вход по аккаунту


Chemical Characterization of Aerosols On-Line and In Situ.

код для вставкиСкачать
Chemical Characterization of Aerosols : On-Line and In Situ
By Reinhard NieSner *
Whereas aerosols were formerly considered to be important in meteorological processes but
in little else, today they play a significant role not only in environmental protection but also
in “new technologies” such as the formation of high performance ceramics, sinter metals,
optical waveguides, and superconductors. Analytical procedures which are capable of a continuous qualitative and quantitative description of processes in which aerosols are formed are,
however, still almost unknown. The present review will discuss initial developments (which will
often appear exotic in nature) such as the use of the condensation properties of ultrafine
particles or time-resolved laser-induced aerosol fluorescence spectroscopy. The progress of
analytical chemistry in the last few years makes it clear that advances in miniaturization, data
processing, and optronics will probably be capable of solving many problems which are
presently regarded as unsolvable.
1. Introduction
1.1. Aerosols
Aerosols are colloidally disperse systems consisting of
ultrafine or finely divided material (particle diameter between ca. 2 nm and a few pm) in a carrier gas. No distinction
is made between liquids and solids in the suspension.
Aerosols play an important role in the atmosphere: they
are responsible for atmospheric turbidity and thus for the
limitation of visibility, they act as a source or sink for gaseous emissions (e.g. SO,, NO,, R,S), and in addition they
provide the condensation nuclei which are of prime importance in the life-giving water cycle. Aerosols are, however,
also of increasing importance in technological processes.
New technologies involve a large number of process stages in
which substances are used or produced in a finely disperse
form as aerosols: a) preparation of ultrapure, monodisperse
and chemically homogeneous dusts as precursors for highperformance ceramics or for ceramic and metallic coatings,“ - 71 b) solvent-free lacquer technologies (electrostatic
coating),‘’’ c) production of optical waveguides,[g1d) microencapsulation of reactive substances,[”. ‘‘I e) manufacture
131 f) preparation of single crysta1s,[l4’151
g) preparation of superconducting materials,[16.1 7 1 h) production of magnetically anisometric
and i) increasing the rate of slow reactions (e.g. aerosol fluorination
of hydrocarbons‘’ 201).
More exotic applications, which document the potential
of aerosols in technical processes, are firstly their use in
fusion experiments (neutron production during the formation of deuterated polyethylene aerosols from an exploding
polyethylene fiber[”’), the development of “chemical lasers”
(production of high concentrations of singlet oxygen for
energy storage[221),or the rapid and selective distribution of
fertilizing and neutralizing mineral aerosols for the rehabilitation of woodland which has suffered damage from “acid”
[*] Prof. Dr. R. NieDner
Institut fur Wasserchemie der Technischen Universitat
Marchioninistrasse 17, W-8000 Miinchen 70 (FRG)
466 0 VCH
Verlagsgesellschuft mbH. W-6940 Weinheim. 1991
Aerosols also cause problems in everyday life. When certain types of emission are involved they are considered a
toxicological risk for humans. The main subjects of discussion in this case are aerosols formed in combustion processes
or at the workplace during production processes. Aerosol
emissions result from incomplete combustion of polycyclic
aromatic hydrocarbons (PAHs), polychlorinated biphenyls
(PCBs), dioxins, and dibenzofurans. Sources include automobiles, refuse incineration, domestic fuel, cigarette smoke,
and coal-fired power stations. These substances are generally
also found at the workplace since pyrolysis processes or direct contact with toxicologically relevant aerosols are often
involved in work processes. A further important toxic class
comprises the heavy metal aerosols formed in grinding,
welding, and casting processes.
Particle size plays a particularly important role. Present
knowledge lists aerosols with aerodynamic particle sizes below 3 pm diameter as particularly dangerous, because they
represent the fraction which can penetrate the lung. However, conflicting reports exist on the metabolism of particles
deposited in the lung and on synergistic effects (for example
how irritant gases and solid aerosol particles act in combination to determine the risk potential), as demonstrated by the
continuing discussion on the risk potential of diesel exhausts
and tobacco smoke aerosols.
1.2. Problems of Measurement in Aerosol
The problem areas mentioned in Section 1.1 make it clear
that a powerful measurement technique would be of great
interest; the following properties of such a technique must be
considered :
- limits of detection: Many substances, including PAHs and
dioxins, are emitted in sub-ppb concentrations.
- selectivity: High selectivity is not necessarily required in
control and regulation procedures, or when screening is to
be carried out as a rapid initial measure. In particular for
process technologies or in warning systems at the work
place, the matrix composition of the environment is welldefined. A final analysis, however, requires the highest
OS7O-0833i9llOSOS-04668 3.50 f ,2510
Angew. Chem. Int. Ed. Engl. 30 (19911 466-476
on-line signaf emission: A continuous determination of the
aerosol composition is required for devices which give a
warning of toxic emissions or immission. Aerosol measurement techniques which form chemically sensitive elements in a measurement and regulation procedure can
also dictate the time constant: thus the response time
should be as short as possible so that signal generation is
as close to continuous as possible.
- in situ concentration determination : The thermodynamic
and kinetic stabilities of many substances are particularly
low when they are present in the form of aerosols. Welldocumented examples are ammonium salt aerosols and
tri- and tetranuclear PAHs produced in combustion processes. If an enrichment step such as filtration collection
on membrane filters is necessary prior to the actual analysis, a substantial amount of the substance will be transferred from the solid to the gas phase. As a consequence
the analysis will give a reading which is much too low
unless the sampling method takes account of both solid
aerosol particles from the filter and gaseous components
which pass through the filter. Other possibilities for falsification are provided by interactions between neighboring
particles at the surface of the collection medium (filter,
impactor) or by an irreversible reaction of the aerosol with
the collector medium itself. It is also necessary to take into
account reactions between the deposited material and reactive trace gases (e. g., HNO,/NO, + pyrene + l-nitropyrene). It is thus desirable to devise a technique for direct
determination of the relevant substance in its state of suspension (i.e., in situ). The technique should subject the
aerosol to as little disturbance as possible.
A complete characterization of an aerosol (see Fig. 1) can
be divided into a chemical and a physical stage. The physical
stage consists of the determination of particle concentration,
particle size distribution, form and structure of single particles, electrical properties of the suspended particles, as well
as of pressure, temperature, and relative humidity. That this
area of aerosol characterization can in general be considered
as already fully d e v e l ~ p e d [ is~ documented
~ - ~ ~ ~ by the large
number of commercially available devices.
