On the performance of an on-line time-of-flight mass spectrometer for aerosols

Pergamon

Plh S0021-8502(96)00067-5

J. Aerosol Sci. Vol. 28, No. 1, pp. 159 171, 1997 Copyright (c~ 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0021-8502/96 $17.00 + 0.00

O N T H E P E R F O R M A N C E O F A N O N - L I N E T I M E - O F - F L I G H T

MASS S P E C T R O M E T E R F O R A E R O S O L S

M. Weiss, P. J. T. Verheijen, J. C. M. Marijnissen and B. Scarlett

Delft University of Technology, Faculty of Chemical Technology and Materials Science, Julianalaan 136, 2628 BL Delft, The Netherlands

(First received 12 January 1996; and in final form 30 May 1996)

Abstract--An instrument has been developed to analyze the size and chemical composition of individual, air-borne particles. The particles are introduced into the vacuum chamber of a time-of- flight mass spectrometer where they are detected and sized using an aerodynamic principle. After sizing, the particles are vaporized by an excimer laser and the resulting ions are analyzed. Experi- ments have shown that polystyrene particles 1/am in diameter are not fully vaporized by the laser pulse. The instrument has been characterized and some examples of measurements on ambient and artificial aerosols are presented. Coyright © 1996 Elsevier Science Ltd

1. I N T R O D U C T I O N

Single-particle analysis of aerosols remains a topic of current interest in aerosol science (Jambers et al., 1995). Besides size, shape and atomic composition, it is i m p o r t a n t for some studies to also identify chemical composition of each particle.

In this context, Marijnissen et al. (1988) proposed an a p p a r a t u s for the single-particle analysis of aerosols by combining laser pyrolysis with time-of-flight (TOF) mass spectro- metry. The idea is quite simple. Off-line analysis with L A M M A and L I M A (Kaufmann, 1986; van Vaeck et al., 1993) had already demonstrated that T O F mass spectra could be taken from individual, stationary particles, but, a m o n g s t other disadvantages, this is a time-consuming procedure because each particle has to be identified manually. The principle is to combine an aerosol b e a m generator with a T O F mass spectrometer and thus to detect only those particles which pass through the sampling volume of the spectrometer. Several groups ( T h o m s o n and M u r p h y , 1993, 1994; Hinz et al., 1994; M u r p h y and T h o m - son, 1995; P r a t h e r et al., 1994; M c K e o w n et al., 1991), including our own (Kievit, 1995), have designed such an a p p a r a t u s and have performed experiments with their own version.

The original p a p e r of 1988 also stressed the possibility of combining these two signals, thus the mass spectrum of each particle with its size, if both can be measured simulta- neously. O u r design strategy has been to measure size and chemical composition as nearly as possible at the same physical position. The present a p p a r a t u s does this with not more than 1 m m between the position where the size is determined and where the particle pyrolysis takes place. This produces m o r e accurate and reproducible size measurements than before combined with 100% hit rate sometimes.

This p a p e r describes the development of the complete instrument for both particle size and mass spectrum determination. It focuses on the performance and limitations which are inherent in this m e t h o d of observation.

2. E X P E R I M E N T A L S E T U P

The aerosol is introduced into the setup by a differentially p u m p e d nozzle/skimmer arrangement. Kievit (1995) reports m o r e details on the design of the aerosol b e a m generator.

The particle b e a m is introduced into the ion source of a T O F laser mass spectrometer (Fig. 1). Approximately 2 cm d o w n s t r e a m of the last skimmer, the particles travel through the focus of a continuous 16 m W H e N e laser. The particles scatter light which is detected by

160 M. Weiss et al.

Beam dump

~ ' ~ x ~

_ ~

Acceleration

~ / ~ . A

,

y

electrOdes

TOF tube

Aerosol beam

U

HeNe laser

,

(continuous:

16 mW)

Dichroic

mirro

---DO

,"

~

Beam splitter

Excimer laser

(pulsed: 5 mJ / 3 ns)

/

i i

4>

I i

,I

i

Ion detector

Fig. 1. Schematic top view of the set-up without the beam generator.

