform and higher temperature than the

release from the wall. The analyte vaporizes after the

tube wall and gas phase reach a steady state

95

• L'vov Platform – Thin graphite plate inserted into

tube with flat trough

– L'vov barely contacts sides of tube

– Sample is heated more uniformly by air

– Platform delays the appearance of analyte until

the furnace has reached atomization temp

96

Functions of the platform

• It resists tndency for sample to saok into surface before or during drying

• It tolerates high acid concnetration

• It gives bigger signal and less background – It is heated gradually by radiation from walls

– When atomization step is reached the temperature of the sample and furnace environment will be similar

– No chance for sudden cooling and recombination forming molecular species inhibiting atomization

I

97

Disadvantages of Graphite Furnace

• Prone to interferences, poor precision

• Must take Step to control Precision:

1. Autosamplers vs. Manual Injection - more precise

2. Duplicate analyses for all samples (repeat if poor agreement)

3. Standard addition method - making standards up in sample

4. Background correction for Molecular Absorption) 1. D2 2. Zeeman Background Correction 3. Matrix Modifiers to control atomization

98

INDUCTIVELY COUPLED

PLASMA

• Main components of the instrument

• Performance of ICP

• ICP-mass spectrometry

• interferences

99

• The device which produces the ICP is commonly

referred to as the ICP torch. It consists of two to four

Argon flows depending on the manufacturer:

• Nebulizer gas (inner Argon flow), at about

1 L/min, carries the analyte aerosol

• Sheath gas for producing a laminar flow to

improve low excitation energy elements eg group

I & II elements

• Auxiliary gas (if present), lifts the plasma above

the injector tube, used when measuring organics

• Plasma gas, at about 12-16 L/min, sets the

plasma conditions, e.g., excitation temperature

Introduction

100

Sample Introduction

• An ICP-AES instrument consists of a sample

delivery system, an IC plasma to generate the signal,

one or more optical spectrometers to measure the

signal, and a computer for controlling the analysis.

• The most common sample delivery system consists

of a peristaltic pump and capillary tube to deliver a

constant flow of analyte liquid into a nebulizer.

101

Inductively Coupled Plasma Torch

102

How does plasma generate?

• RF induction coil powered by RF generator

will generate a fluctuating magnetic field

around the coil

• Ionization of Ar is initiated by a spark

• Ar+ and e- will interact with the magnetic field

thus they will be forced to move in annular

paths

• Omic heating is the result of their resistance

to the forced movement in limited area

103

Inductively Coupled Plasma

Emission Source

104

Inductively Coupled Plasma

Emission Source and Supply

105

106

107

• An ICP torch can be used to atomize and

ionize materials for quantitative trace and

ultratrace elemental analysis

• The compounds in the samples are broken

down into their elemental form and some

fraction of each elemental species is

simultaneously ionized

• The monocationic (+1) form is preferred

for the assay, and conditions are set to

enhance production of that form

• The torch is normally operated at a

temperature of ~3000K

108

Schematic diagram of a typical ICP-OES instrument

109

Schematic diagram showing the different

regions in the IC Plasma

110

Processes that take place in the plasma

•Aerosol vapor is transported to the plasma

•Vapor desolvates

•Atomization occurs within the plasma

•Atoms get excited to atomic and ionic states

•Rich spectra produced because of presence of both

atomic and ionic lines

•Because of their different excitation energies,

different emission lines will have maximum intensities

at different vertical positions in the plasma

111

Typical detection limits in ICP-OES

112

113

Typical detection limits in ICP-OES

114

Sampling

Liquids

• Form an aerosol (<5 µ droplet size) of the liquid sample

• Eliminate larger droplets

• The use of liquids facilitates automatic sampling:

Gases

• May use gasses directly or indirectly, e.g., form hydrides

Solids

• May use slurries (if 95% of particles <5 µ)

• May use spark ablation or laser ablation to generate a small metal vapor

115

Viewing position

• The plasma generated in an ICP can be viewed

by the spectrometer, side-on or end-on.

• These viewing positions are called radial and

axial viewing, respectively.