The chemical part of a comprehensive aerosol characterization should take into account the composition of both the
gas phase and the dispersed material, because they are dependent on each other. A simple example in the area of
I Chemical
Chemical Composition
of the Gas Phase
Chemical Composition
Aerosol Characterization
Physical Characterization
Concentration 1
Particle Shape
I Microstructure 1
Fig. 1. Parameters of a complete aerosol characterization
Even small variations in temperature cause as significant
a change in the amount of solid NH,NO, in the gas phase as
a variation in the concentrations of the gases NH, and
HNO,. Similar effects have been observed in organic aerosol
systems such as fluoranthene.
A further peculiarity of aerosol systems, namely the microstructure of aerosol particles, which is becoming increasingly
important, makes their characterization more difficult.
The substance distribution across the total particle crosssection is not homogeneous (Fig. 2). A typical example of
this behavior is the structure of fly ash particles: as a result
of the stages in the cooling of a combustion gas (see Fig. 3),
the toxicologically relevant substances become attached successively at the particle surface, either by adsorption or by
condensation. In addition, because of the high particle con),
centration near the combustion center ( N > 10’’ ~ m - ~agglomerated submicron particles are found on the surface of
the fly ash, which in turn can be microstructured. Another
model for a microstructured particle, for example, a fog or
cloud droplet, consists of an insoluble particle nucleus which
as the result of a condensation process is covered by a liquid
Reinhard NieJner was born in Tuttlingen (FRG) in 1951 and started his studies in chemistry in
Freiburg in 1972. He obtained his Diplom there in 1976 in analytical chemistry and his doctorate
at Dortmund under Prof. D . Klockow in 1981, where he continued research for his habilitation
till 1986. He was appointedfullprofessor for Hydrogeology, Hydrochemistry and Environmental
Analysis and head of the Institute for Aqueous Chemistry and Chemical Balneology at the
Technische Universitat Miinchen in 1989. Prof. NieJner was awarded the prize of the “Fachgruppe Analytische Chemie” in 1988 and the Heinrich-Emanuel-Merck Prize in 1990. His research
interests lie in the classical analytical chemistry of water, the supervision of disposal sites, the
chemical interaction of Xenobiotics with soil, the development of highly sensitive and selective
immunoassays to detect harmful substances in the environment, analysis of aerosols, and in
particular, the applications of laser spectroscopic techniques.
Angew. Chem. Inr. Ed. Engl. 30 (1991) 466-476
surface film
Polar part of
In the following discussion an attempt will be made to give
a systematic overview of in situ and on-line aerosol characterization methods which either appear feasible or have already been realized. The special features and characteristics
of these procedures will be discussed. A schematic summary
is given in Figure 4.
2. Aerosol Characterization Based on Interactions
with Electromagnetic Radiation
Aqueous electrol y t e solution
A large number of old and new techniques for on-line
measurements based on the interaction of aerosols with electromagnetic radiation is now available. Well-known principles, such as emission, absorption, and scattering, covering
the complete electromagnetic spectrum, can be used. Difficulties are, however, often encountered because of the small
dimensions of the aerosol under study or because of the
influence of the carrier gas, such as for example the production of unwanted new particles during photolysis. On the
other hand, the observation of optical properties is the direct
route to an in situ and on-line analysis.
Fig. 2. Examples of microstructured aerosol particles.
Combustion Zone
(Free radical polymerlzaltonl
Adsorbole Formotion
Fig. 3. A model of aerosol formation during cooling of combustion exhaust
electrolyte layer. Because of the many subsequent evaporation and condensation processes occurring during the "lifetime" of these particles they can also be covered by a film of
volatile organic substances such as terpenes, alkanes, alcohols, and carboxylic acids. This leads to an inverse micelle
microstructure (see Fig. 2). Laboratory studies of artificially
microstructured aerosols of this type have shown that such
films can have a great influence on the phase transition of
gaseous substances into the dissolved liquid phase (e.g.,
NH,(g) into H@/H,0).[27.281
Aerosol analysis of meosuremenl
Physical aerosol
Aerosol ESCA
photon emission
Aerosol mass
Aerosol laser
Laser breokdown
nucleus counter
Tandem differential
aerosol classifier
Raman spectroscopy
IR back-scanering
Fig. 4. In-situ and on-line aerosol determination procedures which are either
already in use or potentially useful.
2.1. Characterization on the Basis of the Light Extinction
by Particles
The majority of procedures used in analytical chemistry
are based on extinction measurements. For aerosols such a
measurement is no longer trivial. Light incident on a particle
can be either absorbed or scattered. Natural aerosols contain
particles whose sizes lie between ca. 2 nm and 20 pm, while
particle concentrations lie between less than 1 per cm3 (e.g.
in clean rooms) and greater than lo9 per cm3 (in combustion
processes). Thus the extremely low effective interaction
cross-section for the extinction measurement can readily be
calculated. At present no simple procedure based on an extinction measurement is known which permits analysis at
such low limits of detection. Improvements could perhaps be
realized in applications involving synchrotron radiation and
by the use of laser radiation within a laser resonator.
Luckily photoacoustic spectroscopy1291is an indirect way
of determining the absorption coefficient of an aerosol. The
optoacoustic interaction is based on the conversion of vibration energy of excited vibration modes of molecular species
into heat. This process causes the formation of a pressure
wave in the surrounding gas. When the exciting energy
source (typically a CW-IR laser) is modulated, the acoustic
signal is subject to a similar modulation and can therefore be
detected above the background noise by a sensitive microphone acting as a pressure sensor.
The absorbed photons thus cause a pressure increase as a
result of the absorbing aerosol particles. If a resonant cell is
used to create a standing wave, the acoustic signal is amplified, and the pressurep within the cell induced by irradiation
with a sinusoidally modulated laser beam of frequency w is
described by Equations (b) and (c).'~']
Equations (b) and (c) represent the relation between the
observed pressure in the resonant cell and the absorption
coefficients p of the irradiated aerosol particles. The integral
Angew. Chem. Int. Ed. Engl. 30 (1991) 466-476
= EAjPj
A . = - io(y
- 1) fi(V,)-’ j p j I d V
w2[1 - (o/wj)’ - i(o/ojQj)]
2.2. Analysis Based on Light Emission from the Particles
Another useful detection principle for quantitative or
qualitative analysis of aerosols is photon emission from selectively stimulated molecules or particles. Relatively inexpensive techniques for the registration of single photons in
the visible region already exist. There are thus a number of
attractive possibilities for on-line and in situ detection of
aerosols, such as laser-induced aerosol fluorescence (LIF),
laser-induced breakdown spectrometry at aerosols (LIBS),
laser-induced Raman spectroscopy at aerosols, IR backscattering at aerosols, and inductively coupled plasma (ICP)
o = frequency of the chopped laser source,
o,= frequency of the jth mode,
y = ratio of the specific heats CJC, at constant pressure and temperature,
= absorption coefficient,
I = spatial variation of the light intensity,
Q, = amplification factor of a mode j at resonance,
p , = vibrational modes,
V,. = cell volume
V,- ‘JpjIdV represents the coupling between the spatial profile of the irradiation intensity and the pressure waves pi. The
final result is that, within certain limits, the induced pressure
p (the signal which is detected!) is a function of the applied
irradiation intensity I.