a photo-multiplier tube (PMT). The photo-multiplier and the collection lens m o u n t e d in front of it are placed in such a way that the light scattered at 45 ° in the forward direction is collected. This is the smallest possible angle within the present configuration since the mass spectrometer electrodes prohibit a view at smaller angles. The optics collects light from 4- 20 ° a r o u n d this angle. A digital oscilloscope (LeCroy model 9354) samples the signal from the P M T at a rate of 500 million readings per second. The oscilloscope also generates a trigger pulse if the P M T signal rises above an adjustable threshold. This pulse triggers an excimer laser which generates ions from the particle. F o r the experiments a 5 mJ excimer laser with a pulse width of 3 ns was used. The diameter of the focus was estimated to be a b o u t 800/tm x 4 0 0 # m (elliptical shape) resulting in an intensity of approximately 6 x 101 z W m - 2. The laser operates at a wavelength of 308 nm and at a m a x i m u m repetition rate of 25 Hz. The total delay time between the detection of a particle and the output of the ionization laser is approximately 1.5/~s. The ionization laser is aligned collinearly but displaced downstream of the detection laser by 50-500/~m. This spacing corresponds to the distance traveled by the particles during the delay time.

The ionization process takes place in the ion source of a T O F mass spectrometer. Generated ions are accelerated towards the ion detector, which are multichannel plates, by a two-stage acceleration system. Typical acceleration potentials are of the order of + 6 kV. So far, only positive ions have been detected, but the electronic setup can be changed to detect negative ions instead. The signal of the ion detector is digitized by the oscilloscope simultaneously with the photo-multiplier signal and both are sent, after acquisition, to a PC via a G P I B bus interface. After receiving the data the PC rearms the system for the next trigger and acquisition. D a t a processing is carried out on a U N I X - b a s e d workstation connected to the PC in an Ethernet-based network.

In this way the instrument offers information on the size ( P M T signal) and the chemical composition (mass spectrum) of single particles. N o sample preparation is necessary, so the instrument can be operated on-line. The m a x i m u m repetition rate is a b o u t 2-10 Hz

On-line time-of-flight mass spectrometer for aerosols 161 depending on the necessary mass range of the mass spectrometer and the resulting a m o u n t of data, because the data transfer between the sampling oscilloscope and the computer is the limiting factor.

3. P A R T I C L E D E T E C T I O N A N D S I Z I N G

The signal of the P M T serves two purposes. Firstly, it is used to trigger the excimer laser that vaporizes the particle. Secondly, the P M T signal can be used to gain information on the size of the particles. This section of the paper explains how the size information is obtained in the Delft on-line aerosol analysis system.

Kievit (1995) investigated the analysis of P M T signals, considering the width of the peaks and ratios of signals obtained from different angles. He failed however, to obtain reliable size information because of the inhomogeneous laser intensity profile. Special diffraction optics can be used to obtain a more "top hat" profile from the laser (Murphy and Thomson, 1995; T h o m s o n et al., 1994; O'Shea, 1985). Since this optics reduce the overall intensity considerably, strong and therefore expensive lasers must be used to achieve reasonable detection limits. T h o m s o n and M u r p h y (1994) used, for example, a 80 mW, diode-pumped, frequency-doubled N d - Y A G laser. F o r the work described here a different approach was implemented that is explained in the next section.

3.1. Aerodynamic particle sizin9

It is well known (Dahneke and Friedlander, 1970; Dahneke and Padliya, 1977; Dahneke, 1978) that the velocity of the particles is size dependent after passing a beam generator assembly such as the one described above. Commercial aerosol spectrometers are available which use this effect for rapid, on-line particle-size distribution monitoring (Baron, 1986; Willeke and Baron, 1993).

In the beam generator, the aerosol stream is expanded and accelerated through a nozzle. Because the pressure ratio across the nozzle is smaller than 0.5 the air reaches supersonic velocity. Since the particles have higher inertia than the gas molecules they cannot fully follow the acceleration and thus attain a size-dependent terminal velocity. The speed of the particles is determined by measuring the transit time of the particle between two laser beams.