• Each has advantages and disadvantages, so

each tends to be used for different applications:

radial viewing for normal analysis and complex

materials, axial viewing for low detection limits

in simpler materials.

116

117

Advantages of Radial View

• Should collect signal from the entire normal analytical zone – no torch adjustment for different elements

– fewer matrix effects and interferences, especially in organics

– less stray light

– better detection limits in difficult matrices, especially in alkalis and organics

– can run any matrix

– less maintenance of the torch

– torches last longer especially with dissolved salts

– lower consumption of argon

118

Advantages of Axial View:

• More intensity

• Better detection limits in simple matrices

119

RF-Generator • ICP's generally require an RF power of 1-2 kW maximum output

to maintain the plasma.

• High efficiency required - especially for organics (not less than 1.5 Kw)

Frequency • Fixed versus Free Running

– Fixed: Crystal controlled, eg at 40 MHz

– Free running: floats eg at 40 MHz ± 2 MHz

• 40 MHz versus 27 MHz

– Because of the skin-effect in RF plasmas, higher frequency gives a thinner plasma with a wider dynamic range (less self-absorption), with lower backgrounds and fewer interferences.

120

121

• > 90% element coverage

• 0.1 to 10 ppb sensitivity for most

elements

• RSD ~ 2% to 4%

122

Typical laser ablation/ICP-MS mass spectrum of a rock sample

123

Comparison of Detection Limits Black = ICP/MS Cross-Hatch = ICP/AES Gray = Furnance AA

124

Comparison of ICP/AES & ICP/MS for the analysis of

Cerium solutions

ICP/AES (100 ppm Ce)

ICP/MS (10 ppm Ce)

• Simple to interpret, especially

for rare elements

• Less interference from

atmospheric contaminants or

molecular bands

Background ions

originated from

the Ar, O, H and

N elements

Nitric acid is commonly used as acidification agent rather than hydrochloric acid, sulfuric acid and

phosphoric acid in sample preparation. Why?

125

• Formation of monocationic ions, doubly-charged ions

and stable metal oxides ions in the torch depends on

the identity of the element and several experimental

conditions, such as the plasma r.f. power and the

nebulizer gas flow rate.

• Spectroscopic interference:

1. Isobaric Interference

- isotopes of different elements (or isobaric interference) e.g. 87Rb+ and 87Sr+; 40Ar+ and 40Ca+; 58Ni+ and 58Fe+

Limitations for ICP/MS Quantitation

126

– Polyatomic Ion Interference

• interactions between species in the plasma and species in the matrix or atmosphere, e.g. molecular ions from the acids of digestion

• e.g. ClO+ and 51V+

Which element(s) does not have isobaric interference caused by the presence of other elements?

127

• Matrix effect:

(a)Limited tolerance for salt concentration

(< 1% is acceptable but < 0.1% is preferred!)

– The deposition of salts leads to a decrease in the aperture diameter, so the sensitivity worsens and the signal gradually decreases as a function of time

(b)Organic solvents tend to quench the plasma

torch

128

Approaches to reduce the occurrence of

polyatomic interferences

• Adjust the plasma conditions to minimize the formation of

polyatomic species

e.g. use 5-10% N2 in Ar, the axial temperature of the plasma increases by ~ 1000 K and the electron density increases by 30%. A marked reduction of MO+ ions

• Minimizing the amount of water vapor introduced into the

plasma in order to limit the signal intensity of oxygen-

containing species

e.g. reduce the spray chamber temperature(?)

• Use He instead of Ar as plasma gas to avoid interference from

polyatomic ions containing argon

• Use electrothermal volatilization for sample introduction

• Use high resolution ICP/MS

129

Approaches to correct for or overcome

matrix effect • Dilution

• Matrix matching

• Use of internal standards

– Addition of a known and preferably an equal concentration of an internal standard to all samples (& standards). The analytical signal is taken as the ratio of the signals for the analyte elements in the sample (& standards) and the signal for the internal standard

• Standard addition

– A known amount of the element(s) of interest is added to the sample and hence suffers the same matrix effect as the analyte