Several groups are at present working on the development
of photoacoustic aerosol sensors, in particular for the monitoring of diesel aerosols. Unsolved problems include the
background signals caused by absorption of the incident
light at cell walls or windows. The wide linear measurement
range is a general advantage of this procedure. The application of photoacoustic spectroscopy to sulfate determination,
initially demonstrated by Rohl et al.[311for aerosol filter
samples, is promising.
Adams et al.[32s331 have recently published details of an
aerosol spectrophotometer (Fig. 5). The photoacoustic cell
Intensity rneosurement
Fig. 5. Construction principle of a photoacoustic aerosol spectrometer.
Preamp = preamplifier, Ar = argon.
consists of a brass cylinder which is constructed in the form
of a resonant cell. The light source is an argon ion laser. The
aerosol is drawn through the cell by a pump at the rate of
0.45 L min- l . The aerosol is first passed through a MnO,
diffusion separator to remove NO,, which interferes in the
measurement; the aerosol fraction is not reduced thereby.
Adurns et al. report a detection limit of 300 ng carbon per m3.
A further improvement can be expected when cells are
constructed within a laser resonator, in analogy with an application as a photoacoustic gas chromatographic detector
as published by Fung and G~ffnney.~~~]
Angew. Chem. In[. Ed. Engl. 30 (1991) 466-476
2.2.1. Analysis by Laser-Induced Fluorescence (LIF)
Laser-induced fluorescence has been known for many
years as an unwanted side effect of Raman spectroscopic
studies in flame experiment^.^^', 361 The first selective experiments designed to use these effects for the characterization of
aerosols were reported by Omenetto and Allegrini,[37.381 who
generated pure polydisperse anthracene and fluoranthene
Fluorescence occurs as a result of excitation of molecular
or atomic systems by the absorption of one or more photons.
Apart from fluorescence, other possibilities for a return to
the ground state are available; their importance depends on
structural considerations and on the nature of the excited
state. Competing processes are quenching, internal relaxation (which leads to warming), and intersystem crossing
(which leads to phosphorescence).
When fluorescence is compared as a detection principle
with other optical principles, the high sensitivity and the
possibility of achieving a high selectivity are immediately
obvious. In contrast to photometry, where the sole available
selectivity parameter is the absorption wavelength, two additional parameters can be used in fluorimetry: the fluorescence emission wavelength and the characteristic decay time
for the fluorescence process. These three fluorimetric parameters should help either to identify and to quantify the various
components in a mixture in the ideal case or at least to
characterize of a group of similar substances.
A typical setup of an in situ and on line analysis of aerosols
by time-resolved fluorimetry is shown in Figure 6.[391It consists of three main components: the tuned laser system
(207 nm-860 nm), the sensor unit (which is connected to the
laser system by fiber optical cable), and the data accumulation and processing module.
The laser light (pulse length ca. 20 nsec, repetition rate up
to 100 Hz) is transmitted via a fiber coupler to the optical
wave-guide fiber and from thence to the sensor unit. Here the
laser light is parallelized and subsequently crosses the aerosol flow. Within the light interaction volume the aerosol flow
is conducted as a free-flowing current in a sheath air arrangement. The fluorescent light from the interaction volume is
led perpendicular to both laser light and aerosol-flow axes
through parallelization optics, an interference filter as a
wavelength selector, and a second set of parallelization optics, and allowed to impede on the window of a fast-response
Beam splitter
Aerosol flow
at right angles
to plane of page
Since the complete system (laser/photomultiplier/interference filter) can be connected via fiber optics, the procedure
in principle enables remote sensing of toxic aerosols by fiber
Work is presently in progress on deconvolution algorithms
for the interpretation of superimposed fluorescence decay
processes. One of the more severe problems in this type of
measurement involves contributions from fluorescing substances in the air and from quench processes. On the other
fiber L
Fiber coupler
Interference filter
Fig. 6. Apparatus for the determination of polycyclic aromatic hydrocarbons
on submicron aerosols by means of laser-induced time-resolved fluorescence.
PM = photomultiplier, HV = high voltage, BBO = barium-gborate crystal.
The data registration is performed by a 400 MHz storage
oscilloscope. Triggering is accomplished by allowing a fraction of the exciting laser light to shine on a fast-response
photodiode. The oscilloscope can now record a time-dependent study of the fluorescence process occurring after a laser
discharge. The monitor image is processed by a CCD camera
and transferred to the computer, where all subsequent processing is carried out.
The analytical possibilities are at present being studied on
test aerosols. Initial results on monodisperse NaCl carrier
aerosols with adsorbed polycyclic aromatic hydrocarbons
(PAHs) gave decay times between 18 and 114 nsec. The absolute limit of detection lies in the femtogram range and is
proportional to the amount of PAH adsorbed (Fig. 7).
Fig. 7. Laser-induced fluorescence of a benzo[u]pyrene (BaP) coated aerosol.
The fluorescence signal S (laser pulse energy normalized) is represented as a
function of the amount of BaP [pg]. d, = 80 nm (constant), T = 90-120°C.
Fluorescence spectra of, for example, NaCl aerosols coated with benzo[a]pyrene or of particles consisting of pure BaP
differ considerably from those of dissolved PAHs. Depending on the degree of surface coverage of the particles, a
fraction of the fluorescence is red-shifted (see Fig. 8), due to
the formation of PAH dimers on the particle surface at
higher vapor pressures (i.e., higher PAH source temperatures).
Fig. 8. Laser-induced aerosol fluorescence as a function of the observed fluorescence emission and the degree of coating (BaP on NaCl aerosol). Fluorescence intensity I [relative units], fluorescence wavelength 1 [nm], at 95"C, A
at 110°C.
hand, the availability of further small lasers (e.g., the 337 nm
N, laser or frequency-doubled semiconductor lasers) makes
it possible to develop mobile measurement systems.