Recently, Prather et al. (1994) implemented such a sizing system into an on-line T O F mass spectrometer for aerosols. To implement this sizing technique in the Delft system the incident H e N e laser beam was split so that two focal points are obtained in the vacuum chamber (Fig. 2). Splitting the laser beam reduces the intensity of each focal point by more than 50% in comparison with the single beam but the alignment is easier with this arrangement than one using a second laser as Prather et al. (1994) did. Furthermore, the spacing between the two lasers is smaller which increases the hit rate and reduces coincid- ence errors (see below). The spacing is fixed for a given prism and therefore easy to reproduce.

F o r the experiments a Melles Griot Inc. beam displacing prism model 03 P P D 301/A was used to split the laser beam. This prism provides a displacement of 2.6 mm at 632.8 nm, is coated with an anti-reflex coating for 632.8 nm, and is of low scatter grade. The transmis- sion efficiency of the prism is 84%, i.e. 42% per exiting beam. Figure 3 shows the resulting photo-multiplier signal for a 1/~m polystyrene latex (PSL) particle. The triggering system was designed to trigger at the occurrence of the second peak in the signal to avoid a vaporization of the particle by the ionization laser before its sizing is completed. This second peak must follow the first within a specific time interval in order to cause a trigger. The time between the two first peaks (7.8 ~s in Fig. 3) is a direct measurement of the velocity of the particles.

F r o m Fig. 2 another advantage of this configuration becomes obvious. Particles with trajectories that are not exactly vertical (i.e., are off-axis) do not pass both detection foci. Therefore, the trigger system does not detect a second peak inside the timing window and

162 M. Weiss et al.

Alignment in normal operation

I I I J . . . T I I 1 I I 501 am ~ ~, 2 . 5 m m 3 0 0 - 500 p m ~ @ .... ! ~ . . . . :' 50 - 500 pm / I 8 5 0 gm

Particle beam generator

Alignment to check for

completeness of evaporation

\ /

O ...

.~0 m m

0

Fig. 2. Side view of the laser alignment with implemented beam splitter.

I

2.5 m m 35 25 > g.

ce

~ 15 I - O.. D e t e c t i o n l a s e r / \ \ TTOF

~l '

'

4

,

i

,

z j I o n i z a t i o n l a s e r L ~ ~ ~ 1'0 lk5 ~ 2~5 ' " 5 1 5 -10 -5 0 5 20 30 35 T i m e [~s]

Fig. 3. P M T signal of a 1 ,urn latex particle with implemented beam splitter.

does not trigger the ionization laser. This leads to an effective horizontal extent of the ionization z o n e that is m u c h smaller than the diameter of the focus of the ionization laser. Shot-to-shot variations in the T O F of the ions are therefore reduced which improves the accuracy of the calibration of the mass spectrometer.

|n order to obtain a calibration curve measurements were made with several sizes of latex particles which were obtained from D u k e Scientific Corp., Palo Alto, California. Figure 4

On-line time-of-flight mass spectrometer for aerosols 163 "6 ._E F- 14000 12000 10000 8000 6000 4000 0.0

- fit: bo(l+b I x pol"s) °s

bo=5265 ns b1=1.14 / ~ - " l ~ m _~-~ 0.5 lun . " 0.364 ~rn S 2.87 Ixm / ~ 2.26 lam 1.94 Ism 4 l.Lm

11o

21o

31o

41o

Particle size [gm]

Fig. 4. Calibration curve for aerodynamic particle sizing.

5.0 80.0 Z 60.0 e ~ 40.0 E 20.0 Z 0.501 ~m 1.01 [,tm 1.94 I.tm 2.87 I, tm 0.0 0.0 1.0 2.0 3.0 4.0 Particle diameter [~tm]

Fig. 5. Size measurement of 4 fractions of latex (each with a uniformity better than 5%).

s h o w s the resulting calibration curve. Sub-micrometer particles as small as 0.5/~m can be measured with very small error. Each point o n the calibration curve represents the median of several hundred particles. This calibration curve is only valid for spherical particles of the same density as polystyrene latex. During the acceleration through the nozzle Stokes law is not applicable because of the high particle Reynolds numbers. The resulting terminal velocity is therefore not a unique function of the a e r o d y n a m i c diameter (Willeke and Baron,

1993).

Figure 5 s h o w s an example of a measurement o f an aerosol consisting of a mixture of four latex fractions. The four sizes are easily resolved by the a e r o d y n a m i c sizing system.

An important source of error in T O F aerosol spectrometers is coincidence of the particles (Heitbrink et al., 1991). W h e n a particle enters the sensing z o n e while another one is already present, a w r o n g measurement results. The probability of this error is a function of the rate

164 M, Weiss et al.

of detected particles, 2, and the residence time of the particles in the sensing zone, T. Using Poisson statistics the coincidence error e(2, T) can be estimated by the following expression:

~ ( 2 T ) n ;,r e - ~ r ,

e(2, T ) = - - e = 1 -

n = l / 2 !

(1)

The 10% coincidence m a r k is reached at detection rates above 10,000 particles per second, assuming a typical T O F of 10ps. Such high detection rates were observed when an ultrasonic nebulizer was used to generate aerosols from solution (estimated n u m b e r concen- tration 10 v cm-3). To overcome this p r o b l e m either the incoming aerosol must be diluted or the distance between the two detection foci, and consequently the T O F , must be reduced. A successful and reproducible reduction of this spacing was facilitated by using a diffraction grating as a beam splitter in place of the b e a m displacing prism. Typical spacings using a grating with 10 lines per m m are of the order of 250/~m. The residence times of the particles in the sensing volume decreases by a factor 10 and thus the tolerable count rate increases by a factor 10.

4. C H A R A C T E R I Z A T I O N O F T H E A E R O S O L BEAM

T w o parameters of the aerosol beam were determined, i.e. the width of the particle beam at the point of detection and the velocity of the accelerated particles.

4.1. Width of the particle beam

The particle beam diameter was measured by scanning the detection laser beam, before splitting, horizontally through the particle b e a m with an adjustable translation table. The rate of detection of the particles (1/~m polystyrene latex) was measured as function of the laser position (Fig. 6).

In order to relate this displacement to that of the beam in the vacuum c h a m b e r the system was opened and a p h o t o sensitive p a p e r m o u n t e d at the focal plane. The scan was now

o == Q. ,$ 200 150 100 50 I I I I I I I I I 0 o o o o o o o o I I I I I I 0.1 0.2 0.3 0.4 0.5 0.6 Beam position in vacuum chamber [mm]

Fig. 6. Concentration profile in the particle beam.

I 0.7 I 0.8 0 0.9

On-line time-of-flight mass spectrometer for aerosols 165 & 13- 550 500 450 400 350 300 250 200 150 100 I I I I i i I i I own measurements - - Pratheret. al. (1994) [] Dahneke (1973) x X I I I I I I I I I 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Particle size [micrometer]

Fig. 7. Particle speed as function of particle size.

repeated over the previously measured range leaving a trail on the p h o t o paper. The length of the trail was then measured by a calibrated microscope. The trail proved to be 850 #m long which consequently is the diameter of the particle b e a m at the height of the detection zone. Since the diameter of the detection laser focus is only a b o u t 50/~m, it is clear that most of the particles exiting the b e a m generator remain undetected.

4.2. Velocity of the particles

Because the relationship between the T O F and the particle size is already k n o w n (Fig. 4), the velocity can be related to the particle size by measuring the distance between the loci. This was done by using an adjustable translation table equipped with a knife edge which was m o u n t e d above the electrodes of the mass spectrometer.