2.2.2. Analysis by Laser-Induced Breakdown
Spectroscopy (LIBS)
The LIBS technique and its application to aerosols have
been reviewed r e ~ e n t l y . [ ~The
~ - ~procedure
is based on a
series of single processes: if a laser beam is focussed on solid
material or on material particles, an aerosol cloud containing evaporated material is formed. If the laser pulse is of high
energy ( E > 100 mJ per pulse, pulse length ca. 10 nsec), the
aerosol cloud forms a microplasma consisting of evaporated
atoms, ions, electrons, and molecular fragments. All the
molecules and particles within the plasma contribute to the
light emission and can thus be determined both qualitatively
and quantitatively. For example, detection limits of ca.
200 pg metal per m3 were observed for aerosols containing
cadmium, lead, and zinc.
2.2.3. Analysis by ZCP-Atomic Emission Spectroscopy
The process of inductively coupled atomic emission has
been applied to ceramic powder aerosols during studies involving materials science problems.t431In ceramics technology, where the highest purity of trace elements is of prime
importance, there would seem to be potential for further
development, The generation of dry ceramic powder aerosols affords advantages over the previously used wet aerosols
derived from slurries,[441since a quench effect due to large
amounts of water in the plasma can no longer occur.
Broekart et al.[451have carried out a detailed study of the
use of this technique in direct trace analysis of A1,0, ceramic
Angew. Chem. Int. Ed. Engl. 30 (1991) 466-476
powder aerosols. Problems exist in the case of larger particles
(> 5 km particle diameter), since when these are present the
material is not completely evaporated.
2.2.4. Analysis by Raman Spectvoseopy
Raman spectroscopy provides an extremely interesting
tool for the characterization of suspended material. The phenomenon of Raman scatttering is caused by interaction between electromagnetic radiation and molecules or atoms in
all aggregate states. Apart from the absorption or emission
of photons, which is irrelevant in Raman spectroscopy, light
scattering is observed within a very short time (ca.
The energy of the incident photon can remain unchanged
(Rayleigh scattering) or change (Raman scattering). The
electric field of an electromagnetic wave induces a dipole
moment proportional to the electric field strength. If vibrations are excited, the molecule’s polarizability changes, and
the prerequisite for observing a Raman spectrum is fulfilled.
The identification of functional groups is possible, as in infrared spectroscopy.
The first Raman spectra of aerosol particles floating freely
in an electrostatic suspension chamber were recorded by
Fung and Tung!46*471who used an argon ion laser (488 nm)
as the excitation source. The scattered light is focussed by a
lens on the entrance slit of a double monochromator.
Photon counting is carried out by means of a cooled photomultiplier and the necessary counting electronics. When this
technique was applied to ammonium sulfate particles in the
micrometer range, the resolution was excellent. However, no
applications to freely flowing aerosols have so far been reported.
The use of resonance Raman spectroscopy in the ultraviolet range can be expected to bring a considerable amplifica*]
this technique
tion of the Raman intensity. A ~ h e r [ ~applied
to deposited PAHs and achieved a remarkable selectivity and
2.2.5. Analysis by IR Laser Back-scattering
A laser beam which is allowed to pass through an aerosol
cloud produces both absorption and scattering effects. This
can be used as the basis for a remote investigation process for
the physical and chemical description of an aerosol, the
DISC process (differential scatter LIDAR system). Wright et
al.[49]and Mudd et al.1501were the first to study sulfuric acid
and ammonium sulfate aerosols by this technique. DISC
uses characteristic differences in the back-scattered IR radiation for quantification and differentiation purposes.
The LIDAR answer signal for a “laser shot” is given by
Equation (d).1501The qualitative information is obtained
from B.
V ( r ) = direct back-scattered signal at a distance r
Eo = transmitted laser pulse energy
b(r) = back-scattering coefficient at a distance r
T(r) = simple transmission of the mass of air from r to the detector
K = calibration constant.
Angew. Chem. Int. Ed. Engl. 30 (1991) 466-476
Fig. 9. Relative back-scattering of some environmental aerosols[50]. -. --.
H,SO, (95%), --- H,SO, (50%), -(NH,)2S0,. ... H,O,
and --ice.
B = relative volume back-scattering coefficient, L = wavelength.
The relative back-scattering of various atmospheric aerosols as a function of the applied wavelength is shown in
Figure 9. The limit of detection which can be achieved lies at
best in the lower mg per m3 range.
3. Aerosol Analysis by Measurement of Electrons,
Ions, or Ionized Particles
Various techniques in analytical chemistry use electrons or
ions as a measure of the amount of the species to be determined. These techniques include X-ray photoelectron spectroscopy (ESCA, UPS, XPS), photoemission, and mass
3.1. Analysis by Determination of the Kinetic Energy
of the Electrons Released
The classical technique for the determination of the kinetic
energy of emitted electrons is photoelectron spectroscopy or
ESCA, which can in principle be used for all the elements of
the periodic table (with the exception of hydrogen) in a quantitative manner. From the kinetic energy of the emitted photoelectrons and the energy of the exciting irradiation, the
bonding energy Eb can be calculated. Detailed measurements
of the bonding energy (chemical shift) provide information
on the bonding states of the various elements.
It is a feature of ESCA that the observed photoelectrons
arise mainly from the surface atom layers (the outer 5 nm or
less, dependent on E,, the material, and the exciting wavelength). ESCA can thus also be regarded as a very promising
tool in the area of aerosol analysis.
One important drawback is that ESCA measurements on
aerosol samples are normally carried out off-line.[51- 531 The
samples (e.g. from an aerosol-generating system) must be
transferred as rapidly as possible from atmospheric pressure
to the high-vacuum conditions of the spectrometer (up to
10- l o Torr) in order to avoid contamination by the pump oil
or similar materials. On the other hand, many studies show
that labile material such as organic substances and volatile
47 1
inorganic materials (e.g. ammonium salts) only survive for
minutes or even seconds under these conditions.[54.5 5 1 An
on-line version of an ESCA spectrometer with a direct coupling between the aerosol and the high-vacuum compartment would thus be extremely useful.
G o l d r n ~ n nwas
~ ~ ~the
] first to develop such an instrument.
As can be seen from the schematic diagram in Figure 10,
expansion of the aerosol under vacuum forms an aerosol
beam comparable with a molecular beam. This aerosol beam
is irradiated by X-rays (AIKeor Mg,,). Particle transfer
through the two nozzles is particularly critical; transfer eficiencies have so far not been determined. The relatively complex pump system consists of Roots and turbomolecular
pumps, which continuously pump off the incursing air.
X-ray source
p=lO1 m b o r
p e 3 - 1 0 - ~ m b o r 20OLs.’
2 LS’
Roots pump-
L 5.’