Using the polarizing b e a m splitter the distance between the focus points was measured as 2.5 _+ 0.1 mm. This coincides to a b e a m displacement of the prism of 2.6 m m at 632.8 nm and a lateral magnification of 0.95 by the imaging lens. Using this information the calibration shown in Fig. 7 is obtained. The error in this curve is mainly due to the inaccuracy of the laser spacing measurement. The velocity measurements agree r e m a r k a b l y well with the ones m a d e by P r a t h e r et al. (1994) although these were made with NaC1 particles. D a h n e k e (1973) measured slightly lower velocities. This is p r o b a b l y due to the different nozzle geometries and pressures.

5. MASS S P E C T R O M E T R Y O F A E R O S O L P A R T I C L E S 5.1. Influence of the laser energy on the appearance of the mass spectra

The intensity and wave length of the ionization laser are i m p o r t a n t parameters for the mass spectra. The ionization laser produces 5 mJ of light at a wavelength 308 nm within 3 ns. To investigate the influence of the ionization laser energy on the mass spectra the laser b e a m was attenuated using glass slides. Since these glass slides are made of conventional glass they a b s o r b significant a m o u n t s of U V radiation. Using several of these slides in series results in an attenuation of the laser beam. The energy output of the laser after attenuation was checked periodically during the experiments with a pyroelectric energy detector.

166 M. Weiss et al.

T h e influence of the laser energy on the mass spectra of 1 # m polystyrene latex particles is s h o w n in Fig. 8. Similar spectra have been reported by Hinz et al. (1994). The spectra show typical clusters of one, two or m o r e c a r b o n atoms. T h e effect of the high energy can clearly be seen. T h e incidence of small fragments tends to increase with higher energy.

T h e influence of the energy is even m o r e m a r k e d in the case of a n t h r a c e n e particles as can be seen in Fig. 9. Below a certain threshold the polycyclic a n t h r a c e n e molecule (molecular

"3 ,g i i i i i i i i i i 1 t 1

mJ

1.6 md I I I I I [ I [ I I 10

20

30

40

50

60

70

80

90

1 O0

110 120

m/e [amu]

Fig. 8. Energy influence on spectra of polystyrene latex particles.

L__

I 0 if} r ; i i 1 i i i i i r i i i ~ i i i t

t

I

k ~

1.8 mJ

I t I I l I I I I I I I I I I I I I I 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 m/e [ainu]

On-line time-of-flight mass spectrometer for aerosols 167

mass 178 ainu) stays intact and only the single peak of the molecular ion can be seen. F o r higher energies the molecule breaks down to carbon clusters.

5.2. H i t rate

An important performance parameter for on-line single particle mass spectrometry systems is the hit rate. Once a particle is detected by the photo-multiplier the ionization laser is triggered. The timing and alignment of this laser must be matched with the particle velocity in order to hit the particle. The excimer laser which was used introduces, together with the triggering electronics, a delay time of approximately 1.5/~s between the external trigger input and the radiation output with a variation of only a few nanoseconds. The width of the operating window can be calculated using the know- ledge of the diameter of the focus, laser delay time and particle velocity as function of particle size.

Obviously, a hit rate of 100% should be attainable for mono-disperse particles assuming a correct alignment and that all the particles have the same velocity. This was confirmed by experiments using mono-disperse latex particles. F o r the polydisperse case however only a certain size fraction will be hit by the laser. Small particles which travel faster are too far downstream at the moment of firing and the large particles are not far enough.

In order to prove this the following experiment was carried out: an aqueous solution of sodium chloride (0.33 wt%) was nebulized using a DeVilbiss Model 40 pneumatic nebulizer. The spray was heated and dried in a diffusion drier resulting in an air stream containing polydisperse sodium chloride crystals. The aerodynamic mean diameter was 0.88 #m with a geometric standard deviation 1.33 as measured with an API Aerosizer. The aerodynamic sizing system was used to measure the particle-size distribution entering the vacuum chamber. Simultaneously, it was noted from which particles spectra were obtained. The result obtained with the default alignment (Fig. 10) clearly shows the operating window. When the spacing between the detection and ionization laser was enlarged the system becomes sensitive to the smaller, faster particles, as expected and as shown in Fig. 11. Z .o "6

E=

Z lOO 90 8o 7o 60 5o 40 30 20 lO o o

,i

F--' i i I- '-~ i ----~,--I 0.5 1 1.5 2 2.5 3

Aerodynamic particle diameter [micrometer]

total measured particles - - spectra obtained . . . .

w~ , I I I

3.5 4

168 M. Weiss et al. 120 100 Z 8O -~ 60 "6 ~ 4o Z 20

|

\

i i

total measured particles - - spectra obtained . . . . !

%_

l ---L i--- r - I r , ~ • 1 1.5 2 2.5

Aerodynamic particle diameter [micrometer]

0 0.5 3 3.5

Fig. 11. Hitting performance for NaCl particles with alternative alignment (laser energy 1.9 m J).

In order to enlarge the width of the operating window several approaches are possible: 1. Use of higher laser power will lead to a larger usable focus diameter and, therefore, to a wider operating window but will also increase fragmentation of fragile molecules.

2. Shorter laser delay times minimize the spatial spread of the particles and, therefore, enlarge the operating window.

3. Active triggering, as reported by Prather et al. (1994), uses the information already obtained to delay the trigger depending on the particle speed. This requires however bigger distances (about a factor 100) between the laser foci which results in reduced hit rates and higher probability of coincidence errors (see above).

4. Scanning of the laser beam or laser delay time as reported by Hinz et al. (1994) allows the measurement of b r o a d size distributions but increases the time to obtain a representa- tive measurement.

5.3. Completeness of evaporation

F o r m a n y applications it is useful to know whether or not the particles are completely vaporized by the ionization laser. If the particles are not completely vaporized, the recorded mass spectrum is not necessarily representative of the whole particle. This is important in the case of inhomogeneous particles which m a y consist of different core and shell materials. In order to investigate whether parts of the particle survive, a post detection laser was installed after the ionization as shown in the right part of Fig. 2.

The electronics was a d a p t e d for this experiment to trigger on the first peak. T w o cases are possible. If the particles are completely evaporated, nothing should be seen on the photo- multiplier after ionization takes place. If, however, the particle partly survives, a second peak should be seen after ionization. The experiments were carried out with a mono- disperse latex aerosol at very low concentration (less than 1 count per second) in order to avoid coincidence errors. Figure 12 shows a typical result when a 1/~m particle was analyzed using a m e d i u m laser energy.

A second peak can clearly be seen after ionization which indicates that the particle was not completely vaporized. The simultaneously acquired mass spectrum ensures that the

On-line time-of-flight mass spectrometer for aerosols 169 ,-'2. o 3

._g

I - " Q - r,.. ~[ Mass spectrum i -2 0 il "~ I 7.8 Its ) PMT signal , , , , 1~4 ' 2 4 6 8 10 12 16 Time [~s]

Fig. 12. Test for complete evaporation of a 1 #m PSL particle (2.9 mJ).

particle was in fact hit by the ionization laser and that it was a latex particle. The time between the two peaks resulting from the scattering of the H e N e laser was approximately 7.8 #s which is exactly the time from the calibration curve (Fig. 4) for a 1/xm particle. This means that the particle did not change its speed during the analysis.

The same experiment carried out with maximum laser energy (5 m J) results in practically the same signals.

This experiment shows that latex particles of the order of 1/~m are not completely vaporized by the laser. The degree of vaporization can, however, not be determined by that experiment. It can only be concluded that the surviving particles are greater than 0.4 #m since this is the detection limit of the sizing system.

5.4. Mass spectra of various particles

In this section some spectra of ambient and artificial aerosols are presented. Figure 13 shows two spectra of NaC1 particles. The aerosol was produced by nebulizing a sodium chloride solution (0.33 wt%). This solution was made with distilled water in the case of the upper spectrum and with tap water in the case of the lower spectrum. Peaks for sodium (23 amu), Na2C1 (81/83 amu) and Na3C12 (139/141 amu) are present in both spectra. The additional peaks in the lower spectrum can be related to the impurities of the tap water. Potassium (39 amu), calcium (40 ainu), CaO (56 amu) and CaC1 (75/77 ainu) can be identified.