L 5-l
Rotary vanepump
pump ,
Diffusion pump
Fig. 10. Schematic diagram of an on-line aerosol ESCA apparatus as proposed
by Goldmann [56].
The application of photoelectron spectroscopy as a quantitative analysis technique is limited in the case of aerosols by
a marked dependence on particle size.[571This is shown
clearly in Figure 11 for a deposited sulfate aerosol. If the
particles are perfectly spherical, the ESCA signal is expected
to be proportional to the square of the particle radius. Since
we are, however, generally interested in obtaining information on particle mass, the expected relationship is propor-
tional to l j u . In Figure 1 1 the area of the S atom’s 2p orbital,
normalized with respect to carbon (the filter matrix) and the
mass of sulfate, is plotted against particle size. The predicted
l j r dependence is reproduced very well.
Photoelectron spectroscopy can therefore only be used for
mass determination of aerosols when an additional particle
size determination is carried out ; the surface-to-mass conversion assumes a homogeneous particle composition.
3.2. Analysis by Mass Spectrometry
The first analysis of airborne particles by mass spectrometry was reported by Davis.[581His method involved focussing a particle beam from the surrounding air onto the
ion source. Just as for the technique discussed in Section 3.1,
the construction of an aerosol-vacuum interface is accompanied by substantial transmission losses. After a detailed
study, Stoffeels[591
suggested a solution in which the particle
stream was passed through a capillary via a skimmer and a
collimator and impacted directly onto a heated rhenium filament. This leads to surface ionization of the particles, giving
charged fragments. A problem with this technique is the
necessity for continuous pumping, which leads to the danger
of contamination by back-diffusion of the pump oil.
The best solution at present is that proposed by Sinha.[601
Single particles are allowed to enter the ion source of a Mattauch-Herzog mass spectrometer in a sequential manner,
where they are completely vaporized and ionized. The mass
fragments are separated in space along a focusing plane of
the magnetic sector. The electro-optical ion detector, which
allows the simultaneous detection of all ions across the focusing plane, makes this apparatus unique. The ions impact
on the active surfaces of a two-dimensional electron multiplier array. The electrons from the array are in turn accelerated
and strike the end of the fiber optics, which is coated with a
luminous material. The procedure generates “light pulse images” for the corresponding ion masses. The intensity of the
bright spots is proportional to the total number of ions of a
particular mass. The position and intensity of the ion images
are registered by a CCD array. The detection limit of this
mass spectrometer, determined using a dibromodiphenyl
g, and the simple
aerosol (particle size 1.8 pm), is ca.
relative standard deviation is 10- 19%.
These results indicate areas of potential development for
the mass spectrometric technique.
3.3. Characterization by Aerosol Photoemission
Fig. 11. ESCA signal S(in relative units) of the S atom’s 2p orbital (normalized
with respect to C Is signal of the filter matrix and the mass of deposited sulfate)
ofmonodispersed Na,SO, aerosol as a function of the particle size (d = particle
diameter). Dashed line: expected l / r dependence.
The principle of aerosol photoemission (APE) is the detection of positively charged aerosol particles.161-631 In contrast to photoelectron spectroscopy, where the actual signal
represents the kinetic energy of the emitted photoelectrons,
APE makes use of the positive particles (their determination
at atmospheric pressure by an aerosol electrometer is relatively easy) as a measure of the photoemitting material at the
surface of the particles [Eq. (e)].
Niejner et al.[“] were able to show that N@ is directly
proportional to the total photoelectrically active surface
Angew. Chem. Inl. Ed. Engl. 30 (1991) 466-476
N @ = number of positively charged particles
N = total number of all photoemitting particles
F = degree of surface coverage by photoemitting material
f = time
d2 = irradiated particle cross-section
Y(hv) = photoelectrical quantum yield at a photon energy hv (dependent
on the material and the light wavelength used)
GVv= photon flux
area, which in turn comprises the surface areas of the various
particles and the frequency of their occurrence in a polydisperse aerosol [Eq. (f)].
S,,,,, = total photoelectrically active surface of all particles
di = particle diameter of the component i
N , = number of the particles of diameter d,
The various components of a mixture of different photoelectrically active substances can be individually determined
when their characteristic quantum yields are known from
prior calibration experiments.[641In analogy to photoelectron spectroscopy the height of the signal is proportional to
the surface area; a mass determination thus shows a l j r
relationship between the signal’s height and the aerosol
mass. Photoemitting materials which can be studied include
metal dust, some heavy metal oxide aerosols, and polycyclic
aromatic hydrocarbons. The technique is at present applied
in the control of diesel exhaust gases,’651the control of waste
incineration,t661cigarette smoke
and workplace
Figure 12 shows, as an example, the aerosol photoemission signals at an irradiation wavelength of 185 nm as a
function of the PAH concentration in the smoke generated
by six different brands of cigarette. There is a good linear
correlation between the photoemission signal and the PAH
concentrations in the aerosol as determined by HPLC analy-
sis. It could be concluded that the PAH profile in the
cigarette smoke varies only in the concentration but not in
the relative ratios of the PAHs. The detection limit of the
photoemission technique is in general in the lower ng m-3
701 The reproducibility of this technique is determined solely by the electronics and the stability of the UV
4. Aerosol Analysis by Measurement
of Physical Properties
4.1. Analysis by Observation of Hygroscopic Properties
Several techniques for aerosol characterization depend on
the observation of the particle size after treatment with moisture or heat.
A good example is provided by the sulfuric acid aerosol
detector proposed by Liu et aLt7’]The process has also been
described as an aerosol mobility chromatograph, since it
depends on a continuous selection of particles which have
different mobilities in an electric field. H,SO, particles are
mixed with air of a well-defined relative humidity, and as a
result they swell. In combination with a differential electrostatic classifier (DMA) it is possible to perform extremely
precise measurements of the particle diameter.
The construction of such an instrument is shown schematically in Figure 13. The aerosol to be analyzed is first passed
y J + =pg&
E x c e s s oir
Aerosol electrometer
Fig. 13. Block diagram of the continuously operating aerosol mobility chromatograph [71].
2 ,ire' /
m lngl
m lngl-
Fig. 12. Correlation between the photoelectric signal Sand the content determined by wet chemical methods (m per cigarette) of a) benzo[u]pyrene,
b) dibenzo[a,h]anthracene, c) benzo[b]fluorene, and d) benzo[b]fluoranthene.
Each point represents the average of 5 analyses (with simple standard deviations) of one of 6 types of cigarettes.
Angew. Chem. Int. Ed. Engl. 30 (1991) 466-476
through a krypton-85 neutralizer in which the particles are
electrically charged according to a Boltzmann distribution.
The charged particles are then allowed to pass into the first
DMA. Particles of a certain polarity and electrical mobility
are continually removed, humidified to a defined extent and
fed into a second identically constructed and operated
DMA. If water absorption and subsequent swelling has
caused the particle diameter to increase and thus changed the
electrical mobility, the ‘‘swollen” aerosol fraction will be
able to pass through the apparatus (as a band pass) at a
different collector voltage. The new particle diameter can be
determined exactly from the voltage difference (see Fig. 14).
At 53 % relative humidity sulfuric acid particles swell, while
47 3
chloride, sulfate etc. remain unchanged. Aerosol electrometers or condensation nucleus counters are used as the on-line
particle detector. The detection limit of this technique is ca.
10 ng H,SO, per m3 with an aerosol electrometer as detector
and ca. 1 ng H,SO, per m3 with a condensation nucleus
modulation frequency and one dependent on the absorption
cross-section of the particles. The first applications involved
flowing ammonium sulfate aerosols (IR absorption band at
1 1 10 cm - ').
4.3. Characterization Based on the Condensation
Properties of Monodisperse Particles
The ability of ultrafine particles to act as condensation
nuclei can be used for the characterization of the surface
properties of such particles.[28,791 Monodisperse particles
with differing surface compositions can thus be characterized by observing the point at which water condensation
occurs. The particles for study are selected by an electrostatic
classifier. Droplet growth is then registered by a multistage
condensation nucleus counter (Fig. 16). Hydrophobic organic layers deposited on hygroscopic nuclei can be studied
especially well.
Fig. 14. Current/voltagecurve for aerosol mobility chromatography of humidified and dry H,SO,/K,SO, mixtures[71]. I = electrometer current, in relative
units; u = voltage at central electrode, in relative units; a) without humidification: relative humidity 8.4%; h) with humidification: relative humidity 53%.
An optothermal procedure in which submicron particles
are irradiated by modulated CO, laser light has been combined with the observation of the particle size-dependent
Mie scattering.r73- '*I Under certain conditions when modulated IR radiation is used, the observed light scattering amplitude is proportional to the IR photon absorption. It is
thus possible to record optothermally induced absorption
spectra when a tunable IR laser is used.
The principle of the method is shown in Figure 15. A
chopped IR source irradiates the aerosol particles. When
light is absorbed, the particle radius of liquid aerosols varies
according to the rhythm of the modulation. Photons from a
second laser (in the visible range) are subjected to continuous
elastic scattering at the aerosol particles. The resulting Mie
scattering thus contains a component dependent on the
The same principle has been applied by Charlsun et a1.r7']
who used a nephelometer to detect particle size changes. The
resolution is, however, much lower than with a DMA, so
that the detection limit lies in the pg range.
4.2. Characterization by Observation
of the Particle Size after IR Absorption
Expansion chambe,
Fig. 16. Block diagram of a multistage condensation nucleus counter for the
detection of the condensation properties of a monodisperse aerosol[79].
MV = magnetic valve; LED = light emitting diode.
Exemplary results, which were obtained by applying various degrees of water vapor supersaturation to monodisperse
aerosols (Fig. 17), clearly show the effect of a small number
Calculated saturation ratio
I .o
"C 0.e
Pressure ratio pIpo
Fig. 15. The principle of optothermal absorption spectroscopy (after Cumpillo
et al. [73]). For further details, see text.
Fig. 17. Condensation spectra of monodisperse H,SO, aerosol at constant
total diameter as a function of water supersaturation. d, = 20 nm; p. =760
Torr; H,SO,, r = 10 nm;organic coating (rH2s04 = 5 nm, Ar = 5 nm): 0 1(hydroxymethy1)adamantine; n-hexadecane, x n-hexadecanol. NcNc/N, =
droplet formation parameter.
Angew. Chem. hi.Ed. Engl. 30 (f99f) 466-476
of monolayers of a hydrophobic material on sulfuric acid
droplets in a condensation process. The molecular structure
of the coating material apparently causes an incomplete
shielding of the nucleus in some cases. Thus a particle which
is coated with n-hexadecanol is apparently wetted only with
difficulty, while an H,SO, droplet coated with l-hydroxyadamantane can be activated even at low degrees of supersaturation.
5. Aerosol Characterization Based
on Chemical Properties
The natural way for a chemist to carry out an aerosol
determination is by chemical identification. Many such detection reactions are now available. Suitable identification
reactions must be well-defined and reproducible, and must
lead to a readily observable and easily quantifiable “signal”.
5.1. Characterization by Means of a Selective
Stoichiometric Reaction and Observation of the Products
A good example of an in situ and on-line process is the
determination of aerosol sulfuric acid by a diffusion separator technique using radioactively labeled Na36C1.[801
A tube
coated with Na36Clis heated to 140°C; though it is used as
a sampler for H,SO, aerosols, it also acts as an analyzer. The
H36CI, which is formed in the stoichiometric reaction
[Eq. (g)] can readily be bound on a filter impregnated with
alkali or in an absorber solution and quantified by means of
radiometric techniques.
+ Na%1
+ H36CI7
Figure 18 shows an automatic system for a quasicontinuous determination of sulfuric acid and ammonium sulfate in
an atmospheric
The air is first freed from gaseous
inorganic and organic sulfur compounds (without any particle loss) by passing it through K,C03 and activated charcoal diffusion tubes. The aerosol to be analyzed then passes
through a Cu/CuO diffusion tube system heated to 120°C or
220 “C. At 120“C only the sulfuric acid is vaporized, while
Kuhnke valve
Activated charcoal
diffusmn seporotor
Cu/CuO diffusion seporotar
N2 exit
the ammonium sulfate is decomposed at 220°C. The fragments (molecular H,SO,) diffuse to the sink at the wall of
the tube, where they are irreversibly sorbed. Thermostable
aerosols pass through the various diffusion tubes without
loss. After a predetermined sampling time the system is
purged with ultrapure gas regulated by computer-controlled
magnetic valves. Each sample tube is separately subjected to
a short heating pulse, during which the sorbed acid is desorbed and converted at the Cu/CuO surface to SO,, and
each reproducible SO, pulse is registered on-line by a flame
photometric detector. During SO, detection, sampling continues in a second system driven in parallel. The detection
limit is ca. 0.1 pg H,SO, per m3 with a sampling time of ca.
5 min.
A similar apparatus has been developed for the determination of nitric acid and ammonium nitrate aerosols.[821In this
case thermolysis gives NO, which is detected via a chemiluminescence reaction with ozone.