Figure 14 shows two typical spectra obtained from an analysis of ambient air in the laboratory. The upper spectrum is the one most often observed during the experiment. The high sodium (23 amu) and potassium (39 amu) peaks indicate a salt particle. The experi- ments were carried out in Delft, The Netherlands, which is located close to the N o r t h Sea (about 20 km). The particle is, therefore, likely to be of maritime origin. A closer look at this spectrum shows that the particle also contains significant amounts of calcium (Ca (40 amu), C a O / C a O H (56/57 amu)).

The second particle (lower part of Fig. 14) is characterized by low sodium and potassium content and a high abundance of calcium. The particle most likely originates from building materials containing C a S O 4.

170 M. Weiss et al. z.

_g

Na Na. CI Pure NaCI Na3CI 2 . L, L 0 Ca Contaminated NaCI

M cao

I l

, L 6~0 8~0 , , , 20 40 100 120 140 160 180 m/e [amu]

Fig. 13, Two spectra of sodium chloride.

Ca

K NaJK rich particle

Ca

~

CaCI i i '

[i

Oa.c. pa ,c,e

J

~JL ,

!,,

'

,t /

/I

i i L 20 4'0 6'0 80 100 120 140 m/e [amu]

Fig. 14. Two spectra from particles of ambient air.

6. C O N C L U S I O N S A N D D I S C U S S I O N

With the T O F mass spectrometry system described here single aerosol particles can be analyzed on-line by means of laser desorption/ionization mass spectrometry. The size of the individual particles can be measured simultaneously. The sizing system is based on an a e r o d y n a m i c principle. Particles in the size range from 0.4 to 10/xm have been analyzed with this system. The particle velocities measured are in good agreement with literature values. Since the width of the particle beam at the detection point is around 850/~m only a small fraction of the particles entering, is detected. Thus, if aerosols are to be measured at low concentrations the time to obtain a representative sample must be increased.

Because of the fixed timing of the triggering electronics and the spread in particle velocities only a certain size fraction can be analyzed at one time. The width of this

On-line time-of-flight mass spectrometer for aerosols 171 o p e r a t i n g w i n d o w is d i f f e r e n t f o r d i f f e r e n t m a t e r i a l s . I n o r d e r to m e a s u r e w i d e size d i s t r i b u - t i o n s , s a y f r o m 0.5 t o 5/~m, e i t h e r t h e l a s e r o r t h e d e l a y t i m e m u s t b e s c a n n e d . T h e 5 m J e x c i m e r l a s e r d o e s n o t v a p o r i z e 1 / ~ m l a t e x p a r t i c l e s c o m p l e t e l y . I f p a r t i c l e s a r e n o t h o m o g e n e o u s t h e i n f o r m a t i o n in t h e m a s s s p e c t r u m m a y t h e r e f o r e n o t be r e p r e s e n t a t i v e o f t h e w h o l e p a r t i c l e . O n t h e o t h e r h a n d , t h i s a l s o m e a n s t h a t t h e s y s t e m c a n be s e n s i t i v e t o s u r f a c e c o n t a m i n a t i o n o n t h e p a r t i c l e . O r g a n i c a n d i n o r g a n i c p a r t i c l e s o f v a r i o u s s u b s t a n c e s h a v e b e e n s u c c e s s f u l l y a n a l y z e d . D u r i n g t h e e x p e r i m e n t s a s i g n i f i c a n t v a r i a t i o n in t h e m a s s s p e c t r a b e c a m e a p p a r e n t e v e n if m o n o - d i s p e r s e a n d c h e m i c a l l y h o m o g e n e o u s p a r t i c l e s w e r e a n a l y z e d . T h e a m o u n t o f v a r i a t i o n a p p e a r s t o b e d e p e n d e n t o n t h e m a t e r i a l s a n a l y z e d a n d t h e i o n i z a t i o n l a s e r e n e r g y used. W o r k is in p r o g r e s s to q u a n t i f y this effect.