5.2. Characterization by Chemiluminescence
StauSfand Stark et a1.[83-861were the first to make use of
chemiluminescence in aerosols during combustion processes
to characterize aerosols. A particularly promising improvement of this technique involves the use of CCD arrays for the
simultaneous observation of luminescence across a wide
wavelength range.
6. Summary and Future Prospects
This review makes it clear that a large number of classical
optical techniques can be applied to aerosol systems. The
procedures presented include not only typical surface-sensitive techniques such as ESCA and photoemission, but also
mass-proportional methods such as LIF, LIBS, and aerosol
mass spectrometry.
In the near future it will be of prime importance to have
available chemometric evaluation methods coupled with
broad-band detection techniques. A great deal of effort will
also have to go into the miniaturization of the methods
presently available. In this respect integrated optical circuits
in combination with frequency-doubled laser diodes as UV
source and CCD arrays as receiver optics have a great future.
Aerosols in the form of selectively prepared substances with
a high specific surface area, a particle size distribution which
is as narrow as possible, and whose formation is continuously controlled will surely take up a permanent place in further
technologies in the future.
The author wishes to thank ail those who were responsible
for stimulating, encouraging, and sustaining his interest in the
analytical chemistry of finely dispersed systems.
SO, diffusion separator
Received: February 2, 1990 [A 814 IE]
German version: Angew. Chem. 103 (1991) 542
Translated by Dr. T. N. Mitchell, Dortmund.
Fig. 18. Block diagram of a microprocessor-controlled H,SO,/ammonium
sulfate aerosol analyzer. FPD = flame photometric detector.
Angew. Chem. Int. Ed. Engl. 30 (1991) 466-476
[l] S. Prochazka, C. Greskovich, Am. Ceram. SOC.Bull. 57 (1978) 579-586.
[2] W. Cannon, S. Danforth, J. Flint, J. Haggerty, R. Marra, J. Am. Ceram.
SOC.65 (1982) 324-330.
[3] W. Cannon, S . Danforth, J. Haggerty, R. Marra, J. Am. Ceram. SOC.65
(1982) 330-335.
[4] M. Khvorov, Y. Khimchenko, A. Chirkov, A. Dudchenko, KoNoidn. Zh.
48 (1986) 597-600.
[5] F. Mayville, R. Partch, E. Matijevic, J. Colloid Interface Sci. I20 (1987)
[6] J. Wu, H. Nguyen, R. Flagan, Langmuir 3 (1987) 266-271.
[7] A. Garg, E. Matijevic, J. Colloid Interface Sci. 126 (1988) 243-250.
[8] S . Harris (Ed.): The Technology of Powder Coatings, Portcullis Press, London 1976.
[9] S. Pratsinis, P. Biswas in Aerosols. Formation and Reactivity (2nd Int.
Aerosol Conf. Berlin 1986). Pergamon, Oxford 1986, p. 931-934.
[lo] P. Guiot, P. Couvreur (Eds.): Polymeric Nanoparticles and Microspheres,
CRC Press, Boca Raton, FL, USA 1986.
[I 11 M. Gutcho (Ed.): Capsule Technology and Microencapsulation. Noyes
Data Corporation, Park Ridge, NJ, USA 1972.
[12] H. Krauss, H. Schmidt, Z . Anorg. Allg. Chem. 392 (1970) 258-270.
[13] J. Otterstedt, H. Carlo, A. Askengren, P. Ahlquist, patent application:
PCT-Anmeldung WO 85/03239 Klasse B 01 J 35/00, 1985.
[14] W. Seifert, DE-A 2742182, Klasse B 01 J 17/24, 1977.
(151 T. Kodas, A. Datye, V. Lee, E. Engler, J. Appl. Phys. 6.5 (1989) 21492151.
[I61 T. Kodas, E. Engler, V. Lee, R. Jacowitz, T. Baum, K. Roche. S . Parkin, W.
Young, S . Hughe, J. Kleder, W. Auser, Appl. Phys. Lett. 52 (1988) 16221624.
1171 T. Kodas, E. Engler, V. Lee, Appl. Phys. Lett. 54 (1989) 1923-1925.
[18] P. Ravindranathan, G. Mahesh, K. Patil, J. Solid State Chem. 66 (1987)
1191 J. Adcock, K. Horita, E. Renk, J. Am. Chem. Soc. 103 (1981) 693776947,
[20] J. Adcock, W. Evans, L. Heller-GroOman, J. Org. Chem. 48 (1983) 40534957.
[21] F. Young, S . Stephanakis, D. Mosher, J. Appl. Phys. 48 (1977) 3642-3650.
[22] J. Blauer, S . Munjee, K. Truesdell, E. Curtis, J. Sullivan, J. Appl. Phvs. 62
(1987) 2508 - 25 17.
[23] U. Krone, W. Kiihn, B. Georgi, A. Hiittermann, Proc. Inr. Pyrotech. Semin. 9th f984.
[24] W. Hollander, VDI-Ber. 429 (1982) 35-57.
[25] K . Leschonski, Ber. Bunsen-Ges. Phys. Chem. 88 (1984) 1112-1123.
[26] D. Pui, B. Liu, Phys. Scr. 37 (1988) 252-269.
[27] B. Daumer, R. NieBner, D. Klockow in W. Asman, H. Diederen (Eds.):
Ammonia and Acidification, EURASAP, Bilthoven, The Netherlands
1987, p. 86-96.
[28] R. NieBner, B. Daumer, D. Klockow in P. Wagner, G. Vali (Eds.): Atmospheric Aerosols and Nucleation, Springer, Berlin 1988. p. 538-541.
[29] P. Hess, Top. Curr. Chem. f f f (1983) 1-32.
[30] C . Patel, A. Tam, Rev. Mod. Phys. 53 (1981) 517-550.
[31] R. Rohl, Z. Childen, M. Palmer, Appl. Spectrosc. 39 (1985) 668-672.
[32] K. Adams, Appl. Opt. 27 (1988) 4052-4056.
[33] K. Adams, C. Davis, S . Japar, W. Pierson, Atmos. Environ. 23 (1989)
693 - 700.
[34] K. Fung, J. Gaffney, J. Chromatogr. 363 (1986) 207-215.
[35] A. Eckbreth, P. Bonczyk, J. Verdieck, Prog. Energy Combust. Sci. 5 (1979)
[36] D. Aeschliman, R. Setchell, Appl. Spectrosc. 29 (1975) 426-429.
[37] N. Omenetto in P. Camagni, S . Sandroni (Eds.): Optical Remote Sensing
of Air Pollution, Elsevier, Amsterdam 1983, p. 329-350.
[38] 1. Allegrini, N. Omenetto, Environ. Sci. Technol. 13 (1979) 349-350.
[39] R. NieBner, W. Robers, A. Krupp, Proc. SPIE 1989 Boston Symp. OE/
Fibers f989, 1172 (1990) 145-156.
[40] A. Biswas, H. Latifi, P. Shah, L. Radziemski, R. Armstrong, Opt. Lett. I2
(1987) 313-315.
[41] M. Essien, L. Radziemski, J. Sneddon, J. Anal. At. Spectrom. 3 (1988)
[42] M. Potschke, H.-P. Sattler, K. Hohla, T. Loree, DE-A 371 8672 A1 (1988).
[43] B. Raeymakers, T. Graule, J. Broekart, F. Adams, P. Tschopel. Spectrochim. Acta Part B 43 (1988) 923-940.
[44] L. Ebdon, J. Wilkinson, J. Anal. A t . Spectrom. 2 (1987) 39-44.
[45] J. Broekart, F. Leis, B. Raeymakers, G. Zaray, Spectrochim. Acta Part B
43 (1988) 339-353.
1461 K. Fung, I. Tang, Chem. Phys. Lett. 147 (1988) 509-513.
[47] K. Fung, I. Tang, Appl. Opt. 27 (1988) 206-208.
[48] S . Asher, Anal. Chem. 56 (1984) 720-724.
[49] M. Wright, J. Pollack, D. Colburn, NBS Spec. Pub/. ( U S ) 464 (1977)
301 -303.
[SO] H. Mudd, C. Kruger, E. Murray, Appl. Opr. 21 (1982) 1146-1154.
1511 R. Schlogl, G. Indlekofer, P. Oelhafen, Angew. Chem. 99 (1987) 312-322;
Angew. Chem. Int. Ed. Engl. 26 (1987) 309-319.
[52] R. Tandon, R. Payling, B. Chenhall, P. Crisp, J. Ellis, R,. Baker, Appl. Surf.
Sci. 20 (1985) 527-537.
[53] J. Campbell, R. Smith, L. Davis, K. Smith, Sci. Total Environ. I2 (1979)
[54] D. Natusch, E. Denoyer, T. Keyser, S . Kirton, D. Taylor, M. Zeller.
AIChE Symp. Ser. 201 (1980) 127-133.
[55] B. Barbaray, J.-P. Contour, G. Mouvier, Environ. Sci. Technol. 12 (1978)
1294- 1296.
[56] A. Goldmann in Stofjl: und Energietransport in Aerosolen, Arbeits- und
Ergebnisbericht des SFB 209. 1986-1988, Universitat Duishurg 1988,
p. 197-207.
[57] F. Miiller, K. Kleinherbers, A. Goldmann, R. NieBner, AerosolSci. Techno/. 7(1987) 109-111.
I581 W. Davis, Environ. SCI.Techno/. 10 (1973) 278.
[59] J. Stoffels, Int. J. Mass Spectrom. Ion Phys. 40 (1981) 217-222, 223-234.
[60] M. Sinha in Aerosols: Formation and Reactivity (2nd Int. Aerosol Conf.
Berlin 1986), Pergamon, Oxford 1986, p. 875-878.
[61] H. Burtscher, L. Scherrer, H. Siegmann, A. Schmidt-Ott, B. Federer, J.
Appl. Phys. 53 (1982) 3787-3791.
[62] R. NieBner, J. AerosolSci. 17 (1986) 705-714.
[63] R. NieOner, Fresenius Z . Anal. Chem. 329 (1987) 406-409.
1641 R. NieBner, B. Hemmerich, U. Panne, Fresenius Z . Anal. Chem. 35 (1989)
[65] N. Pelz, M. Backer, Daimler-Benz AG (1990) personal communication.
[66] A. Zajc, E. Uhlig, H. Hackfort, R. NieBner, J. Aerosol Sci. 20 (1989)
(671 R. NieBner, G. Walendzik, Fresenius Z . Anal. Chem. 333 (1989) 129-133.
[68] R. Nieher, F. Lutz, Polycyclic Aromat. Cmpd., in press.
(691 R. NieBner, W Robers, P. Wilbring, Anal. Chem. 61 (1989) 320-325.
[70] R. NieBner, P. Wilbring, Anal. Chem. 61 (1989) 708-714.
[71] B. Liu, D. Pui, K. Whitby, D. Kittelson, Y Kousaka, R. McKenie, Atmos.
Environ. I2 (1978) 99-104.
1721 R. Charlson, A. Vanderpol, D. Covert, A. Waggoner, N. Ahlquist, Atmos.
Environ. 8 (1974) 1257-1267.
[73] A. Campillo, C. Dodge, H.-B. Lin, Appl. Opt. 20 (1981) 3100-3103.
[74] A. Campillo, S . Petuchowski, S . Davis, H.-B. Lin, Appl. Phys. Lett. 4f
(1982) 327-329.
[75] S . Arnold, Aerosol Sci. Technol. 2 (1983) 194.
1761 H.-B. Lin, A. Campillo, Appl. Opt. 24 (1985) 422-433.
[77] D. Fluckiger, H.-B. Lin, W. Marlow, Appl. Opt. 24 (1985) 1668-1681.
[78] S . Arnold, E. Murphy, G. Sageev, Appl. Opt. 24 (1985) 1048-1053.
[79] R. Nieher, B. Daumer, D. Klockow Aerosol Sci. Technol. 12 (1990) 953963.
[80] R. Nieher, D. Klockow, Int. J. Environ. Anal. Chem. 8 (1980) 163-175.
[81] J. Slanina, C . Schoonebeek, D. Klockow, R. NieBner, Anal. Chem. 57
(1985) 1955-1960.
[82] D. Klockow, R. NieBner, M. Malejczyk, H. Kiendl, B. vom Berg, M.
Keuken, A. Wayers-Ypelaan,J. Slanina, Atmos. Environ. 23 (1989) 11311138.
(831 J. Stauff, H. Fuhr, Angew,. Chem. 87 (1975) 132; Angew. Chenr. Int. Ed.
Engl. 14 (1975) 105.
(841 J. Stauff, H. Fuhr, Staub Reinhalt. Luff 3.5 (1975) 372-373.
[85] G. Stirk, D. Uberschar, J. Stauff, Erdol Kohle Erdgas Petrochem. 39 (1986)
[86] J. Stauff, G. Stark, Z . Naturforsch. B41 (1986) 113-121.
Angew. Chem. h t . Ed. Engl. 30 (1991) 466-476
Без категории
Размер файла
1 146 Кб
chemical, characterization, line, situ, aerosol
Пожаловаться на содержимое документа