Acknowledyments--The authors would like to thank the Netherlands Foundation for Chemical Research (SON) for supporting these investigations. Financial aid was provided by the Netherlands Organization for Scientific Research (NWO) and the Foundation for Technical Sciences (STW).

R E F E R E N C E S

Baron, P. A. (1986) Calibration and use of the aerodynamic particle sizer (APS 3300). Aerosol Sci. Technol. 5, 55-67.

Dahneke, B. (1973) Aerosol beam spectrometry. Nature Phys. Sci. 244, 54-55.

Dahneke, B. (1978) Aerosol beams. In Recent Developments in Aerosol Science (Edited by Shaw, D.), Chap. 9, pp. 187-223. Wiley, New York.

Dahneke, B. and Friedlander, S. (1970) Velocity characteristics of beams of spherical polystyrene particles. Aerosol

Sci. 1, 325-339.

Dahneke, B. and Padliya, D. (1977) Nozzle-inlet design for aerosol beam instruments. In Rarefied Gas Dynamics;

Technical Papers Selected from the lOth International Symposium on Rarefied Gas Dynamics (Edited by Potter, J. L.), July 1976, pp. 1163-1172. American Institute of Aeronautics and Astronautics. New York.

Heitbrink, W. A., Baron, P. A. and Willeke, K. (1991) Coincidence in time-of-flight aerosol spectrometers: phantom particle creation. Aerosol Sci. Technol. 14, 112-126.

Hinz, K.-P., Kaufmann, R. and Spengler, B. (1994) Laser-induced mass analysis of single particles in the airborne state. Anal. Chem. 66, 2071 2076.

Jambers, W., Bock, L. D. and Grieken, R. V. (1995) Recent advances in the analysis of individual environmental particles; a review. Analyst 120, 681-692.

Kaufmann, R. L. (1986) Laser microprobe mass spectrometry (LAMMA) of particulates. In Physical and Chemical

Characterization of Individual Particles (Edited by Spumy, K. R.). Ellis Horwood Limited, Chichester, UK. Kievit, O. (1995) Development of a laser mass spectrometer for aerosols. Ph.D. thesis, Delft University of

Technology, Delft, The Netherlands.

Marijnissen, J., Scarlen, B. and Verheijen, P. (1988) Proposed on-line aerosol analysis combining size determina- tion, laser induced fragmentation and time-of-flight mass spectrometry. J. Aerosol Sci. 19, 1307-1310. McKeown, P., Johnston, M. and Murphy, D. (1991) On-line single-particle analysis by laser desorption mass

spectrometry. Anal. Chem. 63, 2069-2073.

M urphy, D. M. and Thomson, D. S. (1995) Laser ionization mass spectrometry of single aerosol particles. Aerosol

Sci. Technol. 22, 237-249.

O'Shea, D. C. (1985) Elements of Modern Optical Desiyn. Wiley, New York.

Prather, K. A., Nordmeyer, T. and Salt, K. (1994) Real-time characterization of individual aerosol particles using time-of-flight mass spectrometry. Analyt. Chem. 66, 1403 1407.

Thomson, D., Murphy, D. and Kalyuzhny, M. (1994) Particle analysis by laser-ionization mass spectrometry: instrument development. In (Edited by Flagan, R. C.), Fourth International Aerosol Conf., Abstracts, Cincinnati, Ohio, p. 530. American Association for Aerosol Research.

Thomson, D. S. and Murphy, D. M. (1993) Laser-induced ion formation thresholds of aerosol particles in a vacuum. Appl. Opt. 32, 6818 6826.

Thomson, D. S. and Murphy, D. M. (1994) Analyzing single aerosol particles in real time. Chemtech 24, 30-35. van Vaeck, L., van Roy, W., Gijbels, R. and Adams, F. (1993) Laser Ionization Mass Analysis, Chapter Lasers in

mass spectrometry: organic and inorganic instrumentation, pp. 7 126. Wiley, New York.

Willeke, K. and Baron, P. A. (Eds.) (1993) Aerosol Measurement; Principles, Techniques, and Applications. van Nostrand Reinhold, New York.